Characteristics of Nontuberculous Mycobacteria from a Municipal Water Distribution System and Their Relevance to Human Infections

CHARACTERISTICS OF NONTUBERCULOUS MYCOBACTERIA FROM A MUNICIPAL WATER DISTRIBUTION SYSTEM AND THEIR RELEVANCE TO HUMAN INFECTIONS.

Rachel Thomson MBBS FRACP Grad Dip (Clin Epi)

A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy

School of Biomedical Sciences Faculty of Health Queensland University of Technology 2013

Principal Supervisor: Adjunct Assoc Prof Megan Hargreaves (QUT) Associate Supervisors: Assoc Prof Flavia Huygens (QUT)

ii KEYWORDS

Nontuberculous mycobacteria Water Distribution systems Biofilm Aerosols Genotyping Environmental organisms Rep-PCR

iii

iv ABSTRACT

Nontuberculous mycobacteria (NTM) are environmental organisms associated with pulmonary and soft tissue infections in humans, and a variety of diseases in animals. There are over 150 different species of NTM; not all have been associated with disease. In Queensland, M. intracellulare predominates, followed by M. avium, M. abscessus, M. kansasii, and M. fortuitum as the most common species associated with lung disease. M. ulcerans, M. marinum, M. fortuitum and M. abscessus are the most common associated with soft tissue (both community acquired and nosocomial) infections. The environmental source of these pathogens has not been well defined. There is some evidence that water (either naturally occurring water sources or treated water for human consumption) may be a source of pathogenic NTM.

The aims of this investigation were to 1) document the species of NTM that are resident in the Brisbane municipal water distribution system, then 2) to compare the strains of NTM found in water, with those found in human clinical samples collected from Queensland patients. This would then help to prove or disprove whether treated water is likely to be a source of pathogenic strains of NTM for at risk patients. The third aim was to investigate whether shower aerosols contained NTM in respirable particles.

As there are difficulties identifying NTM in water samples using culture methods, a pilot study was performed using spiked samples of sterile and tap water, comparing selected methodological variables. These included centrifugation vs filtration for concentration of samples, the addition of Sodium thiosulphate (routinely used to neutralize chlorine), the addition of Cetylpiridium chloride (CPC) for decontamination, different culture media – including Löwenstein-Jensen (L-J), Middlebrook 7H10 and 7H11, and MGIT Bactec with and without PANTA, and incubation temperature (32°C/35°C/32°C with carbon dioxide (CO2)). The results of this study revealed that sodium thiosulphate was not necessary and might inhibit mycobacterial growth, also that filtration was more effective than centrifugation,

v Middlebrook media were favoured over L-J, and there was no difference in growth at different incubation temperatures. The addition of CPC 0.005% decreased contamination rates, but also significantly reduced the mycobacterial numbers.

From these results a study was designed to test Brisbane water distribution samples for NTM. This is the first Australian study of its kind and one of the largest internationally. Multiple pathogenic and nonpathogenic NTM were identified in 40.2% of sites sampled in summer and 82% sites in winter. The summer samples were marred by bacterial contamination and overgrowth. Factors associated with the isolation of NTM included site variables such as distance from water treatment plants, elevation of sample site above sea level, and certain pipe materials. The species documented to cause disease in humans residing in Brisbane that were also found in water include M. gordonae, M. kansasii, M. abscessus, M. chelonae, M. fortuitum complex, M. intracellulare, M. avium complex, M. flavescens, M. interjectum, M. lentiflavum, M. mucogenicum, M. simiae, M. szulgai, M. terrae. M. kansasii was frequently isolated, but M. avium and M. intracellulare (the main pathogens responsible for disease in QLD) were isolated infrequently.

The main pathogenic species identified in water included M. lentiflavum, M. abscessus, M. kansasii and M. fortuitum. Very few studies have documented M. abscessus in water. The incidence of disease due to this pathogen is increasing in many parts of the world, including QLD. It is particularly difficult to treat, and a significant cause of morbidity and mortality. Patient isolates of M. abscessus referred to the Mycobacterium Reference Laboratory during the same time period as the water sampling, were compared to the water isolates using repetitive unit PCR (Diversilab). Analysis of the water isolates revealed 5 different clusters (Water Patterns (WP) 1-5). Groups of patient isolates clustered with each of these patterns except WP3. A smaller number of patient isolates were unrelated to the water isolates. A high degree of similarity between patient and water isolates, suggests that water may be the source of infection for these patients. This has important implications for those patients at risk for these infections, and for the risk of reinfection following treatment.

Similarity between water and patient isolates was also demonstrated for the less common slow growing M. lentiflavum. This species can cause a variety of infections. Four cases of significant disease due to this pathogen are reported and the international literature summarised in parallel to the molecular epidemiology of water and clinical strains.

In contrast, M. kansasii was frequently isolated from water samples and has been lablelled the ‘tap water bacillus’. Strain types of M. kansasii have previously been typed according to ITS_RFLP patterns as groups I-VI, with group I the most common pathogenic strain in humans in most parts of the world including Brisbane. Very few water isolates were Type I. Types IV and V predominated in water, but were less commonly associated with human infections. Rep-PCR proved to be a more sensitive strain-typing tool for analysis of these isolates, though clusters correlated with ITS_RFLP groups. These results suggest that there may be an alternative source of Type I strains that cause human infection. Despite the high prevalence of other strain types, very few cases of human disease can be attributed to these strains. Hence municipal water may not be the source of infection in most instances.

Strains of M. fortuitum associated with both nosocomial and community acquired infections were compared with municipal water isolates. There was significant strain type heterogeneity, and the clinical and water isolates were diverse. It is postulated then that if water is the vehicle for transmission of nosocomial infections, that point source contamination of water sources within hospitals likely occurs. It is also unlikely that community acquired infections result from exposure to municipal water, and perhaps soil is a more likely source of infection.

To determine whether household water exposure including showering was a risk factor for the acquisition of infection, extensive home sampling was performed on the homes of 20 patients with NTM disease. Mycobacteria isolated from clinical samples from 20 patients included M. avium (5), M. intracellulare (12), M. abscessus (7), M. gordonae (1), M. lentiflavum (1), M. fortuitum (1), M. peregrinum (1), M. vii chelonae (1), M. triplex (1) and M. kansasii (1). Species identified in the home matched that found in the patient in seven (35%) cases – M. abscessus (3), M. avium (1), M. gordonae (1), M. lentiflavum (1), and M. kansasii (1). In an additional patient with M. abscessus infection, this species was isolated from the potable water supply in the vicinity of her home. NTM species grown from aerosols included M. abscessus (3 homes), M. gordonae (2), M. kansasii (1), M. fortuitum complex (4), M. mucogenicum (1) and M. wolinskyi (1). The evidence that patients acquire NTM from their homes appears species specific for M. avium, M. kansasii, M. lentiflavum and M. abscessus. Despite a predominance of disease due to M. intracellulare I found no evidence for acquisition of infection from household water for this species.

In summary, the method for isolation of Mycobacteria from water has been further refined, contributing to the first large scale sampling of the Brisbane Municipal water distribution system. A wide variety of NTM species were found throughout the system, including species known to cause disease in QLD residents. However the strains of these pathogenic NTM were not always the same as the environmental strains. A detailed sampling of NTM patient homes, firstly demonstrated that NTM in water could be aerosolised during showering to a respirable particle size (previously not done). Secondly the finding of strains/species causing human infection within patient homes appears to be species specific. The combined findings of strain comparisons of city wide and patient home sampling, suggests that patients are at risk of infection due to M. abscessus, M. lentiflavum, M. avium and M. kansasii from exposure to municipal water sources and showers. However the divergent findings for M. kansasii and heterogeneity between human and water strains of M. fortuitum suggest there is an alternative environnmental niche for these pathogens. Similarly the striking paucity of M. intracellulare in Brisbane water and patient homes, in an area where disease due to this pathogen predominates, would support the proposal that patients acquire infection with this pathogen from soil/dust/other exposures.

viii

x TABLE OF CONTENTS

KEYWORDS ...... iii ABSTRACT ...... v TABLE OF CONTENTS ...... xi LIST OF TABLES ...... xv LIST OF FIGURES ...... xvii LIST OF ABBREVIATIONS ...... xx ACKNOWLEDGEMENTS ...... xxii STATEMENT OF AUTHORSHIP ...... xxiii LIST OF SUPPLEMENTARY MATERIALS ...... xxiv APPENDICES ...... xxvi CHAPTER 1: INTRODUCTION, HYPOTHESIS AND AIMS ...... 1 INTRODUCTION ...... 1 HYPOTHESIS AND AIMS ...... 3 Hypothesis ...... 3 Aims and Objectives ...... 3 ACCOUNT OF SCIENTIFIC PROGRESS LINKING THE CHAPTERS ...... 5 CHAPTER 2: LITERATURE REVIEW ...... 9 HISTORY ...... 9 PATHOGENIC MYCOBACTERIA IN WATER ...... 12 MYCOBACTERIA IN BIOFILMS ...... 17 MYCOBACTERIA IN POINT OF USE (POU) FILTERS ...... 21 MYCOBACTERIA IN AEROSOLS ...... 23 ANALYTICAL METHODS ...... 25 WATER AS A SOURCE OF PATHOGENIC MYCOBACTERIA CAUSING HUMAN DISEASE ...... 40 REFERENCES ...... 50 CHAPTER 3: A COMPARISON OF METHODS FOR PROCESSING DRINKING WATER SAMPLES FOR THE ISOLATION OF MYCOBACTERIUM AVIUM AND INTRACELLULARE...... 71 ABSTRACT ...... 71 INTRODUCTION ...... 72 METHODS ...... 75 RESULTS ...... 77 DISCUSSION ...... 80 REFERENCES ...... 82 CHAPTER 4 : FACTORS ASSOCIATED WITH THE ISOLATION OF NONTUBERCULOUS MYCOBACTERIA FROM A LARGE MUNICIPAL WATER DISTRIBUTION SYSTEM IN BRISBANE, AUSTRALIA...... 89

xi ABSTRACT ...... 90 BACKGROUND ...... 91 METHODS ...... 92 RESULTS ...... 94 DISCUSSION ...... 101 REFERENCES ...... 106 CHAPTER 5: MYCOBACTERIUM LENTIFLAVUM IN BRISBANE WATER ...... 111 ABSTRACT ...... 111 BACKGROUND ...... 112 AIMS ...... 113 METHODS ...... 113 RESULTS ...... 116 DISCUSSION ...... 123 REFERENCES...... 132 CHAPTER 6: MYCOBACTERIUM ABSCESSUS IN POTABLE WATER: A POTENTIAL SOURCE OF HUMAN INFECTION ...... 137 ABSTRACT ...... 137 INTRODUCTION ...... 138 METHODS ...... 140 RESULTS ...... 141 DISCUSSION ...... 147 REFERENCES...... 150 CHAPTER 7: STRAIN VARIATION AMONGST CLINICAL AND POTABLE WATER ISOLATES OF M. KANSASII USING AUTOMATED REPETITIVE UNIT PCR...... 155 ABSTRACT ...... 155 INTRODUCTION ...... 156 METHODS ...... 158 RESULTS ...... 159 DISCUSSION ...... 164 REFERENCES ...... 168 CHAPTER 8: HETEROGENEITY OF CLINICAL AND ENVIRONMENTAL ISOLATES OF M. FORTUITUM USING REPETITIVE UNIT PCR: MUNICIPAL WATER AN UNLIKELY SOURCE OF COMMUNITY ACQUIRED INFECTIONS...... 173 ABSTRACT ...... 173 INTRODUCTION ...... 174 METHODS ...... 175 RESULTS ...... 177 DISCUSSION ...... 183 REFERENCES ...... 187 CHAPTER 9. ISOLATION OF NTM FROM HOUSEHOLD WATER AND SHOWER AEROSOLS IN PATIENTS WITH NTM PULMONARY DISEASE...... 191 ABSTRACT ...... 191 INTRODUCTION ...... 192 METHODS ...... 193 RESULTS ...... 197 DISCUSSION ...... 210 REFERENCES...... 214

xii CHAPTER 10. CONCLUSIONS AND FUTURE DIRECTIONS ...... 219 REFERENCES ...... 226 SUPPLEMENTARY MATERIAL ...... 228 APPENDIX A ...... 239 APPENDIX B1 ...... 244 APPENDIX B2 ...... 245 APPENDIX C ...... 253 APPENDIX D1 ...... 261 APPENDIX D2 ...... 262 APPENDIX E ...... 269 APPENDIX F ...... 288 APPENDIX G1 ...... 310 APPENDIX G2 ...... 311 APPENDIX H ...... 327

xiii

xiv LIST OF TABLES

Table 2.2. Methods for isolation of mycobacteria...... 31 Table 2.1 Comparison of HPLC, in-house and commercial sequencing and a Line Probe assay for identification of mycobacteria...... 35 Table 3.1. Positive cultures for mycobacteria after concentration by filtration using different culture media after 1 week and 6 weeks (n=66)...... 78 Table 3.2. Comparison of incubation temperatures for culture of mycobacteria in both spiked sterile and tap water samples processed by centrifugation or filtration...... 79 Table 4.1. Summary of NTM positive and negative sampling site variables...... 96 Table 4.2. Presence of pathogenic NTM against different variables ...... 97 Table 4.3. Species of NTM identified in water samples collected in winter and summer...... 99 Table 4.4. Effect of Decontamination on Culture results for different media ...... 100 Table 4.5. Number of NTM isolates cultured using different media...... 101 Table 5.1 ...... 114 American Thoracic Society/ Infectious Diseases Society of America Diagnostic Criteria for Nontuberculous Mycobacterial Lung Disease (Griffith et al., 2007) ...... 114 Table 5.2. Summary of patients with isolates of M. lentiflavum in Queensland 2001- 2008 ...... 117 Table 5.3 Summary of published cases of clinically significant M. lentiflavum isolates† ...... 128 Table 6.1: Sites of isolation of M. abscessus from human samples 2001-2010...... 142 Table 6.2. Clinical details of patients with M. abscessus isolates used for strain typing comparison...... 142 Table 9.1. Clinical details of patients whose homes were sampled...... 198 Table 9.S1. Home sampling results ...... 200 Table 9.2. Supplementary Table of patient isolates, home grown isolates and those found in local distribution system samples...... 206

xvi LIST OF FIGURES

Figure 1.1 Outline of Chapters according to aims and objectives...... 4 Figure 2.1. Flow chart of the analysis process (From Stinear et al., (2004)) ...... 27 Figure 3.1. Flow chart for processing of sterile water samples ...... 75 Figure 3.2. Boxplot demonstrating Median CFU/L (black bar), middle 2 quartiles (box) and range (extent of bar) of CFU/L Mycobacteria in spiked sterile water concentrated by filtration processed with and without the addition of NaThiosulphate...... 77 Figure 5.1. Bar chart of number of patients with isolates of M. lentiflavum per year...... 116 Figure 5.2. Map of Brisbane urban catchment area, human isolates and potable water isolates of M. lentiflavum...... 118 Figure 5.3. Dendrogram and virtual gel images representing rep-PCR fingerprint patterns of 16 human and 7 water isolates of M.lentiflavum (CI clinical isolate, W potable water isolate, BAL bronchoalveolar lavage, * clinically significant isolate) ...... 122 Figure 6.1. rep-PCR dendrogram of water isolates of M. abscessus...... 144 Figure 6.2. rep-PCR dendrogram of examples of main clinical strain patterns 17 (P1=pattern 1; green) and 17 (P2= pattern 2; orange) demonstrating similarities between water and clinical isolates. (Full dendrogram in Additional material 1) ...... 145 Figure 7.2b. Scatterplot of M. kansasii clinical isolates with gridline spacing 5% similarity. Of 14 unique strain types (patterns) the majority of clinical isolates clustered together as Clinical Pattern (CP) 11. Clusters CP1-4 and 11-12 were similar (≥95% similiarity) ...... 161 Figures 7.3a and 3b. Dendrogram and scatterplot of clinical isolates from cluster CP11 with Extended Jaccard analysis. The coloured bars correlate to the different patterns in the dendrogram. The numbered isolates in the scatterplot are coloured according to geographic regional distribution...... 162

xvii Figure 7.4. Dendrogram and scatterplot of 28 different patterns of M. kansasii water isolates. The gridline spacing on the scatterplot correlates with 5% similarity...... 163 Figure 7.5. Scatterplot showing differences between clinical and water isolates of M. kansasii. The colouring of the numbers correlates to the site of origin of the isolate as shown in the side legend (eg brown-potable water, Navy – pulmonary significant isolates)...... 164 Figure 8.3. PFGE patterns of 10 M. fortuitum isolates associated with laparoscopic gastric band procedures (same isolates as for Figure 1). Lanes 1-3: isolates from patient AM, lanes 4-5, patient KG; lanes 6-7 patient RB; lane 8 patient PN; lane 9 patient ML and lane 10 wild type control strain...... 179 Figure 8.5. Scatterplot of numbered M. fortuitum isolates, colour coded according to site of infection, with water isolates in pink (side legend). Gridline spacing correlates with 5% similarity...... 180 Figure 8.4a. Full rep-PCR dendrogram of M. fortuitum water and clinical isolates. Red interrupted line of similarity at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into similar groups (≥95% similarity)...... 181 Figure 8.4b. Full rep-PCR dendrogram of M. fortuitum water and clinical isolates Red interrupted line of similarity at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into similar groups (≥95% similarity)...... 182 Figure 8.4c. Full rep-PCR dendrogram of M. fortuitum water and clinical isolates. Red interrupted line of similarity at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into similar groups (≥95% similarity)...... 183 Figure 9.1: Rep-PCR dendrograms of A. M. kansasii isolated from patient sputum (DP) and samples from the patients home; B. M. lentiflavum isolated from patient (#4:MC)), her home water (#1), and surrounding water distribution sites (#2,3,5), demonstrating >90% similarity; C. M. abscessus subs bolletii from one patient patient (#4 & 5) and M. abscessus subs abscessus grown from shower water (# 1 & 2). M. abscessus control #3. D. M. abscessus isolated from

xviii 2 patients (#2: NF; #3: NW) and from water supply close to their homes (#1 & 4)...... 205 Figure 10.1. Map of Queensland statistical local areas, showing prevalence of M. kansasii infections 2001-2010...... 225

xix LIST OF ABBREVIATIONS

AOC assimilable organic carbon BOM biodegradable organic matter BDOC biodegradable organic carbon COPD chronic obstructive pulmonary disease CPC cetylpyridium chloride DS distribution systems FLA free living amoebae HIV human immunodeficiency virus HPC heterotrophic plate count GORD gastroesophageal reflux disease HPLC high performance liquid chromatography HRM high resolution melt ITS internal transcribed spacer LJ Löwenstein-Jensen MAC Mycobacterium avium complex MGIT mycobacterial growth indicator tube M7H11 Middlebrook 7H11 agar NaOH sodium hydroxide NTM nontuberculous mycobacteria PANTA polymixin azlocillin nalidixic acid trimethoprim amphotericin PCR polymerase chain reaction PFGE pulsed field gel electrophoresis POU point-of-use Rep-PCR repetitive unit PCR RFLP restriction fragment length polymorphism SDW sterile distilled water WP water pattern CP clinical pattern

xxi

ACKNOWLEDGEMENTS

I would like to thank a number of people who have been instrumental in the initiation and completion of this thesis. I am grateful to Prof Joe McCormack for first planting the seed many years ago, that I should consider this journey. Dr David Looke first encouraged me to embark on a career as a clinician researcher in the area of NTM, when completing my Physician training and pregnant with my first child. It wasn’t until Dr Kevin Fennelly from Florida first introduced me to Dr Megan Hargreaves (four children later) that prompted action. Megan, together with Dr Flavia Huygens, has provided supervision and support and without them, this thesis would not have reached fruition.

I am also grateful to the staff of the QLD Mycobacterial reference laboratory including the director Dr Chris Coulter and Chief Scientists Chris Gilpin and Sushil Pandey, for accommodating the multitude of water samples, and the work I peformed there. I would also like to thank Carla Tolson, Robyn Carter, Matthew Torbey, Sharri Minion and Hanna Sidjabat for their assistance with laboratory work.

Finally I would like to thank my patient and understanding family - Richard, Harry, Phoebe, Georgia and Sam, who have supported me throughout this journey, particularly during the many family holidays spent with a laptop!

xxii

STATEMENT OF AUTHORSHIP

The work contained in this thesis has not been previously submitted to meet the requirements for an award at this or any other higher education institution. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person, except where due reference is made.

QUT Verified Signature

Date: 1/12/2013

xxiii LIST OF SUPPLEMENTARY MATERIALS

Additional Material 4.1: Characteristics of the Brisbane Water distribution network. (From National Performance Report 2007-2008: urban water utilities. Downloaded 9/1/2012 from www.nwc.gov.au/publications Australian Government National Water Commission (page last updated 25/11/11)

Additional Material 4.2. Pipe materials and characteristics of different sample types.

Additional material 4.3: Species of NTM isolated from different sample types.

Additional material 6.1. Full Diversilab report of M. abscessus analysis

xxiv

xxv APPENDICES

Appendix A: Thomson, R., Carter, R., Gilpin, C., Coulter, C., & Hargreaves, M. (2008). Comparison of methods for processing drinking water samples for the isolation of Mycobacterium avium and Mycobacterium intracellulare. Applied and Environmental Microbiology, 74(10): 3094-3098.

Appendix B1: Thomson, R., Carter, R., Tolson, C., Huygens, F., & Hargreaves, M. (2011). Pathogenic nontuberculous mycobacteria in the Brisbane potable water distribution system. American Journal of Respiratory and Critical Care Medicine 183: A2395. (Conference poster)

Appendix B2: Thomson, R. M., Carter, R., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2013) Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia. BMC Microbiology, 13:89

Appendix C: Marshall, H., Carter, R., Sidjabat, H., Huygens, F., Hargreaves, M., & Thomson, R. (2011). Mycobacterium lentiflavum in drinking water supplies, Australia. Emerging Infectious Diseases, 17(3).

Appendix D1: Thomson, R. M., Tolson, C., Sidjabat, H., Coulter, C., Huygens, F., & Hargreaves, M. (2012). Mycobacterium abscessus infection and potable water. (2012). American Journal of Respiratory and Critical Care Medicine, 185, A4022. (conference poster)

Appendix D2: Thomson, R. M., Tolson, C., Sidjabat, H., Coulter, C., Huygens, F., & Hargreaves, M. (2013). Mycobacterium abscessus isolated from municipal water – a potential source of human infection. BMC Infectious Diseases, 13:241

Appendix E: Thomson, R.M., Tolson,C.E., Huygens, F., & Hargreaves, M. (2013) Strain variation amongst clinical and potable water isolates of M. kansasii using

xxvi automated rep-PCR. PDF submitted to the Internation Journal of Medical Microbiology.

Appendix F: Thomson, R.M., Tolson, C.E., Huygens, F., & Hargreaves, M. (2013) Heterogeneity of clinical and environmental isolates of Mycobacterium fortuitum using repetitive unit-PCR: municipal water an unlikely source of community acquired infections. PDF (R3) submitted to Epidemiology and Infection, November 2013.

Appendix G1: Thomson, R. M., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2012) Isolation of NTM from household water and shower aerosols in patients with NTM pulmonary disease. American Journal of Respiratory and Critical Care Medicine, 185, A4030. (conference poster)

Appendix G2: Thomson, R. M., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2012) Isolation of NTM from household water and shower aerosols in patients with NTM pulmonary disease. Journal of Clinical Microbiology. 2013, 51(9):3006

Appendix H: Thomson, R. (2010). Changing epidemiology of pulmonary nontuberculous mycobacteria infections. Emerging Infectious Diseases, 16(10): 1576-1582.

xxvii

CHAPTER 1: INTRODUCTION, HYPOTHESIS AND AIMS

INTRODUCTION The environmental mycobacteria (also called atypical mycobacteria, nontuberculous mycobacteria (NTM) or mycobacteria other than TB (MOTT)) are a group of human and animal pathogens that have significant impacts on the morbidity and mortality of humans and important economic impacts on agriculture. They are normal inhabitants of a variety of environmental reservoirs including natural and municipal water, soil, aerosols, protozoans, animals and humans. The incidence of pulmonary disease due to environmental mycobacteria is increasing. Notified cases in Queensland rose from 0.63 (1985) to 1.21 (1994), 2.2 (1999) and 3.2 (2006) per 100 000 population (Thomson, 2010). Reasons for this increase include increased awareness of mycobacteria as pulmonary pathogens, improvements in methods of detection and culture, and an ageing population (as this is often a disease of the elderly).

An increase in exposure is potentially responsible as the organisms are everywhere – in water, biofilms, soil and aerosols and it is likely that people encounter them on a daily basis. Humans are exposed to mycobacteria in water through drinking, swimming and bathing. Aerosols generated during these activities can be inhaled, and water can be aspirated when swallowed. Drinking water has been shown to be at least one source of human infection in AIDS patients – where contaminated water is ingested and infection acquired through the intestinal wall (von Reyn CF, 1994). In a case-control study I have shown that patients with NTM disease are more likely to have Gastrooesophageal reflux disease (GORD) or swallowing difficulties, postulating that NTM enter the lungs via the aspiration of contaminated water. (Thomson, 2007)

Outbreaks of pulmonary disease due to mycobacteria have been linked to the isolation of M.xenopi and M.kansasii from drinking water samples (McSwiggan, 1974; Slosarek, 1993; Wright, 1985). Mycobacterial cervical lymphadenitis in children was previously predominantly due to M.scrofulaceum, however changes in

1 chlorination and fluoridation are thought to have caused the shift to predominantly M.avium disease, as M.scrofulaceum is more chlorine sensitive than M.avium (Wolinsky, 1995). Whilst disinfection has reduced a number of human diseases, it may also be responsible for the increased incidence of mycobacterial disease. It is postulated that disinfection may also contribute in part to the persistence of M. avium and M.intracellulare in drinking water distribution systems (Falkinham III JO, 2001; Covert, 1999; von Reyn, 1994). These organisms are many times more resistant to chlorine, chloramine, chlorine dioxide and ozone than other water borne micro-organisms such as E.coli. As a result disinfection selects for mycobacteria, as competition for growth is lacking.

Whilst NTM DNA has been isolated from showerheads (Feazel, 2009), the isolation of pathogenic strains of mycobacteria from aerosols generated by showers has not been demonstrated. If proven, this finding would have relevance not only to the individual patient, in whom behavior modification may enhance the likelihood of treatment success and reduce reinfection after treatment, but may have wider public health implications for water management.

Previous investigators have documented the presence of mycobacteria in soil and house dusts in QLD and have shown approximately 50% of environmental species were also found in human specimens such as sputum (Reznikov M, 1971; Reznikov, 1971; Reznikov, 1971; Dawson, 1971). Evidence from studies done in the US and Finland, shows that water is an important environmental reservoir of mycobacteria causing human disease. There have been no published studies examining the presence of mycobacteria in water distribution systems in Australia, and routine sampling/monitoring is not performed as part of public health practice.

It is predicted that interaction between humans and mycobacteria will increase, resulting in more cases of disease. Public and environmental health efforts must focus on actions that will specifically remove mycobacteria from habitats where susceptible humans are exposed. However, the evidence that pathogenic strains of NTM are present in Australian drinking water, and that these strains are the same as those that are causing disease in Australian patients, currently does not exist.

2 Hence, public health authorities do not believe the presence of mycobacteria in potable water poses a significant risk.

HYPOTHESIS AND AIMS Hypothesis My hypothesis is that environmental mycobacteria, found in water and air of patients’ homes, will prove to be a source of infection in at-risk patients. To prove this hypothesis, I have developed a series of aims and objectives as follows:

Aims and Objectives Aim 1 To determine that species/strains of environmental mycobacteria causing human pulmonary disease can be found in water distribution systems in Brisbane.

Objective 1: To refine the method for identification of mycobacteria from potable water samples.

Objective 2: To document the species of environmental mycobacteria in Brisbane water distribution systems.

Aim 2: To show that the strains of mycobacteria found in potable water are the same as those found in patients living in the area supplied by that water.

Objective 3: To compare the location of positive water sampling sites with the geographic distribution of disease.

Objective 4: To compare, using molecular typing methods, strains of mycobacteria found in water with those responsible for pulmonary disease in Brisbane patients.

Aim 3: To demonstrate that strains of environmental mycobacteria known to cause human pulmonary disease can be aerosolized during normal household activities.

3 Objective 5: To test air, water and biofilm in patient homes for the presence of environmental mycobacteria known to cause pulmonary disease, particularly focussing on aerosol producing sites such as showers.

Chapter1 Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1: Objective 1 Aim 3: Objective 5 Aim 2: Objectives 3&4 Chapter 3: Chapter 9: Isolation Chapter 5: Development of of NTM from homes method for isolation Mycobacterium of patients with NTM lentilavum of NTM from water disease

Aim 1:Objective 2 Chapter 4: Isolation Chapter 6: of NTM from M. abscessus Brisbane water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

Figure 1.1 Outline of Chapters according to aims and objectives.

4 ACCOUNT OF SCIENTIFIC PROGRESS LINKING THE CHAPTERS

The thesis consists of seven papers for publication (four published, two submitted and under initial review, one reviewed and revision submitted). The relationship between the various chapters/papers is shown in Figure 1.1

Chapter 3 relates to Aim 1, objective 1. This was presented in abstract (poster form) at the Australian Society of Microbiology meeting in Adelaide (July 2007) and then published in 2008.

Thomson, R., et al. (2008). Comparison of Methods for Processing Drinking Water Samples for the Isolation of Mycobacterium avium and Mycobacterium intracellulare. Applied and Environmental Microbiology, 74(10): 3094-3098. Appendix A – published version.

Chapter 4 relates to Aim 1, objective 2 and reports the isolation of NTM from the Brisbane water distribution system. This was presented in poster form at the American Thoracic Society ASM Denver 2011 and published in 2013.

Thomson, R., C. Tolson, et al. (2011). Pathogenic nontuberculous mycobacteria in the Brisbane potable water distribution system. American Journal of Respiratory and Critical Care Medicine 183: A2395 (Appendix B1 – conference poster)

Thomson, R. M., Carter, R., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2013) Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia. BMC Microbiology, 13:89 (Appendix B2- published version)

Chapters 5-8 relate to Aim 2. Objectives 3 and 4 were performed for each of the main pathogenic species found in Brisbane water, and each species reported separately.

Chapter 5: Mycobacterium lentiflavum (presented in abstract (poster) form at the European Respiratory Society meeting in Barcelona (September 2011) and published the same year.

Marshall, H., R. Carter, et al. (2011). Mycobacterium lentiflavum in drinking water supplies, Australia. Emerging Infectious Diseases, 17(3). (Appendix C – published version)

Chapter 6: M. abscessus

“Mycobacterium abscessus infection and potable water” was presented in abstract form at the Thoracic Society of Australia and New Zealand (Canberra 2012). An oral presentation was allocated and won the best presentation for the Infectious Diseases SIG. It was also presented at the American Thoracic Society meeting (San Francisco 2012). 5

Thomson, R. M., Tolson, C., Sidjabat, H., Coulter, C., Huygens, F., & Hargreaves, M. (2012). Mycobacterium abscessus infection and potable water. (2012). American Journal of Respiratory and Critical Care Medicine, 185, A4022. Appendix D1 – conference poster.

Thomson, R. M., Tolson, C., Sidjabat, H., Huygens, F., & Hargreaves, M. (2013) Mycobacterium abscessus isolated from municipal water – a potential source of human infection. BMC Infectious Diseases 13;241 (Appendix D2 – published version)

Chapter 7: Mycobacterium kansasii

Thomson, R., Tolson, C., Huygens, F., & Hargreaves, M. (2013). Strain variation amongst clinical and potable water isolates of M. kansasii using automated repetitive unit PCR. Submitted to International Journal of Medical Microbiology Feb 14th, 2013, currently under review. (IJMM-D-13-00047) Appendix E.

Chapter 8: Mycobacterium fortuitum

Thomson, R., Tolson, C., Carter, R., Huygens, F., & Hargreaves, M. (2013) Heterogeneity of clinical and environmental isolates of M. fortuitum using repetitive unit PCR: Municipal water an unlikely source of community acquired infections. This manuscript was submitted to Epidemiology and Infection in June and R3 submitted November 2013. Appendix F.

Chapter 9 relates to Aim 3, objective 5 and reports the results of sampling of patients’ homes. This was presented in abstract form at the TSANZ (Canberra 2012) and ATS (San Francisco 2012)

Thomson, R., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2012). Isolation of Nontuberculous mycobacteria from household water and shower aerosols in patients with NTM Pulmonary disease. American Journal of Respiratory and Critical Care Medicine, 185, A4030. Appendix G1 – conference poster.

The manuscript was published in the Journal of Clinical Microbiology 2013, 51(9):3006 (ahead of print 10th July, 2013) Appendix G2

Chapter 10 Summary and Future Directions

Chapter1 Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1: Objective 1 Aim 3: Objective 5 Aim 2: Objectives 3&4 Chapter 3: Chapter 9: Isolation Development of Chapter 5: of NTM from homes Mycobacterium method for isolation of patients with NTM lentilavum of NTM from water disease

Aim 1:Objective 2 Chapter 4: Isolation Chapter 6: of NTM from M. abscessus Brisbane water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

CHAPTER 2: LITERATURE REVIEW

Falstaff: “What says the doctor to my water?” Page: “He said the water itself was good healthy water; but for the party that owned it, he might have more disease than he knew for.” Shakespeare, Henry IV, part II

HISTORY

After the tubercle bacillus was discovered in 1885, there were initially sporadic reports of isolates of Nontuberculous Mycobacteria –NTM, (“Atypical TB”, “Atypical mycobacteria”). For the following 50 years little attention was given them as serious pathogens in humans. By the early 1950’s, the concept of disease due to NTM (mycobacteriosis) had become established. Since then however, the significance of isolating NTM from respiratory specimens has been unclear. The pathogenic potential of the different species varies greatly; colonization of an individual without infection or invasion occurs, and because of its ubiquitous presence coincidental isolation of NTM can occur.

NTM have traditionally been classified according to the Runyon system. Rapid growing mycobacteria (RGM) are characterised by growth on solid media (e.g. Middlebrook 7H11) within 7-10 days, compared with the slow growing species where growth usually appears within 10-21 days, but may take six weeks. Mycobacterium avium complex is a group of slow-growing species of mycobacteria most commonly associated with human pulmonary disease. The complex includes M. intracellulare and M. avium – the former being the most common species isolated from pulmonary specimens in Queensland (Thomson, 2010). MAC

9 organisms are associated with different diseases in the lung. The most traditional and well-recognized condition is cavitary upper lobe disease, seen commonly in middle-aged smoking males. Increasingly patients are being identified with middle lobe and lingula bronchiectasis, seen more in elderly females with no obvious clinical risk factors. A third type of disease is Hypersensitivity pneumonitis, in which there is a reaction to inhaled mycobacterial antigen often without evidence of invasive disease.

Mycobacteria have been isolated from environmental sources for over 50 years. They were first reported in water in 1922, by Maie (reviewed by Collins (Collins, Grange & Yates, 1984)). Kubica published an abstract in 1961, describing the identification of mycobacteria from soil and water samples in the state of Georgia (USA) (Kubica, Beam, Palmer & Rigdon, 1961). Goslee and Wolinsky (1976) evaluated 321 water samples – including natural waters, treated waters, and waters in contact with animals. They identified mainly slow growing NTM, including species of Mycobacterium avium complex known to be pathogenic to humans.

During the 1970s, Dawson and Reznikov documented the presence of mycobacteria in soil and house dusts in Queensland and showed approximately 50% of environmental species were also found in human specimens such as sputum (Dawson, 1971; Reznikov, Leggo, Dawson, 1971; Reznikov & Dawson, 1971). In the Adelaide area--where pulmonary mycobacterial disease was rarely encountered at the time- only 17% of environmental soil samples were positive for serotypes known to be disease-associated (Reznikov & Dawson, 1980). These findings suggest that the disease-associated MAC serotypes are significantly more common in the environment of the Brisbane area.

Given the close association between water and soil, Tuffley and Holbeche (1980) tested water samples from rainwater tanks and identified disease-associated serotypes of M. intracellulare and M. avium from 32 of 141 rainwater tanks in Central Queensland, 7/32 tanks in the hinterland of Rockhampton and 2/32 sampled repetitively near Toowoomba. The humans who consumed the water from

10 these tanks were free of symptoms at the time, but not formally medically assessed. The authors felt that the mycobacteria adhered to dust particles disturbed by mechanical cultivation and subsequently contaminated the tanks.

Skin testing for evidence of tuberculosis was commonly performed during this period, however it was recognized that cross reactivity between tuberculin antigens and environmental mycobacterial antigens occurred. This resulted in false positive tests in areas of high exposure to environmental mycobacteria. Falkinham sampled widely in the US (Gruft, Loder, Osterhout, Parker & Falkinham, 1979), from both soils from riverbeds, and water from rivers, lakes and other natural sources, and reported distinct geographic variations in the isolation of mycobacteria. He was able to correlate this with skin test reactivity in naval recruits in the US in some areas but not all (Brooks, Parker, Gruft & Falkinham, 1984; Falkinham, Parker & Gruft, 1980). Abrahams performed a number of studies of dual skin testing in Queensland School children in the 1960s and found that exposure to environmental mycobacteria was common (measured by reactivity to antigens of M. avium)(Abrahams, 1970; Abrahams & Harland, 1967; Abrahams & Harland, 1968; Abrahams & Silverstone, 1961).

From the studies in the USA a distinct geographic variability in the isolation of mycobacteria from environmental sources has been documented (Falkinham et al., 1980; Falkinham, 1996; Kirschner, Parker & Falkinham, 1992). Similarly there are geographic areas of high incidence of disease documented around the world (Marras & Daley, 2002; Martin-Casabona et al., 2004). The optimum temperature for growth of M. avium is between 10°C and 45°C (Falkinham, 2002), a range which includes typical ambient temperatures experienced in Queensland. Few publications have come from Australia regarding the epidemiology of disease, though from laboratory reports of the frequency of isolation of NTM (Haverkort & Australian Mycobacterium Reference Laboratory Network, 2003) and from anecdotal reports, disease is most common in QLD, Western Australia and the Northern Territory (O'Brien, Currie & Krause, 2000; Pang, 1991).

PATHOGENIC MYCOBACTERIA IN WATER Mycobacteria in Drinking water distribution systems

It has been shown that mycobacteria are resident in drinking water distribution systems (DS) (Covert, Rodgers, Reyes, & Stelma, 1999; Engel, Berwald & Havelaar, 1980; Falkinham, Norton, & Le Chavallier, 2001; Gangadharam, Lockhardt, Awe & Jenkins, 1976; Le Dantec et al., 2002b; Mankiewicz & Majdaniw, 1982; McSwiggan & Collins, 1974; Pryor et al., 2004; September, Brozel, & Venter, 2004; Torvinen et al., 2004). They have also been found in hospital water distribution systems (du Moulin, Stottmeier, Pelletier, Tsang & Hedley-White, 1988; Fox et al., 1992; Galassi et al, 2003; Tobin-D'Angelo et al., 2004; Fernandez-Rendon et al., 2012) and domestic tap water (Chang, Wang, Liao & Huang, 2002; Gangadharam, et al., 1976; Goslee & Wolinsky, 1976; Peters, et al., 1995; von Reyn, Marlow, Arbeit, Barber, & Falkinham, 1994). Mycobacteria were isolated from 38% (16/42) of drinking water DS in the USA (Covert et al., 1999), from 21.3% (42/197) in Greece (Tsintzou, Vantarakis, Pagonopoulou, Athanassuadou, & Papapetropoulou, 2000) and 72% (104/144) in Paris (Le Dantec et al., 2002b). Mycobacterial numbers were similar in DS that used groundwater compared to surface water (Covert et al., 1999). Mycobacteria were found in Finnish DS samples – from 35%, up to 80% at sites more distal in the network (Torvinen et al., 2004).

Seasonal variations in the prevalence of mycobacteria in environmental samples has been explored. Kubalek (Kubalek & Komenda, 1995; Kubalek & Mysak, 1996) sampled potable water from 16 localities in Czechoslovakia -340 running water and 346 tap swabs or scrapings between 1984 and 1989. Forty two percent of samples were positive for mycobacteria, with significantly more positive in Spring than Autumn. No MAC was identified, however Löwenstein-Jensen (L-J) medium was used and the cultures were read at 1 week. L-J is a slower medium for identification of MAC and hence the cultures were probably not incubated for long enough.

12 Species identified included M. gordonae, M. flavescens, group IV, M. aurum, M. scrofulaceum, and M. fortuitum.

The effect of different water treatment processes and water sources on NTM presence has also been investigated. Le Dantec (Le Dantec et al., 2002b) compared the frequency of isolation of mycobacteria from two different treatment plants in Paris (France) one that utilized slow (gravity fed) and the other rapid (pressure applied), sand filtration. Overall 72% of samples were positive for mycobacteria, with no differences between groundwater samples (68%) and treated surface water (69%) and water from other (usually mixed) origins (83%). They reported that slow sand filtration was more effective at removing mycobacteria than rapid sand filtration. Saprophytic mycobacteria (present in 41.3% of positive samples), potentially pathogenic mycobacteria (16.3%), and unidentifiable mycobacteria (54.8%) were isolated from 12 sites within the Paris water distribution system. M. intracellulare was detected in only one sample, which may be a reflection of the decontamination method used (discussed later). They also found that mycobacteria could accumulate and grow on granular activated carbon used as a step in the treatment process. The mycobacterial species that were isolated at treatment plants were different to those isolated from distribution samples, suggesting that mycobacteria grew or entered downstream in the system.

Similarly Falkinham (Falkinham, et al., 2001) sampled eight distribution systems over an 18-month period (528 water, 55 biofilm samples); overall recovery of mycobacteria was low at 15%, in the range of 10-700000 CFU/L. He demonstrated significantly higher mycobacterial numbers in distribution samples (average 25000 fold) than those collected immediately downstream from treatment plants, indicating that mycobacteria grew in the distribution system. Mycobacterial numbers correlated with assimilable organic carbon (AOC) and biodegradable organic carbon (BDOC), and M. avium correlated with turbidity. M. intracellulare was rarely found in water samples but isolated from six of eight biofilm systems, and in high numbers (average 600CFU/cm3). Falkinham concluded from these

13 findings that the environmental niches for M. avium and M. intracellulare were different.

The theory that mycobacteria replicate within water systems is further supported by the finding of greater numbers of mycobacteria in recirculating hot water systems, than in input systems (du Moulin, et al., 1988). One mechanism by which mycobacteria may survive and replicate is through the intracellular survival within amoebae. There has been increasing interest in the observation that free-living amoebae (FLA) may increase both the numbers and virulence of water based organisms such as mycobacteria (Thomas & Ashbolt, 2011; Adekambi, Salah, Khlif, Raoult, & Drancourt, 2006). FLA colonization and regrowth within reservoirs and storage tanks has been documented (Thomas, Loret, Jousset, & Greub, 2008). They have been consistently detected at point of use, in up to 45% of samples (Marciano- Cabral, Jamerson, & Kaneshiro, 2010). Species identified within FLA in water include human pathogens M. chelonae, M. fortuitum and M. chimaera, (Pagnier, Merchat, Raoult & La Scola, 2009 ) and M. gordonae, M. kansasii, M. xenopi (Thomas, Herrera-Rimann, Blanc, & Greub, 2006).

It has been postulated that amoebal cysts may protect intracellular mycobacteria against adverse conditions and may act as a vector for mycobacteria (Salah, Ghigo, & Drancourt, 2009). In vitro, Cirillo et al. have shown that M. avium can enter A. castellani and replicate (Cirillo, Falkow, Tompkins, & Bermudez, 1997). Compared to bacteria grown in broth, those grown within the amoebae had enhanced entry and intracellular growth, implying enhanced virulence. Bacterial proliferation in laboratory cultures was also enhanced in the presence of amoebae in a study by Marciano-Cabral et al (2010). This was further assessed in the beige mouse model of infection. Furthermore Miltner and Bermudez (2000) assessed the susceptibility of M. avium grown with A. castellani to antibiotic (rifabutin, azithromycin, clarithromycin) in vitro, demonstrating a reduced effectiveness of these antibiotics. Further research is needed to determine the importance of amoebae in the acquisition and pathogenesis of NTM infection in humans.

14 Disinfection resistance

The persistence of NTM in water systems is also related to their relative resistance to disinfection. The mycolic acid concentration, slow growth, biofilm formation and hydrophobicity of NTM contribute to their resistance to chemical disinfection. Mycobacteria are resistant to varying concentrations of chlorine at concentrations used for disinfection in distribution systems (Le Dantec et al., 2002a; Payment, 1999; Pelletier, et al., 1988). Pelletier et al., (1988) measured chlorine resistance of M. fortuitum, M. gordonae and MAC strains from surface water and clinical isolates of M. chelonae, M. kansasii and M. intracellulare. Free chlorine concentrations of 1mg/L eliminated all of the strains tested within 8 hours of exposure, whereas 0.15mg/L had virtually no effect. Free chlorine residuals measured at sites within the Boston distribution system were between <0.1mg/L and 0.6mg/L. This system used halogen treatment to disinfect by chloramination. The theory is that this combined chlorine degrades at a slower rate than free chlorine and as a result remains in the system for longer periods of time resulting in greater contact times with water borne bacteria. There may also be lower levels of regulated disinfection- byproducts and a reduction in Legionella species (Pryor, 2004). In this Boston study free chlorine concentrations gradually decreased as water travelled down the system because of interactions with aging pipe surfaces. Isolation rates of mycobacteria from the system were observed in an inverse relationship to decreasing chlorine concentrations.

The time (minutes) to 99.9% inactivation (Ct99.9%) of M. avium strains has been reported as 700-3000 times greater than that for E. coli. (Taylor, Falkinham III, Norton, & LeChevallier, 2000). High levels of resistance of M. avium strains to chlorine, monochloramine, chlorine dioxide and ozone for both water grown and medium grown strains have been demonstrated. Le Dantec et al., (2002a) also investigated the chlorine resistance of mycobacteria in distribution systems. Relative chlorine resistance was reported for different species of mycobacteria, however M. avium and M. intracellulare were not included. Mycobacteria grown in DS appear to be more resistant to chlorine than those grown in culture. For M.

15 gordonae, low nutrient conditions increased the resistance to chlorine, but with increasing temperature and lower pH, this resistance reduced.

Pryor et al (2004) measured changes in Legionella sp. and mycobacteria before and after change from chlorine to chloramine disinfection. Mycobacteria were found throughout the DS main lines, backflow valve, home water meters and hose bibs and where residual chlorine was >0.3mg/L. Following the switch to chloramination, mycobacteria appeared to be more common, as shown by predominant bands from DGGE gels, possibly due to reduced competition, as many other species were reduced or absent. M. avium was isolated only once; M. intracellulare was the most frequently isolated. Mycobacterial populations in biofilms and water samples from showerheads appeared to increase when the disinfectant changed to chloramines – 19.1% to 42.2%, in biofilm samples from 17.7 to 27.1%, and in bulk water samples – 20.8 to 57.3%. Many species were found in the DS, however at point of use only four mycobacterial species were detected (M. gordonae and M. intracellulare being dominant). Pretreatment with a relatively high concentration of CPC (0.8%) may have led to an underestimation of mycobacterial numbers in this study.

More recently Lee et al (Lee, Yoon, Lee, Han, & Ka, 2010) explored the effect of low (10W) and medium (1kW) pressure UV irradiation in the inactivation of M. avium, M. fortuitum, M. intracellulare and M. lentiflavum – all human pathogens. Whilst the sensitivity to UV was found to be species specific (M. fortuitum being most resistant), all species were found to be resistant at both UV levels. The susceptibility of M. abscessus, M. massiliense, M. fortuitum and M. smegmatis to quaternary ammonium, commonly used in different commercial disinfectants, was studied by Cortesia et al., (Cortesia, Lopez, de Waard, & Takiff, 2010) and there were phenotypic changes in those strains that persisted after exposure to disinfection. M. smegmatis is often selected as a testing strain and was shown to be more susceptible to these disinfectants than the other pathogenic NTM tested. The persistence of high numbers resistant strains of NTM even after 24 hours of exposure to disinfectant, led the authors to recommend that these agents not be

16 relied upon for adequate disinfection in surgical settings, as the NTM they tested are often associated with nosocomial infection.

The isolation of NTM from hot water systems has also been shown to relate to the type of disinfection used. Sebakova et al., (2008) compared the hot water systems of 4 hospital buildings using different types of disinfection – hydrogen peroxide and silver, thermal disinfection, chlorine dioxide and none. Samples collected over a 3- month period revealed pathogenic NTM including M. kansasii, M. xenopi, and M. fortuitum and the ‘often contaminating’ M. gordonae. The highest concentrations of NTM were found in the system using silver and hydrogen peroxide (97% samples positive) and those with no additional disinfection (70%). Only 20% samples from the chlorinated system and none from the thermal treated system were positive. Water temperatures greater than 50 °C were associated with significantly less NTM.

MYCOBACTERIA IN BIOFILMS

As mentioned above several authors have demonstrated that mycobacteria grow in distribution systems and are able to form biofilms inside pipes (Falkinham et al., 2001; Lehtola, Juhna, Miettinen, Vartiainen, & Martikainen, 2004; Lehtola, et al., 2004b; Lehtola et al., 2005; Lehtola, et al., 2006a; Lehtola, Torvinen, Miettinen, & Keevil, 2006b; Lehtola et al., 2007; Norton, LeChevallier, & Falkinham III, 2004; Schulze-Röbbecke, Janning & Fischeder, 1992; September et al., 2004).

Formation of biofilm is likely to dependant on several variables. Norton and Chevallier (2000) studied the compositions of biofilms in DS supplied with water that was conventionally treated (high levels Biodegradable organic matter (BOM)) and biologically treated (low levels BOM). They found that iron pipes stimulated the rate of biofilm development and bacterial levels on disinfected iron pipes exceeded those for chlorinated PVC pipes. Mycobacterium spp were found in 75% of the biofilm samples from conventionally treated water, but none were isolated from the biologically treated system with free chlorine used as a post disinfectant.

17 (Species identified included M. intracellulare, M. fortuitum and M. phlei). This supports previous findings that limiting BDOC can limit mycobacterial growth in water; by confirming it can also decrease growth in biofilms. It is postulated that the iron pipe surface influences the composition, activity and disinfection resistance of biofilm bacteria.

In Falkinham’s study (Falkinham et al.,2001) biofilm samples were collected from water meters and portions of pipe material wrapped in plastic wrap to prevent drying. There were 55 biofilm samples – 40 (69%) yielded 450 individual acid-fast isolates – 267 Mycobacterium spp of which 131 were M. intracellulare and 4 M. avium. M. intracellulare was found in 13/40, one from every site except two yielded M. intracellulare (mean CFU 600/cm2 (1-2850)). The material supporting biofilms did not seem to make a difference in frequency of recovery (bronze or brass 63%, galvanized 100%, plastic 64% - M. intracellulare: brass/bronze 25%, galvanized 0% and plastic 43%), though the numbers in each category were small. However as the authors state -“proof that surface composition influences mycobacterial numbers will require comparison of different surfaces in the same system”. There was no apparent effect of residual disinfectant type or concentration on recovery of mycobacteria from DS or biofilm samples in this study.

Biofilm analysis from chloraminated and chlorinated DS in South Africa, however, failed to reveal any MAC (September, 2004).

In order to assess the contribution of different pipe surfaces to biofilm formation, Lehtola et al (2004a; 2004b) studied changes in water quality and formation of biofilms in a pilot-scale water distribution system with copper and polyethylene (PE) pipes. The formation of biofilms was slower in copper pipes than in the PE pipes, but after 200 days there was no difference in microbial numbers between the two. Copper ion led to lower microbial numbers in water during the first 200 days but thereafter there were no differences between the two pipe materials. They found PE pipes released phosphorous during the initial phase of use, possibly increasing microbial growth potential. Previously they had found that phosphate (PO4) could

18 increase the formation of biofilms in PO4 limited water (Lehtola, Miettinen, Vartiainen, Martikainen, 2002).

There are several bacterial species growing in biofilms of copper pipes that can increase the copper concentration in drinking water. Copper has been shown to be toxic to bacteria in water and in the latter study (Lehtola at al, 2004b) copper in pipes decreased the heterotrophic plate count (HPC) in water during the first 200 days. After 200 days there were almost equal numbers in PE and copper pipes, and in inlet and outlet water. At the same time the number of bacteria growing on copper pipes reached the levels in biofilms growing in PE pipes. Thus the reason for the increase in bacterial numbers in outlet water of copper pipes is probably attributable to bacteria detaching from biofilms where the bacterial population had adapted or been selected to live on the copper surface. Cell densities in biofilms growing on copper were lower than those growing on plastic in previous studies – possibly as only young biofilms were studied and these authors suggested the formation of biofilm on copper requires more time than the corresponding formation on plastic.

Flow velocity of water in this first study was constant and low, thus not exactly representing the real situation in DS networks, and this could have influenced biofilms. In a subsequent study, the same authors found that the formation of biofilms increased with the flow velocity of water. The increase in microbial numbers and ATP content was clearer in PE than copper pipes; this was also seen as increased consumption of microbial nutrients in the pipeline system. This indicates that the mass transfer of nutrients plays a major role in the growth of biofilms. However the increased biomass of biofilms did not affect microbial numbers in the water. Rapid changes in water flow rate resuspended biofilms and sediments, which increased the concentrations of bacteria and copper in water. Transient pressure events that occur in DS may therefore be associated with transient changes in the concentration of mycobacteria in water due to release from biofilms. LeChevallier (Le Chevallier, Gullick, Karin, Friedman & Funk, 2003) also suggested that during

19 negative pressure events pipeline leaks might provide a potential portal for entry of groundwater into treated drinking water.

To examine these issues further Norton et al., (2004) looked at the survival of M. avium in a model distribution system and found that it was dependent upon a complex interaction between pipe surface, nutrient levels and disinfectants. They compared chlorinated PVC (cPVC), copper, galvanized and black iron pipes under varying temperature and pH conditions and with the addition of different disinfectant chemicals and M. avium organisms grown in broth. Whilst previously it had been shown that M. avium levels increased with higher levels of assimilable organic carbon (AOC), in this study they demonstrated growth in relatively low nutrient levels – lower than those present in water utilities of North America. When biofilms were grown on non-corroded surfaces (copper or PVC) free chlorine was more effective for controlling heterotrophs and M. avium, but monochloramine controlled bacterial levels better on corroded iron pipe surfaces. M. avium biofilm levels were higher on iron and galvanized pipe surfaces than on copper or cPVC surfaces. When copper surfaces were exposed to disinfectants, M. avium constituted nearly 100% of organisms identified. Therefore during pressure transients in buildings that utilize copper pipes there may be significant release of M. avium from biofilms into water.

M. avium can survive in biofilms for at least several weeks, even under conditions of high shear turbulent flow (Lehtola et al., 2007; Torvinen, Lehtola, Martikainen, & Miettinen, 2007). During a series of experiments comparing effects of flow velocity, temperature and availability of phosphorous, culturability of M. avium increased, as did the numbers of culturable M. avium cells in outlet water. They attributed this to an adaptation of M. avium to live in water and biofilms; however other authors have suggested that the organisms replicate within biofilms over time (Falkinham et al., 2001). It has also been shown that temperature is a more important factor than the availability of nutrients like phosphorous (Torvinen et al., 2007).

20 In contrast, heat treatment has been suggested as a method of minimizing M. avium. Schulze-Robbecke and Buchholtz, (1992) reported a 90% reduction in M. avium levels by heat treatment at 60°C for 4 minutes under laboratory conditions. Du Moulin et al., (1988) however, recovered M. avium from a hospital hot water system with temperatures between 52° and 57°C. In Norton’s study (Norton et al., 2004) the effectiveness of heat treatment was dependent upon the pipe material, the level of nutrients and the effluent temperature. Treatment was most effective for low-nutrient hot water (50°) in copper pipes.

The ability of mycobacteria to form biofilms is not only important for their survival within water distribution systems, but also in their ability to colonise invasive patient access devices (such as intravascular catheters) and the small terminal alveoli within the lung with progression to infection and disease. The ability of 168 strains of rapidly growing mycobacteria (including 41 clinically significant isolates) to form biofilm was explored by Martin-de-Hijas et al., (2009). Clinically significant isolates were more likely to form biofilms, suggesting that this characteristic is not only important for survival but also for virulence.

To minimize M. avium biofilm formation in drinking water DS therefore, it is necessary to reducing the organic material in drinking water, control corrosion in pipes, maintain an effective disinfectant residual and manage hot water temperatures.

MYCOBACTERIA IN POINT OF USE (POU) FILTERS

The ability of mycobacteria to form biofilms allows them to readily colonize point of use water filters. Several studies have assessed the ability of these filters to remove mycobacteria.

Rodgers et al (Rodgers, Blackstone, Reyes, & Covert, 1999) looked at colonization of POU water filters by silver resistant NTM, particularly pour-through filtration devices that use granular activated carbon to remove chlorine and contain silver as

21 a bactericidal agent to improve the taste of water. They found that NTM were able to multiply within the carbon filters impregnated with silver and were released into the water passing through. M. avium was able to grow in the presence of high concentrations of silver; M. fortuitum and M. mucogenicum were inhibited at low concentrations. Over a period of 8 weeks, M. avium levels peaked at weeks 4-5 and tapered by week 8.

Sheffer et al (Sheffer, Stout, Wagener, & Muder, 2005) evaluated a sterile disposable POU filter that contains a Pall Nylon 6.6 Posidyne filter membrane rated and validated at 0.2µm (Pall-Aquasafe Water Filter, Pall Corp Pall Medical, Ann Arbor, MI), designed to eradicate Legionella, Heterotrophs and mycobacteria from hot water samples. These filters are designed to be used for a maximum of seven days following initial installation and must be replaced after this time. The investigators collected 120ml hot water samples at 40-45°C; 1 sample was collected immediately on turning on the tap (‘Immediate’); another after a 1-minute flush (‘Post’); and all cultures were performed in duplicate. No M. gordonae was isolated from POU filtered water, but it was found in 10.3% of control samples. No mycobacterial species were isolated from the postflush water samples from either the control taps or POU filter faucets. Unfortunately the authors don’t report other mycobacterial species in control samples; therefore one can’t assess the efficacy of the filter for other species. The need to change the filters weekly suggests there is strong potential for biofilm formation and a subsequent increase in bacteria if the filters were left too long.

Chaidez and Gerba (2004) compared water quality from a tap mounted POU water activated carbon treatment device vs tap water with POU connections vs tap water without POU devices. The highest concentration of bacteria was found in POU- treated water; with acid-fast organisms in 43.8% of samples; followed by tap water with POU then tap water without a POU device. These tap mounted units were installed 1 week before first samples collected and each site sampled weekly for 6 weeks. The presence of Pseudomonas aeruginosa and NTM were similar in POU treated (NTM 43.8%) and POU connected (NTM 35%) – with the occurrence in tap

22 water being half of that observed in other samples (NTM 16.6%). These findings again imply that POU connections and treatment devices promote biofilm formation, and amplify numbers of bacteria.

The ability of mycobacteria to form biofilm is important for their survival within distribution systems and household plumbing, and as sloughing into water from biofilm can occur intermittently, the levels of NTM detected in water alone may not be a true reflection of the potential for NTM dissemination from water sources. Hence biofilm sampling must be considered when investigating sources of NTM infection.

MYCOBACTERIA IN AEROSOLS

The ability of mycobacteria to aerosolise from natural waters was first demonstrated in 1979 (Parker, Ford, Gruft & Falkinham, 1983; Wendt, George, Parker, Gruft, & Falkinham, 1980) and Gruft et al (1979). Natural transfer of mycobacteria from seawater to air occurs as a result of binding of mycobacterial cells to air bubbles and the ejection of water droplets after the air bubbles reach the liquid surface interface (Parker et al., 1983). Aerosolisation can result in >1000 fold increase in the numbers of viable mycobacterial cells per ml of water droplets ejected from water. It was concluded from these studies, that cell surface hydrophobicity was the major determinant of enrichment in ejected droplets (as opposed to surface charge), and that the rate of transfer of mycobacteria from water to air was subject to prevailing physicochemical conditions. Salts or detergents reduce the rate of transfer of mycobacteria from water to air. The mycobacteria in these natural aerosols were also found in particles and droplets of a respirable size (ie <5µm) that could enter the lung at the alveolar level.

Rautiala et al (2004) demonstrated that potentially pathogenic mycobacteria were released into workplace air during the remediation of buildings. Air samples were taken during the remediation/dismantling of 3 buildings in Finland, generally within 5 metres of the workers. They compared 3 methods of air sampling - the 6-stage

23 Anderson impactor, an all-glass impinger, and filter sampling. Impactor samples were collected onto Petri dishes containing M7H11 supplemented with OADC enrichment (100ml/L), malachite green (25mg/L) and cycloheximide (500mg/L); plates were sealed and incubated at 30 degrees for 6 months. It was found that 33% of samples were positive (5-160CFU/m3); with the highest yield from the Anderson impactor, and from mouldy buildings with longstanding water damage. Of the 43 isolates recovered, 95% were slow growers and 24 (58.1%) were MAC. In agreement with Falkinham’s findings, the mycobacteria isolated from Anderson stages 1-5 indicate the size of the particles was in the respirable range.

Air sampling for NTM has also been performed in peat moss processing plants. Cayer et al (2007) applied molecular techniques to air samples for direct non- culture based detection of NTM. Real-time PCR was performed on air samples obtained using the AGI-30 (All-glass impinger) – which sampled at a flow rate of 12.5L/min for 16 minutes. The authors detailed the process by which DNA was extracted from the air samples; 16S rRNA amplification and cloning was used to identify species of mycobacteria and real-time PCR used to quantify the species. Mycobacteria were also isolated by culture from the peat moss samples and identified using 16S rRNA sequencing, and compared using the GenBank database.

The results were significant in that 40 mycobacterial clones were evident in air samples from 2 plants; these being 19 from plant 1 (M. intracellulare 9/19, M. graecum 1/19, 6 actinomycetes, 3 not identified) and 30 clones from plant 2 (M. intracellulare 12/30, M. interjectum 7/30, M. bohemicum 1/30, M. graecum 1/30, M. smegmatis 1/30, 7 actinomycetes, 1 not identified). From the peat moss cultures, species identified included M. avium, M. palustre, M. terrae, M. malmoense (Plant 1) and M. avium, M. intracellulare, M. interjectum, and M. bohemicum (Plant 2). Unfortunately the authors did not acknowledge or discuss other techniques of airsampling for mycobacteria. Rautiala et al., (2004) had previously compared the yield of mycobacteria from air samples using the AGI, Andersen impactor and filter sampling, and found the greatest yield using the Andersen impactor. However the application of molecular techniques directly to air

24 samples avoids the problem of fungal and bacterial overgrowth that occurs with culture based techniques, and the reduced yield of mycobacteria that occurs as a result of decontamination required to reduce this overgrowth.

Hot tubs and water therapy pools are another potential source of aerosolised NTM. Glazer et al (2007) collected aerosols from these sources onto M7H11 agar plates using the 6th stage only of the Andersen impactor (hence of a respirable size). No agents were added to the agar to reduce bacterial or fungal overgrowth, and whist the authors admit overgrowth was a problem they do not provide these details in the paper. The water samples were sent to a central laboratory and the details of how they were processed are not provided. Spiked control samples were also sent, and wide variability of NTM concentrations and species from the same site, and a 23 % false negative rate was observed. For purposes of comparison with previous studies – standard laboratory culture methods as opposed to PCR or genetic staining techniques were selected. Nevertheless, NTM were cultured from air and water samples, a correlation between the presence of NTM and type of disinfection was found, with significantly higher concentrations were found in pools that used nonhalogen disinfection (ozone or hydrogen peroxide). There was a also an association between the managers’ reports of problems maintaining water clarity and reports of biofilm formation on walls and maximum NTM concentrations in water. Higher turnover rates (the number of times the entire pool volume passes through the filter in a given time period) were associated with lower concentrations of NTM in air and water.

ANALYTICAL METHODS

Whilst a waterborne transmission mode for mycobacteria has been suggested, there are problems that arise when investigating this link. Infections in humans are usually sporadic; there are often a number of exposures that may coexist and the methods used for typing organisms to discriminately determine whether the

25 environmental isolates are the same as patients’ isolates, have been rapidly evolving in recent times. Environmental mycobacteria are generally not included in standard microbiological analysis of water in distribution systems or treatment plants. Most are virtually never included in HPC as plates are usually incubated for 48 hours, 72 hours or 7 days and the composition of routine media such as R2A agar and plate count agar does not support the growth of mycobacteria.

Widely varying growth rates, specific growth requirements and differing natural ecologies are factors which affect the ability to isolate mycobacteria from environmental specimens. General mycobacterial methods must be tailored to detect specific pathogens. “Because of their slow growth, pre-treatment methods for samples to detect mycobacteria are critical to being able to detect the target bacteria. However the pretreatment used may have a) prevented the detection of some particular mycobacterial species and b) also minimized the number of samples found to be positive and the number of colonies seen.” (Pryor et al., 2004). This quote from Pryor sums up the feelings of many authors who have tried to identify mycobacteria in environmental samples.

The analysis of water specimens for the identification of mycobacteria has been outlined by Stinear et al (Stinear, Ford & Vincent, 2004) Figure 1. There are five distinct phases: 1) sample collection; 2) sample storage; 3) sample preparation; 4) detection; 5) result interpretation.

26 1. Sample collection • Sample type: water, air, soil, biological • Collection method: grab sample, swab • Statistical considerations: frequency, volume

2. Sample Storage • Temperature, Time • Other conditions

3. Sample preparation • Concentration: centrifugation, iltration • Puriication and decontamination: chemical IMS

4. Detection and analysis • First level detection: culture, DNA (PCR, ISH) • Second level detection, microscopy, DNA (PCR) • Third level detection: Genotyping

5. Result interpretation

Figure 2.1. Flow chart of the analysis process (From Stinear et al., (2004))

Isolation of MAC from water requires consideration of the methods for concentration and disinfection. Water samples have been analysed after concentration of the mycobacteria by filtration or centrifugation. Decontamination is often required to reduce the background flora of other bacteria and fungi and relies upon the relative resistance of mycobacteria to disinfectants and detergents. It has been suggested that if filtration is used for concentration, the sample should be disinfected before filtration (Falkinham, 2004). However, several authors have decontaminated after concentration. Different combinations of concentration and disinfection for isolation of mycobacteria have been reviewed (Brooks, Parker, Falkinham, & Gruff, 1984; Iivanainen, Martikainen, & Katila, 1997; Kamala, Paramasivan, Herbert, Venkatesan & Prabhakar, 1994; Neumann, Schulze-

27 Robbecke, Hagenau & Behringer, 1997; Schulze-Robbecke, Weber, & Fischeder, 1991) and several protocols have been described but there is no standard established method. These have been summarized in Table 2.1.

Decontaminating agents include 1% NaOH, 1% Oxalic acid, 1% HCl and varying concentrations of CPC (cetylpyridinium chloride). However, in addition to killing contaminating organisms, these agents also kill mycobacteria to a lesser extent. Therefore the number of organisms reported in studies of environmental sampling for mycobacteria using culture-based methods, represent only a minority of the total population of waterborne mycobacteria. In 1984 Brooks et al., reported 5% survival after decontamination. Le Dantec et al., (2002b) performed experiments to determine the detection limit of their decontamination method, by using cultures of M. gordonae. One liter of tap water, sterilized by filtration, was inoculated with a culture of M. gordonae to obtain final concentrations of 10, 102, 103, and 104 CFU · liter-1. After filtration the membrane was decontaminated by adding sodium dodecyl sulfate and NaOH and was shaken at room temperature. The pH was adjusted to 7. The membrane was removed, and the remaining neutralized solution was centrifuged after which most of the supernatant was removed. The exact number of cells in each water culture was checked by plating on Middlebrook 7H11 plates. Colony counting showed that the decontamination process reduced the number of mycobacteria to 1% of the original number.

Du Moulin and Stottmeier (1978) first described the use of cetylpyridinium chloride. They applied 0.04% to 1L samples of seeded distilled water and compared to a control group of samples processed without CPC treatment. Another 54 x 1L water samples obtained from hospital water supplies were similarly processed. Survival of mycobacteria in spiked specimens varied from 1 to 100%, depending on the species - M. kansasii 18.4%, M. gordonae 8.4%, M. intracellulare 100%, M. fortuitum 1.1% and M. bovis 39.9%. Twenty-nine (52%) of the hospital water samples were positive for mycobacteria (25 M. chelonae and 9 M. intracellulare), 20 were negative and 5 became contaminated with molds. M. chelonae appeared within 7 days, M. intracellulare required three weeks. The authors stressed that these were

28 laboratory experiments carried out on strains of mycobacteria from stock cultures in nutrient free distilled water and that these organisms may not accurately reflect the survival characteristics of environmental organisms. It was also suggested that the exposure time could be reduced to 5 hours and concentration of CPC reduced to 0.004% to increase yield of mycobacteria without compromising contamination levels.

CPC, sodium hydroxide (NaOH) and formaldehyde (HCHO) were compared for their efficacy as decontamination substances for the isolation of mycobacteria from drinking water samples (Schulz-Robbecke et al, 1991). CPC 0.005% ensured the highest recovery of mycobacteria and the lowest contamination rates, using both spiked samples and environmental samples. This finding was confirmed by Neumann et al (1997) who compared 12 methods defined by combination of decontamination method (CPC 0.05% or 0.005% vs TSB (tryptic soy broth)), growth medium (L-J vs Ogawa egg yolk (OEY) vs Ogawa egg yolk supplemented with ofloxacin and ethambutol (OEOE)) and incubation temperature (30°C vs 37°C) applied to surface and treated water samples cultured for 12 weeks. CPC 0.005% was too strong for the rapidly growing mycobacteria M. chelonae and M. fortuitum, and NaOH may be better (Schulz-Robbecke et al., 1991). However in studies that have used NaOH, the yield for MAC has been very low (Stinear et al., 2004).

In a study of the Lisbon water distribution system (Santos et al., 2005) five litre water samples (1L bottles each containing 20mg Sodium thiosulphate) from cold water taps inside buildings were filtered (0.22µm), then agitated, decontaminated by Petroff’s method (NaOH added 1:1, incubated for 15 minutes), then neutralized with HCl. Samples were inoculated into BACTEC TM (Becton Dickinson) liquid media; and PANTA (polymyxin, azlocillin, nalidixic acid, trimethoprim, amphotericin B) and incubated for 42 days; L-J tubes were inoculated and incubated at 22°C, 37°C and 44°C for 60 days. Mycobacteria were identified in 90.5% of water samples -15.78% considered potentially pathogenic (5% M. intracellulare- 2 samples, 63% M. gordonae); 20.05% of species were not identified at time of publication. No samples

29 were inoculated into BACTECTM + PANTA that were not first decontaminated and there is no information provided about contamination rates.

Falkinham (2001; 2004) has suggested that for drinking water samples, decontamination may not be required. In his study published in 2001, he processed samples initially without decontamination but if plates were overgrown, reprocessed them using CPC. Three-litre samples were collected with 0.1ml 1% Na thiosulphate. 300ml samples were then centrifuged, the pelleted cells suspended in 1 ml SDW, and 0.1ml inoculated on M7H10 agar, each sample onto 5 plates, incubated at 37 °C, and examined at 21 days. If high numbers of contaminants were present the process was repeated after decontamination with CPC 0.005%. Unfortunately it is not stated in the paper how often decontamination was necessary. 15% of samples grew slow-growing mycobacteria (3% M. avium, 1% M. intracellulare) and there were 2% rapid-growers. In the same study biofilm scrapings from pipes were examined and CPC decontamination was “frequently required”.

Because of the limitations of culture based techniques, the use of more direct detection methods is appealing. Chang et al (2002) investigated whether NTM could be identified by direct PCR on hospital tap water samples concentrated by centrifugation, and compared this to culture with L-J medium and indirect PCR on positive samples. None of the 49 tap water samples was positive for NTM by direct PCR_RFLP. However after culture on LJ medium, colonies grew from 13 samples – 10 of which were identified as NTM using indirect PCR. Quantitative PCR was used to study the prevalence of NTM in cooling towers in Barcelona (Adrados et al., 2011) using a Q-PCR assay with genus specific primers with a detection limit of 500cells/L. They were able to document NTM in 56% of samples and then confirmed different pathogenic species by sequencing of the intergenic spacer region. Cultures were not performed in parallel so relative yield could not be ascertained. Nor could viability of the mycobacteria, as potential infectious agents, though they were the first to show that the technique could be applied to water

30 samples. Given the limitation of culture-based techniques this method of detection and quantification appears to be promising.

Table 2.2. Methods for isolation of mycobacteria. Adapted from Stinear et al (Analytical Methods (Stinear T, 2004)) Sample Type Method Summary Reference

Water (raw and Sampling: 300ml water samples, 4cm2 biofilm samples Falkinham et al., treated): surface and Concentration: centrifugation. Decontamination: CPC, 0.005%, 2001 groundwater sources. 30min. Isolation:7H10 agar (OADC),37ºC. Detection: Biofilms: pipes, water representational colony selection, ZN staining, sub-culture to meters purity,16sRNA PCR and sequencing Water (raw and Sampling: 1000ml water samples. Concentration: filtration. Le Dantec et al., treated): surface and Decontamination: SDS/NaOH. Isolation: Löwenstein- 2002b groundwater sources Jensen,37º. Detection: ZN staining, subculture to purity, 16sRNA PCR and sequencing, hsp65 PRA and sequencing Water: hospital and Sampling: 500ml water samples. Concentration: 0.45µm Peters et al., consumer taps filtration. Decontamination: CPC, 0.005% 15 min. Isolation: 1995 BACTEC and LJ, 37ºC. Detection: AFB staining, subculture, biochemical ID. Water: various sources Sampling: 25ml water samples Concentration: centrifugation. Von Reyn et al., Decontamination: none used. Isolation: Tsukamura minimal- 1993 Tween80-cycloheximide agar, 37ºC.Detection:AFB staining, subculture to purity, DNA probes for MAC Water (raw and Sampling: 500ml samples. Concentration: 0.45µm filtration. Covert et al., treated): cold taps, hot Decontamination: CPC. Isolation: 7H10 agar (cycloheximide), 1999 taps, ice, showers, 37ºC, 5% CO2. Detection: colony morphology, subculture, bottled water 16sRNA PCR and sequencing Swimming pools: water, Sampling: 500ml samples, swabs (20 cm2) Concentration: Leoni, et al., biofilm 0.45µm filtration. Decontamination: CPC 0.04%, 30 1999

min.Isolation:7H10 agar, 30 ºC, 5%CO2.Detection: colony morphology, subculture, biochemical ID Water (raw and treated) Concentration: 0.45µm filtration. Decontamination: CPC Neumann et al., 0.05%-CPC 0.005% 30 min. Isolation: LJ, 30 ºC. Detection: 1997 subculture based on colony type on 7H10 agar, TLC and hsp65 PRA. Hospital tap water Sampling: 200ml samples. Concentration: centrifugation. Chang et al., Decontamination: not specified. Isolation: LJ, 37 ºC. Detection: 2002 AFB, subculture, hsp65 PRA. Surface water, Concentration: centrifugation, filtration. Decontamination: Stinear et al., vegetation, soil IMS, CPC 0.005% 30 min, Isolation: Brown & Buckle agar, 30 2000 ºC, microaerophilic. Detection: subculture, AFB, 16sRNA PCR and sequencing

31 Water –production Samples: 1L water, biofilm scrapings vortexed in source water. Pryor et al., 2004 wells, distribution Concentration: Filtration (0.45µm) Decontamination CPC 0.8%. system sites. Biofilms- Isolation 7H10 agar at 22,30,37, and PVC pipe sections, 42°C.Detection:subculture, PCR using 439bp fragment of the water meters,end cap hsp 65 gene, RA using BST E II and Hae III scrapings. Dam water Samples: 1L. Concentration: centrifugation. Decontamination: Whittington et ? 0.9% HPC; Isolation: Bactec 460 radiometric culture al.,1998; Detection: IS 900 PCR Whittington, Marsh, & Reddacliff,2005; Whittington, Marshall, Nicholls, Marsh, &Reddacliff,2004 Drinking water, tap Samples: 1L-10L. Concentration: filtration, filters placed in Kubalek & swabs, tap scrapings 20ml PBS and centrifuged. Decontamination: Lauryl sulfate+1% Mysak, 1996 NaOH Hot water through POU Samples: 100ml Concentration: filtration. Decontamination: Sheffer et al., filters none. Isolation: 7H10 at 35-37°(5% CO2) for 6 weeks 2005 Detection: Gen-Probe accuprobe Tap water ± POU Samples: 1L + NaThio. Concentration: Filtration. Chaidez & connections and Decontamination: 0.04% CPC. Isolation: 7H11 at 37° (5% CO2) Gerba, 2004 activated carbon for 4 weeks. Detection: Kinyoun staining treatment devices Aerosols Samples: 6 stage Anderson impactor, an all-glass impinger, and Rautiala, 2004 by filter sampling. Impactor samples- collected onto Petri dishes containing M7H11 +OADC enrichment (100ml/L), malachite green(25mg/ml) and cycloheximide (500mg/L) (Decont); plates sealed with parafilm, packed in a plastic bag and incubated at 30 degrees for 6 months. Filter samples: Conc by centrifugation. Decontamination: 1% NaOH. Isolation: various egg media + cycloheximide Detection: Biochemical methods, MAC DNA accuprobe, 16srDNA Cold Tap Water Samples: 5L (1L bottles +NaThio). Concentration: Filtration. Santos et al., Decontamination: NaOH Isolation: BACTEC +PANTA,LJ at 22,37 2005 & 44°C 60 days. Detection: Genotyping Wet soils, natural Samples 50ml. Concentration: Centrifugation. Parashar et al., waters (Johar et al, Decontamination: 3% Na dodecyl sulfate + 4% NaOH, 2% 2004 2007) cetrimide treatment. Isolation: LJ at 30 and 37°. Detection: to species level, method not stated

Identification of mycobacteria.

The number of recognised NTM species continues to increase as genotypic methods provide fresh understanding of previously indistinguishable environmental species. The International Journal of Systematic and Evolutionary Biology currently lists 128 Mycobacterium species and 11 subspecies. Traditional identification techniques have not clearly resolved slow-growing species and many biochemical tests are non- specific or equivocal. High Performance Liquid Chromatography (HPLC) provides rapid identification, although several of the rapid growing species are not resolvable.

Sequencing based methods are becoming more widespread – the most common target genes include 16S rRNA (Cloud et al., 2002), heat shock protein (McNabb, Adie, Rodrigues, Black, & Isaac-Renton, 2006), and rpoB (Adekambi, Berger, Raoult, & Drancourt, 2006; Adekambi, Colson, & Drancourt, 2003). Other techniques include restriction fragment length polymorphisms of the heat shock protein (Hafner, Haag, Geiss, & Nolte, 2004) or 16S-23S rDNA intergenic spacer region (Gurtler, Harford, Bywater, & Mayall, 2006), pyrosequencing (Tuohy, Hall, Sholtis, & Procop, 2005) and microchip array (Fukoshima et al., 2003; Tobler, Pfunder, Herzog, Frey, & Altwegg, 2006), which have been in development. These are still time consuming and in some cases cumbersome. Recent advances in molecular biology techniques have resulted in a number of publications, and were reviewed by Neonakis et al (Neonakis, Gitti, Krambovitis, & Spandidos, 2008).

DNA pyrosequencing was further assessed by Heller (Heller, Jones, & Widen, 2008). Sequencing is performed by DNA synthesis. This method appropriately identified >90% of clinical isolates. In particular, sequencing of isolates identified as M. avium- M. intracellulare complex and M. interjectum resulted in several possible identifications. Because shorter sequences are generated, the method is not as discriminating as more traditional sequencing. The principal limitation however, is the quality of the available databases.

Real-time PCR has been used for the detection and identification of mycobacterium species however usually a single species only. Lim et al (Lim, Kim, Lee, & Kim, 2008) have developed a real-time PCR based method for rapid identification and differentiation of many clinically important mycobacterial species. This technique is based on melting curve analysis using hybridization probes. Real-time PCR during the amplification cycles, detected all the 18 NTM species at the genus level, and all of the targeted species at the species or complex level, but two non-target species cross-reacted. M. smegmatis cross-reacted with M.intracellulare and M. haemophilum cross-reacted with M. avium. Those species identical using 16S rRNA sequencing (M. kansasii type 1 and M. gastri, M. fortuitum type 1 and M. peregrinum and M. ulcerans and M. marinum) were differentiated by using different mastermixes. The real-time PCR melting curve analysis using five mastermixes detected all of the NTM species tested at the genus level and all of the targeted Mycobacterium species at the species or complex level without cross reaction with the other species tested.

The technique was limited by lack of validation using clinical isolates, which may result in a variation of the target sequences, potentially leading to detection failure or misidentification. Different strains of species were not tested, and for the identification of some species additional probes were not used. Further work is needed to improve on this method, however it is certainly promising.

Daley et al., (2008) compared in-house and commercial 16S rRNA sequencing with HPLC and genotype AS (additional species) and CM (common) (Line probe – Hain Life science, Germany). The objectives of the study were to compare three sequence databases (GenBank, RIDOM and MicroSeq) for resolution of common and rare species of mycobacteria, to compare resolution of different lengths of the 16SrRNA fragment, to compare costs between commercial and in-house systems, and to compare sequencing with line probe assay, which uses a second locus for identification, the 23S RNA gene. They collected 55 clinical mycobacterial strains, ten ATCC strains, four unique mycobacterial strains not yet assigned a species

34 name, and three related nonmycobacterial genera. Consensus of identification methods was considered the only possible reference standard. Overall 72/73 strains were definitively identified by consensus identification.

Table 2.1 Comparison of HPLC, in-house and commercial sequencing and a Line Probe assay for identification of mycobacteria.

Method Correct Provided correct Incorrect identification (cf but less specific identification (cf consensus) information consensus) HPLC 21/50 (42%) 12/50 (24%) 17/50 (34%) In House 36/51 (70.6%) 10/51 (19.6%)* 5/51 (9.8%) sequencing Commercial 37/46 (80.4%) 7/46 (15.2%) ** 2/46 (4.3%) Sequencing Line probe assay 32/44 (72.7%) 12/44 (27.3%)#

* 7 of which due to known identical 16S sequences ** 6 of which due to known identical 16S sequences # all were species not contained in the line probe assay

Importantly genotype CM was able to distinguish species pairs with identical 16S rRNA sequences (M. abscessus from M. chelonae, M. gastri from M. kansasii, and M. marinum from M. ulcerans). The authors provide a comprehensive discussion of the advantages and limitations of the different sequence databases.

The diversity of environmental rapidly growing mycobacteria in haemodialysis water was assessed using a multigene sequencing approach (Gomila, Ramirez, & Lalucat, 2007). Twenty isolates taken from a previous study of Restriction Fragment Length Polymorphism (RFLP) of the 16S rRNA gene fragments after digestion with Taq1 and HaeIII, were selected for further analysis, together with seven mycobacteria from clinical specimens, classified phenotypically as atypical mycobacteria. The fragments of seven genes were sequenced - the 16S rRNA gene, ITS1, gyrB, hsp65, recA, rpoB, and sodA, and analysed them for the 27 strains and 11

35 related type strains. The concatenation or consensus analysis of several genes allowed increased discriminatory power and provided a more robust phylogenetic tree. This technique may be useful in environmental studies where 16S rRNA sequencing is insufficient.

Molecular identification using partial sequencing of rpoB, hsp65 and secA gene fragments and strain typing using repPCR (Diversilab) has been used in order to differentiate M. abscessus from subspecies formerly named M. massiliense and M. bolletii (Zelazny et al., 2009). The majority of M. abscessus complex clinical isolates clustered with the ATCC strain 19977, with a degree of diversity as low as 80% similarity to the reference strain.

The appropriate identification of species of NTM is a rapidly changing area. It is important to consider this evolution in the comparison of studies done at different times, particularly when trying to ascertain the environmental source of a particular pathogenic species. For example, many studies have identified M. chelonae in water, but now that M. abscessus is a separate species from M. chelonae, there is no way of knowing whether M. abscessus itself has previously been detected in water but not identified as such.

36 Strain Typing

A robust typing method with strain level resolution is necessary when comparing patient and environmental isolates in order to identify the source of infection and associate a human health risk to that source, especially when the rate of genomic evolution of that organism is not known. Pulsed Field Gel Electrophoresis (PFGE) is considered the gold standard. It been used to retrospectively analyse nosocomial M. abscessus outbreaks (Wallace, et al., 1993; Zhang, Rajagopalan, Brown, & Wallace, 1997; Zhang et al., 2004). More recently the molecular epidemiology of M.abscessus was assessed by Jonsson et al., (2007) with a focus on three groups: patients with cystic fibrosis in Sweden and respiratory infection/transient colonization; a group of isolates from patients with soft tissue infections acquired during the tsunami in Thailand in 2004; and a group of other isolates thought to be environmental contaminants. After culture on M7H10 or horse blood agar, smooth and rough colony morphology was observed. Among the rough isolates, 81% (13/16) were isolated from patients with chronic colonization/disease in the airways, while only 10% (1/10) of the smooth ones were from this patient group (p=0.0014, Fisher’s exact test). Isolates were compared using PFGE, after digestion by AseI to determine strain identity. All isolates from a single patient yielded the same PFGE pattern. Interestingly, smooth and rough isolates from the same patient had identical PFGE patterns suggesting perhaps a limitation in discriminatory power for PFGE. There was one cluster identified in 5 patients (4CF, one non-CF). A single strain colonising a 6th patient differed by a single band from this cluster. Two of the four CF patients with this cluster were siblings, however there was no connection found between the remaining patients. The older of the siblings had repeated isolation of M. abscessus and disease due to the organism, but the younger sibling had only a single positive isolate consistent with transient colonisation. The results of this study suggest that patients acquire M. abscessus from the environment, and that spread between patients is uncommon.

37 PFGE takes approximately eight days to complete and is labour intensive. Hence a number of alternative methods have been studied. Amplified fragment length polymorphism (AFLP) analysis is a random amplified polymorphic DNA technique in which genomic DNA is digested with restriction enzymes and fragments are randomly amplified to generate a pattern of bands. It is reported to be highly sensitive for discriminating between closely related strains of bacteria and takes two days to complete once DNA is obtained from pure culture. Automated electrophoresis allows for high throughput.

AFLP was applied to 159 patient and environmental MAC isolates from Southern California (Pfaller, Aronson, Holtzman, & Covert, 2007) and accurately identified strains belonging to M. avium and M. intracellulare, and differentiated between strains within each species. Strains from patient and hospital water sources that were previously thought to be identical based on RFLP-PFGE analysis, were shown to be different using AFLP. None of the patient isolates shared identical AFLP patterns with the environmental isolates. The drinking water isolates were closely related, patient isolates were closely related, but there was little overlap between the two sources. The authors report that the threshold for assigning peaks can be adjusted to maximise the number of bands and improve reproducibility, and other restriction enzymes and primer sets can be included to reveal additional polymorphisms, thereby increasing the discriminatory power of the method. Modifications of the AFLP technique have been described to discriminate other species of mycobacteria including M. kansasii. Gafar et al., (2003) were able to differentiate strains belonging to one of five subspecies of M. kansasii with a simplified AFLP technique using a single enzyme, single adaptor and single primer. This approach revealed a high level of heterogeneity among a subspecies that was believed to be clonal.

Alvarez et al., (2008) also looked at the genetic diversity of M. avium isolates from clinical samples and from the environment. They assessed 47 clinical samples from patients in different sections of a hospital in Spain, and compared them to 17 environmental isolates collected within the hospital, using commercial probes,

38 phenotypic characteristics, PCR for detection of IS1245 and IS901, sequencing of the hsp65 gene, and PFGE. The same strain (Pattern A – lacking IS1245, code 2 hsp65 sequevar and a certain PFGE profile) was found in all of the environmental isolates, and in the isolates cultured from urine samples that were considered contaminants (due to the nature of the collection of these specimens). This strain was also found in 5/19 (26.3%) of the respiratory specimens but was believed to cause disease in these patients. This strain was best characterized as M. avium subs hominissuis, with an unusual finding that 81.4% lacked the IS1245 element. RFLP using IS1245 as a probe has been widely employed, however it is limited in the case of M. avium strains that harbour a low number of IS1245 copies. PFGE is the only applicable typing technique for these isolates. The hsp65 sequencing was a useful tool for confirmation of species identification of IS1245-negative M. avium isolates. The molecular characterization correlated with clinical significance both in those considered not to have disease (described above, Pattern A) but also in the two patients with M. avium that harboured IS1245 and had code 1 of the hsp65 sequevar.

ERIC (enterobacterial repetitive intergenic consensus) PCR has been used successfully to differentiate strains of mycobacteria associated with outbreaks of disease in mesotherapy clinics due to M. abscessus and M. chelonae, and in post mammoplasty patients with M. fortuitum infections. (Sampaio et al., 2006; Sampaio, Viana-Niero, de Freitas, Hfling-Lima, & Leao, 2006). A high concordance with PFGE was demonstrated. More recently rep-PCR (Diversilab) was used in comparison to PFGE in a suspected outbreak of M. abscessus in a lung transplantation and cystic fibrosis centre (Aitken et al., 2012). Isolates involved were strain typed using both PFGE and rep-PCR (Diversilab). The isolate of the index case and outbreak cases were reported to be “genetically identical” using PFGE. The rep-PCR patterns of the same isolates showed 90% genetic similarity, again suggesting inferior discriminatory power of PFGE.

Rep-PCR was shown to concur with strain typing using VNTR (variable number tandem repeats) (Harris et al., 2012). Forty-one M. abscessus complex strains from

39 17 pediatric cystic fibrosis patients were typed using VNTR and compared with rep- PCR (Diversilab). As with the Aitken (2012) paper, isolates with 90% similarity were regarded as indistinguishable. The discovery of six distinct profiles among 12 patients with M. abscessus complex suggests that the typing methods have sufficient discriminatory power. Automated rep-PCR has significant advantages in cost and time saving over other methods and has demonstrated high discriminatory power.

WATER AS A SOURCE OF PATHOGENIC MYCOBACTERIA CAUSING HUMAN DISEASE

It has long been assumed that exposure to environmental mycobacteria is universal, disease being uncommon and requiring a host defect for it to become established. Yet there are well-documented geographic variations in disease incidence, suggesting variations in environmental loads of these organisms. The recent increase in reporting of cases of hypersensitivity pneumonitis associated with hot tub use, where large numbers of organisms are aerosolized in a close space, raises the possibility that the intensity of exposure may also be relevant in pulmonary infection and disease. An informal assessment of patients with mycobacterial infection (as opposed to hypersensitivity pneumonitis) in Brisbane has identified a number of patients with hot tubs and one who regularly attended water aerobics in an indoor heated pool (unpublished data). In one of the hot tub cases, the water and filter were cultured and the same organism was isolated from these specimens as from the patient. One could postulate the possibility of a “dose response” whereby pulmonary disease (often slowly progressive over many years) is associated with lower intensity exposure than the hot tub-hypersensitivity scenario, yet the exposure is greater than that encountered by individuals without infection. Women particularly are over represented in those with nodular bronchiectasis, a type of disease often not associated with pre-existing structural lung disease. Women also are reported to shower for longer periods of time than men, and it is

40 traditionally women who spend more time washing up at a kitchen sink. (unpublished data)

Humans are exposed to mycobacteria in water through drinking, swimming and bathing. Aerosols generated during these activities can be inhaled, and water can be aspirated when swallowed (Thomson, Armstrong, & Looke, 2007). In the era of HIV and AIDS (post 1980), the incidence of disseminated disease due to mycobacteria increased and resulted in significant morbidity and mortality for these patients. Drinking water has been shown to be at least one source of human infection in AIDS patients – where contaminated water is ingested and infection acquired through the intestinal wall. This finding prompted researchers to further investigate the environmental sources of these organisms. Hot water systems in hospitals were shown to be one such source (du Moulin et al.,1988) as did home hot water systems - shown to be colonized with M. avium (Montecalvo, Forester, Tsang, du Moulin, & Wormser, 1994; von Reyn et al., 1994). DNA fingerprints have shown that water isolates of M. avium were identical to those recovered from AIDS patients who either drank or were exposed to the tested waters (von Reyn et al., 1994).

Often patients without NTM disease have sputum specimens that are positive for NTM organisms. It has been postulated that they are transiently contaminated or colonized by these organisms and that ingested water or inhaled aerosols are the source of the organisms. Tobin D’angelo et al., (2004) compared respiratory isolates of MAC from 131 patients (27% with MAC disease; 72% with HIV) and 13 MAC isolates for hospitals hot water systems. They found ten different clusters using PFGE. Of 110 MAC isolates that were recovered from patients without MAC disease – 65% were part of a single large cluster that contained isolates the same as those recovered from the hospital hot water system, 25% of the 51 isolates from patients with MAC disease were also in this cluster.

Outbreaks of pulmonary disease due to mycobacteria have been linked to the presence of M. xenopi and M. kansasii, isolated from drinking water samples

41 (McSwiggan & Collins, 1974; Slosarek, Kubin, & Jaresova, 1993; Wright, Collins & Yates, 1985). Mycobacterial cervical lymphadenitis in children was previously predominantly due to M. scrofulaceum, however changes in chlorination and fluoridation are thought to have caused the shift to predominantly M. avium disease, as M. scrofulaceum is more chlorine sensitive than M. avium (Wolinsky, 1995). This finding has also been observed in Queensland (Dr W. Oliver, Dr A. Konstantinos –personal communication). Whilst disinfection reduced a number of human diseases, it may also be responsible for the increased incidence of mycobacterial disease, as most mycobacteria are inherently resistant to disinfectants. It is postulated that disinfection may also contribute in part to the persistence of M. avium and M. intracellulare in drinking water distribution systems (Covert et al., 1999; Falkinham III et al., 2001; von Reyn et al., 1994). As discussed previously, these organisms are many times more resistant to chlorine, chloramine, chlorine dioxide and ozone than other water borne micro-organisms such as E. coli. As a result, disinfection results in positive selection of mycobacteria, as competition for growth is lacking.

There have been several reports of hypersensitivity pneumonitis as a consequence of exposure to aerosols that contain mycobacteria. These include occupational settings (metalworkers), lifeguards at indoor pools, and the in-home use of hot tubs. Hypersensitivity pneumonitis was reported in lifeguards and pool attendants, who worked in an indoor swimming pool that featured waterfalls and sprays (Rose et al., 1998). The authors suggested a dose response effect, as the affected lifeguards worked longer hours than those unaffected. The number of respirable particles was increased fivefold as a result of aerosolisation by the waterfalls and sprays. However based on the presence of endotoxin in the aerosol samples the authors suggested that this was the cause of the pneumonitis in the lifeguards. Other authors have documented mycobacteria in swimming pools and whirlpools and the environment surrounding swimming pools (Embil et al., 1997; Havelaar, Berwald, Groothuis and Baas, 1985; Iivanainen et al., 1999; Kahana & Kay, 1997; Leoni, Legnani, Mucci, & Pirani, 1999; Mangione et al., 2001).

42 Hypersensitivity pneumonitis has also been reported in workers exposed to metalworking fluid. These fluids are used to lubricate industrial metal grinding operations, cool the instruments and working surfaces. They are oil-water emulsions that contain paraffins, pine oils, polycyclic aromatic hydrocarbons and heavy metals. An outbreak of disease was reported following disinfection of metalworking fluid (Shelton, Flanders, & Morris, 1999; Bernstein, Lummus, Santilli, Siskorsky & Bernstein, 1995). Mycobacteria were isolated more frequently from metalworking fluid samples from facilities where hypersensitivity pneumonitis (HP) was found, compared to facilities where HP was not found (Shelton et al., 1999). Although the metalworking fluid was contaminated by micro-organisms, it was only after disinfection that symptoms developed in the workers. It is likely that the isolation of mycobacteria from the metalworking fluid was caused by their selection following disinfection.

There have been numerous reports of HP in hot tub users as a result of inhalation of aerosolised mycobacteria (Embil et al., 1997; Kahana, Kay, Yakrus & Wasserman, 1997b; Lumb et al., 2004; Mangione et al., 2001; Rose et al., 1998). Embil et al., (1997) reported five cases of respiratory illness in healthy people who had used hot tubs contaminated with M. avium complex. The illness consisted of bronchitis, fever and “flu-like” symptoms, which came on and were exacerbated within hours of using the hot tub. Radiology and pathology (where available) was consistent with HP, and all recovered without treatment for the MAC. M. avium was isolated from the hot tub water samples. The strength of the relationship between mycobacteria and disease in these cases relies on the isolation of the same clonal strain of mycobacteria in patient specimens as in water sources to which they are exposed. In such cases the load of mycobacteria was high, and exposure was close and intense.

Marras et al., (2005) has since reported a case of HP in a patient and postulated that it was caused by inhalation of mycobacteria from showering. The patient had histological evidence of HP and sputum cultures were positive for MAC. Multiple

43 respiratory samples, and shower and bathtub specimens grew MAC that reportedly had the same PFGE patterns. No aerosol sampling was performed.

As a result of these reports some health professionals are advising patients that they should bathe rather than shower, if they have pulmonary disease due to MAC – including the nodular bronchiectatic type or the traditional cavitary disease (NTMinfo.com). The possibility of lower intensity exposure over a longer period of time, causing pulmonary disease, has not been explored. There have been no studies examining the role of aerosols in these more common forms of MAC pulmonary disease and it remains to be shown whether these patients inhale or aspirate the organisms that cause their disease. In a cohort study of patients with pulmonary mycobacterial disease in QLD, 36.5% (19/52) of MAC patients had gastro-oesophageal reflux disase (GORD) documented by their treating doctor. Of these, 55.8% (24/43) took some form of acid suppression regularly and six patients were thought to aspirate for other reasons (Thomson et al., 2007). Hence the possibility of GORD and micro- or macro-aspiration also needs further exploration.

The evidence linking water exposure to disease is probably stronger for infections due to the rapidly growing mycobacteria. M. fortuitum is a non-pigmented rapidly growing mycobacterium, often associated with skin and soft tissue infections, though pulmonary disease can occur (De Groote & Huitt, 2006). Wound infections (Kuritsky et al., 1983), post-injection abscesses, and post pedicure infections (Vugia et al, 2005) are well described. M. fortuitum is also a recognised cause of nosocomial infections (Burns et al., 1991) and pseudo-outbreaks (Wallace, Brown, & Griffith, 1998). It is usually considered that municipal and hospital water are the sources of these outbreaks, though M. fortuitum is also found in soil and house dust (Dawson, 1971). M. fortuitum has been found in municipal water (Carson et al., 1988; Castillo-Rodal et al., 2012; Le Dantec et al., 2002b) and is relatively resistant to chlorine and other disinfectants. (Cortesia et al., 2010; Le Dantec et al., 2002a)

Whilst there have been many outbreaks reported where water has been the suspected vehicle of transmission, very few have been realised in a timely fashion

44 that would allow full environmental sampling to adequately document the source of the infection (Kuritsky et al., 1983; Quiñones et al., 2010; Sampaio et al., 2006a; Wallace et al., 1989; Wallace et al., 1998). Quiñones et al., (2010) reported one such outbreak, in which a single physician, performing mesotherapy injections, was using sterilized syringes and needles. Material being injected could not be tested because it came from single use vials, and had been discarded by the time infections were manifested. The investigators were not able to identify any environmental source within the salon and postulated that the infections occurred during the injection process- possibly from the physician who may have been carrying mycobacteria on the skin.

One of the first reports of nosocomial transmission of M. fortuitum was published by Burns et al (1991) who investigated a cluster of cases of positive sputum cultures in patients in an alcoholic rehabilitation ward. The common factor between cases was usage of two ward showers. M. fortuitum was isolated from the tap water connecting to the showers but not from the showers themselves. Using PFGE the 16 case isolates were found to be identical to the water isolate. However, there were no other parts of the hospital where such cases were identified and no further cases were reported after the showers had been disconnected and decontaminated. This would suggest that the shower water was indeed the source of the M. fortuitum.

Wallace investigated the sources of M. fortuitum associated with infections following cardiac bypass surgery (Wallace et al., 1989) across 12 states in the USA. In the analysis of a particular outbreak in Texas in 1981, one isolate of M. fortuitum was recovered from an operating room waterbath and had a similar phenotype and enzyme electrophoretic pattern as a cardiac associated isolate. A similar isolate was also recovered from the municipal water coming into the hospital and from an ice machine on one of the hospital floors. Two patients with other surgical wound infections (neck and abdominal) were infected with the same strain. In the five years following that outbreak an additional 21 clinical isolates of M. fortuitum were recovered from patients in the same hospital. Six were studied. All differed in phenotypic markers or plasmid profile from the outbreak isolates.

In a separate outbreak in Colorado in 1976, an isolate of M. fortuitum recovered from a settle plate in an operating room had the same phenotype and electrophoretic pattern (PFGE) as four of the epidemic strains.

Winthrop et al (2002) reported 110 cases of furunculosis due to M. fortuitum acquired following pedicures at a single nail salon in California. All patients had their feet and lower legs soaked in a whirlpool footbath. Large amounts of hair and skin debris were found behind the inlet suction screens of the whirlpool footbath and M. fortuitum was cultured from these areas, from all ten footbaths. The authors concluded that M. fortuitum had entered via the salon tap water, seeded the accumulated organic debris behind the footbath inlet screens, then multiplied and circulated within the footbath basin. They felt that because all of the footbaths cultured rapid-growers including multiple strains of M. fortuitum that it was unlikely a client contaminated the baths. The salon owner reported that these inlet suction areas were never cleaned.

A community outbreak of M. porcinum infections in Texas was reported in 2011 (Brown-Elliott et al., 2011). This rather uncommon, rapidly growing mycobacteria was first described to cause tuberculous lymphadenitis in pigs. In Galveston, Texas, it was isolated from 24 patients and considered to be a pathogen in 46%, though only four were categorized as being community acquired. M. porcinum was also isolated from municipal water, household water, hospital ice and tap water. Patient isolates were highly clonal (22/24 forming a single clone) using PFGE, and 8/10 environmental M. porcinum belonged to two closely related clones. Twenty-six of 29 clinical isolates were clonal with the water isolate patterns, including patients with community-acquired disease. Sequencing of hsp65, secA, and rpoB revealed the PFGE outbreak clones to have identical sequences, while epidemiologically unrelated strains varied.

Hilborn et al., (2008) also compared M. avium isolates from POU water taps and patients within a single county in the USA. Twenty-seven M. avium isolates from

46 four drinking water POU sites were typed using Multilocus enzyme electrophoresis (MEE). PFGE yielded four identical or closely related groups. They analysed 156 clinical isolates (79% respiratory) using MEE – of these 70% were identified as M. avium. 74 of these isolates came from patients within the county served by the municipal water supply, and one isolate was closely related to an environmental M. avium isolate. In this case the zip code of the patient’s residence was adjacent to the drinking water site. Three other clinical isolates were possibly related to another M. avium isolate, but there was no geographic association between these isolates. The study was limited by the lack of patient specific information, the ‘convenience’ sample of clinical isolates and the limited number of environmental isolates tested.

M. kansasii has readily been grown from tap water, though rarely from natural waters. It has been referred to as the ‘tap water bacillus’, however a definitive link between human infection and tap water has not been established. There are reports of high incidence areas of M. kansasii disease associated with the mining industry in South Africa (Engel, et al., 1980; Falkinham, 2010), industrial zones of Czechoslovakia -particularly mining and iron manufacture, (Chobot et al.,1997; Kubin, Švanclová, Medek, Chobot, & Olšovský, 1980) and the Japanese coastal industrial areas (Iinuma et al., 1997). In the Karvina district of Czechoslovakia, M. kansasii was isolated from 7% of drinking water samples, and 43.7% industrial water samples and scrapings between 1971 and 1995, during which the annual incidence of disease was 12.03/100000 with a very high incidence of disease in miners (Chobot, 1997).

Engel applied phage typing of M. kansasii isolates to compare strains from clinical and environmental specimens from The Netherlands, Britain and Czechoslovakia (Engel, Berwald, Grange, & Kubin, 1980). Of the Czech strains, phage type 13 accounted for 25/28 (89.3%) of the clinical isolates from the Karvina coalmine district. This phage type was also found in 28 of 31 (90.4%) environmental isolates (water from the distribution system of the mines), and 12/18 (66.6%) clinical isolates from the rest of the country. This phage type also dominated clinical isolates from the UK, though only 1 of 7 environmental isolates was of this type, and

47 this was isolated from soil in Uganda. Given developments in the field of strain typing, it has been suggested that this technique did not have sufficient discriminatory power to adequately compare the clinical and environmental strains. Phage typing has not been widely applied because it is labor intensive, and technical problems can result in poor reproducibility and interpretation of results (Picardeau, Prod’hom, Raskine, Le Pennec, & Vincent, 1997).

M. kansasii Type I was not present among 113 environmental strains isolated in Germany (Alcaide et al., 1997), though was found in water in France (Picardeau et al., 1997). Picardeau characterized five subspecies of M. kansasii using a variety of strain typing methods, including RFLP with the major polymorphic tandem repeat (MPTR) and IS1652 probe, PFGE, AFLP and PRA of the hsp-65 gene. Both clinical (38) and water (24) isolates from France were included in this study. The clinical isolates came from patients from eight different French towns. The water isolates were recovered from tap water from six different hospitals in and around Paris. Whilst there were water isolates and clinical isolates that had the same strain pattern using different methods of typing, only one set of isolates was epidemiologically related. This set consisted of identical isolates from patients from a single family living in the same house, and a water isolate recovered from their home. Five other sets of isolates were considered epidemiologically related but there was no evidence of nosocomial infection or water transmission.

Other investigators have explored the possibility that patients with NTM disease (predominantly M. avium and M. intracellulare) may have acquired their infection from water sources in the home. A Japanese group (Nishiuchi et al., 2007) looked at the isolation of NTM from the homes of 49 patients with MAC pulmonary disease and 43 healthy volunteers without MAC disease. From each home, three 200ml water samples (kitchen tap, showerhead, and used bathtub water) were collected. In addition, scale on the surface of showerheads, slime from three drains and dust from airconditioners was sampled. Eleven samples contained MAC – ten were from 9/49 patients residences and one from the 43 volunteers residences. Eight isolates were M. avium and three were M. intracellulare – a similar ratio to the species

48 causing pulmonary disease in the Japanese population. The isolates from the patients’ bathrooms were compared with those from their sputum samples and two had identical PFGE patterns and serotype. For four patients, the isolates were the same species but had different RFLP or PFGE profiles. Interestingly, only bathroom samples were positive – MAC was not isolated from kitchen tap water, yet the same system was supplying water to the whole house. The authors postulate that favourable temperature and humidity allow MAC to preferentially colonise the bathroom. However, the frequency of use of kitchen taps may allow more flushing out of mycobacteria from pipes (including sloughing of biofilm) and the possibility that expectorating patients may have contaminated bathroom drains should not be discounted. In the single patient who had M. intracellulare infection, where the pathogen was grown only from a drain site, this may certainly have been the case. Given the small numbers in this study, and the low yield of mycobacteria (possibly technique related), the results cannot sufficiently prove that the bathrooms are an infection source of MAC disease.

Summary

Evidence from multiple studies from around the world, shows that water is an important environmental reservoir of mycobacteria causing human disease. There have been no published studies examining the presence of mycobacteria in water distribution systems in Australia to date, and routine sampling/monitoring is not performed as part of public health practice. There are methodological issues at different levels, which need to be considered in both the interpretation and conduct of water sampling for mycobacteria. The means by which patients may acquire infection from water must also take into account the various means of exposure to NTM in their environment, including municipal water (that may be aspirated or may contaminate wounds), and aerosols that are generated during showering, bathing, garden hosing etc. Confirming a link between water exposure and infection requires the demonstration of the same strain of NTM in both human and environmental samples, in an appropriately related setting. A variety of strain typing techniques

49 exist and the advantages and limitations of these needs to be considered in the interpretation of the evidence.

REFERENCES

Abrahams, E.W. (1970). Original mycobacterial sin. Tubercle, 51(3), 316-321. Abrahams E. W., & Harland R. D. (1967). Sensitivity to avian and human PPD in Brisbane school children. Tubercle, 48, 79-94. Abrahams E. W., & Harland R. D. (1968). Studies with the purified protein derivative of human and avian tuberculins in south Queensland. Tubercle, 49, 192-209. Abrahams E. W., & Silverstone H. (1961). Epidemiological evidence of the presence of non-tuberculous sensitivity to tuberculin in Queensland. Tubercle, 42, 487-499. Adekambi, T., Berger, P., Raoult, D., & Drancourt, M. (2006) rpoB gene sequence- based characterization of emerging non-tuberculous mycobacteria with descriptions of Mycobaterium bolletii sp. nov., Mycobacterium phocaicum sp. nov. and Mycobacterium aubagnense sp. nov. International Journal of Systematic and Evolutionary Microbiology, 56(Pt 1), 133-143. Adekambi, T., Colson, P., & Drancourt, M. (2003) rpoB-based identification of nonpigmented and late-pigmenting rapidly growing mycobacteria. Journal of Clinical Microbiology, 28, 5699-5708. Adekambi, T., Salah, S.B., Khlif, M., Raoult, D., Drancourt, M. (2006). Survival of environmental mycobacteria in Acanthamoeba polyphaga. Applied and Environmental Microbiology, 72(9), 5974-5981. Adrados, B., Julian, E., Codony, F., Torrents., Luquin, M., & Morato, J. (2011). Prevalence and concentration of non-tuberculous mycobacteria in cooling towers by means of quantitative PCR: a prospective study. Current Microbiology, 62(1), 313-319 Aitken, M. L., Limaye, A., Pottinger, P., Whimbey, E., Goss, C. H., Tonelli, M.R., ... Wallace, R. J. (2012) Respiratory Outbreak of Mycobacterium abscessus subspecies massiliense in a lung transplant and cystic fibrosis center. American Journal of Respiratory and Critical Care Medicine, 185(2), 231-232.

50 Alcaide, F., Richter, I., Bernasconi, C., Springer, B., Hagenau, C., & Telenti, A. (1997) Heterogeneity and clonality among isolates of Mycobacterium kansasii:Implications for epidemiological and pathogenicity studies. Journal of Clinical Microbiology, 35(8), 1959-1964. Alvarez, J., Garcia, I. G., Aranaz, A., Bezos, J., Romero, B., de Juan, L., ... Dominguez, L. (2008). Genetic diversity of Mycobacterium avium isolates recovered from clinical samples and from the environment: Molecular characterization for diagnostic purposes. Journal of Clinical Microbiology, 46(4), 1246-1251. Bernstein, D. I, Lummus, Z. L., Santilli, G., Siskorsky, J., Berstein, I. L. (1995) Machine operator's lung, a hypersensitivity pneumonitis disorder associated with exposure to metalworking fluid aerosols. Chest, 108, 636-641. Brooks, R. W., George, K. L., Parker, B. C., Falkinham, J. O. 3rd, & Gruff H. (1984) Recovery and survival of nontuberculous mycobacteria under various growth and decontamination conditions. Canadian Journal of Microbiology., 30(9), 1112-1117. Brooks, R.W., Parker, B.C., Gruft, H., & Falkinham, J.O. 3rd. (1984) Epidemiology of infection by nontuberculous mycobacteria. V. Numbers in eastern United States soils and correlation with soil characteristics. American Review of Respiratory Diseases, 130(4), 630-633. Brown-Elliott, B. A., Wallace, R. J., Tichindelean, C., Sarria, J. C., McNulty, S., Vasireddy, R., ... Loeffelholz, M. (2011). Five-year outbreak of community- and hospital-acquired Mycobacterium porcinum infections related to public water supplies. Journal of Clinical Microbiology, 49(12), 4231-4238. Burns, D. N., Wallace, R. J. Jr, Schultz, M. E., Zhang, Y. S., Zubairi, S. Q., Pang, Y. J., ... Gordin, F. M. (1991). Nosocomial outbreak of respiratory tract colonization with Mycobacterium fortuitum: demonstration of the usefulness of pulsed- field gel electrophoresis in an epidemiologic investigation. American Review of Respiratory Disease, 142(2), 468-470. Carson, L. A., Bland, L. A., Cusick, L. B., Favero, M. S., Bolan, G. A., Reingold, A. L., Good, R. C. (1988) Prevalence of nontuberculous mycobacteria in water supplies of hemodialysis centers. Applied and Environmental Microbiology, 54(12), 3122-3125.

51 Castillo-Rodal, A., Mazari-Hiriart, M., Lloret-Sánchez, L., Sachman-Ruiz, B., Vinuesa, P., López-Vidal, Y. (2012). Potentially pathogenic nontuberculous mycobacteria found in aquatic systems. Analysis from a reclaimed water and water distribution system in Mexico City. European Journal of Clinical Microbiology and Infectious Diseases, 31(5), 683-694. Cayer, M-P., Veillette, M., Pageau, P., Hamelin, R., Bergeron, M-J., Mériaux, A., ...Duchaine, C. (2007). Identification of mycobacteria in peat moss processing plants: application of molecular biology approaches. Canadian Journal of Microbiology, 53(1), 92-99. Chaidez, C., & Gerba, C. P. (2004). Comparison of the microbiologic quality of point- of-use (POU)-treated water and tap water. International Journal of Environmental Health Research, 14(4), 253-260. Chang, C-T., Wang, L-Y., Liao, C-Y., & Huang, S-P. (2002). Identification of nontuberculous mycobacteria existing in tap water by PCR-restriction fragment length polymorphism. Applied and Environmental Microbiology, 68(6), 3159-3161. Chobot, S., Malis, J., Sebakova,H., Pelikan,M., Zatloukal,O., Palicka,P., & Kocurova,D. (1997) Endemic incidence of infections caused by Mycobacterium kansasii in the Karvina District in 1968-1995. Central European Journal of Public Health 5, 5(4), 164-173. Cirillo, J. D., Falkow, S., Tompkins, L. S., & Bermudez, L. E. (1997) Interaction of Mycobacterium avium with environmental amoebae enhances virulence. Infection and Immunity, 65(9), 3759-3767. Cloud, J. L., Neal, H., Rosenberry, R., Turenne, C. Y., Jama, M., Hillyard, D. R., & Carroll, K. C. (2002). Identification of Mycobacterium spp. by using a commercial 16S ribosomal DNA sequencing kit and additional sequencing libraries. Journal of Clinical Microbiology, 40, 400-406. Collins CH, Grange JM, & Yates MD. (1984). Mycobacteria in water. Journal of Applied Bacteriology, 57, 193-211. Cortesia, C., Lopez, G. J., de Waard, J. H., Takiff, H.E. (2010). The use of quaternary ammonium disinfectants selects for persisters at high frequency from some species of non-tuberculous mycobacteria and may be associated with

52 outbreaks of soft tissue infections. Journal of Antimicrobial Agents and Chemotherapy, 65(12), 2574-2581. Covert, T. C., Rodgers, M. R., Reyes, A. L., & Stelma, G. N. (1999). Occurrence of nontuberculous mycobacteria in environmental samples. Applied and Environmental Microbiology, 65(6), 2492-2496. Daley, P., Petrich, A., May, K., Luinstra, K., Rutherford, C., Chedore, P., .... Smieja, M. (2008). Comparison of in-house and commercial 16SrRNA sequencing with high-performance liquid chromatography and genotype AS and CM for identification of nontuberculous mycobacteria. Diagnostic Microbiology and Infectious Disease, 61, 284-293. Dawson, D.J. (1971). Potential pathogens among strains of mycobacteria isolated from house-dusts. Medical Journal of Australia, 1(13), 679-681. De Groote, M.A., & Huitt, G. (2006) Infections due to rapidly growing Mycobacteria. Clinical Infectious Diseases, 42(12), 1756-1763. du Moulin, G. , Stottmeier, K., Pelletier, P., Tsang, A., Hedley-Whyte, J. (1988). Concentration of Mycobacterium avium by hospital hot water systems. Journal of the American Medical Association, 260(11), 1599-1601. du Moulin, G. C., & Stottmeier, K. D. (1978). Use of cetylpyridinium chloride in the decontamination of water for culture of mycobacteria. Applied and Environmental Microbiology, 36(5), 771-773. Embil, J., Warren, P., Yakrus, M., Stark, R., Corne, S., Forrest, D., & Hershfield, E. (1997). Pulmonary illness associated with exposure to Mycobacterium-avium complex in hot tub water. Hypersensitivity pneumonitis or infection? Chest, 111, 813-816. Engel, H. W. B., Berwald, L. G., Havelaar, A. H. (1980). The occurrence of Mycobacterium kansasii in tapwater. Tubercle, 61(1), 21-26. Engel H. W. B, Berwald, L. G., Grange, J. M.,& Kubin M. (1980). Phage typing of Mycobacterium kansasii. Tubercle, 61(1):11-19. Falkinham III, J. O., Norton, C. D., Le Chavallier, M. W. (2001). Factors influencing numbers of Mycobacterium avium, Mycobacterium intracellulare and other mycobacteria in drinking water distribution systems. Applied and Environmental Microbiology, 67(3), 1225-1231.

53 Falkinham, J. (1996). Epidemiology of infection by nontuberculous mycobacteria. Clinical Microbiology Reviews, 9(2), 177-215. Falkinham, J. O. III (2002). Nontuberculous mycobacteria in the environment. Clinics in Chest Medicine, 23, 529-552. Falkinham III, J. O, Parker, B. C., & Gruft, H. (1980). Epidemiology of infection by nontuberculous mycobacteria. I. Geographic distribution in the eastern United States. American Review of Respiratory Disease, 121(6), 931-937. Falkinham, J. O. (2004). Environmental sources of Mycobacterium avium linked to routes of exposure. In B. J. Pedley S, Rees G, Dufour A, Cotruvo J (Ed.), Pathogenic Mycobacteria in Water (pp. 26-38). London: IWA Publishing. Falkinham, J. O. (2010). Impact of human activities on the ecology of nontuberculous mycobacteria. Future Microbiology, 5(6), 951-960 Fernandez-Rendon, E., Cerna-Cortes, J. F., Ramirez-Medina, M. A., Helguera- Repetto, A. C., Rivera-Gutierrez, S., Estrada-Garcia, T., & Gonzalez-y- Merchand, J. A. (2012). Mycobacterium mucogenicum and other nontuberculous mycobacteria in potable water of a trauma hospital: a potential source for human infection. Journal of Hospital Infection, 80(1), 74-76. Fox, C., Smith, B., Brogan, O., Rayner, A., Harris, G., & Watt, B. (1992). Non- tuberculous mycobacteria in a hospital's piped water supply. Journal of Hospital Infection, 21(2), 152-154. Fukoshima, M., Kakinuma, K., Hayashi, H., Nagai, H., Ito, K., Kawaguchi, R. (2003). Detection and identification of Mycobacterium species isolates by DNA microarray. Journal of Clinical Microbiology, 41, 2605-2615. Gaafar, A., Unzaga, M., Cisterna, R., Clavo, F.E., Urra, E., Ayarza, R., & Martin G. (2003) Evaluation of a modified single-enzyme amplified-fragment length polymorphism technique for fingerprinting and differentiating of Mycobacterium kansasii type I isolates. Journal of Clinical Microbiology, 41, 3846-3850. Galassi, L., Donato, R., Tortoli, E., Burrini, D., Santianni, D., Dei, R. (2003). Nontuberculous mycobacteria in hospital water systems: application of HPLC for identification of environmental mycobacteria. Journal of Water and Health, 1(3), 133-139.

54 Gangadharam, P.R., Lockhart, J.A., Awe, R.J., & Jenkins, D.E. (1976). Mycobacterial contamination through tap water. American Review of Respiratory Disease, 113(6), 894. Glazer, C. S., Martyny, J. W., Lee, B., Sanchez, T. L., Sells, T. M., Newman, L. S., ... Rose, C. S. (2007). Nontuberculous mycobacteria in aerosol droplets and bulk water samples from therapy pools and hot tubs. Journal of Occupational and Environmental Hygiene, 4(11), 831 - 840. Gomila, M., Ramirez, A., & Lalucat, J. (2007) Diversity of environmental mycobacterium isolates from hemodialysis water as shown by a multigene sequencing approach. Applied and Environmental Microbiology, 73(12), 3787-3797. Goslee, S., & Wolinsky, E. (1976). Water as a source of potentially pathogenic mycobacteria. American Review of Respiratory Diseases, 113(3), 287-292. Gruft,H., Loder, A., Osterhout, M., Parker, B. D., & Falkinham, J. O., III. (1979). Postulated sources of Mycobacterium intracellulare and Mycobacterium scrofulaceum infection: isolation of mycobacteria from estuaries and ocean waters. American Review of Respiratory Diseases, 120(6), 1385-1388. Gurtler, V., Harford, C., Bywater, J., & Mayall, B.C. (2006). Direct identifiation of slowly growing Mycobacterium species by analysis of the intergenic 16S-23S rDNA spacer region (ISR) using a GelCompar II database containing sequence based optimization for restriction fragment site polymorphisms (RFLPs) for 12 enzymes. Journal of Microbiological Methods, 64, 185-199. Hafner, B., Haag, H., Geiss, H.K., & Nolte, O. (2004). Different molecular methods for the identification of rarely isolated non-tuberculous mycobacteria and description of new hsp65 restriction fragment length polymorphism patterns. Molecular and Cellular Probes, 18, 59-65. Harris, K. A., Kenna, D. T. D., Blauwendraat, C., Hartley, J. C., Turton, J. F., Aurora, P., & Dixon, G. L. J. (2012). Molecular fingerprinting of Mycobacterium abscessus strains in a cohort of pediatric cystic fibrosis patients. Journal of Clinical Microbiology, 50(5), 1758-1761.

55 Havelaar, A.H., Berwald, L.G., Groothuis, D.G., & Baas, J.G. (1985). Mycobacteria in Semi-public swimming pools and whirlpools. Zentralblatt fur Bakteriologie, Mikrobiologie und Hygiene, 180(5-6), 505-514. Haverkort, F., Australian Mycobacterium Reference Laboratory Network. (2003). National atypical mycobacteria survey, 2000. Communicable Diseases Intelligence, 27, 180-189. Heller, L. C., Jones, M., & Widen, R. H. (2008). Comparison of DNA pyrosequencing with alternative methods for identification of mycobacteria. Journal of Clinical Microbiology, 46(6), 2092-2094. Hilborn, E. D., Yakrus, M., Covert, T. C., Harris, S. I., Donnelly, S. F., Schmitt, M. T., ... Stelma Jnr, G. N. (2008). Molecular comparison of Mycobacterium avium isolates from clinical and environmental sources. Applied and Environmental Microbiology, 74(15), 4966-4968. Iinuma, Y., Ichiyama, S., Hasegawa, Y., Shimokata, K., Kawahara, S., & Matsushima, T. (1997). Large-restriction-fragment analysis of Mycobacterium kansasii genomic DNA and its application in molecular typing. Journal of Clinical Microbiology, 35(3), 596-599. Iivanainen, E., Martikainen, P. J., & Katila, M. L. (1997). Comparison of some decontamination methods and growth media for isolation of mycobacteria from northern brook waters. Journal of Applied Microbiology, 82(1), 121- 127. Iivanainen, E., Northrup, J., Arbeit, R. D., Ristola, M., Katila, M-L., & Von Reyn, C. F. (1999). Isolation of mycobacteria from indoor swimming pools in Finland. APMIS : acta pathologica, microbiologica, et immunologica Scandinavica, 107, 193-200. Johar, M., Manning, T., Tse, C., Desroches, N., Agrawal, B., Kunimoto, D.Y., & Kumar,R. (2007). Growth Inhibition of Mycobacterium bovis, Mycobacterium tuberculosis and Mycobacterium avium In Vitro: Effect of 1--D-2'- Arabinofuranosyl and 1-(2'-Deoxy-2'-fluoro--D-2'-ribofuranosyl) Pyrimidine Nucleoside Analogs. Journal of Medicinal Chemistry, 50(15), 3696-3705.

56 Jonsson, B. E., Gilljam, M., Lindblad, A., Ridell, M., Wold, A.E., & Welinder-Olsson, C. (2007). Molecular epidemiology of Mycobacterium abscessus, with focus on cystic fibrosis. Journal of Clinical Microbiology, 45(5), 1497-1504. Kahana, L.M., & Kay, J.M. (1997) Pneumonitis due to Mycobacterium avium complex in hot tub water: infection or hypersensitivity? Chest, 112(6), 1713- 1714. Kahana, L. M., Kay, J. M., Yakrus, M. A., & Waserman, S. (1997) Mycobacterium avium complex infection in an immunocompetent young adult related to hot tub exposure. Chest, 111, 242-245. Kamala, T., Paramasivan, C. N., Herbert, D., Venkatesan, P., & Prabhakar, R. (1994). Evaluation of procedures for isolation of nontuberculous mycobacteria from soil and water. Applied and Environmental Microbiology, 60(3), 1021-1024. Kirschner, R. A., Parker, B., & Falkinham III, J. (1992). Mycobacterium avium, Mycobacterium intracellulare, and Mycobacterium scrofulaceum in acid, brown-water swamps of the southeastern United States and their association with environmental variables. American Review of Respiratory Disease, 145, 271-275. Kubalek, I., & Komenda, S. (1995). Seasonal variations in the occurrence of environmental mycobacteria in potable water. APMIS : acta pathologica, microbiologica, et immunologica Scandinavica 103(5), 327-330. Kubalek, I., & Mysak, J. (1996) The prevalence of environmental mycobacteria in drinking water supply systems in a demarcated region in Czech Republic, in the period 1984-1989. Eur J Epidemiol, V12(5), 471-474. Kubica, G. P., Beam, R. E., Palmer, J. R., & Rigdon, A. L. (1961). The isolation of unclassified (atypical) acid fast bacilli from soil and water samples collected in the state of Georgia. American Review of Respiratory Disease, 84, 131352. Kubin, M., Švanclová, E., Medek, B., Chobot, S., Olšovský, Ž. (1980). Mycobacterium kansasii infection in an endemic area of Czechoslovakia. Tubercle, 61(4), 207-212. Kuritsky, J. N., Bullen, M. G., Broome, C. V., Silcox, V. A., Good, R. C., & Wallace, R. J. (1983) Sternal wound infections and endocarditis due to organisms of the

57 Mycbobacterium fortuitum complex. Annals of Internal Medicine, 98(6), 938- 939. Le Dantec, C., Duguet, J-P., Montiel, A., Dumoutier, N., Dubrou, S., & Vincent, V. (2002a) Chlorine disinfection of atypical mycobacteria isolated from a water distribution system. Applied and Environmental Microbiology, 68(3), 1025- 1032. Le Dantec, C., Duguet, J-P., Montiel, A., Dumoutier, N., Dubrou, S., Vincent, V. (2002b) Occurrence of mycobacteria in water treatment lines and in water distribution systems. Applied and Environmental Microbiology, 68(11), 5318- 5325. LeChevallier, M. W., Gullick, R. W., Karim, M. R., Friedman, M., & Funk, J.E. (2003) The potential for health risks from intrusion of contaminants into the distribution system from pressure transients. Journal of Water and Health, 1(1), 3-14. Lee, E-S., Yoon, T-H., Lee, M-Y., Han, S-H., & Ka, J-O. (2010). Inactivation of environmental mycobacteria by free chlorine and UV. Water Research, 44(5), 1329-1334. Lehtola, M. J., Juhna, T. Ä, Miettinen, I. T., Vartiainen, T., Martikainen, P. J. (2004a). Formation of biofilms in drinking water distribution networks, a case study in two cities in Finland and Latvia. Journal of Industrial Microbiology and Biotechnology, V31(11), 489-494. Lehtola, M. J., Miettinen, I. T., Keinanen, M. M., Kekki, T. K., Laine, O., Hirvonen, A., ...Martikainen, P.J. (2004b). Microbiology, chemistry and biofilm development in a pilot drinking water distribution system with copper and plastic pipes. Water Research, 38(17), 3769-3779. Lehtola, M. J., Miettinen, I. T., Lampola, T., Hirvonen, A., Vartiainen, T., Martikainen, P. J. (2005). Pipeline materials modify the effectiveness of disinfectants in drinking water distribution systems. Water Research, 39(10), 1962-1971. Lehtola, M. J., Laxander, M., Miettinen, I. T., Hirvonen, A., Vartiainen, T., Martikainen, P. J. (2006a). The effects of changing water flow velocity on the formation of biofilms and water quality in pilot distribution system

58 consisting of copper or polyethylene pipes. Water Research, 40(11), 2151- 2160. Lehtola, M. J., Torvinen, E., Miettinen, I. T., Keevil, C. W. (2006b). Fluorescence in situ hybridization using peptide nucleic acid probes for rapid detection of Mycobacterium avium subsp. avium and Mycobacterium avium subsp. paratuberculosis in potable-water biofilms. Applied and Environmental Microbiology, 72(1), 848-853. Lehtola, M. J., Torvinen, E., Kusnetsov, J., Pitkanen, T., Maunula, L., von Bonsdorff, C-H., ... Miettinen, I. T. (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(9), 2854-2859. Lehtola, M. J., Miettinen, I. T., Vartiainen, T., Martikainen, P. J. (2002). Changes in content of microbially available phosphorus, assimilable organic carbon and microbial growth potential during drinking water treatment processes Water Research, 36(15), 3681-3690. Leoni, E., Legnani, P., Mucci, M. T., & Pirani, R. (1999). Prevalence of mycobacteria in a swimming pool environment. Journal of Applied Microbiology, 87(5), 683-688. Lim, S. Y., Kim, B. J., Lee, M. K, & Kim, K. (2008). Development of a real-time PCR- based method for rapid differential identification of Mycobacterium species. Letters in Applied Microbiology, 46(1), 101-106. Lumb, R., Stapledon, R., Scroop, A., Bond, P., Cunliffe, D., Goodwin, A., ... Bastian, I. (2004). Investigation of spa pools associated with lung disorders caused by Mycobacterium avium complex in immunocompetent adults. Applied and Environmental Microbiology, 70(8), 4906-4910. Mangione, E.J., Huitt, G., Lenaway, D., Beebe, J., Bailey, A., Figoski, M., ... Yakrus, M.A. (2001) Nontuberculous Mycobacterial Disease Following Hot Tub Exposure. Emerging Infectious Diseases, 7(6), 1039-1042. Mankiewicz, E., & Majdaniw, O. (1982) Atypical mycobacteria in tapwater. Canadian Journal of Public Health, 73, 358-360.

59 Marciano-Cabral, F., Jamerson, M., & Kaneshiro, E. S. (2010). Free-living amoebae, Legionella and Mycobacterium in tap water supplied by a municipal drinking water utility in the USA. Journal of Water and Health, 8(1), 71-82. Marras, T.K., Wallace, R.J., Jr, Koth, L.L., Stulbarg, M.S., Cowl, C.T., & Daley, C.L. (2005) Hypersensitivity Pneumonitis Reaction to Mycobacterium avium in Household Water. Chest, 127(2), 664-671. Marras, T. K, & Daley, C. L. (2002). Epidemiology of human pulmonary infection with nontuberculous mycobacteria. Clinics in Chest Medicine, 23(3), 553-567. Martin-Casabona, N., Bahrmand, A. R., Bennedsen, J., Osltergaard Thomsen, V., Curcio, M., Fauville-Dufaux, M., ...Watt, B for the Spanish Group for Nontuberculous mycobacteria. (2004) Non-tuberculous mycobacteria: patterns of isolation. A multi-country retrospective survey. International Journal of Tuberculosis and Lung Disease, 8, 1186-1193. MartÌn-de-Hijas, N.Z., GarcÌa-Almeida, D., Ayala, G., Fernandez-Roblas, R., Gadea, I., Celdran, A., ... Esteban, J. (2009). Biofilm development by clinical strains of non-pigmented rapidly growing mycobacteria. Clinical Microbiology and Infection, 15(10), 931-936. McNabb, A., Adie, K., Rodrigues, M., Black, W.A., Isaac-Renton, J. (2006) Direct identification of mycobacteria in primary liquid detection media by partial sequencing of the 65-kilodalton heat shock protein gene. Journal of Clinical Microbiology, 44(1), 60-66. McSwiggan, D. A., & Collins, C. H. (1974). The isolation of M. kansasii and M. xenopi from water systems. Tubercle, 55, 291-297. Miltner, E.C., & Bermudez, L.E. (2000). Mycobacterium avium grown in Acanthamoeba castellani is protected from the effects of antimicrobials. Jouranl of Antimicrobial Agents and Chemotherapy, 44(7), 1990-1994. Montecalvo, M. A., Forester, G., Tsang, A. Y., du Moulin, G., & Wormser, G.P. (1994). Colonisation of potable water with Mycobacterium avium complex in homes of HIV-infected patients. Lancet, 343(8913), 1639. Neonakis, I. K., Gitti, Z., Krambovitis, E., & Spandidos, D. A. (2008) Molecular diagnostic tools in mycobacteriology. Journal of Microbiological Methods, 75(1), 1-11.

60 Neumann, M., Schulze-Robbecke, R., Hagenau, C., & Behringer, K. (1997). Comparison of methods for isolation of mycobacteria from water. Applied and Environmental Microbiology, 63(2), 547-552. Nishiuchi, Y., Maekura, R., Kitada, S., Tamaru, A., Taguri, T., Kira, Y., ...Ito, M. (2007). The recovery of Mycobacterium avium-intracellulare Complex (MAC) from the residential bathrooms of patients with pulmonary MAC. Clinical Infectious Diseases, 45(3), 347-351. Norton, C. D., LeChevallier, M. W., & Falkinham III, J. O. (2004). Survival of Mycobacterium avium in a model distribution system. Water Research, 38(6), 1457-1466. Norton, C. D., & LeChevallier, M. W. (2000). A pilot study of bacteriological population changes through potable water treatment and distribution. Applied and Environmental Microbiology, 66(1), 268-276. O'Brien, D. P., Currie, B. J., Krause, V. L. (2000). Nontuberculous mycobacterial disease in northern Australia: a case series and review of the literature. Clinical Infectious Diseases, 31(4), 958-967. Pagnier, I., Merchat, M., Raoult, D., La Scola, B. (2009). Emerging Mycobacteria spp. in cooling towers [letter]. Emerging Infectious Diseases, 15(1), 121-122. Pang, S.C. (1991). Mycobacterium kansasii infections in Western Australia. Respiratory Medicine, 85, 213-218. Parashar, D., Chauhan, D.S., Sharma, V.D., Chauhan, A., Chauhan, S.V.S., Katoch, V.M. (2004). Optimization of Procedures for Isolation of Mycobacteria from Soil and Water Samples Obtained in Northern India. Applied and Environmental Microbiology, 70(6), 3751-3753. Parker, B. C., Ford, M. A., Gruft, H., & Falkinham III, J.O. (1983). Epidemiology of infection by nontuberculous mycobacteria. IV. Preferential aerosolization of Mycobacterium intracellulare from natural waters. American Review of Respiratory Disease, 128(4), 652-656. Payment, P. (1999). Poor efficacy of residual chlorine disinfectant in drinking water to inactivate water borne pathogens in distribution systems. Canadian Journal of Microbiology, 45(8), 709.

61 Pelletier, P. A., Du Moulin, G. C., & Stottmeier, K.D. (1988). Mycobacteria in public water supplies: comparative resistance to chlorine. Microbiological sciences, 5(5), 147-148. Peters, M., Muller, C., Rusch-Gerdes, S., Seidel, C., Gobel, U., Pohle, H. D., & Ruf, B. (1995). Isolation of atypical mycobacteria from tap water in hospitals and homes: Is this a possible source of disseminated MAC infection in AIDS patients? Journal of Infection, 31(1), 39-44. Pfaller, S. L., Aronson, T. W., Holtzman, A. E., & Covert, T.C. (2007). Amplified fragment length polymorphism analysis of Mycobacterium avium complex isolates recovered from southern California. Journal of Medical Microbiology, 56(9), 1152-1160. Picardeau, M., Prod'hom, G., Raskine,L., LePennec, M.P., & Vincent,V. (1997) Genotypic characterization of five subspecies of Mycobacterium kansasii. Journal of Clinical Microbiology, 35(1), 25-32. Pryor, M., Springthorpe, S., Riffard, S., Brooks, T.,Huo, Y., Davis ,G., Sattar, S.A. (2004) Investigation of opportunistic pathogens in municipal drinking water under different supply and treatment regimes. Water Science and Technology, 50(1), 83-90. Quiñones, C., Ramalle-Gómara, E., Perucha, M., Lezaun, M.E., Fernández-Vilariño, E., García-Morrás, P., & Simal, G. (2010). An outbreak of Mycobacterium fortuitum cutaneous infection associated with mesotherapy. Journal of the European Academy of Dermatology and Venereology, 24(5), 604-606. Rautiala, S., Torvinen, E., Torkko, P., Suomalainen, S., Nevalainen, A., Kalliokoski, P., Katila, M.L. (2004) Potentially pathogenic, slow-growing mycobacteria released into workplace air during the remediation of buildings. Journal of Occupational and Environmental Hygiene, 1(1), 1-6. Reznikov, M., & Dawson, D. J. (1971). Serological investigation of strains of Mycobacterium intracellulare ("battey" bacillus) isolated from house-dusts. Medical Journal of Australia, 1(13), 682-683. Reznikov, M., & Dawson, D. J. (1980). Mycobacteria of the intracellulare- scrofulaceum group in soils from the Adelaide area. Pathology, 12(4), 525- 528.

62 Reznikov, M., Leggo, J. H., & Dawson, D. J. (1971). Investigation by seroagglutination of strains of the Mycobacterium intracellulare-M. scrofulaceum group from house dusts and sputum in Southeastern Queensland. American Review of Respiratory Disease, 104(6), 951-953. Rodgers, M. R., Blackstone, B. J., Reyes, A. L., & Covert, T. C. (1999). Colonisation of point of use water filters by silver resistant nontuberculous mycobacteria. (Letter) Journal of Clinical Pathology, 52(8), 629 Rose, C. S., Martyny, J. W., Newman, L. S., Milton, D. K., King, T. E., Beebe, J. L., ... Kreiss, K. (1998). Lifeguard Lung: endemic granulomatous pneumonitis in an indoor swimming pool. American Journal of Public Health, 88(12), 1795- 1800. Salah, I. B., Ghigo, E., & Drancourt, M. (2009). Free-living amoebae, a training field for macrophage resistance of mycobacteria. Clinical Microbiology and Infection, 15(10), 894-905. Sampaio, J. L. M, Chimara, E., Ferrazoli, L., da Silva Telles, M. A., Del Guercio, V. M. F., Jerico, Z. V. N. .… & Leao, S. C. (2006). Application of four molecular typing methods for analysis of Mycobacterium fortuitum group strains causing post-mammaplasty infections. Clinical Microbiology and Infection, 12(2), 142-149. Sampaio, J. L. M., Viana-Niero, C., de Freitas, D., Hfling-Lima, A. L., & Leao, S. C. (2006). Enterobacterial repetitive intergenic consensus PCR is a useful tool for typing Mycobacterium chelonae and Mycobacterium abscessus isolates. Diagnostic Microbiology and Infectious Disease, 55(2), 107-118. Santos, R., Oliveira, F., Fernandes, J., Gonçalves, S., Macieira, F., & Cadete, M. (2005). Detection and identification of mycobacteria in the Lisbon water distribution system. Water Science and Technology, 52(8), 177-180. Schulze-Röbbecke, R., Janning, B., & Fischeder, R. (1992). Occurrence of mycobacteria in biofilm samples. Tuberculosis and Lung Disease., 73(3), 141- 144. Schulze-Robbecke, & R., Buchholtz, K. (1992) Heat susceptibility of aquatic mycobacteria. Applied and Environmental Microbiology, 58(6), 1869-1873.

63 Schulze-Robbecke, R., Weber, A., & Fischeder, R. (1991). Comparison of decontamination methods for the isolation of mycobacteria from drinking water samples. Journal of Microbiological Methods, 14(3), 177-183. Sebakova, H., Kozisek, F., Mudra, R., Kaustova, J., Fiedorova, M., Hanslikova, D., ...Sasek, J. (2008). Incidence of nontuberculous mycobacteria in four hot water systems using various types of disinfection. Canadian Journal of Microbiology, 54(11), 891-898. September, S. M., Brozel, V. S., Venter, S. N. (2004). Diversity of nontuberculoid mycobacterium species in biofilms of urban and semiurban drinking water distribution systems. Applied and Environmental Microbiology, 70(12), 7571- 7573. Sheffer, P. J., Stout, J. E., Wagener, M. M., & Muder, R. R. (2005) Efficacy of new point-of-use water filter for preventing exposure to Legionella and waterborne bacteria. American Journal of Infection Control, 33(5, Supplement 1), S20-S25. Shelton, B. G., Flanders, W. D., & Morris, G. K. (1999). Mycobacterium sp. as a possible cause of hypersensitivity pneumonitis in machine workers. Emerging Infectious Diseases, 5(2), 270-273. Slosarek, M., Kubin,M., & Jaresova, M. (1993). Water-borne household infections due to Mycobacterium xenopi. Central European Journal of Public Health, 1(2), 78-80. Stinear, T., Ford, T., Vincent, V. (2004). Analytical methods for the detection of waterborne and environmental pathogenic mycobacteria. In B. J. Pedley S, Rees G, Dufour A, Cotruvo J (Ed.), Pathogenic Mycobacteria in Water: a Guide to Public Health Consequences, Monitoring and Management (pp. 55- 73). London: World Health Organisation, IWA Publishing. Stinear, T., Davies, J.K., Jenkin, G.A., Hayman, J.A., Oppedisano, F., Johnson, P.D.R. (2000). Identification of Mycobacterium ulcerans in the Environment from Regions in Southeast Australia in Which It Is Endemic with Sequence Capture-PCR. Applied and Environmental Microbiology, 66(8), 3206-3213. Taylor, R. H., Falkinham III, J. O., Norton, C. D., & Le Chevallier, M. W. (2000). Chlorine, chloramine, chlorine dioxide, and ozone susceptibility of

64 Mycobacterium avium. Applied and Environmental Microbiology, 66(4), 1702-1705. Thomas, J. M., & Ashbolt, N. J. (2011). Do free-living amoebae in treated drinking water systems present an emerging health risk? Environmental Science and Technology, 45(3), 860-869. Thomas, V., Herrera-Rimann, K., Blanc, D.S., Greub, G. (2006). Biodiversity of amoebae and amoeba-resisting bacteria in a hospital water network. Applied and Environmental Microbiology, 72(4), 2428-2438. Thomas, V., Loret, J-F, Jousset, M., Greub, G. (2008). Biodiversity of amoebae and amoebae-resisting bacteria in a drinking water treatment plant. Environmental Microbiology, 10(10), 2728-2745. Thomson, R. (2010). Changing epidemiology of pulmonary nontuberculous mycobacteria infections. Emerging Infectious Diseases, 16(10), 1576-1582. Thomson, R. M., Armstrong, J. G., & Looke, D. F. (2007). Gastroesophageal reflux disease, acid suppression, and Mycobacterium avium complex pulmonary disease. Chest, 131, 1166-1172. Tobin-D'Angelo, M. J., Blass, M. A., del Rio, C., Halvosa, J. S., Blumberg, H. M., & Horsburgh Jr., C. R. (2004) Hospital water as a source of Mycobacterium avium complex isolates in respiratory specimens. Journal of Infectious Diseases, 189, 98-104. Tobler, N. E., Pfunder, M., Herzog, K., Frey, J. E., & Altwegg, M. (2006) Rapid detection and species identification of Mycobacterium spp. using real time PCR and DNA-microarray. Journal of Microbiological Methods, 66, 116-124. Torvinen, E., Lehtola, M. J., Martikainen, P. J., & Miettinen, I. T. (2007). Survival of Mycobacterium avium in drinking water biofilms as affected by water flow velocity, availability of phosphorus, and temperature. Applied and Environmental Microbiology, 73(19), 6201-6207. Torvinen, E., Suomalainen, S., Lehtola, M. J., Miettinen, I. T., Zacheus, O., Paulin, L., ... Martikainen, P. J. (2004) Mycobacteria in water and loose deposits of drinking water distribution systems in Finland. Applied and Environmental Microbiology, 70(4), 1973-1981.

65 Tsintzou, A., Vantarakis, A., Pagonopoulou, O., Athanassuadou, A., & Papapetropoulou, M. (2000). Environmental mycobacteria in drinking water before and after replacement of the water distribution network. Water Air and Soil Pollution, 120, 273-282. Tuffley, R. E., & Holbeche, J. D. (1980). Isolation of the Mycobacterium avium-M. intracellulare-M. scrofulaceum complex from tank water in Queensland, Australia. Applied and Environmental Microbiology, 39(1), 48-53. Tuohy, M. J., Hall, G. S., Sholtis, M., & Procop, G. W. (2005) Pyrosequencing as a tool for the identification of common isolates of Mycobacterium species. Diagnostic Microbiology and Infectious Disease, 51, 245-250. von Reyn, C. F., Waddell, R. D., Eaton,T., Arbeit, R. D., Maslow, J. N., Barber, T. W., ... Lahdevirta, J. (1993). Isolation of Mycobacterium avium complex from water in the United States, Finland, Zaire, and Kenya. Journal of Clinical Microbiology, 31(12), 3227-3230. von Reyn, C. F., Marlow, J. N., Arbeit, R. D., Barber, T. W., Falkinham, J. O. (1994). Persistent colonisation of potable water as a source of Mycobacterium avium infection in AIDS. The Lancet, 343(8906), 1137-1141. Vugia, D.J., Jang, Y., Zizek, C., Ely, J., Winthrop, K.L., & Desmond, E. (2005). Mycobacteria in nail salon whirlpool footbaths, California. Emerging Infectious Diseases, 11(4), 616-618. Wallace, R. J., Musser, J. M., Hull, S.I ., Silcox, V. A., Steele, L. D., Forrester, G. D., ... Selander, R. K. (1989). Diversity and sources of rapidly growing mycobacteria associated with infections following cardiac surgery. Journal of Infectious Diseases, 159(4), 708-716. Wallace Jr.,R. J., Zhang, Y., Brown, B. A., Fraser, V., Mazurek, G. H., Maloney, S. (1993) DNA large restriction fragment patterns of sporadic and epidemic nosocomial strains of Mycobacterium chelonae and Mycobacterium abscessus. Journal of Clinical Microbiology, 31(10), 2697-2701. Wallace, R. J., Brown, B. A., & Griffith, D. E. (1998) Nosocomial Outbreaks/pseudooutbreaks caused by nontuberculous mycobacteria. Annual Reviews in Microbiology, 52(1), 453-490.

66 Wendt, S. L., George, K. L., Parker, B. C., Gruft, H., Falkinham III, J. O. (1980) Epidemiology of infection by nontuberculous Mycobacteria. III. Isolation of potentially pathogenic mycobacteria from aerosols. American Review of Respiratory Disease., 122(2), 259-263. Whittington, R. J., Marsh, I., Turner, M. J., McAllister, S., Choy, E., Eamens, G. J., Marshall, D. J., Ottaway, S. (1998). Rapid Detection of Mycobacterium paratuberculosis in clinical samples from ruminants and in spiked environmental samples by modified BACTEC 12B radiometric culture and direct confirmation by IS900 PCR. Journal of Clinical Microbiology, 36(3), 701-707. Whittington, R. J., Marsh, I. B., Reddacliff, L. A. (2005). Survival of Mycobacterium avium subsp. paratuberculosis in dam water and sediment. Applied and Environmental Microbiology, 71(9), 5304-5308. Whittington, R. J., Marshall, D. J., Nicholls, P. J., Marsh, I. B., Reddacliff, L. A. (2004) Survival and dormancy of Mycobacterium avium subsp. paratuberculosis in the environment. Applied and Environmental Microbiology, 70(5), 2989- 3004. Winthrop, K., Abrams, M., Yakrus, M., Schwartz, I., Ely, J., Gillies, D., Vugia, D. (2002) An outbreak of mycobacterial furunculosis associated with footbaths at a nail salon. New England Journal of Medicine, 364(18), 1366-1371. Wolinsky, E. (1995) Mycobacterial lymphadenitis in children: a retrospective study of 105 nontuberculous cases with long term followup. Clinical Infectious Diseases, 20, 954-963. Wright, E.P., Collins, C.H., Yates, M.D. (1985) Mycobacterium xenopi and Mycobacterium kansasii in a hospital water supply. Journal of Hospital Infection, 6, 175-178. Zelazny, A. M., Root, J. M., Shea, Y. R., Colombo, R. E., Shamputa, I. C., Stock, F., ... Sampaio, E.P. (2009). Cohort study of molecular identification and typing of Mycobacterium abscessus, Mycobacterium massiliense, and Mycobacterium bolletii. Journal of Clinical Microbiology, 47(7), 1985-1995. Zhang, Y., Rajagopalan, M., Brown, B. A., & Wallace, R. J., Jr. (1997). Randomly amplified polymorphic DNA PCR for comparison of Mycobacterium

67 abscessus strains from nosocomial outbreaks. Journal of Clinical Microbiology, 35(12), 3132-3139. Zhang, Y., Yakrus, M.A., Graviss, E.A., Williams-Bouyer, N., Turenne, C., Kabani, A., & Wallace, R.J., Jr. (2004) Pulsed-field gel electrophoresis study of Mycobacterium abscessus isolates previously affected by DNA degradation. Journal of Clinical Microbiology, 42(12), 5582-5587.

Chapter 1:Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1: Objective 1 Aim 3: Objective 5 Aim 2: Objectives 3&4 Chapter 3: Chapter 9: Isolation of Chapter 5: Development of NTM from homes of method for isolation Mycobacterium patients with NTM lentilavum of NTM from water disease

Aim 1:Objective 2 Chapter 4: Isolation of Chapter 6: NTM from Brisbane M. abscessus water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

70 CHAPTER 3: A COMPARISON OF METHODS FOR PROCESSING DRINKING WATER SAMPLES FOR THE ISOLATION OF MYCOBACTERIUM AVIUM AND INTRACELLULARE.

The following chapter has been published as a paper as follows: Thomson, R., Carter, R., Gilpin, C., Coulter, C., & Hargreaves, M. (2008). Comparison of methods for processing drinking water samples for the isolation of Mycobacterium avium and Mycobacterium intracellulare. Applied and Environmental Microbiology, 74(10): 3094-3098.

Author contributions:

RT (70%) designed the study, performed the processing of water samples, collated and analysed the data, and wrote the manuscript.

RC (15%) assisted with the processing of the water samples, collating of the results and reviewed the manuscript.

CG (2.5%) Intellectual and design support

CC (2.5%) Organisational support

MH (10%) intellectually contributed to the study design and methodology, and contributed to the writing of the manuscript.

ABSTRACT

Several protocols for isolation of mycobacteria from water exist but there is no established standard method. This study compared methods of processing potable water samples for the isolation of M.avium and M.intracellulare using spiked sterilized water and tap water decontaminated using Cetylpyridinium Chloride (CPC) 0.005%. Samples were concentrated by centrifugation or filtration, and inoculated onto Middlebrook 7H10 and 7H11 plates, Löwenstein-Jensen (LJ) slopes and into Mycobacterial Growth Indicator tubes (MGIT) ± PANTA (polymyxin, azlocillin, nalidixic acid, trimethoprim, amphotericin B). The solid media were incubated at

32º, 35º and 35ºC with CO2 and read weekly. Results suggest filtration of water for the isolation of mycobacteria is a more sensitive method of concentration than

71 centrifugation. The addition of Sodium Thiosulphate may not be necessary and may reduce the yield. M7H10 or 7H11 were equally sensitive culture media. CPC decontamination, whilst effective in reducing growth of contaminants also significantly reduces mycobacterial numbers. There was no difference at 3 weeks between the different incubation temperatures.

INTRODUCTION

The isolation of Mycobacteria from both environmental and treated drinking water samples was first reported in the early 1900s. However it has only been in the last 3-4 decades that these environmental mycobacteria have been recognized as pathogens of human disease. As compared to Mycobacterium tuberculosis, these organisms are generally of low virulence and require a host defect for the establishment of disease (e.g. disseminated disease in AIDS patients, pulmonary disease in patients with underlying structural lung disease). However there is a subset of patients who develop pulmonary disease without an obvious immune defect or one that is yet to be defined. Nontuberculous Mycobacteria (NTM) have been demonstrated in drinking water (Chang, Wang, Liao, & Huang, 2002; Falkinham III, Norton, & Le Chavallier, 2001; Goslee & Wolinsky, 1976; Hilborn et al., 2006; I Kubalek & Komenda, 1995; Kubalek & Mysak, 1996; Le Dantec et al., 2002b; Pryor, 2004; von Reyn, Marlow, Arbeit, Barber, & Falkinham, 1994), drinking water distributions systems (Le Dantec et al., 2002a, 2002b; Norton, LeChevallier, & Falkinham III, 2004; Tobin-D'Angelo et al., 2004; Torvinen et al., 2004; Vaerewijck, Huys, Palomino, Swings, & Portaels, 2005), hot water systems (du Moulin, Stottmeier, Pelletier, Tsang, & Hedley-Whyte, 1988), spas (Embil et al., 1997) and pools (Iivanainen, 1999; Leoni, Legnani, Mucci, & Pirani, 1999). However several authors have failed to identify NTM in water samples, often because of unsuitable isolation techniques. Variable growth rates, specific growth requirements and different sources of water samples (eg treated/surface/natural) are all variables that will affect the choice of method for identification. Because of their slow growth, pre-treatment methods are necessary to limit bacterial and fungal overgrowth and hence detect mycobacteria. However, the pretreatment method chosen may also

72 prevent the detection of certain species of mycobacteria, reducing the rate of positive samples and number of colonies seen. A number of different protocols have been described (Stinear, Ford, & Vincent, 2004) but a standard protocol has not yet been established.

Du Moulin and Stottmeier (1978) first described the use of Cetylpyridinium chloride. They applied 0.04% to 1L samples of distilled water seeded with dilutions of 5-day- old cultures of mycobacteria grown in M7H9 broth and allowed them to stand for 24 hours, prior to filtration, rinsing and applying membrane to M7H10 agar plates.

Plates were incubated at 37°C (5-10%CO2) for 60 days. A control group of samples were processed the same way but without CPC treatment. Survival of mycobacteria in spiked specimens varied from 1 to 100%, depending on the species - M. kansasii 18.4%, M. gordonae 8.4%, M. intracellulare 100%, M. fortuitum 1.1% and M. bovis 39.9%. These authors actually reported greater survival of M. intracellulare in treated (7400 viable units (vu)/L) vs untreated samples (440 vu/L) perhaps by the disruption of mycobacterial aggregates, releasing more cells. Schulz-Robbecke et al (Schulze-Robbecke, Weber, & Fischeder, 1991) compared cetylpyridinium chloride (CPC), sodium hydroxide (NaOH) and formaldehyde (HCHO) for their efficacy as decontamination substances for the isolation of mycobacteria from drinking water samples. They found that 0.005% CPC had the highest recovery of mycobacteria and the lowest contamination rates, using both spiked samples and environmental samples. This finding was confirmed by Neumann et al (Neumann, Schulze- Robbecke, Hagenau, & Behringer, 1997). Glover et al (1994) found that 0.04%CPC decontamination of tap water samples resulted in less contamination than 1- 3%NaOH or 5% OA, but also the highest number of samples with no growth. CPC was applied at this concentration for 24 hours to sterile water seeded with MAC to a final concentration of 1.5X103 CFU/500ml. This resulted in a reduction of 89% in viable mycobacteria. NaOH 1% and OA 5% resulted in reductions of 64% and 59% respectively.

Le Dantec (2002b) used membrane filtration followed by decontamination with Na dodecyl sulfate and NaOH, adjusting pH with 40% phosphoric acid. Using M.

73 gordonae spiked sterile tap water they showed that this decontamination method reduced mycobacterial numbers to 1% of the original number.

Falkinham (Falkinham III et al., 2001; Falkinham, 2004) has suggested that for drinking water decontamination may not be required. In his study published in 2001 (Falkinham III et al., 2001) he processed samples initially without decontamination but if plates were overgrown, reprocessed them using CPC. Unfortunately it is not stated in the paper how often decontamination was necessary. Only 15% samples grew slow growing mycobacteria (3% M. avium, 1% M. intracellulare) and there were 2% rapid growers.

Other variables that may affect the yield of mycobacteria from environmental water samples include sample volume, the use of Sodium Thiosulphate to neutralize chlorine based disinfectants, method of concentration (eg filtration vs centrifugation), culture media and incubation temperature.

In Queensland the main mycobacterial pathogen associated with pulmonary disease is M. intracellulare followed by M. avium, M. abscessus and M. kansasii. It has been postulated that patients acquire disease by inhaling aerosols containing mycobacteria from environmental water sources and water outlets in their homes (Marras et al., 2005). Patients may also aspirate contaminated water as a result of swallowing disorders or severe gastro-esophageal reflux disease (Thomson, Armstrong, & Looke, 2007).

This pilot study was undertaken to try and identify the best method of processing water samples for the isolation of mycobacteria prior to a larger environmental survey. The aim of this study was to compare different methods of processing drinking water samples for the isolation of species of mycobacteria pathogenic to humans, particularly M. intracellulare and M. avium, with regard to concentration (centrifugation vs filtration), culture media (LJ, 7H10, 7H11, MGIT and MGIT+PANTA), and incubation temperature (32°C, 35°C, and 35°C+CO2).

METHODS

M. avium (ATCC 35765) and M. intracellulare (ATCC 13950) were inoculated in 7H9 broth (0.5 McFarland), correlating to concentration 1.5x108 CFU/ml and diluted to a concentration of 100 CFU/500ml water.

Control samples. Organisms (M. avium and M. intracellulare separately) were added to 8x 500ml samples of sterile water (sterilized by filtration to preserve chlorination using MediaKap-2 - Hollow Fibre Media Filter 0.2µm - Spectrum Laboratories Inc.) to a final concentration of 100CFU/500ml. Sodium thiosulphate (0.5ml of 10% solution) was added to 4 of the samples (2 M. avium and 2 M. intracellulare). Half of the samples were processed by filtration and half processed by centrifugation. (Figure 3.1)

Filtration was performed through 0.45µm cellulose nitrate filters (Sartorius AG 37070 Goettingen, Germany). Filters were then rinsed with 2ml sterile distilled water (SDW) and macerated in 3 ml SDW. From this 3ml, 0.1ml aliquots were then

Figure 3.1. Flow chart for processing of sterile water samples transferred in triplicate to LJ slants, M7H10 and M7H11 plates , sealed in gas permeable plastic bags for incubation at 32°, 35° and 35°C +CO2. 0.5 ml aliquots

75 were transferred to 2 MGIT tubes, 1 containing PANTA (polymixin, azlocillin, nalidixic acid, trimethoprim, amphotericin B).

Centrifugation: Four - 500ml samples (2 containing Na Thiosulphate -1 M.avium, 1 M. intracellulare) were centrifuged in 250ml sterile bottles at 5000gx 20min at 25 °C. The pelleted cells were rinsed twice with Phosphate Buffered Saline (PBS)(Murray, 1999). The resulting suspension was added to sterile diluent to make 3ml and 0.1ml aliquots were used to inoculate in triplicate each of the following: LJ slants, M7H10 and M7H11 plates. Plates were sealed in gas permeable plastic bags and incubated as previously indicated. Additional 0.5 ml aliquots were used to inoculate 2 MGIT tubes, with and without PANTA.

Tap water. Tap water samples (4x 500ml) were collected after flushing for 2 minutes from a single tap within the laboratory. These tap water samples were spiked with M. avium (x2) and M. intracellulare (x2) to give a final concentration of 100CFU/500ml. Samples were then decontaminated with 0.005% CPC and incubated at room temperature for 30 minutes. Two samples (1 M. avium, 1 M. intracellulare) were then processed by filtration and 2 processed by centrifugation, as described for sterile samples.

All plates were read weekly. At three weeks all plates were photographed digitally and colonies counted. Colonies from plates demonstrating growth were stained to confirm acid fast bacilli, and morphologically different colonies were subcultured on M7H10 agar and incubated at 35°C. Subcultured organisms were then identified to the species level using multiplex PCR as described by Wilton and Cousins (1992). All colonies grown from the tap water samples were treated similarly.

Data were analyzed using SPSS v12.0 for Windows 2003 (Apache Software Foundation). Tests of association were performed using Fishers exact test for Chi Squared 2x2 tables. Statistical significance was defined as a 2-sided p value<0.05.

76 Colony counts were also compared using the Mann Whitney U test as the values were not normally distributed.

RESULTS

There were 88 spiked sterile water cultures, and 44 spiked tap water samples. 83.3% of all filtered samples grew mycobacteria compared to 12.1% of all centrifuged samples (p<0.0001).

Of spiked sterile samples not treated with Sodium thiosulphate, 52.3% grew mycobacteria compared to 43.2% of samples that were treated (p=0.223). For filtered samples the addition of Sodium thiosulphate did not affect recovery. However, for centrifuged samples, 4.5% of treated samples were positive compared to 22.7% of untreated samples (p=0.058). Colony counts were lower in filtered sterile samples with Sodium thiosulphate added (mean ± SD: 151.7±169.8 vs 259.0 ± 352.8 CFU/L; however this was not statistically significant (p=0.178) Mann Whitney U test (p=0.709) (Figure 3.2).

Concentration: Filtration, Water Type: sterile

1000.00

800.00

600.00 CFU/L

400.00

200.00

0.00

No Yes Na Thiosulphate

Figure 3.2. Boxplot demonstrating Median CFU/L (black bar), middle 2 quartiles (box) and range (extent of bar) of CFU/L Mycobacteria in spiked sterile water concentrated by filtration processed with and without the addition of NaThiosulphate.

There was no overall difference between Middlebrook 7H10 and 7H11 with 12 and 13 of 18 filtered samples showing positive growth respectively after 1 week. The LJ slants initially appeared less sensitive, but there was no difference between it and the Middlebrook media at 3 weeks (Table 3.1).

For filtered samples CPC decontamination did not appear to affect the number of positive cultures at 3 weeks - 86.4% of filtered samples treated with CPC were positive at final reading, compared to 81.8% of those untreated. However colony counts were significantly reduced in CPC treated spiked tap water samples (Mean ± SD CFU/L 7.4±8.5) compared to untreated sterile samples (205.4±262.4; p=0.0001). This CPC effect equates to a mean reduction to 3.6% of original numbers.

Table 3.1. Positive cultures for mycobacteria after concentration by filtration using different culture media after 1 week and 6 weeks (n=66).

Culture Week 1 Final Culture Result media Contaminated Negative Positive Overgrown Negative Positive 7H10 (n=18) 1 5 12 0 2 16

7H11 (n=18) 0 5 13 1 0 17

LJ (n=18) 2 16 0 0 1 17

MGIT (n=6) 0 6 0 0 5 1

MGIT+PANTA 0 3 3 0 2 4 (n=6)

78 There was no difference overall between the different incubation temperatures (Table 3.2).

Table 3.2. Comparison of incubation temperatures for culture of mycobacteria in both spiked sterile and tap water samples processed by centrifugation or filtration.

Concentration Incubation Temperature Total

32°C 35°C 35°C+CO2 BACTEC Centrifugation Contaminated 2 2 3 0 7 Negative 14 14 12 11 51 Positive 2 2 3 1 8 Total 18 18 18 12 66

Filtration Contaminated 0 0 1 0 1 Negative 0 0 0 7 7 Positive 18 18 17 5 58 Total 18 18 18 12 66

At 3 weeks, 3 samples not treated with CPC were overgrown compared to none of those treated. Nine of 88 (10.2%) spiked sterile samples grew contaminants in addition to mycobacteria compared to 13/44 (29.5%) tap water samples. (p=0.012). These contaminants did not affect the ability to isolate mycobacteria. Of the spiked sterile samples, in 2 of these the plates had fungal overgrowth at week 4 – a week after they had been photographed, and this was likely aerial contamination when the plates were inspected for photography. Of the remaining 7 - 4 plates had single non-buff colonies, 2 had 2 colonies and 1 had 3. Whilst these were not formally identified it is presumed they entered the system during the processing of samples.

A number of samples grew morphologically different colonies on Middlebrook plates. These were subcultured and then identified to the species level using multiplex PCR and found to be the same organism. PFGE was not performed on these isolates. All colonies grown from the tap water samples were similarly processed. No other mycobacteria (other than the spiked organisms) were identified from the tap water samples.

DISCUSSION

In this study we have demonstrated that filtration is a more effective method than centrifugation for isolating mycobacteria from water samples. Apart from having a far greater yield it was also simpler and more time efficient. To our knowledge there have been no published direct comparative studies, but previous authors have been able to isolate mycobacteria from water samples processed by centrifugation. Perhaps it was our technique, but alternatively the success of previous authors maybe related to much higher concentrations of mycobacteria in the water being sampled. In this study low concentrations of target organisms were used as may be expected to exist in suburban, treated, water supplies (Covert, Rodgers, Reyes, & Stelma, 1999; Falkinham III et al., 2001; Galassi, 2003; Hilborn et al., 2006; Le Dantec et al., 2002b).

The majority of reported studies have processed samples with sodium thiosulphate to neutralize residual chlorine (Clesceri, Greenberg, & Eaton, 1999). It is not known whether neutralizing residual chlorine interferes with the ability to isolate mycobacteria by increasing bacterial overgrowth, or whether the presence of residual chlorine reduces the yield and diversity of species of mycobacteria subsequently isolated. As most opportunistic pathogenic NTM are relatively resistant to chlorine (Le Dantec et al., 2002b; Payment, 1999; Pelletier, Du Moulin & Stottmeier, 1988; Steed & Falkinham, 2006; Taylor, Falkinham III, Norton, & LeChevallier, 2000) the addition of sodium thiosulfate may not be necessary and may increase contamination rates.

The thiosulfate anion characteristically reacts with dilute acids to produce sulfur, 2− + sulfur dioxide and water: S2O3 (aq) + 2H (aq) → S(s) + SO2(g) + H2O(l). Thiosulfate reduces the hypochlorite and in so doing becomes oxidized to sulfate. The complete reaction is: 4NaClO + Na2S2O3 + 2NaOH → 4NaCl + 2Na2SO4 + H2O (Holleman & Wiberg, 2001).

80 From our results it would appear that sodium thiosulphate may have some antibacterial properties in water, perhaps by generation of sulfur, as contamination rates and mycobacterial colony counts were less in those treated samples. Though not statistically significant, this is an interesting observation. Of importance, it would seem that for the purposes of isolating mycobacteria from treated water, the addition of sodium thiosulphate is unnecessary.

The addition of CPC to tap water samples spiked with M. avium and M. intracellulare resulted in 3.6% survival of organisms, but did not affect the number of positive samples using this concentration of organisms. The organisms used in our study were grown in 7H9 broth. It has been shown that antecedent growth conditions may affect susceptibility to chlorine-based disinfectants. Water grown strains of M. avium were shown to be significantly more chlorine resistant than those grown in medium (Taylor et al., 2000). The magnitude of reduction of growth we have shown may not necessarily apply to water grown organisms from environmental or distribution system samples. There were no differences between the temperatures tested nor between the different solid media overall. However the Middlebrook media were more sensitive at one week and provided the advantage of quantitation of growth over LJ. The MGIT system has recently been introduced for the culture of clinical specimens, and has not been used widely for the processing of water samples. Supplementation with PANTA (Polymyxin, Amphotericin B, Nalidixic acid, Trimethroprim and Azlocillin) is used to reduce contamination. A further study utilizing raw tap water samples (ie no decontamination) and the MGIT system (+PANTA to control contamination) is currently underway. The MGIT without PANTA used in this study did not contain OADC enrichment which may explain the lower yield using this system.

There have been a number of studies published using different methods to isolate mycobacteria from water samples, and no established standard. We have demonstrated that sodium thiosulphate may not be necessary and may interfere with growth. We have confirmed the findings of previous authors that CPC controls contamination, but significantly reduces mycobacterial growth also. Whilst it would

81 be appealing to process samples without decontamination, the utility of this method would depend on the origin of the samples.

This study has added refinement to concentration and culture techniques for the isolation of mycobacteria from water; however the major challenge remains the need for decontamination to reduce bacterial and fungal overgrowth. We and others have demonstrated that the addition of CPC is effective for this purpose; however we have quantified the reduction in yield of M. intracellulare and M. avium, two of the main pathogens associated with lung disease and found it is significant. Given that that major environmental niche for M. intracellulare is in biofilms (Falkinham III et al., 2001) and only small numbers are found in water samples, the detection of low concentrations of organisms is important. Perhaps a metagenomic study may obviate the need for any decontamination and culture method, and developments in this area are awaited with interest.

REFERENCES

Chang, C-T., Wang, L-Y., Liao, C-Y., & Huang, S-P. (2002). Identification of nontuberculous mycobacteria existing in yap water by PCR-restriction fragment length polymorphism. Applied and Environmental Microbiology, 68(6), 3159-3161. Clesceri, L. S., Greenberg, A. E., & Eaton, A. D. (Eds.). (1999). Standard Methods for the examination of water and wastewater. (20th ed.). Washington DC: American Public Health Association. Covert, T. C., Rodgers, M. R., Reyes, A. L., & Stelma, G. N., Jr. (1999). Occurrence of nontuberculous mycobacteria in environmental samples. Applied and Environmental Microbiology, 65(6), 2492-2496. Du Moulin, G., Stottmeier, K., Pelletier, P., Tsang, A., & Hedley-Whyte, J. (1988). Concentration of Mycobacterium avium by hospital hot water systems. Journal of the American Medical Association, 260(11), 1599-1601.

82 Du Moulin, G. C., & Stottmeier, K. D. (1978). Use of cetylpyridinium chloride in the decontamination of water for culture of mycobacteria. Applied and Environmental Microbiology, 36(5), 771-773. Embil, J., Warren, P., Yakrus, M., Stark, R., Corne, S., Forrest, D., & Hershfield, E. (1997). Pulmonary illness associated with exposure to Mycobacterium-avium complex in hot tub water. Hypersensitivity pneumonitis or infection? Chest, 111, 813-816. Falkinham III, J. O., Norton, C. D., & Le Chavallier, M. W. (2001). Factors influencing numbers of Mycobacterium avium, Mycobacterium intracellulare and other mycobacteria in drinking water distribution systems. Applied and Environmental Microbiology, 67(3), 1225-1231. Falkinham, J. O. (2004). Environmental sources of Mycobacterium avium linked to routes of exposure. In B. J. Pedley S, Rees G, Dufour A, Cotruvo J (Ed.), Pathogenic Mycobacteria in Water (pp. 26-38). London: IWA Publishing. Galassi, L, Donato, R., Tortoli, E., Burrini, D. Santianni, D., & Dei, R. (2003). Nontuberculous mycobacteria in hospital water systems: application of HPLC for identification of environmental mycobacteria. Journal of Water and Health, 1(3), 133-139. Glover, N., Holtzman, A., Aronson, T., Froman, S., Berlin, O. G. W., Dominguez, P, . . . Yakrus, M. (1994). The isolation and identification of Mycobacterium avium complex (MAC) recovered from Los Angeles potable water, a possible source of infection in AIDS patients. International Journal of Environmental Health Research, 4, 63-72. Goslee, S, & Wolinsky, E. (1976). Water as a source of potentially pathogenic mycobacteria. American Review of Respiratory Disease, 113(3), 287-292. Hilborn, E. D., Covert, T. C., Yakrus, M. A., Harris, S. I., Donnelly, S. F., Rice, E. W., . . . Stelma, G. N., Jr. (2006). Persistence of nontuberculous mycobacteria in a drinking water system after addition of filtration treatment. Applied and Environmental Microbiology, 72(9), 5864-5869. Holleman, A. F, & Wiberg, E. (2001). Inorganic Chemistry: Academic Press: San Diego.

83 Iivanainen, E., Northrup, J., Arbeit, R. D., Ristola, M., Katila, M-L., Von Reyn, C. F. (1999). Isolation of mycobacteria from indoor swimming pools in Finland. APMIS, 107, 193-200. Kubalek, I., & Komenda, S. (1995). Seasonal variations in the occurrence of environmental mycobacteria in potable water. APMIS, 103(5), 327-330. Kubalek, I., & Mysak, J. (1996). The prevalence of environmental mycobacteria in drinking water supply systems in a demarcated region in Czech Republic, in the period 1984-1989. European Journal of Epidemiology, V12(5), 471-474. Le Dantec, C., Duguet, J-P., Montiel, A., Dumoutier, N., Dubrou, S., & Vincent, V. (2002a). Chlorine disinfection of atypical mycobacteria Isolated from a water distribution system. Applied and Environmental Microbiology, 68(3), 1025- 1032. Le Dantec, C., Duguet, J-P., Montiel, A., Dumoutier, N., Dubrou, S., & Vincent, V. (2002b). Occurrence of mycobacteria in water treatment lines and in water distribution systems. Applied and Environmental Microbiology, 68(11), 5318- 5325. Leoni, E., Legnani, P., Mucci, M. T., & Pirani, R. (1999). Prevalence of mycobacteria in a swimming pool environment. Journal of Applied Microbiology, 87(5), 683-688. Marras, T. K., Wallace, R. J., Jr, Koth, L. L., Stulbarg, M. S., Cowl, C. T., & Daley, C. L. (2005). Hypersensitivity Pneumonitis reaction to Mycobacterium avium in household water. Chest, 127(2), 664-671. Murray, P. et al. (1999). Manual of Clinical Microbiology (7th ed.). Washington DC: American Society for Microbiology. Neumann, M., Schulze-Robbecke, R., Hagenau, C., & Behringer, K. (1997). Comparison of Methods for Isolation of Mycobacteria from Water. Applied and Environmental Microbiology, 63(2), 547-552. Norton, C. D., LeChevallier, M. W, & Falkinham III, J. O. (2004). Survival of Mycobacterium avium in a model distribution system. Water Research, 38(6), 1457-1466.

84 Payment, P. (1999). Poor efficacy of residual chlorine disinfectant in drinking water to inactivate water borne pathogens in distribution systems. Canadian Journal of Microbiology, 45(8), 709. Pelletier, P. A, Du Moulin, G. C, & Stottmeier, K.D. (1988). Mycobacteria in public water supplies: comparative resistance to chlorine. Microbiological sciences, 5(5), 147-148. Pryor, M, Springthorpe, S., Riffard, S., Brooks, T., Huo, Y., Davis, G., Sattar, S.A. (2004). Investigation of opportunistic pathogens in municipal drinking water under different supply and treatment regimes. Water Science and Technology, 50(1), 83-90. Schulze-Robbecke, R., Weber, A., & Fischeder, R. (1991). Comparison of decontamination methods for the isolation of mycobacteria from drinking water samples. Journal of Microbiological Methods, 14(3), 177-183. Steed, K. A., & Falkinham, J. O., III. (2006). Effect of growth in biofilms on chlorine susceptibility of Mycobacterium avium and Mycobacterium intracellulare. Applied and Environmental Microbiology, 72(6), 4007-4011. Stinear, T., Ford, T., & Vincent, V. (2004). Analytical Methods for the detection of waterborne and environmental pathogenic mycobacteria. In B. J. Pedley S, Rees G, Dufour A, Cotruvo J (Ed.), Pathogenic Mycobacteria in Water: a Guide to Public Health Consequences, Monitoring and Management (pp. 55- 73). London: World Health Organisation, IWA Publishing. Taylor, R. H, Falkinham III, J. O, Norton, C. D, & LeChevallier, M. W. (2000). Chlorine, chloramine, chlorine dioxide, and ozone susceptibility of Mycobacterium avium. Applied and Environmental Microbiology, 66(4), 1702-1705. Thomson, R. M., Armstrong, J. G., & Looke, D. F. (2007). Gastroesophageal reflux disease, acid suppression, and Mycobacterium avium complex pulmonary disease. Chest, 131(4), 1166-1172. Tobin-D'Angelo, M. J., Blass, M. A., del Rio, C., Halvosa, J. S., Blumberg, H. M., & Horsburgh Jr., C. R. (2004). Hospital water as a source of Mycobacterium avium complex isolates in respiratory specimens. Journal of Infectious Diseases, 189, 98-104.

85 Torvinen, E., Suomalainen, S., Lehtola, M. J., Miettinen, I. T., Zacheus, O., Paulin, L., . . . Martikainen, P. J. (2004). Mycobacteria in water and loose deposits of drinking water distribution systems in Finland. Applied and Environmental Microbiology, 70(4), 1973-1981. Vaerewijck, M. J. M., Huys, G., Palomino, J. C., Swings, J., & Portaels, F. (2005). Mycobacteria in drinking water distribution systems: ecology and significance for human health. FEMS Microbiology Reviews, 29(5), 911-934. von Reyn, C. F., Marlow, J. N., Arbeit, R. D., Barber, T. W., & Falkinham, J. O. (1994). Persistent colonisation of potable water as a source of Mycobacterium avium infection in AIDS. The Lancet, 343(8906), 1137-1141. Wilton, S, & Cousins, D. (1992). Detection and identification of multiple mycobacterial pathogens by DNA amplification in a single tube. PCR Methods and Applications, 4, 269-273.

87 Chapter1 Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1:Objective 2 Chapter 4: Isolation of Chapter 6: NTM from Brisbane M. abscessus water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

88 CHAPTER 4 : FACTORS ASSOCIATED WITH THE ISOLATION OF NONTUBERCULOUS MYCOBACTERIA FROM A LARGE MUNICIPAL WATER DISTRIBUTION SYSTEM IN BRISBANE, AUSTRALIA.

The following chapter was presented as a poster paper at the American Thoracic Society ASM Denver 2011 and accepted for publication by BMC Microbiology on April 9th 2013.

Author contributions: RT (65%) designed the study, coordinated the collection of samples, participated in the processing of water samples, collated and analysed the data, and wrote the manuscript. RC (12.5%) coordinated, received and processed the water samples (including subculturing and sequencing), collated the results and reviewed the manuscript. CT (10%) processed water samples, performed sequencing and collated results. CC (2.5%) organisational and intellectual support FH (5%) intellectually contributed to the study design and methodology and the writing of the manuscript. MH (5%) intellectually contributed to the study design and methodology, liaised with Brisbane Water, and contributed to the writing of the manuscript.

89 ABSTRACT

Background: Nontuberculous mycobacteria (NTM) are normal inhabitants of a variety of environmental reservoirs including natural and municipal water. The aim of this study was to document the variety of species of NTM in potable water in Brisbane, QLD, with a specific interest in the main pathogens responsible for disease in this region and to explore factors associated with the isolation of NTM. Methods: One-litre water samples were collected from 189 routine collection sites in summer and 195 sites in winter. Samples were split, with half decontaminated with CPC 0.005%, then all concentrated by filtration and cultured on 7H11 plates and MGIT tubes (winter only). Results: Mycobacteria were grown from 40.21% sites in summer (76/189) and 82.05% sites in winter (160/195). The winter samples yielded the greatest number and variety of mycobacteria, as there was a high degree of subculture overgrowth and contamination in summer. Of those samples that did yield mycobacteria in summer, the variety of species differed from those isolated in winter. The inclusion of liquid media increased the yield for some species of NTM. Species that have been documented to cause disease in humans residing in Brisbane that were also found in water include M. gordonae, M. kansasii, M. abscessus, M. chelonae, M. fortuitum complex, M. intracellulare, M. avium complex, M. flavescens, M. interjectum, M. lentiflavum, M. mucogenicum, M. simiae, M. szulgai, M. terrae. M. kansasii was frequently isolated, but M. avium and M. intracellulare (the main pathogens responsible for disease is QLD) were isolated infrequently. Distance of sampling site from treatment plant and elevation above sea level in some cases was associated with isolation of NTM. Pathogenic NTM (defined as those known to cause disease in QLD) were more likely to be identified from sites with narrower diameter pipes, predominantly distribution sample points, and from sites with asbestos cement or modified PVC pipes. Conclusions: NTM responsible for human disease can be found in large urban water distribution systems in Australia. Based on our findings, additional point chlorination, maintenance of more constant pressure gradients in the system, and the utilisation of particular pipe materials should be considered.

90 BACKGROUND

The environmental or nontuberculous mycobacteria (NTM) other than TB are a group of human and animal pathogens that have significant impacts on the morbidity and mortality of humans and important economic impacts on agriculture (Falkinham, 2003). They are normal inhabitants of a variety of environmental reservoirs including natural and municipal water, soil, aerosols, and protozoans. The incidence of pulmonary disease due to environmental mycobacteria is increasing in many parts of the world including Queensland (Thomson, 2010). Clinically significant cases represent approximately one-third of all NTM pulmonary patient- isolates processed by laboratories in the state. Postulated reasons for this increase include increased awareness of mycobacteria as pulmonary pathogens, improvements in methods of detection and culture, and an ageing population (as this is often a disease of the elderly).

It has been shown that mycobacteria are resident in drinking water distribution systems (Martin-Casabona et al, 2004; Engel, Berwald, & Havelaar, 1980; Mankiewicz & Majdaniw, 1982; Carson et al., 1988; Covert, Rodgers, Reyes & Stelma, 1999; Le Dantec et al., 2002). They have also been found in hospital water distribution systems (du Moulin, Stottmeier, Pelletier, Tsang, & Hedley-White, 1988; Tobin-D’Angelo et al.,2004; Fox et al., 1992) and domestic tap water (Gangadharam, Awe & Jenkins, 1976; Petres et al, 1995; von Reyn, Marlow, Arbeit, Barber & Falkinham, 1994; Chang, Wang, Liao & Huang, 2002; Gosley & Wolinsky, 1976). In 2007, Brisbane Water managed the supply of potable water to the population of Brisbane. Details of the system supplying water to approximately 1 million residents are tabled in additional material 4.1. Water is treated by chloramination at the treatment plants and by point chlorination at points of entry into the system. There are 45 Reservoirs within the network.

91 There have been no published studies examining the presence of mycobacteria in water distribution systems in Australia, and routine sampling/monitoring is not mandated as part of public health practice.

The aim of this study was to determine if mycobacteria that currently cause disease in Queensland residents and those that are found as contaminants in human samples, are present in Brisbane drinking water and to explore system factors that may be associated with the presence or absence of these NTM.

METHODS

Water authorities routinely sample approximately 220 sites across Brisbane as part of water quality maintenance. These sites are a mixture of Trunk Main (TM) samples, Reservoir (R) samples, and Distribution point (D) samples. For this study an extra litre (sample) was collected from each site to allow filtration and culture for mycobacteria. Samples were collected in 1Litre sterile bottles and transported at 4°C to the QLD Mycobacterial Reference Laboratory.

Samples were collected over a 3-week period in both Winter (July-August 2007) and Summer (December 2007-January 2008). Each sample was halved, with 500ml treated with 0.005% Cetylpyridinium chloride (CPC) for 30 minutes. Filtration was performed through 0.45µm cellulose nitrate filters (Sartorius AG 37070 Goettingen, Germany). Filters were then rinsed with 2ml sterile distilled water (SDW) and macerated and then 0.1ml aliquots were then transferred in triplicate to Middlebrook 7H11 plates, which were sealed in gas permeable plastic bags for incubation at 32°C. For the winter samples 0.5 ml aliquots were also transferred to two Mycobacterial Growth Indicator Tubes (MGITs), one containing PANTA (polymixin, azlocillin, nalidixic acid, trimethoprim, amphotericin B) and incubated using the Bactec 960 system (Becton Dickinson, North Ryde, NSW). As a result, each sample collected in winter resulted in 10 processed cultures and each sample in summer resulted in six processed cultures.

92 Plates were inspected weekly and a representative selection of each morphological type of Ziehl Nielsen (ZN) positive colonies from each site sample were subcultured onto 7H11 plates. Multiplex PCR (Wilton & Cousins, 1992) was performed followed by 16S-rRNA sequencing of mycobacterial isolates and compared using RIDOM and GenBank database (Harmsen, Frosch & Albert, 2002;Benson, Karsch-Mizrachi, Lipman, Ostell & Sayers, 2008). Sequence homology of ≥97% was accepted. Those identified as M. abscessus/M. chelonae underwent hsp65 and rpoB sequencing for more definitive identification. Because of the widely varying growth rates of different NTM species, and the presence of multiple different colony types and species in samples from each site, determination of concentrations of individual NTM species in CFU/ml was not determined. The number of different species/strains from each site was determined and expressed per site (1L sample) for each season.

Information regarding the different sampling sites was obtained and included mains age, pipe material, distance from nearest reservoir, and elevation above sea level. Distances between main treatment plants and sampling sites were calculated from latitude and longitude values provided for each site.

Statistics: Statistical analysis was performed using IBM SPSS v 20. Sampling site variables were analysed against individual site culture results using a one-way ANOVA with post hoc Bonferroni correction. Culture method variables were analysed against results of individual replicates. Where variables were not normally distributed, the non parametric Kruskal Wallis test was used. Elevation above sea level was transformed (square root) and analysed using one way ANOVA. Categorical variables were analysed using Chi Square rxc tables. Values are represented as mean +/- SD or median (where non normal distribution); significance level p=0.05.

93 RESULTS

Sampling Site Analysis Of a total of 217 sites, 1L-samples from 189 sites in summer and 195 sites in winter were received. Because of the drought conditions experienced in QLD at the time of the study and subsequent water restrictions, 17 of the sampling sites were dry during summer and not able to be sampled. An additional 11 sites were therefore recruited that had not been part of the sampling routine during the preceding winter. Overall mycobacteria were identified in 61.5% samples. Mycobacteria were grown from 40.2% sites in summer (76/189) and 82.1% sites in winter (160/195). The lower yield in summer was due to higher rates of contamination, including that of subculture plates. Of the colonies subcultured and sequenced, 236 colonies were subsequently identified as NTM. Winter yields were greater (Mean 2.59±1.62 colonies per site sample; range 1-10) compared with summer (1.70±0.84; 1-4).

For those sites that were supplied water from Mt Crosby (152 sites in summer, 158 sites in winter), the distance of the sampling site from the treatment plant was associated with culture result particularly in summer; the mean distance from plant to site was 81.75 ± 6.99 km for negative sites, 82.50 ± 6.17km for contaminated/overgrown sites and 85.40 ± 6.46 km for positive sites (p=0.015). In winter the distances were similar (negative 84.95± 6.77km; contaminated/overgrown 82.49 ± 6.77 km; positive 83.34± 6.65 km; p=0.581)

For those 17 sites receiving water from the Pine treatment plant or from both treatment plants (19 summer, 17 winter), the distance of sampling site from the treatment plant didn’t correlate with culture result.

Type of sample Samples came from distribution points (D), reservoirs (R) or trunk mains (TM). By their nature, the samples differed significantly according to differences in pipe diameter, and pipe material. The characteristics of the different type of samples are shown in Additional material 4.2. The majority of Trunk Main samples (also larger

94 diameter) were of Mild Steel Cement Lined (88%), the remainder were Cast iron spun lined (6.6%), cast iron cement lined (3.3%) or Mild steel unlined black piping (2.1%). Reservoir samples similarly came mostly from Mild Steel Cement lined pipes (75.5%), with the remainder from Cast Iron spun lined (13%), Cast Iron Cement Lined (4.3%), Asbestos Cement (2.2 %) or Ductile Iron Cement Lined (2.2%) In contrast the majority of distribution samples came from Asbestos cement or Cast Iron Spun lined pipes.

There was no statistically significant difference between culture results overall and sample type (ie Distribution/Reservoir/Trunk Main), mains age, pipe material or distance of sampling site from nearest reservoir. (Table 4.1) This held true when winter and summer samples were analysed separately, though there was a trend towards more positive sites that were distribution samples (p=0.074) with narrower diameter pipes in winter (p=0.114). Whilst there were differences in the culture results from different pipe materials the numbers in some categories were too small to be statistically meaningful.

Trunk Main samples grew M. kansasii, M. gordonae, M. mucogenicum, M. abscessus, M. chelonae, M. lentiflavum, M. simiae, M. szulgai, M. fortuitum complex, and hence these species are also potentially present in more distal sites. Some species relevant to humans, namely M. intracellulare, and M. flavescens were grown from reservoir samples though may not have been detected more distally in distribution point samples because of the limitations of culture techniques (overgrowth, contamination etc). (Additional material 4.3: Species of NTM isolated from different sample types)

All variables were examined between different species of NTM. Pathogenic NTM (defined as those that had been found in human samples in QLD and known to cause disease) were more likely to be identified from sites with narrower diameter pipes, predominantly distribution sample points, and from sites with asbestos cement or modified PVC pipes. No other variables were found to be significant. (Table 4.2)

Table 4.1. Summary of NTM positive and negative sampling site variables. NTM Negative NTM Positive Significance (p value) Sampling Site Factor (Mean ± SD) Site elevation 44.75 ±40.12 43.78 ±39.99 0.977 (meters above sea level)* Summer 44.94±41.92 44.88±38.86 0.680 Winter 43.51±26.54 43.26±40.63 0.751

Pipe Diameter(cm) 438.01±459.91 435.21±461.92 0.954 Summer 403.23±417.56 489.15±513.25 0.211 Winter 553.94±571.58 409.59±434.81 0.103

Mains Age (years) 46.56±19.53 48.94±19.15 0.246 Summer 46.15±19.83 50.97±17.74 0.091 Winter 47.91±18.71 47.97±19.77 0.987

Pipe material Asbestos cement 28 (30.8% 63 (69.2) Cement lined† 77 (41.8) 107 (58.2) PVC 6 (42.9) 8 (57.1) Cast iron spun 30 (35.7) 54 (64.3) 0.166 lined Other‡ 7 (63.3) 4 (36.4)

Sample type n (%) Distribution 86 (37.1) 146 (62.9) Reservoir 36 (39.1) 56 (60.9) 0.668 Trunk Main 26 (43.3) 34 (56.7)

Surface water source n (%) Mt Crosby 120 (38.6) 191 (61.4) Pine 14 (37.8) 23 (62.2) 0.995 Mixed 14 (38.9) 22 (61.1)

* Elevation non normally distributed, square root transformation to analyse † Cast iron, ductile iron or mild steel cement lined ‡ Steel unlined/ polyethylene/unknown

Table 4.2. Presence of pathogenic NTM against different variables

Variable Pathogenic NTM Non pathogenic NTM P value Sample type Distribution 203 129 Reservoir 56 75 0.001 Treatment Plant 33 41

Surface water source Crosby 231 195 Mixed 25 25 0.695 Pine 36 26

Distance to nearest reservoir 4.46 (5.01) 4.85 (6.18) 0.423 (km) Mean (±SD)

Age of water mains (yrs) 49.45 (19.50) 49.03 (19.11) 0.646 Mean (±SD)

Pipe diameter (mm) 360.82 (414.90) 509.74 (503.47) <0.0001 Mean (±SD)

Site elevation 43.26 (45.50) 44.97 (37.17) 0.638

Pipe material Asbestos cement 91 (62.3) 55 (37.7) Cast iron cement lined 26 (56.5) 20 (43.5) Cast iron spun lined 68 (59.1) 47 (40.9) Ductile Iron cement lined 14 (50) 14 (50) Mild steel cement lined 75 (44.4) 95 (55.9) Mild steel unlined black 3 (42.9) 4 (57.1) 0.046 Modified PVC 5 (88.3) 1 (16.7) Polyethylene 0 1 (100) Unplasticized PVC 8 (61.5) 5 (38.5)

Location The reservoir zones that cluster around the Central Brisbane District (CBD) appeared to contribute more positive sites than those in more peripheral zones (ie had more positive sites relative to the proportion of sites sampled) however this did not meet statistical significance. Of the sites within an approximate 5-kilometre radius of the CBD, 64.8% grew NTM, compared to 59.9% of sites outside this area (p=0.431 Fisher’s exact test).

97 Methodological Factors associated with positive culture results To assess the effect of decontamination and the relative contribution of the different media to positive results and species variety, the individual results of each culture taken per site was analysed. The results were analysed for summer and winter separately as contamination issues in summer would have confounded the results.

In winter, there were 10 cultures per site, and in summer 6 cultures per site. Hence, there were 1176 plates and 784 MGITs processed in winter (with PANTA added to half of these) and 1140 7H11 plates were processed in summer. For funding reasons, MGITs were not used in summer. Overall 65.3% of cultures were positive for mycobacterial growth, though there were statistically significant differences between summer and winter (p<0.0001).

Winter Of 1960 cultures processed during winter, 528 (26.9%) failed to grow any colonies and 188 (9.6%) were overgrown to the extent that mycobacteria could not be detected, if they were present; 847 (43.2%) of cultures had positive growth and 397 (20.3%) were positive but with contaminants (presumed fungal on the basis of plate morphology, but not formally identified). The winter cultures yielded the greatest number and variety of mycobacteria (Table 4.3). This held true even if MGIT samples were excluded, though there were some specific contributions of the liquid media discussed below.

Summer Of 1140 cultures processed in summer, only 1.6% were negative, 30.1% were overgrown, 50.2% were positive and 18.2% were positive with contaminants. Unfortunately, of the positive plates that were subcultured, a large percentage became contaminated and the mycobacterial yield was disappointing.

There was a wide variety of species identified using 16s rRNA sequencing. (Table 4.3) Those isolates identified as M. abscessus/M. chelonae underwent subsequent

98 hsp65 and rpoB gene fragment sequencing for further differentiation. Exhaustive speciation was not performed as only potentially pathogenic mycobacteria were of interest. Overall there were more species identified in winter. All of the M. intracellulare, MAC, M. lentiflavum, M. simiae, M. chelonae isolates were found in winter, along with the majority of other pathogenic species such as M. abscessus, M. kansasii, and M. mucogenicum. M. poriforae and M. fluoranthenivorans were predominantly found in summer samples and M. fortuitum and M. mucogenicum were found equally in winter and summer.

Table 4.3. Species of NTM identified in water samples collected in winter and summer.

NTM species SUMMER WINTER NTM species SUMMER WINTER M. abscessus 2 11 M. interjectum 1 7 M. abscessus/chelonae 1 M. intracellulare 2 M. angelicum/szulgai 1 M. kansasii 5 133 M. arupense 4 5 M. lentiflavum 19 M. austroafricanum 1 M. mageritense 1 4 M. bolletii/M. 1 M. moriokaense 1 massiliense M. chelonae 2 M. mucogenicum 31 42 M. cookii 1 1 M. poriforae 18 6 M. cosmeticum 1 1 M. rhodesiae 1 M. diernhoferi 1 1 M. sengalense 2 M. farcinogenes 1 2 M. simiae 2 M. flavescens 2 1 M. species NFI 5 4 M. fluoranthenivorans 11 4 M. terrae 2 M. fortuitum complex 13 14 M. tilburgii 2 1 M. gadium 1 4 M. triplex 1 M. gilvum 1 M. wolinsky 1 M. gordonae 24 120 MAC 3 TOTAL 130 408

Decontamination Decontamination made a statistically significant difference to culture results for all media used (p<0.0001 for all). Overall decontamination did decrease the overgrowth and contamination of positive plates, and increased the yield from positive plates. (Table 4.4)

Table 4.4. Effect of Decontamination on Culture results for different media

Media Decontam Culture result Significance

ination n (%)

Negative Positive Positive + Overgrown

contaminants

MGIT No 43 (21.9) 35 (17.9) 115 (58.7) 3 (1.5) p<0.0001*

Winter Yes 99 (50.8) 49 (25.1) 46 (23.6) 1 (0.5)

MGIT + No 4 (0.7) 48 (24.7) 82 (42.3) 3 (1.5) p<0.0001*

PANTA Yes 14 (2.5) 64 (32.8) 19 (9.7) 1 (0.5)

Winter

7H11 No 4 (0.7) 234 (41.1) 145 (25.4) 187 (32.8) p<0.0001

Summer Yes 14 (2.5) 338 (59.3) 62 (10.9) 156 (27.4)

7H11 No 46 (7.9) 313 (53.5) 115 (19.7) 111 (19)

Winter Yes 161 (27.5) 335 (57.3) 20 (3.4) 69 (11.8)

The use of the MGIT tubes in winter increased the yield for certain species of mycobacteria. There were 13 isolates of M. abscessus – 10 of these were only grown using liquid media (7 MGIT+PANTA, 3MGIT). However the three isolates that grew on solid media were from sites that were not picked up by liquid media. M. lentiflavum was only identified in winter samples. Eight sites grew M. lentiflavum from MGIT only (1 MGIT, 7 MGIT+PANTA). It was grown on solid media from 6 sites – 4 of these sites also had positive MGITs (2) and MGIT + PANTA (3). For the majority of sites from which M. gordonae was identified, it was detected using M7H11. However, from ten sites it was only grown from MGIT tubes (4 MGIT+PANTA). Twenty-four sites grew M. kansasii only from MGIT (11 MGIT+PANTA only). Several of the pathogenic species shown in Table 4.5, were not grown in liquid media and only grew on solid media (eg M. fortuitum, M. intracellulare, M. avium). 100

Table 4.5. Number of NTM isolates cultured using different media.

Winter Media Summer Media

Species MGIT MGIT+PANT 7H11 7H11

M. abscessus 3 7 3

M. abscessus/chelonae 1 2

M. bolletii/M. massiliense 1

M. fortuitum complex 14 13

M. gordonae 14 12 94 24

M. kansasii 34 45 52 5

M. mucogenicum 6 4 32 31

M. terrae 2

M. intracellulare 2 1

M. lentiflavum 3 10 6

MAC 3

M. flavescens 1 2

M. interjectum 1 6 1

M. simiae 2

M. szulgai 1 9 MGIT: Mycobacterial Growth Indicator Tube; PANTA: polymixin, azlocillin, naladixic acid, trimethroprim, amphotericin B

DISCUSSION

This is the first study to document the presence of potentially pathogenic NTM in an Australian drinking water distribution system (DS). The incidence of disease is increasing (Thomson, 2010) and water as a potential source of infection needs to be addressed. NTM have been reported in potable water studies from other countries. Mycobacteria were isolated from 38% (16/42) of drinking water DS in the USA, (Covert et al, 1999) from 21.3% (42/197) in Greece (Tzintzou, Vantarakis, Pagonopoulou, Athanassuadou, & Papapetropoulou, 2000) M and 72% (104/144) in Paris (Le Dantec). Mycobacteria were found in Finnish DS samples – from 35%, up to

101 80% at sites more distal in the network. Torvinen et al., 2004) Mycobacterial numbers reported are similar in DS that used groundwater compared to surface water. (Covert et al, 1999) In our study we identified NTM in samples from 82.1% of sites tested in winter and 40.2% sites in summer.

Kubalek & Komenda (1995) demonstrated seasonal variations in the occurrence of environmental mycobacteria in potable water in the Czech Republic between 1984 and 1989. Forty two percent of samples were positive for mycobacteria, with significantly more positive in Spring than Autumn. We have similarly shown differences in seasonal isolation of NTM, and differences in the species isolated between seasons.

Factors associated with the isolation of pathogenic NTM included distance of sampling points from the main treatment plant, diameter of the pipes at point of sampling, and certain pipe materials. Pelletier (1988) found that free chlorine concentrations gradually decrease as water travels down the distribution system. From previous studies (Torvinen et al., 2004; Falkinham, Norton & Chevallier, 2001) one would expect that mycobacterial growth would be greater the further from disinfection. Du Moulin (1985) found that communities in Massachusetts were more likely to have patients with MAC isolates if they lived further away from water treatment plants, and if they lived in more densely populated areas (du Moulin, Sherman, Hoaglin & Stottmeier, 1985). This can be explained by more complex water distribution systems in urban areas, with increased numbers of smaller diameter pipes, coupled with greater transit time of water in the system allowing for degradation of disinfection products.

By their hydrophobic nature, mycobacteria have the ability to form biofilms in pipes of distribution networks, contributing to their proliferation and survival (Falkinham, 2002) Smaller pipes have a greater surface area of biofilms to contribute NTM into the water, and are usually present at more distal points of the system where

102 complex bends occur. During pressure transients at point of turbulence such as the bends in pipes, release of biofilms occurs (sloughing). Falkinham et al (2001) demonstrated significantly higher mycobacterial numbers in distribution samples (average 25000 fold) than those collected immediately downstream from treatment plants, indicating that mycobacteria actively grow within the distribution system. Whilst we didn’t find that smaller diameter pipes were more likely to yield NTM, pathogenic species more certainly more likely to come from sites with smaller diameter pipes.

Some pipe materials have been shown to contribute to biofilm formation particularly Iron pipes (compared to chlorinated PVC) (Norton & LeChevallier, 2001). However the survival of mycobacteria in DS is dependent upon a complex interaction between pipe surface, nutrient levels and disinfectants. In one study, (Norton, LeChevallier & Flakinham, 2004) when biofilms were grown on non- corroded surfaces (Cu or PVC) free chlorine was more effective for controlling HPC and M. avium, but monochloramine controlled bacterial levels better on corroded iron pipe surfaces. M. avium biofilm levels were higher on iron and galvanized pipe surfaces than on Cu or cPVC surfaces. In this study we were unable to assess the relative contribution of disinfectant concentrations, and nutrient levels, however there did seem to be some pipe surfaces (such as asbestos cement or modified PVC) associated with a greater yield of pathogenic mycobacteria at point of sampling. These results were consistent for both summer and winter, when chlorine concentrations may have been different (due to heat inactivation).

There was a wide variety of species isolated from water, many of which have been documented to cause disease in QLD patients (Thomson, 2010). M. intracellulare is the main pathogen associated with pulmonary disease in many parts of the world (including Australia and the United States) (Hoefsloot et al, 2013) In our study, the isolation of M. intracellulare from water distribution samples was disappointing and similar to previous investigators. This has been attributed to the difficulties associated with culturing this organism from environmental samples as high

103 concentrations have been found in biofilm samples from water meters or pipes. (Falkinham et al., 2001) However as disease associated serotypes of M. intracellulare have been found in soil and house dust, (Dawson, 1971; Reznikov, Leggo & Dawson, 1971) and rainwater tanks, (Tuffley & Holbeck, 1980) the environmental niche for M. intracellulare may not necessarily be potable water, rather soil and dust contaminates water supplies through breaches in distribution systems (e.g. cracked underground pipes).

It has long been recognised that M. kansasii can be found in potable water. (Engel et al, 1980; Wright, Collins & Yates, 1985; McSwiggan & Collins, 1974) Disease due to this organism is not common in Queensland (approximately 20 cases of significant pulmonary disease per year), yet this species was readily isolated from potable water.

M. abscessus has become an increasing clinical problem in the last 5-10 years, and its presence in potable water has not previously been emphasized. In the majority of published studies looking for NTM in water, no M. abscessus was documented. There have been taxonomical changes, which led to M. abscessus being recognised as independent from M. chelonae, so older studies reporting M. chelonae may have included M. abscessus. But in studies done since 2000 M. abscessus has been rarely reported. (Falkinham et al, 2001; Septbember, Brozel & Venter, 2004) The inclusion of liquid media in our study may have increased the yield for M. abscessus.

The universal problem with studies of environmental samples has been the difficulty in culturing these slow growing organisms in the presence of fungal and other bacterial contaminants. (Falkinham, 2002; Pedley, Ress, Dufour & Cotruvo, 2004). Direct detection using PCR probes or a metagenomic approach is appealing however positive results may indicate the presence of mycobacterial DNA, but not necessarily viable organisms. This is especially relevant in the presence of disinfection, such as with potable water.

104 A major study examining showerheads in the USA using such an approach, (Feazel et al., 2009) did find M. avium and M. gordonae in multiple samples. M. abscessus was not reported.

It is predicted that the interaction between humans and mycobacteria will increase, resulting in more cases of disease. (Pedley et al, 2004) Factors driving this increase include disinfection of drinking water with chlorine, selecting mycobacteria by reducing competition and the increasing percentage of our population with predisposing conditions, especially age and immunosuppression. Public and environmental health efforts must therefore focus on actions that will specifically remove mycobacteria from habitats where susceptible humans are exposed. Based on our findings, additional point chlorination, maintenance of more constant pressure gradients in the system, and the utilisation of particular pipe materials should be considered.

Funding: The study was funded by grants from the The Prince Charles Hospital Foundation and the Gallipoli Medical Research Foundation of Greenslopes Private Hospital.

Acknowledgements: The authors would like to acknowledge the contribution from Brisbane Water in providing water samples. Urban Utilites provided a map of the distribution network and data on the individual site points. We are grateful also to the QLD Mycobacterial Reference Laboratory for assistance and accommodation of this work in the MRL.

105 REFERENCES

Benson D, Karsch-Mizrachi I, Lipman D, Ostell J, & Sayers E. (2008) Genbank. Nucleic Acids Research, 37 (database issue):D26-31. Carson, L. A., Bland, L. A., Cusick, L. B., Favero, M. S., Bolan, G. A., Reingold, A. L., & Good, R. C. (1988) Prevalence of nontuberculous mycobacteria in water supplies of hemodialysis centers. Applied and Environmental Microbiology, 54:3122-3125. Chang, C., Wang, L., Liao, C., & Huang, S.(2002). Identification of nontuberculous mycobacteria existing in tap water by PCR-Restriction fragment length polymorphism. Applied and Environmental Microbiology, 68:3159-3161. Covert, T. C., Rodgers, M. R., Reyes, A. L., & Stelma, G. N., Jr. (1999). Occurrence of nontuberculous mycobacteria in environmental samples. Applied and Environmental Microbiology, 65:2492-2496. Dawson D: Potential pathogens among strains of mycobacteria isolated from house- dusts. (1971). Medical Journal of Australia, 1:679-681. du Moulin, G. C., Sherman, I. H., Hoaglin, D. C., & Stottmeier, K. D. (1985). Mycobacterium avium complex, an emerging pathogen in Massachusetts. Journal of Clinical Microbiology, 22:9-12. du Moulin, G., Stottmeier, K., Pelletier, P., Tsang, A., & Hedley-Whyte, J. (1988) Concentration of Mycobacterium avium by hospital hot water systems. Journal of the Americal Medical Association, 260:1599-1601. Engel, H. W. B., Berwald, L. G., & Havelaar, A. H. (1980). The occurrence of Mycobacterium kansasii in Tapwater. Tubercle, 61:21-26. Falkinham III, J. Nontuberculous mycobacteria in the environment. (2002).Clinics in Chest Medicine, 23: 529-551. Falkinham III, J., Norton, C., & Le Chavallier, M. (2001). Factors influencing numbers of Mycobacterium avium, Mycobacterium intracellulare and other mycobacteria in drinking water distribution systems. Applied and Environmental Microbiology, 67:1225-1231.

106 Feazel, L., Baumgartner, L., Peterson, K., Frank, D., Harris, J., & Pace, N. (2009) Opportunistic pathogens enriched in showerhead biofilms. Proceedings of the National Academy of Science, 106:16393-16399. Fox, C., Smith, B., Brogan, O., Rayner, A., Harris, G., & Watt, B. (1992) Non- tuberculous mycobacteria in a hospital's piped water supply. Journal of Hospital Infection, 21:152-154. Gangadharam, P. L. J., Awe, R. J., & Jenkins, D. E. (1976) Mycobacterial contamination through tap water. American Review of Respiratory Disease, 113:894. Goslee, S., & Wolinsky, E. (1976) Water as a source of potentially pathogenic mycobacteria. American Review of Respiratory Disease, 113:287-292. Harmsen, D., Rothganger, J., Frosch, M., & Albert, J. (2002). RIDOM: Ribosomal differentiation of medical microorganisms database. Nucleic Acids Research, 30:416-417. Hoefsloot, W., van Ingen, J., Andrejak, C., Ängeby, K., Bauriaud, R., Bemer, P., ... & D. Wagner for NTM-NET. (2013). A snapshot of the geographic diversity of nontuberculous mycobacteria isolated from pulmonary samples: An NTM- net collaborative study. European Respiratory Journal (In Press) Huang, W-C., Chiou, C-S., Chen, J-H., & Shen, G-H. (2010). Molecular epidemiology of Mycobacterium abscessus in a subtropcal chronic ventilatory setting. Journal of Medical Microbiology, 59:1203-1211. Kubalek, I., & Komenda, S. Seasonal variations in the occurrence of environmental mycobacteria in potable water. (1995). Acta Pathologica Microbiologica et immunologica Scandinavica, 103:327-330. Le Dantec, C., Duguet, J-P., Montiel, A., Dumoutier, N., Dubrou, S., & Vincent, V. (2002) Occurrence of mycobacteria in water treatment lines and in water distribution systems. Applied and Environmental Microbiology, 68:5318- 5325. Mankiewicz, E. M., & Majdaniw, O. (1982). Atypical mycobacteria in tapwater. Canadian Journal of Public Health, 73:358-360 McSwiggan, D. A., & Collins, C. H. (1974) The isolation of M. kansasii and M. xenopi from water systems. Tubercle, 55:291-297.

107 Norton, C. D., & LeChevallier, M. W. (2000) A pilot study of bacteriological population changes through potable water treatment and distribution. Applied and Environmental Microbiology, 66:268-276. Norton, C. D., LeChevallier, M. W., & Falkinham III, J. O. Survival of Mycobacterium avium in a model distribution system. Water Research, 38:1457-1466. Pedley, S. Bartram, J., Rees, G., Dufour, A., & Cotruvo, J. (2004) Pathogenic Mycobacteria in Water. London: IWA Publishing. Pelletier, P., Du Moulin, G., & Stottmeier, K. D. (1988). Mycobacteria in public water supplies: comparative resistance to chlorine. Microbiological Sciences, 5:147-148. Reznikov, M., Leggo, J. H., & Dawson, D. J. (1971). Investigation by seroagglutination of strains of the Mycobacterium intracellulare-M. scrofulaceum group from house dusts and sputum in Southeastern Queensland. American Review of Respiratory Disease, 104:951-953.

108

109 Chapter1 Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1:Objective 2 Chapter 4: Isolation Chapter 6: of NTM from M. abscessus Brisbane water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

110 CHAPTER 5: MYCOBACTERIUM LENTIFLAVUM IN BRISBANE WATER

This chapter has been presented in poster form at the European Respiratory Society meeting in Barcelona (September 2011) and published as a paper as follows: Marshall, H., Carter, R., Torbey, M., Minion, S., Tolson, C., Sidjabat, H., Huygens, F., Hargreaves, M., Thomson, R. Mycobacterium lentiflavum in drinking water supplies, Australia (2011) Emerging Infectious Diseases 2011, 17, 395-402

Author Contributions HM (35%) collected information on patients, collated data and wrote the clinical aspects of the manuscript MT, SM, CT (2.5% each) assisted with processing of water samples RC (7.5%) regrew clinical and water isolates, assisted with rep-PCR experiments, contributed to writing of the manuscript HS (5%) performed Diversilab rep-PCR experiments, constructed dendrogram, contributed to writing of manuscript MH (2.5%) intellectual design and support, reviewed the manuscript FH (2.5%) intellectual design and support, reviewed the manuscript RT (40%) designed the study, sourced patients for review, contributed clinical data, contributed water isolates and analysed rep-PCR results, wrote non-clinical aspects of the paper, and revised the clinical aspects.

ABSTRACT

M. lentiflavum, a slow-growing non-tuberculous mycobacterium, is a rare cause of human disease. It has been isolated from environmental samples worldwide. We assessed the clinical significance of M. lentiflavum isolates reported to the Queensland Tuberculosis Control Centre between 2001-2008. We explored the genotypic similarity and geographical relationship between human and potable water isolates in the Brisbane metropolitan area. Forty-seven isolates from thirty-six patients were reported; four had clinically significant disease. Thirteen of 206

111 drinking water sites grew M. lentiflavum. These sites overlapped geographically with the patients with clinically significant isolates. Automated rep-PCR genotyping revealed a dominant environmental clone closely related to clinical strains. This suggests potable water as a possible source of M. lentiflavum infection in humans.

BACKGROUND

M. lentiflavum is a slow-growing non-tuberculous mycobacterium (NTM) first identified in 1996 (Springer et al., 1996). M. lentiflavum was found to be slow growing between temperatures of 22-37oC, had yellow pigmentation, negative tests for Tween 80 hydrolysis, nicotinic acid, nitrate reductase and urease, distinct fatty and mycolic acid patterns and unique 16S rRNA and 65 kDa heat shock protein gene sequences. It shared phenotypic features with M. avium, but was more closely related to M. simiae and M. genavense. Because of similarities to M. avium complex (MAC), differentiation can be difficult without molecular identification, hence possible misclassification in the past (Griffith et al., 2007).

As with other NTM, M. lentiflavum has been isolated from soil and water samples around the world, however, links between environmental sources and human disease have not yet been demonstrated. In Queensland (population 4.28 million) NTM disease remains notifiable. A central reference laboratory performs speciation of all positive isolates. In 2008, approximately 900 isolates of NTM were reported.

Strain variation within mycobacterial species is well known. Whilst epidemiologic studies provide useful information, molecular strain typing can be invaluable, especially if a single clone can be linked to an outbreak source. Pulsed field gel electrophoresis (PFGE) has been considered the gold standard for mycobacterial strain typing but is time and labor-intensive and requires expensive dedicated equipment. Also, DNA degradation can occur during electrophoresis generating uninterpretable banding patterns (Zhang, et al., 2004). Repetitive-sequence-based PCR has been used to differentiate mycobacterial strains associated with disease outbreaks in mesotherapy clinics (M. abscessus and M. chelonae) (Sampaio, Viana-

112

Niero, de Freitas, Hofling-Lima & Leao, 2006) and in post-surgical patients (M. fortuitum) (Sampaio et al., 2006b). An automated rep-PCR system (Diversilab®) showed high concordance with PFGE (Cangelosi et al, 2004) in identifying mycobacterial strain clusters and had significant time advantages over PFGE.

AIMS

1) To describe the clinical significance and outcomes of M. lentiflavum infection in Queensland. 2) To explore the genotypic and geographical relationship between patient isolates and potable water isolates in the Brisbane area.

METHODS

We reviewed the records of all patients with M. lentiflavum isolates between July 2001 and November 2008. Attending physicians were contacted in order to establish clinical significance according to American Thoracic Society (ATS)/ Infectious Diseases Society of America (IDSA) criteria (Griffith et al, 2007), (Table 5.1). In 2007-8 potable water was collected from 206 sites in the Brisbane drinking water system.

Laboratory identification

Human samples were digested and decontaminated using 4% NaOH, neutralised with phosphoric acid and centrifuged to concentrate the acid-fast bacilli (AFB). Smears were prepared from the sediment and stained by the Ziehl-Niehlsen (ZN) method. One Löwenstein-Jensen slope (± pyruvate) and 7ml Mycobacterial Growth Indicator Tube (MGIT) were inoculated and incubated at 35oC until growth was detected. ZN staining of colonies confirmed AFB. Multiplex PCR (Wilton & Cousins, 1992) was performed to discriminate between M. tuberculosis, M. avium, M. intracellulare, M. abscessus and other Mycobacterium spp. Other Mycobacterium spp were further speciated using Hain Life Sciences GenoType® Mycobacterium AS (additional species) kit (2004-7 only) and/or 16S rRNA sequencing in conjunction with phenotypic characteristics.

113

Table 5.1

American Thoracic Society/ Infectious Diseases Society of America Diagnostic Criteria for Nontuberculous Mycobacterial Lung Disease (Griffith et al., 2007) Clinical • Pulmonary symptoms • Nodular or cavitary opacities on chest radiograph or • Multifocal bronchiectasis with multiple small nodules on high- resolution computerised tomography and • Appropriate exclusion of other diagnoses

Microbiological • Positive culture results from at least two separate expectorated sputum samples or • Positive culture results from at least one bronchial wash or lavage or o Biopsy# showing granulomatous inflammation or acid fast bacilli and positive culture or • Biopsy# showing granulomatous inflammation or acid fast bacilli and one or more culture-positive sputum or bronchial washings • Risk-benefit of therapy should be considered for each patient before institution of therapy • Expert consultation should be obtained when NTM are recovered that are either infrequently encountered or that usually represent environmental contamination • Patients suspected of having NTM lung disease but who do not meet the diagnostic criteria should be followed until the diagnosis is firmly established or excluded

# Transbronchial or other lung biopsy

114

Water sampling

Water was collected from routine sampling sites across Brisbane and processed according to described methods (Thomson, Carter, Gilpin, Coulter, & Hargreaves, 2008). Each 1000ml sample was transported at 4 oC and processed within 24 hours. Half of each sample was decontaminated using 0.005% cetylpiridinium chloride, and each 500ml filtered separately using 45µm cellulose nitrate filters (Sartorius AG, Gottingen, Germany). The filters were rinsed and macerated in 3 ml SDW. Aliquots (0.1 ml) were transferred in triplicate to M7H11 plates, sealed in gas-permeable plastic bags and incubated at 32°C. Aliquots (0.5 ml) were transferred to two MGIT tubes, one of which contained PANTA (polymyxin, azlocillin, nalidixic acid, trimethoprim, amphotericin B). ZN staining of colonies confirmed AFB and these were subcultured on M7H11plates. Multiplex PCR was performed (Wilton & Cousins, 1992) followed by 16S-rRNA sequencing of mycobacterial isolates and compared using RIDOM and GenBank databases (Harmsen, Rothganger, Frosch & Albert, 2002; Benson, Karsch-Mizrachi, Lipman, Ostell, & Sayers, 2009).

Automated rep-PCR strain typing

The similarity of 16 clinical and 7 water isolates was determined using a rep-PCR based method (Diversilab® system, bioMerieux, Melbourne). DNA was extracted from clinical and water isolates using the Ultraclean Microbial DNA Isolation Kit (Mo Bio Laboratories, CA, USA). PCR mixture was prepared using AmpliTaq polymerase and PCR buffer (Applied Biosystems, New Jersey, USA) and Mycobacterium Diversilab®- primer mix according to the manufacturer’s instructions. Separation and detection of rep-PCR products was performed by microfluidic chips of the Diversilab® System. Fingerprints were analysed with Diversilab® software v.3.4.38 using the Pearson correlation co-efficient and unweighted pair group method with arithmetic means to compare isolates and determine clonal relationships.

115

RESULTS

Clinical isolates

Figure 5.1. Bar chart of number of patients with isolates of M. lentiflavum per year.

Forty-seven isolates of M. lentiflavum were reported from 36 patients (Figure 5.1). The significance and source of the isolates is summarised in Table 5.2. Full clinical information was available for 32 patients (89%). Four patients (eight isolates) had clinically significant disease. Seven patients were taking treatment or under surveillance for MAC (one to two isolates each); none had a change of treatment in light of the new isolate and thus were considered non-significant. Twenty-one other patients (18 adults, three children) had clinically non-significant isolates. Four patients had probable non-significant isolates but sufficient clinical information was lacking. No cases demonstrated positive AFB smears by ZN staining. Of the 32 patients with uncertain or non-significant disease, 26 had a single positive specimen; two had two positive specimens from the same time period; three had two positive specimens separated by three months; one had two positive specimens separated by 11 months. Two isolates (Case 1 and Case 2 below) had antibiotic susceptibility tests performed. Both were sensitive to clarithromycin 4.0 µg/ml, and resistant to isoniazid 0.4 µg/ml, ethambutol 5.0 µg/ml and streptomycin 1.0 µg/ml. The Case 1 isolate was sensitive to ofloxacin 2.0 µg/ml; the Case 2 isolate was resistant to rifampicin 1.0 µg/ml.

116

Table 5.2. Summary of patients with isolates of M. lentiflavum in Queensland 2001- 2008

Source of isolate (number) Age, years Wound Gender median bronchial swab/ n m/f (range) washings Sputum aspirate Other Adult cases:

Significant 3 0/3 49 (42-85) 2 0 0 1 (blood) Non-significant 18 12/6 67 (22-88) 9 4 4 1 (blood) Probable non- 1 significant 4 3/1 74 (59-81) 1 2 0 (ascites) Non-significant with 7 4/3 66 (49-75) 0 7 0 0 MAC

Paediatric cases:

Significant 1 0/1 1.6 0 0 1 0 12 (1.6 - Non-significant 3 1/2 17) 0 2 1 0

Environmental isolates Mycobacteria were grown from 70% of water sites. The predominant isolates were M. gordonae and M. kansasii. M. lentiflavum was isolated from 13 sites (6.3%), two of which were reservoirs, one a treatment plant and the remainder were points in the distribution system. Eleven sites shared the same ground water source but were distributed amongst ten different reservoir zones. For 12 patients living within 20 km of Brisbane CBD, the mean distance between their residential address and nearest positive water site was 3.49 km (range 0.9 to 9.8 km). The four clinically significant cases lived a mean of 2.7 km from a positive water site (Figure 5.2).

117

Figure 5.2. Map of Brisbane urban catchment area, human isolates and potable water isolates of M. lentiflavum.

Significant isolates: Case descriptions The four patients who met the ATS/IDSA criteria for disease are described below. All specimens were ZN stain negative.

Case 1: Disseminated infection A 43 year old female smoker had a background of intravenous drug use and HIV. In 1998 she developed granulomatous hepatomegaly, thought to be a reaction from injecting methadone mixed with orange juice, which settled after ceasing this activity. A tunnelled intravenous access device was placed February 2006. In April 2007 she presented with hepatosplenomegaly and mild pancytopenia. Liver and gastric lymph node biopsies revealed granulomata. Two bone marrow biopsies six weeks apart showed initially scant, but then more marked, granulomata. All specimens were culture negative for AFB. A working diagnosis of sarcoidosis was made and prednisone with HAART cover (tenofovir, emtricitabine and efavirenz) was commenced. Azathioprine was introduced and prednisone ceased by April

118

2008. In June she was admitted with massive hepatosplenomegaly, weight loss and fevers. CD4+ count was 140 cells/mL and viral load <50 copies/mL. Over the next month all four blood cultures grew M. lentiflavum; after 15 days mycobacteria were apparent and M. lentiflavum was confirmed on day 22. Bone marrow biopsy showed granulomata and grew M. lentiflavum. Urine and faecal samples were negative. She did not produce any sputum. Chest radiograph showed extensive miliary nodules, CT revealed peribronchial thickening and bronchiolitis but no lymphadenopathy. Empiric treatment consisted of isoniazid, rifampicin, pyrazinamide, clarithromycin and ethambutol. Oral prednisone (25mg od) improved symptoms, liver biochemistry and decreased splenic size. She was discharged on 15mg prednisone daily, isoniazid 300mg daily, ethambutol 400mg bd and clarithromycin 500mg bd. Her organomegaly improved over the next six months. The intravenous port was removed. She remains well and compliant with treatment.

Case 2: Chronic pulmonary nodules and bronchiectasis An 85 year old female presented in December 2007 with lobar pneumonia. She had never smoked and had no previous lung disease or immunosuppression. At follow- up she was lethargic with a persistent cough but no weight loss or fever. CT thorax confirmed bilateral well-defined nodules up to 1cm diameter. Bronchoscopic washings grew mycobacteria but the organism could not be speciated. Percutaneous nodule biopsy was non-diagnostic. Surgical biopsy of the right lung found caseating granulomata but culture was negative. At seven months a CT thorax found no change in the nodules but mild bronchiectasis had developed. Bronchoscopic lavage grew M. lentiflavum for the first time. She commenced ethambutol 800mg, rifampicin 450mg daily and clarithromycin 500mg bd in February 2009. Her symptoms improved. Treatment continues.

Case 3: Bronchiectasis A 49 year old female Taiwanese never smoker presented in 1998 with haemoptysis. She had moved to Australia five years earlier. Thoracic CT revealed a right middle lobe infiltrate. Three sputum samples were culture negative for AFB. Transbronchial

119

lung biopsies showed peribronchial granulomata but were culture negative. She received empirical quadruple therapy for TB. The cough continued but without haemoptysis. In 2004 the chest radiograph demonstrated middle lobe and lingular bronchiectasis. Three sputum samples were AFB culture negative. Bronchoscopic washings were ZN negative but grew M. lentiflavum, thought to represent colonisation. In 2007 one of three sputum specimens grew an unspeciated NTM. By January 2009 she was well, with no exacerbations in the previous year and stable radiographic appearance.

Case 4: Cervical lymphadenitis A 20-month old female presented with a four week history of bilateral cervical lymphadenopathy. The largest node (20x24mm) was excised. Necrotizing granulomata were seen. M. lentiflavum was cultured. No antimycobacterial therapy was administered; she made a full recovery.

Non-significant isolates A 29 year old female underwent bilateral lung transplantation. Routine post- transplant bronchial washings grew M. lentiflavum. Despite immunosuppressive therapy, no further AFB have been cultured from multiple samples in the subsequent 2.5 years.

Seven patients (four male, three female, mean age 62 years) grew one to two isolates of M. lentiflavum from sputum in the context of MAC disease or colonisation. Four were concurrently treated for MAC; one had recently completed treatment; two received no treatment for NTM and continue under surveillance. All seven had underlying lung disease (two cavitatory, five bronchiectatic). In no case was M. lentiflavum specifically treated. In addition, three patients’ sputum also grew one of M. interjectum, M. fortuitum or M. abscessus respectively.

Three otherwise well patients (40 year male, psoas abscess; 2 year female, cervical lymphadenitis; 54 year male, neck abscess) grew M. lentiflavum and S. aureus. All made good recoveries after flucloxacillin with or without drainage. No histology

120

samples were taken but cytological examination of lymph node aspirate from the 2 year old found lymphocytes, macrophages, neutrophils and fragments of epithelioid histiocytes but no well-formed granulomas. Two other patients (35 year female, chronic leg ulcer; 59 year female, post-thyroidectomy wound abscess) grew M. lentiflavum without S. aureus and were treated with debridement and flucloxacillin. Biopsies showed no granulomata.

Most other isolates were cultured from respiratory samples. One isolate was grown from ascitic fluid and one from blood. Three patients with cystic fibrosis (two with mild disease, one lung transplant recipient) had one or two isolates each but no evidence of disease.

Strain typing

Diversilab® patterns were grouped into seven rep-PCR Profiles, A–G (Figure 5.3). The eight clinical isolates of Profile A showed 97-99% similarity. This profile included two clinically significant isolates (Cases 1 and 3) and six non-significant isolates (respiratory samples n =3, soft tissue n = 2 and ascites n = 1). Two further pulmonary isolates (Profiles A1 and A2) were approximately 90% similar to the Profile A isolates. The isolate from Case 2 was contaminated and could not be analysed. The isolate from Case 4 (Profile B) had 94% similarity to a non-significant isolate from soft tissue. These two isolates were from patients who lived in locations 1800 km apart.

121

Figure 5.3. Dendrogram and virtual gel images representing rep-PCR fingerprint patterns of 16 human and 7 water isolates of M.lentiflavum (CI clinical isolate, W potable water isolate, BAL bronchoalveolar lavage, * clinically significant isolate)

Profile D comprised a pair of non-significant pulmonary isolates of 97% similarity. These isolates came from patients who lived within 80 km of each other, 450 km north of Brisbane. Profiles C and E were non-significant isolates and distinct from other rep-PCR profiles. Five water sample isolates (Profile A3) had 97-99% similarity and shared 90% similarity with the clinical isolates of Profiles A, A1 and A2. The other two water isolates (Profiles F and G) were distinct from all other clinical and water isolates.

122

Global case reports

Table 5.3 summarises 30 cases with clinically significant disease published in English. The disease spectrum ranges from cervical lymphadenitis (eight of nine cases in children), through to acute or chronic disease usually affecting lungs and pleura (infiltrates, cavities, nodules, effusions) but also arthritis/discitis, bone lesions, skin ulcers and hepatosplenomegaly. The rapid onset of cervical lymphadenitis has been noted in many reports with usually an excellent outcome from excision alone. The mean age of adults with non-lymphadenitis disease (20 cases) was 56 (range 87) and evenly split between the sexes. Eleven cases had associated immunocompromise. Eleven were reportedly stable or improved at follow-up, six died and three had uncertain outcomes.

DISCUSSION

The diagnosis of M. lentiflavum disease can be difficult to establish, exemplified by the cases in this paper. The clinical information we gathered was largely retrospective, which poses certain limitations; however both Case 1 and 2 are current patients undergoing active treatment at the time of writing. We note the occurrence of occasional M. lentiflavum isolates from patients colonised or undergoing treatment for MAC and in patients with S. aureus soft-tissue infections. Certainly it is recognised that some patients can grow multiple NTM at the same or different times, and M. lentiflavum may be no different in this respect. It has been reported from sputum alongside MAC and M. tuberculosis but these cases may represent colonization/contamination rather than infection (Suffys, et al., 2006; Shin, Yoon, Song, & Kim, 2007). Co-isolation of M. lentiflavum with S. aureus has not been reported as far as we are aware. This most probably represents contamination or colonization. Co-infection of S. aureus and M. tuberculosis has been reported, possibly as superadded staphylococcal infection in tuberculous tissue (Lee & Liu, 2007; Franco-Paredes & Blumberg, 2001). Although no histology samples were taken, cytological examination of lymph node aspirate from the 2 year old with lymphadenitis is intriguing as the inflammatory cells were

123

predominantly lymphocytes/ macrophages with epithelioid histiocytes. Treatment using flucloxacillin with or without drainage effected a complete cure in all cases.

M. lentiflavum is a rare isolate and an unusual cause of disease in humans. As with other NTM it can be isolated from contaminated samples: clinical significance should be assessed before any treatment is considered (Griffith et al., 2007). In 2005, of 488 patients with pulmonary NTM isolates in Queensland, only 26.6% were considered to have significant disease (Thomson, 2010). The proportion was higher for M. intracellulare (39.4%), M. avium (33.3%) and M. kansasii (52.6%), and much lower for species traditionally thought to be more likely contaminants, eg M. gordonae (11.1%). In our series, four of 36 cases (11%) had clinically significant isolates, similar to published estimates of 10% to 21% (Tortoli, et al., 2002; Safdar & Han, 2005). This may be an underestimate given that we could not determine clinical significance in four patients. Worldwide M. lentiflavum may be under- reported and incorrectly identified as other, more familiar species, especially if access to molecular identification is limited.

M. lentiflavum in the environment M. lentiflavum has been isolated from water distribution samples previously. Torvinen isolated NTM from up to 80% of sites across Finland (Torvinen, et al., 2004) with M. lentiflavum the second commonest species (38% of sites). Laboratory isolation of M. lentiflavum from clinical specimens in Finland has increased independently of speciation methods, but details of patients with disease are lacking (Tsitko, et al., 2006). In Korea, Lee found 26% of 84 drinking water sites grew mycobacteria. Sixty-five percent of isolates were M. lentiflavum (Lee, Lee, Han & Ka, 2008). In our study, mycobacteria were isolated from 70% of sites, but M. lentiflavum from only 6.3%. The difficulties in isolating mycobacteria from potable water are well recognised and relate to mycobacterial growth characteristics and the need for specimen decontamination to reduce bacterial and fungal overgrowth. Decontamination reduces mycobacterial yields hence the prevalence of mycobacteria in potable water samples is considered a significant underestimation of the true figure (Thomson et al., 2008). Culture-based techniques may be less

124

sensitive than direct PCR, however detecting mycobacterial DNA does not necessarily prove the presence of viable organisms able to cause infection; detection of M. lentiflavum using culture-based methods is significant with respect to human health. Case 1 had long term intravenous access which may have allowed direct exposure to contaminated water through illicit drug administration. In this report we have geographically associated culture-positive water samples and clinical disease.

Mycobacterial Genotyping Diversilab® strain typing showed that Profiles A and A3 were most prevalent amongst clinical and water isolates and shared approximately 90% similarity. The criteria for interpreting rep-PCR-based typing results have been established for some mycobacterial species. For example Cangelosi et al (2004) found high concordance between RFLP and rep-PCR reporting 93% similarity as the cut-off value for clustered M. tuberculosis isolates and 92% for M avium. Zelazny’s analysis of M. abscessus, the largest study of rep-PCR in NTM to date, used rep-PCR to successfully cluster M. abscessus strains that were clonally related by PFGE analysis (Zelazny et al., 2009). Four of the water samples comprising Profile A3 and one unrelated strain (Profile G) came from sites that shared a ground water source. These findings suggest there may be a dominant environmental strain closely related (90%), but not identical, to strains found in human specimens and as a cause of human disease.

The theory of dominant local environmental strains is supported by the finding of a different strain type from two patients living in proximity to each other but 450km away from Brisbane (Profile D).

Profile A contained clinically significant and insignificant isolates. Profile B also contained a pair of highly similar isolates (94%) of which one was clinically significant. Whilst the residential addresses of these patients were 1800km apart, nothing is known about the duration of residence or travel/work habits of these cases. As such the infection may not have been acquired locally. Conclusions

125

cannot be drawn regarding the pathogenicity of different strains; a larger study is required to address this question.

The finding of different, less common strain types (Profiles E, F, G) confirms the validity of using automated rep-PCR (Diversilab®) as a tool for strain typing this species. Variation in M. lentiflavum strain type has been previously demonstrated. Buijtels (Buijtels, Petit, Verbrugh, van Belkum & van Soolingen, 2005) reported 55 M. lentiflavum isolates from 149 specimens obtained from 38 Zambian patients at a single hospital. Two patients definitely fulfilled ATS criteria for significant disease. As this species is a rare cause of disease, the authors performed molecular identification on a subset of 12 isolates to investigate the possibility of laboratory contamination. Six strain types were identified; the Zambian strains were clearly different from comparator Dutch strains. The finding of a dominant strain probably represented the local endemic strain, but laboratory or point-of-collection contamination could not be entirely excluded. As our isolates came from multiple laboratories state-wide at different time-points, it is unlikely that contamination would explain their presence in multiple clinical specimens.

Treatment of Mycobacterium lentiflavum The optimal treatment for M. lentiflavum disease is not established; a wide range of regimens has been used in previous case series. Although there is no evidence base to support any specific regimen, we have achieved symptomatic and radiological improvement in Cases 1 and 2 above with rifampicin/ ethambutol/ clarithromycin at twelve months and isoniazid/ethambutol/clarithromycin at seven months respectively. More detailed reporting of treatment regimes and outcomes will help establish optimum therapy. Summary In summary, we have added four cases to the growing body of evidence that M. lentiflavum is a rare but definite cause of infection and that the source of this infection may be potable water.

126

Acknowledgements We thank Chris Macdermott for NTM database assistance, Chris Coulter for advice, David L. Paterson for providing the Mycobacterium kit and LabChip (bioMerieux), bioMerieux Australia for use of the Agilent 2100 Analyzer and fingerprint analysis software and the following clinicians for providing clinical data: John Aukes, Michael Bint, Allan Finnimore, Paul Griffin, Peter Hopkins, Joseph McCormack, William Oliver, Eric Pertnikovs, Sandra Rizzo, David Sowden, and Shyam Sunder.

127

Table 5.3 Summary of published cases of clinically significant M. lentiflavum isolates† Referenc Se Countr Organ/ tissue Immu Comorbid Treatment Outcome e x, y involved ne conditions Ag com- e promi se Soft tissue/ skin (Springe F Germa 5m history No DM, CCF [INH, RMP, Marked improve r Et al., 85 ny thoracic discitis PZA] 3m then -ment at 2M 1996) T9, T10. Biopsy [INH / RMP] showed 6m granulomata, C+ve after 22d. (Ibanez F Spain Arthritis, skin Yes, Antisynthe [INH, RMP, Condition et al., 52 lesions, synovial GC, tase EMB, PZA] worsened 2002) fluid was sterile; CYC syndrome then with weight synovial biopsy [FA, LFX, CLR] loss, found for 1wk before synovitis. granulomata, death Died after 4m ZN+ve, C+ve after 4wk (Montej M Spain 2 skin ulcers, Yes HIV for [HAART, INH, Lost to follow o, et al., 48 2yr history. 15yr, CD4+ EMB, LFX] up 2006) Biopsy ZN+ve, 46, IVDU C+ve

Cervical lymphadenitis (Haase M Germa 10d history, No No Excision Well at 18m et al., 42 ny ZN+ve, C+ve, 1997) m granulomata present (Haase, M Germa 14d history. ZN- No No Excision Well at 18m 1997) 33 ny ve, C+ve, m granulomata present (Tortoli, F 4 Italy 15d history, No No [INH, RMP] Well at 2yr 1997) biopsy scanty duration NS; ZN+ve, C+ve excision (Cabria M Spain 10d history, No Asthma [INH, RMP, Recovered et al., 19 ZN+ve, C+ve, PZA] duration fully 2002) m granulomata NS; then present excision (Tortoli, M Italy No details No No Excision Well at 1yr 2002) 4 (Piersim F Italy 4wk history. ZN- No Severe Excision Well at 18m oni, et 45 ve, C+ve at 10d, periodonta al., granulomata l disease 2004) present (Tortoli, M Italy 4wk history. No No Excision Recurrent et al., 18 ZN+ve, C+ve lymphadenop 2006) m after 3 wk; athy after 3m recurrent treated with lymphadenopat excision hy at 3m in same position

128

showed granulomata QLD F Austra 4wk history, No No Excision Full recovery 20 lia biopsy ZN-ve, m granulomata present, C +ve

Pulmonary (Tortoli M Italy LUL pulmonary Yes RA [INH, RFB, Stable at 4m. et al., 58 infiltrates, (RA) EMB, PZA] No 1997) pleural effusion, improvement low-grade fever, in CT thorax weight loss. appearance Pleural fluid and sputum C-ve; pleural biopsy showed granulomata, ZN-ve, C+ve after 3 wk (Tortoli, F Italy Cough, fever, No Bronchiect [RMP, INH, 5yr follow- et al., 61 weight loss, RUL asis since PZA] for up, sputum 2002) nodular age 30 ‘several intermittentl pulmonary months’, then y ZN +ve, infiltrates and [RFB, EMB, C+ve; no adenopathy. CLR, CIP] improvement Sputum ZN+ve, in radiology C+ve (Iwamot M Japan Haemoptysis, No20 2 yr [RMP, EMB, 3yr follow- o, et al, 71 bilateral 05 treatment INH, PZA] 1 yr, up, sputum 2003) pulmonary for CLR also remained ZN infiltrates, pulmonary added, +ve, C+ve, cavities RUL. TB aged duration NS symptoms Sputum ZN+ve, 30; continued, C+ve after 35 smoker CXR slowly days (multiple progressed specimens) (Buijtels, F Zambi 4wk history of No No [RMP, INH, Improved. et al., ) 35 a cough, pleural PZA, EMB] Duration of (case 6) effusion fluid duration NS follow-up NS C+ve. (Molteni F Italy Haemoptysis, No Previous [INH, PZA, 3yr follow- , et al., 67 low-grade fever, pulmonary EMB, RMP] up. Poor 2005) weight loss. TB, with 3m: no effect: compliance Sputum ZN+ve, fibrosis - ceased. with drugs, C+ve RUL, COPD 2 yrs later, intermittent started CLR haemoptysis, monotherapy, weakness, improved by dyspnoea, 3m, but sputum sputum still remained ZN+ve [EMB, ZN+ve, CXR RFB, CIP] unchanged added for 2wk (Safdar M USA Fever, RUL Yes MF on pIF [CIP 500bd, ‘Stable & Han, 49 pulmonary non- α2 AZM 500 od, disease’, lost 2005) cavitary EMB 1g od] to follow up;

129

nodules, 2x unknown duration of sputum C+ve at duration follow-up NS 27d, BAL C-ve (Safdar M USA Fever, Yes HSCT for [CLR 500 bd, Died after 12 & Han, 66 neutropenia, CLL, GvHD EMB 1g od] wk, septic 2005)* necrotizing shock due to pneumonia, pseudomona single sputum s aeruginosa. C+ve at 28d Uncertain significance of M.lentiflavu m (Suffys, M Brazil Cough, fever, Yes Newly [Streptomycin Died a few et al., 28 reticulonodular diagnosed ] duration NS months later, 2006)* infiltrate, HIV, PCP (? 1m) GC, no further PCP+ve, 1x sulphamethox clinical sputum C-ve. azole details 3wk later high /trimethoprim fever returned – , zidovudine, 1x blood C+ve didanosine (Suffys, F Brazil Pulmonary Yes Chronic Various Sputum et a., 64 cavities, treated pulmonary including CLR, persistently 2006)* for ZN+ve tuberculosi EMB, C+ve: 12/15 pulmonary s clofazimine, sputum tuberculosis for RMP and samples grew 6 years doxycycline. MAC, 2 grew No details M.lentiflavu given. m. Died of uterine cancer (Tortoli, M Italy Fever, chest Yes ALL, [CLR, AMK, Improved, et al., 14 pain, pleural chemother CFT] 5m CXR almost 2006) effusion, apy returned to pneumothorax, normal multiple RUL and RML nodules. Pleural fluid ZN+ve, C+ve at 4 wk (Tortoli, M Italy Bilateral pleural No No [LFX] After 1m, et al., 87 thickening, left monotherapy pulmonary 2006) pleural effusion, for 3 wk, then opacities pulmonary [CLR, LFX , resolved, opacities, RFB] effusion pleural fluid ZN- remained ve, C+ve at 3 wk (Neonaki M Greec Pleural effusion, NS NS [INH, RMP, Recovered s, et al., 23 e sputum C+ve PZA, EMB] 2007) duration NS (? (Case 2) 6m) QLD F Austra Cough, No No [EMB 800mg, Stable at 7m 85 lia pulmonary RMP 450mg, nodules, mild CLR 500mg bronchiectasis, bd] bronchial washings ZN-ve,

130

C+ve

QLD F Austra Haemoptysis, No No No specific Currently 49 lia bronchiectasis, treatment well, stable bronchial given chest washings ZN-ve, radiograph at C+ve 11 yr

Disseminated (Niobe, M France Fever, Yes HIV for 7 yr [CLR, RFB, Fully et al., ) 49 pulmonary EMB] and recovered, infiltrates, 3x antiretrovirals, died 3 yr blood culture at least 4m later of heart and BAL C+ve treatment, failure after 4 wk probably 9 m (Tortoli, M Italy Basal Yes HIV, NHL [Antiretroviral Well at 1 yr. et al., 45 pulmonary s, RFB, CLR], Hepatic 2002)* infiltrate, T4 then lesion was vertebra [RFB, CIP, aspirated involvement, EMB, CLR] C+ve but hepatic lesion. then [RFB, continued to Empirical CLR] grow. It was treatment given later excised with clinical and proved improvement. to be NHL Liver biopsy C+ve, no other samples reported (Buijtels. F Zambi 8 wk duration No No No treatment NS Et al., 67 a hepato- given 2005) splenomegaly, (Case 1) axillary lymphadenopat hy; LN biopsy ZN+ve, C+ve; 2x Sputum C+ve. QLD F Austra 4x Blood Yes HIV, hep C, [INH 300, EMB Compliant 43 lia cultures C+ve IVDU, SLE 400 bd, CLR with after 15 days, 500 bd] with treatment, 1x bone reducing improving at marrow ZN-ve, course of 12m granulomata prednisone present, C+ve; hepatosplenom egaly, pulmonary nodules ALL-acute lymphoblastic leukaemia, BAL- bronchoalveolar lavage, C+ve -culture positive, CCF- congestive cardiac failure, CLL- chronic lymphocytic leukaemia, COPD- chronic obstructive pulmonary disease, CXR chest radiograph, CYC cyclophosphamide, d days, DM diabetes mellitus, GC glucocorticoids, GvHD graft versus host disease, HAART highly active antiretroviral therapy, hep C hepatitis C, HSCT haematopoetic stem cell transplant, IVDU intravenous drug user, LFTs liver function tests, LN lymph node, MF myelofibrosis, m month, NHL non-Hodgkins lymphoma, NS information not stated or unknown, PCP pneumocystis pneumonia, pIF α2 pegylated interferon-α2, QLD Queensland cases in this paper, RA rheumatoid arthritis, SLE systemic lupus erythematosus, TB tuberculosis , wk week, yr year, ZN+ve/-ve Ziehl-Neelsen smear stain positive/ negative

131

AMK amikacin, AZM azithromycin, CIP ciprofloxacin, CFT ceftriaxone, CLR clarithromycin, EMB ethambutol, FA fusidic acid, INH isoniazid, LFX levofloxacin, PZA pyrazinamide, RFB rifabutin, RMP rifampicin† Two further cases from articles written in Spanish have not been included in this table (Galarraga, Torreblanca, & Jimenez, 2002; Uria, Garcia, Menendez, & Jimenez, 2003)* Not definite cases of disease REFERENCES.

Benson , D. A, Karsch-Mizrachi, I., Lipman, D. J., Ostell, J., Sayers, E. W. (2009) GenBank. Nucleic Acids Research, p. D26-31. Buijtels, P. C., Petit, P. L., Verbrugh, H. A., van Belkum, A., & van Soolingen, D. (2005). Isolation of nontuberculous mycobacteria in Zambia: eight case reports. Journal of Clinical Microbiology, 43(12):6020-6. Cabria, F., Torres, M. V., Garcia-Cia, J. I., Dominguez-Garrido, M. N., Esteban, J., Jimenez, M. S. (2002). Cervical lymphadenitis caused by Mycobacterium lentiflavum. Pediatric Infectious Diseases Journal, 21(6):574-5. Cangelosi, G., Freeman, R., Lewis, K., Livingston-Rosanoff, D., Shah, K., Milan, S. J., ...Goldberg, S. V.(2004). Evaluation of a high-throughput repetitive-sequence- based PCR system for DNA fingerprinting of Mycobacterium tuberculosis and Mycobacterium avium complex strains. Jouurnal of Clinical Microbiology, 42(6):2685-93. Franco-Paredes, C., & Blumberg, H. M. Psoas muscle abscess caused by Mycobacterium tuberculosis and Staphylococcus aureus: case report and review. American Journal of Medical Science, 321(6):415-7. Galarraga, M., Torreblanca, A., & Jimenez, M. (2002). Isolation of Mycobacterium lentiflavum in a case of suspected lung cancer. Enfermedades Infecciosas y Microbiologica Clinic, 20:93-4. Griffith, E., Aksamit, T., Brown-Elliott, B., Catanzaro, A., Daley, C., Gordin, F., ... Winthrop, K. (2007) An Official ATS/IDSA Statement: Diagnosis, Treatment, and Prevention of Nontuberculous Mycobacterial Diseases. American Journal of Respiratory and Critical Care Medicine, 175:367-416. Haase, G., Kentrup, H., Skopnik, H., Springer, B., Bottger, E. C. (1997). Mycobacterium lentiflavum: an etiologic agent of cervical lymphadenitis. Clinical Infectious Disease, 25(5):1245-6. Harmsen, D., Rothgänger, J., Frosch, M., Albert, J. (2002). RIDOM: Ribosomal differentiation of medical microorganisms database. Nucleic Acids Research, 30:416-7.

132

Ibanez, R., Serrano-Heranz, R., Jimenez-Palop, M., Roman, C., Corteguera, M., & Jimenez, S. (2002). Disseminated infection caused by slow-growing Mycobacterium lentiflavum. European Journal of Clinical Microbiology and Infectious Diseases, 21(9):691-2. Iwamoto, T., Sonobe, T., Hayashi, K., Okazaki, M., Umeda, B. ( 2003). A chronic pulmonary disease caused by Mycobacterium lentiflavum in a patient with a history of pulmonary tuberculosis. Clinical Microbiology Newsletter, 25(10):79. Lee, I. K., & Liu, J. W. (2007). Osteomyelitis concurrently caused by Staphylococcus aureus and Mycobacterium tuberculosis. Southern Medical Journal, 100(9):903-5. Lee, E. S., Lee, M. Y., Han, S. H., & Ka, J. O. (2008). Occurrence and molecular differentiation of environmental mycobacteria in surface waters. Journal of Microbiology and Biotechnology, 18(7):1207-15. Molteni, C., Gazzola, L., Cesari, M., Lombardi, A., Salerno, F., Tortoli E,... Gori, A. (2005). Mycobacterium lentiflavum infection in immunocompetent patient. Emerging Infectious Diseases, 11(1):119-22. Montejo, M., Goicoetxea, J., Agesta, N., Gil, A., Urra, E., & Jimenez, M. S. (2006). Cutaneous infection by Mycobacterium lentiflavum in a patient with HIV. Dermatology, 213(2):173-4. Neonakis, I. K., Gitti, Z., Kourbeti, I. S., Michelaki, H., Baritaki, M., Alevraki, G., ... Spandidos, D. A. (2007). Mycobacterial species diversity at a general hospital on the island of Crete: first detection of Mycobacterium lentiflavum in Greece. Scandinavian Journal of Infectious Diseases, 39(10):875-9. Niobe, S. N., Bebear, C. M., Clerc, M., Pellegrin, J. L., Bebear, C., & Maugein, J. (2001). Disseminated Mycobacterium lentiflavum infection in a human immunodeficiency virus-infected patient. Journal of Clinical Microbiology, 39(5):2030-2. Piersimoni, C., Goteri, G., Nista, D., Mariottini, A., Mazzarelli, G., Bornigia, S. (2004). Mycobacterium lentiflavum as an emerging causative agent of cervical lymphadenitis. Journal of Clinical Microbiology, 42(8):3894-7. Safdar, A., & Han, X. Y. (2005) Mycobacterium lentiflavum, a recently identified slow-growing mycobacterial species: clinical significance in

133

immunosuppressed cancer patients and summary of reported cases of infection. European Journal of Clinical Microbiology and Infectious Diseases, 24(8):554-8. Sampaio, J., Viana-Niero, C., de Freitas, D., Höfling-Lima, A., & Leão, S. Enterobacterial repetitive intergenic consensus PCR is a useful tool for typing Mycobacterium chelonae and Mycobacterium abscessus isolates. (2006a). Diagnostic Microbiology and Infectious Diseases, 55(2):107-18. Sampaio, J., Chimara, E., Ferrazoli, L., da Silva Telles, M., Del Guercio, V., Jericó, Z.,... Liao, S.C. (2006b). Application of four molecular typing methods for analysis of Mycobacterium fortuitum group strains causing post-mammaplasty infections. Clinical Microbiology and Infection, 12(2):142-9. Shin, S., Yoon, J.H., Song, S. H., & Kim, E. C. (2007).Isolation of Mycobacterium lentiflavum from a patient with a lung destroyed by tuberculosis. Korean Journal of Laboratory Medicine, 27(2):124-7. Springer, B., Wu, W. K., Bodmer, T., Haase, G., Pfyffer, G. E., Kroppenstedt, R.M., ... Bottger, E. C. (1996). Isolation and characterization of a unique group of slowly growing mycobacteria: description of Mycobacterium lentiflavum sp. nov. Journal of Clinical Microbiology, 34(5):1100-7. Suffys, P., Rocha Ada, S., Brandao, A., Vanderborght, B., Mijs, W., Jannes, G., ... Saad, M. H. (2006). Detection of mixed infections with Mycobacterium lentiflavum and Mycobacterium avium by molecular genotyping methods. Journal of Medical Microbiology, 55(Pt 1):127-31. Thomson, R., Carter, R., Gilpin, C., Coulter, C., Hargreaves, M. (2008). Comparison of methods for processing drinking water samples for the isolation of Mycobacterium avium and Mycobacterium intracellulare. Applied and Environmental Microbiology, 74(10):3094-8. Thomson, R. Changing epidemiology of pulmonary nontuberculous mycobacteria infections. (2010). Emerging Infectious Diseases, 16(10), 1576-1582 Tortoli, E., Piersimoni, C., Kirschner, P., Bartoloni, A., Burrini, C., Lacchini, C., Bottber, E. C. (1997). Characterization of mycobacterial isolates phylogenetically related to, but different from Mycobacterium simiae. Journal of Clinical Microbiology, 35(3):697-702.

134

Tortoli, E., Bartolonib, A., Erba, M. L., Levre, E., Lombardi, N., Mantella, A., ...Ghittinoc, C. (2002). Human infections due to Mycobacterium lentiflavum. Journal of Clinical Microbiology, 40(2):728-9. Tortoli, E., Mattei, R., Russo, C., Scarparo, C. (2006). Mycobacterium lentiflavum, an emerging pathogen? Journal of Infection, 52(6):e185-7. Torvinen, E., Suomalainen, S., Lehtola, M. J., Miettinen, I. T., Zacheus, O., Paulin L., ... Martikainen, P. J. (2004). Mycobacteria in water and loose deposits of drinking water distribution systems in Finland. Applied and Environmental Microbiology, 70(4):1973-81. Tsitko, I., Rahkila, R., Priha, O., Ali-Vehmas, T., Terefework, Z., Soini, H., & Salkinoja- Salonen, M. S. (2006). Isolation and automated ribotyping of Mycobacterium lentiflavum from drinking water distribution system and clinical specimens. FEMS Microbiology Letters, 256(2):236-43. Uria, M., Garcia, J., Menendez, J., Jimenez, M. (2003). Mycobacterium lentiflavum infection: case history and review of the medical literature. Enfermedades Infecciosas y Microbiologica Clinica, 21:274-5. Wilton, S., & Cousins, D. (1992) Detection and identification of multiple mycobacterial pathogens by DNA amplification in a single tube. PCR Methods and Applications, 4:269-73. Zelazny, A., Root, J., Shea, Y., Colombo, R., Shamputa, I., Stock, F.,...Sampaio, E. P. (2009). Cohort study of molecular identification and typing of Mycobacterium abscessus, Mycobacterium massiliense, and Mycobacterium bolletii. Journal of Clinical Microbiology, 47(7):1985-95. Zhang, Y., Yakrus, M., Graviss, E., Williams-Bouyer, N., Turenne, C., Kabani, A., .. Wallace, R. J. (2004). Pulsed-field gel electrophoresis study of Mycobacterium abscessus isolates previously affected by DNA degradation. Journal of Clinical Microbiology, 42(12):5582-7.

135

Chapter 1:Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1: Objective 1 Aim 3: Objective 5 Aim 2: Objectives 3&4 Chapter 3: Chapter 9: Isolation Chapter 5: Development of of NTM from homes Mycobacterium method for isolation of patients with NTM lentilavum of NTM from water disease

Aim 1:Objective 2 Chapter 4: Isolation Chapter 6: of NTM from M. abscessus Brisbane water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

136

CHAPTER 6: MYCOBACTERIUM ABSCESSUS IN POTABLE WATER: A POTENTIAL SOURCE OF HUMAN INFECTION

This paper was published as a conference poster and paper as follows: Thomson, R. M., Tolson, C., Sidjabat, H., Coulter, C., Huygens, F., & Hargreaves, M. (2012). Mycobacterium abscessus infection and potable water. (2012). American Journal of Respiratory and Critical Care Medicine, 185, A4022. (Appendix D)

Thomson, R.M., Tolson, C., Sidjabat, H., Huygens, F., and Hargreaves, M. (2013) Mycobacterium abscessus isolated from municipal water – a potential source of human infection. BMC Infectious Diseases, 13; 241

Author Contributions

RT (70%) designed the study, obtained clinical details of patients, analysed the Diversilab results, produced the analysis reports and graphics, wrote the manuscript. CT (10%) sourced and regrew clinical and water isolates, extracted DNA and assisted with rep-PCR strain typing HS (7.5%) assisted with strain typing and analysis of rep-PCR results CC (2.5%) provdied intellectual and organisational support FH (5%) provided intellectual support, review of the manuscript MH (5%) provided intellectual and design and support, contributed to writing and review of the manuscript

ABSTRACT

Mycobacterium abscessus is a rapidly growing mycobacterium responsible for progressive pulmonary disease, soft tissue and wound infections. The incidence of disease due to M. abscessus has been increasing in Queensland. In a study of Brisbane drinking water, M. abscessus was isolated from ten different locations. Aim: To compare genotypically the M. abscessus isolates obtained from water to

137

those obtained from human clinical specimens. Methods: Between 2007 and 2009, eleven isolates confirmed as M. abscessus were recovered from potable water, one strain was isolated from a rainwater tank and another from a swimming pool and two from domestic taps. Seventy-four clinical isolates referred during the same time period were available for comparison using rep-PCR strain typing (Diversilab). Results: The drinking water isolates formed two clusters with ≥97% genetic similarity (Water patterns 1 and 2). The tankwater isolate (WP4), one municipal water isolate (WP3) and the pool isolate (WP5) were distinctly different. Patient isolates formed clusters with all of the water isolates except for WP3. Further patient isolates were unrelated to the water isolates. Conclusion: The high degree of similarity between strains of M. abscessus from potable water and strains causing infection in humans from the same geographical area, strengthens the possibility that drinking water may be the source of infection in these patients.

INTRODUCTION

Nontuberculous mycobacteria are environmental organisms that can cause progressive lung disease in susceptible patients. M. abscessus is a significant problem as it is highly resistant to antimicrobial agents and usually requires prolonged treatment (>six months) with intravenous and oral antibiotics in combination (Jeon et al., 2009; Thomson & Yew, 2009; Griffith et al., 2007). The infection is often relentless despite treatment, and even if treatment is apparently effective, relapse is common. Rapidly growing mycobacteria such as M. abscessus are also well-documented causes of skin and soft tissue infections, especially complicating surgical procedures and injection sites (De Groote & Hewitt, 2006).

Water is an important environmental reservoir of mycobacteria causing human disease (Covert, Rodgers, Reyes & Stelma, 1999; Fox et al., 1992; Le Dantec et al., 2002). Humans are exposed to waterborne mycobacteria through drinking, swimming and bathing. Also aerosols generated during these activities may be inhaled (Falkinham, 2003), potentially resulting in disease. There is a recognized association between pulmonary infection with rapid growing mycobacteria (M. abscessus and M. fortuitum) and esophageal disorders (Hadjiliadis, Adlakha, &

138

Prakash, 1999; Hutchins & Boitnott, 1978). It is possible that patients acquire infection by aspirating contaminated water (Thomson, Armstrong & Looke, 2007). Outbreak investigations have identified M. abscessus in hospital water, dialysis (Carson, et al., 1988; Gomila, Ramirez, & Lalucat, 2007), and surgical equipment and endoscopy cleaning equipment (Gubler, Salfinger & von Graevenitz, 1992; Dawson, Armstrong & Blacklock, 1982). In many of these instances it is not clear whether the source of these outbreaks was an infected patient or water.

The taxonomy of Mycobacterium abscessus has been a topic of debate in the literature. M. massiliense and M. bolletii cannot be separated from M. abscessus on the basis of extensive phenotypic analysis and genotypic studies have lead to the proposition that the three taxa represent a single species with internal variability (Leao, Tortoli, Euzeby & Garcia, 2011). Strain variation amongst species of mycobacteria is well known and Pulsed Field Gel Electrophoresis (PFGE) has been considered the “Gold standard” for strain typing of many mycobacteria. ERIC (enterobacterial repetitive intergenic consensus) PCR has been used successfully to differentiate strains of mycobacteria associated with outbreaks of disease in mesotherapy clinics due to M. abscessus and M. chelonae, and in post mammoplasty patients with M. fortuitum infections (Sampaio, Viana-Niero, de Freitas, Hfling-Lima, & Leao, 2006; Sampaio, et al., 2006). A high concordance with PFGE was shown. More recently rep-PCR (Diversilab) was used in comparison to PFGE in a suspected outbreak of M. abscessus in a cystic fibrosis clinic (Aitken et al., 2012). Isolates that were identical on PFGE shared inly 90% similarity with rep-PCR, suggesting that the latter method may be more discriminatory. Rep-PCR was similarly shown to concur with strain typing using variable number tandem repeats (VNTR)(Harris et al., 2012). This method has significant advantages in cost and time saving over other methods and has demonstrated high discriminatory power.

In Queensland NTM disease remains notifiable and a central reference laboratory performs speciation of all culture positive isolates, providing a unique opportunity to capture isolates from around the state. The incidence of disease due to NTM has been increasing (Thomson, 2010) and between 2007-2008 water sampling was

139

conducted to investigate the presence of pathogenic NTM in potable water (Thomson et al., 2013). This paper reports the comparison of these water isolates with human isolates collected from patients in South east Queensland during the same time period.

METHODS

Water isolates obtained in a previous study (Thomson et al., 2013) had been stored in Dubos broth at -20°C and were thawed and subbed onto 7H11 plates as well as Löwenstein-Jensen slopes and incubated at 35°C until sufficient growth was available.

Those identified using 16S rDNA sequencing as M. abscessus/M. chelonae underwent hsp65 and rpoB gene fragment sequencing for more definitive identification. According to the proposed changes by Leao et al (2011) in line with the Bacteriological code, for the purposes of this study all isolates identified as M. abscessus, M. bolletii, M. massiliense were included as M. abscessus species.

Human samples were digested and decontaminated using 4% NaOH, neutralized with phosphoric acid and centrifuged at 3000g to concentrate the acid-fast bacilli (AFB). Smears were prepared from the sediment and stained by the Ziehl-Niehlsen (ZN) method. One Löwenstein-Jensen slope (±pyruvate) and 7ml Mycobacterial Growth Indicator Tube (MGIT) were inoculated and incubated at 35oC until growth was detected. ZN staining of colonies confirmed AFB. Multiplex PCR was performed to discriminate between M. tuberculosis, M. avium, M. intracellulare, M. abscessus and other Mycobacterium spp. Isolates identified as other Mycobacterium spp were further speciated using Hain Life Sciences GenoType® Mycobacterium CM kit (2004-7 only) and/or 16S rDNA sequencing in conjunction with phenotypic characteristics. All clinical isolates underwent gene fragment sequencing for hsp65 and rpoB.

140

Strain typing using automated Rep-PCR

The clonality of clinical and water M. abscessus isolates and a control strain (ATCC 19977) was determined using a rep-PCR based method (Diversilab® system, bioMerieux, Melbourne). DNA was extracted from clinical and water isolates using the Ultraclean Microbial DNA Isolation Kit (Mo Bio Laboratories, CA, USA). PCR mixture was prepared using AmpliTaq polymerase and PCR buffer (Applied Biosystems, New Jersey, USA) and Mycobacterium Diversilab®- primer mix according to the manufacturer’s instructions. Separation and detection of Rep-PCR products was performed by micro-fluidic chips of the Diversilab® System. A laboratory control strain of M. abscessus was included, and two chips were repeated because of low intensity banding. In most cases when the intensity improved the pattern didn’t change from that obtained with low intensity. An additional chip was run twice to confirm reproducibility of strain patterns. Any aberrances were excluded from the analysis. Fingerprints were analysed with Diversilab® software v.3.4.38 using the Pearson correlation co-efficient and unweighted pair group method with arithmetic means to compare isolates and determine clonal relationship. A similarity index of ≥97% was used to define isolates that were indistinguishable, based on the manufacturers recommendations despite previously published comparisons of rep- PCR with both PFGE and Variable Number Tandem Repeats, where a cut off of 90% was used. Isolates that shared 95% similarity were considered similar (and formed groups), and those <90% considered ‘different/unrelated’

The study protocol was approved by the Human Research Ethics Committee of The Prince Charles Hospital (EC-2617).

RESULTS

During the 10 years from 2001-2010, there were 486 patient notifications: 68.1% were pulmonary, 22.2% cutaneous or soft tissue. (Table 6.1)

141

Table 6.1: Sites of isolation of M. abscessus from human samples 2001-2010.

Site of isolate Frequency (%) Blood 14 (2.9) Bones and joints 5 (1) Cutaneous/Soft tissue 108 (22.2) Eye 1 (0.2) Lymph node + Other 1 (0.2) Lymph nodes 3 (0.6) Peritoneal 2 (0.4) Post surgical/Medical access device 19 (3.9) Pulmonary 331 (68.1) Unknown 2 (0.4) Total 486 (100)

Table 6.2. Clinical details of patients with M. abscessus isolates used for strain typing comparison.

142

Seventy-four clinical isolates collected during 2007-8 were available for comparison with fifteen water isolates. The mean age of patients was 55.94 yrs (SD 22.25; median 58.5; range 2-90yrs). Fifty percent were male. Forty-four patients had clinically significant pulmonary disease (according to clinician notification and the ATS/IDSA criteria), 11 patients had pulmonary isolates that were considered contaminants or not clinically significant and 19 patients had isolates from extra- pulmonary sites. (Table 6.2)

Eleven patients (14.9%) were included from other parts of Australia who lived >500 km outside the Brisbane water distribution network, and would have received potable water supply to their homes from different networks. Of the remaining patients, 31/63 (49.2%) patients lived in reservoir zones that contained sample points that grew M. abscessus.

Of 19 water isolates identified as M. abscessus/M. chelonae using 16s rDNA sequencing, 17 underwent successful hsp65 and rpoB sequencing. This identified 14 isolates as M. abscessus and one as M. bolletii – all considered M. abscessus subs abscessus according to the proposed taxonomy by Leao. (2009) There was insufficient DNA in two samples. Fifteen water isolates of M. abscessus were strain typed using rep-PCR. Eleven isolates came from six different potable water distribution system sampling sites; two came from domestic taps in the same dwelling, one came from an unrelated rainwater tank and the other from a suburban domestic swimming pool.

The 15 water isolates formed 6 different patterns (Figure 6.1). Municipal water isolates formed two clusters of indistinguishable isolates (≥97% similarity) and were classified as Group WP1 (eight isolates) and Group WP2 (two isolates) and a single isolate WP3. The rainwater tank isolate (Group WP4) and the swimming pool isolate (Group WP5) were distinctly different. The two isolates from a patient’s domestic taps were indistinguishable (WP 6). (Figure 6.1)

143

When the clinical isolates were analysed with the water isolates there were 28 patterns/clusters that shared >97% similarity (indistinguishable). Thirteen of these clusters contained more than 1 isolate, and the remaining patterns were single isolates only. The indistinguishable clusters that were similar (>95% similarity) were then grouped into six main groups. Full report and graphics in additional material 6.1.

Figure'6.1.'rep+PCR'dendrogram'of'water'isolates'

Figure 6.1. rep-PCR dendrogram of water isolates of M. abscessus. P=pattern' s sharing 97% similarity. The 15 water isolates formed 6 differed patterns (Water Patterns (WP) 1-6); P7=M. abscessus control strain. P 1 and 2 were similar (i.e. shared 95% similarity). The two home bathroom isolates (WP3) were indistinguishable. The tank (WP4) and swimming pool (WP5) isolates were different (<95% similarity).

144

There were two large similar clusters of 21 (Pattern 15) and 22 (Pattern 17) isolates respectively. Examples demonstrated in Figure 6.2. P15 included one of the municipal water isolates from WP1 and P17 contained the remaining seven isolates from WP1. A further three clusters (P12-14) contained 5 clinical isolates was similar to cluster P15 and a single isolate P16 was similar to P17.

Figure'6'2:"Combined"P15"and"P17"examples"demonstra6ng"similari6es"between"clinical"and"water"isolates."

Figure 6.2. rep-PCR dendrogram of examples of main clinical strain patterns 17 (P1=pattern 1; green) and 17 (P2= pattern 2; orange) demonstrating similarities between water and clinical isolates. (Full dendrogram in Additional material 1)

145

Figure'6.3."rep#PCR"dendrogram"showing"examples"of"diﬀerent"strain"pa9erns"where"there"were"similari:es" between"clinical"and"water"strains.""

Figure 6.3. rep-PCR dendrogram showing examples of different strain patterns where there were similarities between clinical and water strains. P=indistinguishable patterns 1-5 (≥97% similarity); G= 1 Group of isolates with similar patterns (≥95% similarity)

The two water isolates from WP2 also clustered with six clinical isolates (P20-22;Gp 4: Additional material). Five clinical isolates were indistinguishable from the tank water isolate (P2). A further three clinical isolates and the two domestic tap water isolates were similar to this cluster (Group 1). One clinical isolate was indistinguishable from the swimming pool isolate. Figure 6.3. The isolate from water site 320 (WP3) differed from all the clinical isolates.

There were no significant differences between the characteristics of groups of patients in each strain cluster (age, gender, type of disease, site of infection, ground water source or reservoir zone). Some of the isolates from more remote areas did have unique strain patterns, but equally so did some of the Brisbane area isolates. 146

Some of the remote area isolates were similar to those of patients from Brisbane. However as is often the case with NTM infections, the precise location of where infection is acquired can be difficult to pinpoint.

DISCUSSION

M. abscessus has become an increasingly important clinical problem in the last 10 years, and its presence in potable water has not previously been emphasized.

In the majority of published studies looking for NTM in water, no M. abscessus was documented (van Ingen, Boeree, Dekhuijzen, van Soolingen, 2009). However there have been taxonomical changes that have led to M. abscessus being recognized as independent from M. chelonae, so older studies reporting M. chelonae may not necessarily have differentiated M. chelonae subsp abscessus, particularly as both species have identical 16s rDNA sequences. But in studies of potable water done since 2000 M. abscessus has been rarely reported.

A study done in Pretoria, South Africa in 2004 (September, Brozel & Venter, 2004) reported analysis of 78 samples – 14 had NTM – no MAC, one had M. abscessus (16s rDNA sequencing only). In a multi-hospital outbreak investigation in Taiwan, M. abscessus was grown from one water sample from one of the hospitals involved however this differed on strain typing from the isolates found in patients Huang, Chiou, Chen & Shen, 2010). Falkinham (Falkinham, Norton & Chevallier, 2001) sampled 8 different water treatment systems (raw, treated and Distribution System (DS) samples). M. abscessus was only found in one raw water sample from one system. There were no rapid growers in treated water, two rapid growers were found in the DS samples but the paper doesn’t clarify the species. In a more recent study (2009), van Ingen et al (2010) reported no M. abscessus in shower or tap water in the Netherlands. We have found that the inclusion of liquid media increased the yield for M. abscessus from potable water samples (Thomson et al, 2013). However three of the 13 isolates were only grown on solid media – suggesting that for future studies using culture-based techniques, both media forms should be included.

147

It has long been recognized that outbreaks of disease due to rapid growers may have originated in hospital and tap water, particularly for M. chelonae and M. fortuitum (Wallace, Brown & Griffith, 1998) though in some studies the authors have acknowledged that speciation separating M. abscessus has not occurred. Outbreak investigations have linked M. abscessus infections and pseudo infections to hospital tap water, and endoscope cleaning fluids (Maloney et al, 1994; Zhang, Rajagopalan, Brown & Wallace, 1997). In all of these hospital outbreaks, the potential for contamination of hospital sources and equipment by infected patients or other environmental sources (such as dust or dirt) exists; hence the origin of the culprit strain remains unclear.

Zelazny et al (2009) compared partial sequencing of rpoB, hsp65 and secA gene fragments and rep-PCR (Diversilab) in order to differentiate M. abscessus from subspecies formerly named M. massiliense and M. bolletii. They were able to group the majority of M. abscessus complex clinical isolates into a group similar to the ATCC strain 19977, with a degree of diversity as low as 80% similarity to the reference strain. There were two other strain groups neither of which resembled the strain patterns obtained in our study. The geographic origin of the patients included in this study was not reported nor examined, but as the study was performed by a tertiary referral centre many came from patients around the United States (A. Zelazny, personal communication). In our study the strains obtained both from patients and from water were closely related, and differed from the strain patterns reported in the US, suggesting geographical differences in strain variation. The diversity of the US strains, may be due to the diverse origins of the isolates reported, reflecting differences in the dominant environmental strains. There was a degree of geographic diversity of our strains with unique strain types identified from central and far north Queensland (>500km and > 1000km from Brisbane respectively) and two unique strains from the bayside of Brisbane (> 30km from the CBD).

Whilst rep-PCR using the Diversilab has only recently been used for strain typing mycobacteria, published studies support its performance as equivalent or perhaps

148

even more discriminatory than PFGE. Aitken et al (2012) reported an outbreak of M. abscessus infections in a lung transplantation and cystic fibrosis center. Isolates involved were strain typed using both PFGE and rep-PCR (Diversilab). The isolate of the index case and outbreak cases were reported to be “genetically identical” using PFGE. The rep-PCR patterns of the same isolates showed only 90% genetic similarity, suggesting in fact that these strains are different, and that rep-PCR may be more discriminatory that PFGE. Harris (2012) describes a novel VNTR (variable number tandem repeats) method to type 41 M. abscessus complex strains from 17 pediatric cystic fibrosis patients and compares this with rep-PCR (Diversilab). As with the Aitken paper, isolates with 90% similarity were regarded as indistinguishable. The finding of six distinct profiles among 12 patients with M. abscessus complex suggests that the typing methods have sufficient discriminatory power. We have analysed both patient and water isolates from quite diverse and epidemiologically unrelated settings across a geographical area of >1367km2 and found 7 distinct clusters. We have more definitively identified M. abscessus strains in potable water, as strains that are indistinguishable from those found in samples from patients known to have disease. Whilst point source investigations of outbreaks have linked M. abscessus infection and hospital water and equipment, the present study tightens the link between potable water, to which patients are regularly exposed, and the acquisition of infection. This has important implications for those patients at risk (such as those with cystic fibrosis and bronchiectasis, swallowing disorders, or immunosuppression). Given the difficulty in treating these infections and the devastating effects they have for patients, efforts to identify effective disinfection methods for M. abscessus should be a high priority.

149

Acknowledgements: This work was supported by research grants from The Gallipoli Medical Research Foundation and The Prince Charles Hospital Foundation, Brisbane, Australia.

REFERENCES.

Aitken, M. L., Limaye, A., Pottinger, P., Whimbey, E., Goss, C. H., Tonelli, M. R.,… & Wallace, R. J. (2012). Respiratory outbreak of Mycobacterium abscessus subspecies massiliense in a lung transplant and cystic fibrosis center. American Journal of Respiratory and Critical Care Medicine, 185(2): 231-2. Carson, L. A., Bland, L. A., Cusick, L. B., Favero, M. S., Bolan, G. A., Reingold, A. L., & Good, R. C. (1988). Prevalence of nontuberculous mycobacteria in water supplies of hemodialysis centers. Applied and Environmental Microbiology, 54(12): 3122-5. Covert, T. C., Rodgers, M.R., Reyes, A.L., & Stelma, G.N., Jr. (1999) Occurrence of Nontuberculous Mycobacteria in Environmental Samples. Applied and Environmental Microbiology, 65(6): 2492-6. Dawson, D. J., Armstrong, J., & Blacklock, Z. M. (1982). Mycobacterial cross- contamination of bronchoscopy specimens. (1982). American Review of Respiratory Diseases.126(6): 1095-7. De Groote, M. A., & Huitt, G. Infections due to rapidly growing Mycobacteria. Clinical Infectious Diseases, 42(12): 1756-63. Falkinham, III J., Norton, C., & Le Chavallier, M. (2001). Factors influencing numbers of Mycobacterium avium, Mycobacterium intracellulare and other mycobacteria in drinking water distribution systems. Applied and Environmental Microbiology, 67(3): 1225-31. Falkinham, III J. (2003). Mycobacterial aerosols and respiratory disease. Emerging Infectious Diseases, 2003; 9: 763-7. Fox, C., Smith, B., Brogan, O., Rayner, A., Harris, G,, & Watt B. (1992). Non- tuberculous mycobacteria in a hospital's piped water supply. Journal of Hospital Infection, 21(2): 152-4.

150

Gomila, M., Ramirez, A., & Lalucat, J. (2007). Diversity of environmental Mycobacterium isolates from hemodialysis water as shown by a multigene sequencing approach. Applied and Environmental Microbiology, 73(12): 3787-97. Griffith, D. E., Aksamit, T., Brown-Elliott, B. A., Catanzaro, A., Daley, C., Gordin, F., … & Winthrop, K. on behalf of the ATS Mycobacterial Diseases Subcommitte. (2007). An Official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. American Journal of Respiratory and Critical Care Medicine, 175: 367-416. Gubler, J., Salfinger, M., & von Graevenitz, A. (1992). Pseudoepidemic of nontuberculous mycobacteria due to a contaminated bronchoscope cleaning machine. Report of an outbreak and review of the literature. Chest, 101(5): 1245-9. Hadjiliadis, D., Adlakha, A., & Prakash, U. B. (1999). Rapidly growing mycobacterial lung infection in association with esophageal disorders. Mayo Clinic Proceedings, 74: 45-51. Harris, K. A., Kenna, D. T. D., Blauwendraat, C., Hartley, J. C., Turton, J. F., Aurora, P., & Dixon, G. L. J. (2012). Molecular fingerprinting of Mycobacterium abscessus strains in a cohort of pediatric cystic fibrosis patients. Journal of Clinical Microbiology, 50(5): 1758-61. Huang, W-C., Chiou, C-S., Chen, J-H., & Shen, G-H. (2010). Molecular epidemiology of Mycobacterium abscessus infections in a subtropical chronic ventilatory setting. Journal of Medical Microbiology, 59(10): 1203-11. Hutchins, G. M., & Boitnott, J. K. (1978). Atypical mycobacterial infection complicating mineral oil pneumonia. Journal of the American Medical Association, 240: 539-41. Jeon, K., Kwon, O. J., Lee, N. Y., Kim, B-J., Kook, Y-H., Lee, … & Koh,W-J. (2009). Antibiotic treatment of Mycobacterium abscessus lung disease: a retrospective analysis of 65 Patients. American Journal of Respiratory and Critical Care Medicine, 180(9): 896-902. Leao, S. C., Tortoli, E., Euzeby, J. P., & Garcia, M. J. (2011). Proposal that Mycobacterium massiliense and Mycobacterium bolletii be united and

151

reclassified as Mycobacterium abscessus subsp. bolletii comb. nov., designation of Mycobacterium abscessus subsp. abscessus subsp. nov. and emended description of Mycobacterium abscessus. International Journal of Systematic and Evolutionary Microbiology, 61: 2311-3. Le Dantec, C., Duguet, J-P., Montiel, A., Dumoutier, N., Dubrou, S., & Vincent, V. (2002). Occurrence of Mycobacteria in Water Treatment Lines and in Water Distribution Systems. Applied and Environmental Microbioloy, 68(11): 5318-25. Maloney, S., Welbel, S., Daves, B., Adams, K., Becker, S., Bland, L., … & Jarvis WR. (1994). Mycobacterium abscessus pseudoinfection traced to an automated endoscope washer: utility of epidemiologic and laboratory investigation. Journal of Infectious Diseases, 169: 1166-9. Sampaio, J. L. M., Viana-Niero, C., de Freitas, D., Hfling-Lima, A. L., & Leao, S. C. (2006). Enterobacterial repetitive intergenic consensus PCR is a useful tool for typing Mycobacterium chelonae and Mycobacterium abscessus isolates. Diagnostic Microbiology and Infectious Diseases, 55(2): 107-18. Sampaio, J. L. M., Chimara, E., Ferrazoli, L., da Silva Telles, M. A., del Guercio, V. M. F., Jerico, Z. V. N.,… & Leao SC. (2006). Application of four molecular typing methods for analysis of Mycobacterium fortuitum group strains causing post-mammaplasty infections. Clinical Microbiology and Infection, 12(2): 142-9 September, S. M., Brozel, V. S., & Venter, S. N. (2004). Diversity of nontuberculoid Mycobacterium species in biofilms of urban and semiurban drinking water distribution systems. Applied and Environmental Microbiology, 70(12): 7571-3. Thomson, R., Armstrong, J., & Looke, D. (2007). Gastroesophageal reflux disease, acid suppression, and Mycobacterium avium complex pulmonary disease. Chest, 131: 1166-72. Thomson, R. M., Carter, R., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2013) Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia. BMC Microbiology, 13;89

152

Thomson, R., & Yew, W-W. (2009). When and how to treat pulmonary nontuberculous mycobacterial diseases. Respirology, 14(1): 12-26. Thomson, R. Changing epidemiology of pulmonary nontuberculous mycobacteria infections. (2010). Emerging Infetious Diseases, 16(10): 1576-82. Van Ingen, J., Boeree, M., Dekhuijzen, P., & Van Soolingen, D. (2009). Environmental sources of rapid growing nontuberculous mycobacteria causing disease in humans. Clinical Microbiology and Infection, 15: 888-892. Van Ingen, J., Blaak, H., de Beer, J., de Roda Husman, A. M., & Van Soolingen, D. (2010). Rapidly growing nontuberculous mycobacteria cultured from home tap and shower water. App Environ Microbiol, 76(17): 6017-6019. Wallace, R. J., Brown, B. A., & Griffith, D. E. (1998). Nosocomial Outbreaks/pseudooutbreaks caused by nontuberculous mycobacteria. Annual Reviews in Microbiology, 52(1): 453-90. Zhang, Y., Rajagopalan, M., Brown, B. A., & Wallace, R. J., Jr. (1997). Randomly amplified polymorphic DNA PCR for comparison of Mycobacterium abscessus strains from nosocomial outbreaks. Journal of Clinical Microbiology, 35(12): 3132-9. Zelazny, A. M., Root, J. M., Shea, Y. R., Colombo, R. E., Shamputa, I. C., Stock, F., …Conlan S, & Sampaio, E. P. (2009). Cohort study of molecular identification and typing of Mycobacterium abscessus, Mycobacterium massiliense, and Mycobacterium bolletii. Journal of Clinical Microbiology, 47(7): 1985-95.

153

Chapter 1:Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1: Objective 1 Aim 3: Objective 5 Aim 2: Objectives 3&4 Chapter 3: Chapter 9: Isolation of Development of Chapter 5: NTM from homes of method for isolation of Mycobacterium patients with NTM NTM from water lentilavum disease

Aim 1:Objective 2 Chapter 4: Isolation of Chapter 6: NTM from Brisbane M. abscessus water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

154

CHAPTER 7: STRAIN VARIATION AMONGST CLINICAL AND POTABLE WATER ISOLATES OF M. KANSASII USING AUTOMATED REPETITIVE UNIT PCR.

This chapter has been submitted as a paper for publication: Thomson, R., Tolson, C., Huygens, F., & Hargreaves, M. (2013). Strain variation amongst clinical and potable water isolates of M. kansasii using automated repetitive unit PCR. Submitted to International Journal of Medical Microbiology Feb 14th, 2013, currently under review. (IJMM-D-13-00047)

Author Contributions RT (65%) designed the study, coordinated the collection of samples, participated in the processing of water samples, collated and analysed the clinical and strain typing data, and wrote the manuscript. CT (15%) processed water samples, performed rep-PCR strain typing and collated results. FH (10%) intellectually contributed to the study design and methodology and the writing of the manuscript. MH (10%) intellectually contributed to the study design and methodology, and contributed to the writing of the manuscript.

ABSTRACT

Mycobacterium kansasii is a pulmonary pathogen that has been grown readily from municipal water, bur rarely isolated from natural waters. A definitive link between water exposure and disease has not been demonstrated and the environmental niche for this organism is poorly understood. Strain typing of clinical isolates has revealed seven subtypes with Type 1 being highly clonal and responsible for most infections worldwide. The prevalence of other subtypes varies geographically. In this study 49 water isolates are compared with 83 patient isolates from the same geographical area (Brisbane, Australia), using automated repetitive unit PCR. The clonality of the dominant clinical strain type is again demonstrated but with this 155

method strain variation within this group is evident comparable with other reported methods. There is significant heterogeneity of water isolates and very few are similar or related to the clinical isolates. This suggests that if water or aerosol transmission is the mode of infection, then point source contamination likely occurs from an alternative environmental source.

INTRODUCTION

Mycobacterium kansasii is a pulmonary pathogen found in tap water (Engel, Berwald & Havelaar, 1980). It is the second most common cause of Nontuberculous Mycobacterial (NTM) disease in the United States, South America and areas of high incidence occur around the world – particularly Poland, Slovakia, Paris, (Hoefsloot, van Ingen* et al., 2013) South-east England and Wales. In Queensland, Australia, it is the fourth most common cause of NTM pulmonary disease (behind M. intracellulare, M. avium and M. abscessus) (Thomson, 2010).

The typical presentation of disease is a cavitary upper lobe pulmonary disease, affecting middle-aged men, often with underlying COPD and heavy alcohol intake. Underlying malignancy, HIV co-infection and silicosis are known risk factors for disease. The nodular and bronchiectatic form of pulmonary disease seen with M. avium complex has also been described with M. kansasii. Soft tissue infection can also occur, though this is rare (Evans, Colville, Evans, Crisp, Johnsto, n., 1996; Bloch et al., 1998; Maliwan & Zvetina, 2005)

The reservoir of M. kansasii is still not well understood, though it has been postulated that water is the natural habitat. Whilst M. kansasii has been readily isolated from tap water, (McSwiggan & Collins, 1974; Engel et al., 1980b; Steadham, 1980) it has only been rarely isolated from natural water sources such as rivers or lakes. It has been rarely recovered from soil samples or animals. A definitive link between an environmental source of infection and human disease has not been shown, nor has human-to-human transmission.

156

Seven subtypes of M. kansasii have been described using amplification of the 16s- 23s r RNA spacer region, and PCR-restriction fragment length polymorphism (RFLP) analysis of the hsp65 gene (Iinuma et al., 1997; Picardeau, Prod’hom, Raskine, LePennec & Vincent, 1997). Type I is the most prevalent clinical isolate worldwide (Alcaide et al., 1997; Taillard et al., 2003; Santin et al., 2004). In Queensland typing was performed on 140 stored clinical isolates of M. kansasii. Type I was again the most dominant, with other types only evident in the last 2 decades. (Psaltis, 2005, Conference presentation) A summary of the genotypes found in Queensland over this 20 year period is shown in Figure 7.1.

Genotypes 1975-2004

15 Type VI Type V 10 Type IV Number 5 Type I

1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Year

Figure 7.1. ITS-RFLP Strain typing of M. kansasii isolates from Queensland patients between 1975 and 2004. (Psaltis 2005, conference presentation)

In Queensland, the incidence of disease due to NTM has been increasing (Thomson, 2010) and between 2007-2008 water sampling was conducted to investigate the presence of pathogenic NTM in potable water (Thomson et al., 2013). A study of NTM patients’ homes was also conducted (Thomson et al 2013b, manuscript under review). The aim of this study was to compare M. kansasii strains found in potable water and patient houses to clinical isolates from QLD patients. In Queensland NTM disease remains notifiable and a central reference laboratory performs speciation of all culture positive isolates, providing a unique opportunity to capture patient isolates from the whole state. This paper reports the comparison of these water

157

isolates with human isolates collected from patients in Queensland during the same time period.

METHODS

During 2007-2008 citywide water sampling of approximately 220 sites (Trunk Main, Reservoir and Distribution samples) was performed and 49 isolates were identified as M. kansasii using 16s rRNA and hsp 65 gene fragment sequencing. A home sampling study of water, swabs and aerosols from NTM patients’ houses was also conducted in 2009-10 and 16 isolates of M. kansasii were recovered. M. kansasii ATCC strain 12478 was included for comparison. Water isolates had been stored in Dubos broth at -20°C and were thawed and sub-cultured onto 7H11 plates as well as Löwenstein-Jensen slopes and incubated at 35°C until sufficient growth was available.

A total of 83 isolates were recovered from 71 patients– 10 patients had 2 isolates each, and 1 patient had 3 isolates. Human samples submitted to the MRL had been digested and decontaminated using 4% NaOH, neutralised with phosphoric acid and centrifuged at 3000g to concentrate the acid-fast bacilli (AFB). Smears were prepared from the sediment and stained by the Ziehl-Niehlsen method. One Löwenstein-Jensen slope (±pyruvate) and 7ml Mycobacterial Growth Indicator Tube (MGIT) were inoculated and incubated at 35oC until growth was detected. ZN staining of colonies confirmed Acid Fast Bacilli (AFB). Multiplex PCR (Wilton and Cousins, 1992) was performed to discriminate between M. tuberculosis, M. avium, M. intracellulare, M. abscessus and other Mycobacterium spp. Isolates identified as other Mycobacterium spp were further speciated using Hain Life Sciences GenoType® Mycobacterium AS (additional species) kit (2004-7 only) and/or 16S rRNA sequencing in conjunction with phenotypic characteristics. Clinical information on the patients was obtained from the NTM database at the QLD Tuberculosis Control unit.

158

The clonality of clinical and water M. kansasii isolates was determined using a rep- PCR based method (Diversilab® system, bioMerieux, Melbourne, Australia). DNA was extracted from clinical and water isolates using the Ultraclean Microbial DNA Isolation Kit (Mo Bio Laboratories, CA, USA). PCR mixture was prepared using AmpliTaq polymerase and PCR buffer (Applied Biosystems, New Jersey, USA) and Mycobacterium Diversilab®- primer mix according to the manufacturer’s instructions. Separation and detection of Rep-PCR products was performed by micro-fluidic chips of the Diversilab® System. Fingerprints were analysed with Diversilab® software v.3.4.38 using the Pearson correlation co-efficient and unweighted pair group method with arithmetic means to compare isolates and determine clonal relationship. This method emphasizes peak intensities rather than than peak presence/absence and was selected based on the appearance of the virtual gel images generated and as recommended by the manufacturer (bioMerioux, Australia). A cutoff of 97% strain similarity was used to define indistinguishable strains, those with 95% similarity were considered related, and those <95% different. A large cluster of indistinguishable isolates determined using the Pearson analysis, was further analysed using the Extended Jaccard method that weighs the presence or absence of peaks more so than intensities.

RESULTS

The mean age of the 71 patients was 64.40 years (SD 18.47; Range 5-95 years). Forty-one (57.7%) were male. The majority (90.14%) of patients had pulmonary infection, four had soft tissue isolates and one isolate came from a child’s lymph node.

Using a similarity cut off of 97%, there were 14 strain types identified amongst all clinical isolates. Figure 7.2a The majority of clinical strains were of type CP (Clinical pattern) 11 – a highly clonal strain (presumed Type I on ITS_RFLP). Four isolates were labeled strain type CP4, three CP12, two CP1 and two CP3 with the remaining strain types consisting of single isolates only (including the ATCC strain, CP6). The strains belonging to clusters CP1-4 were related (95% similarity), as were the strains of CP 11 and 12. The remaining clusters differed.

159

C1 C2 C1b

C3 C4 C5 C6 C7 C8 C9

CP11

CP12 CP13 CP14

Figure 7.2a. Rep-PCR dendrogram of clinical isolates of M. kansasii. Strains are grouped according to genetic similarity of >97%. The majority of isolates (16-79 – CP11) cluster together in a highly clonal group.

160

The relationship between clusters is demonstrated on the scatterplot in Figure 7.2b

Figure 7.2b. Scatterplot of M. kansasii clinical isolates with gridline spacing 5% similarity. Of 14 unique strain types (patterns) the majority of clinical isolates clustered together as Clinical Pattern (CP) 11. Clusters CP1-4 and 11-12 were similar (≥95% similiarity)

Of the four soft tissue isolates – two were strain type CP11, one CP12 and one CP13. The lymph node isolate was CP11 and the remaining less common strain types were pulmonary isolates. There was no evidence of geographical clustering of strains across the state of Queensland.

The dominant clinical cluster CP11 was further analysed using the Extended Jaccard method. (Figure 7.3a and 7.3b) Using a similarity index of 97%, 13 strain patterns were observed. The dominant patterns were P1 containing 26 isolates, P3 (n=17), P4 (n=13), P2 (n=2). The remaining patterns consisted of single isolates only. The

161

Figures 7.3a and 3b. Dendrogram and scatterplot of clinical isolates from cluster CP11 with Extended Jaccard analysis. The coloured bars correlate to the different patterns in the dendrogram. The numbered isolates in the scatterplot are coloured according to geographic regional distribution. geographic distribution of these isolates is shown in the scatterplot. All three isolates from West Moreton (Ipswich) were P1 but isolates from all other areas were distributed across different groups.

162

There was significant heterogeneity amongst the water isolates, forming 28 patterns (WP1-28) Figure 7.4. The majority of water isolates were different to the clinical isolates, best appreciated on the scatterplot Figure 7.5. Of the water isolates

WP5

WP2

WP6 WP3 WP7

WP8

Figure 7.4. Dendrogram and scatterplot of 28 different patterns of M. kansasii water isolates. The gridline spacing on the scatterplot correlates with 5% similarity.

There were 2 isolates (Cluster WP5) that matched the dominant clinical strain type CP11. A dominant cluster of water isolates (WP2) matched a single patient isolate (CP9). The remaining water and clinical isolates were unrelated.

163

WP2 CP9

CP11 WP5

Figure 7.5. Scatterplot showing differences between clinical and water isolates of M. kansasii. The colouring of the numbers correlates to the site of origin of the isolate as shown in the side legend (eg brown-potable water, Navy – pulmonary significant isolates).

DISCUSSION

Several typing methods have revealed significant heterogeneity of M. kansasii strains, yet a high degree of clonality of Type I strains exists; this is the most common clinical strain in most parts of the world. (Alcaide et al., 1997; Iinuma et al., 1997) In this study we have shown that automated rep PCR has sufficient discriminatory power to similarly demonstrate heterogeneity of clinical and environmental isolates. We have also shown the majority of clinical isolates of M. kansasii found in Queensland are highly clonal using rep PCR, and most likely correlate to the Type 1 strain identified using ITS_RFLP in a previous study (Psaltis 164

2005, conference presentation). Whilst the study conducted on M. kansasii clinical isolates from QLD using ITS_RFLP was from an earlier time period, it would appear that either rep-PCR is more discriminatory as a typing tool, or that there has been further evolution/acquisition of clinical strain types over time.

Using the Extended Jaccard analysis tool it is possible to identify strain variation within this dominant highly related group of clinical isolates. PFGE of M. kansasii DNA digested with DraI revealed limited polymorphism within Type I isolates from Spain and Switzerland, though did reveal significant polymorphism amongst Type II isolates. Large restriction fragment PFGE (Iinuma, et al. 1997) digested with VspI revealed 21restriction patterns when applied to clinical isolates from Japan. Eighteen of the 21 types formed a cluster showing genetic relatedness with a similarity coefficient of ≥0.87. The other three were divergent from the cluster and from each other. Amplified-fragment length polymorphism (AFLP)(Gaafar et al., 2003) of 253 Type 1 clinical strains, using the combination of patterns generated by three primers, identified 12 clusters. The majority of isolates however were members of 5 clusters (C122, n=121; C111, n=59, C222; n=51; C221; n=12; C112, n=3). The remaining seven clusters contained only single isolates. In this study automated rep-PCR was comparable in discriminating strain patterns within the clinical strains assumed to be Type I.

In comparison to clinical isolates, potable water isolates showed greater diversity. Relative to the number of strains of M. kansasii that were isolated from potable water, there were very few that matched clinical isolates. M. kansasii is not a common infection in Queensland (186 patient notifications between 2001-2010, Population 4 million; 0.465/100 000/year) and this may be a reflection of the low burden of pathogenic strains within the water supply. However clinical isolates were included from a variety of locations across the state, and water was only sampled from the distribution system of the capital city. The possibility that the water supply in other locations contains pathogenic strains is still possible. It is also possible that at certain points the water becomes contaminated by another

165

environmental source and hence there may be another environmental niche for this pathogen.

This possibility is supported by reports of high incidence areas of M. kansasii disease associated with the mining industry in South Africa, (Engel et al., 1980a; Falkinham, 2010), industrial zones of Czechoslovakia (Chobot et al., 1997) (particularly mining and iron manufacture) (Kubin, Svanclova, Medek, Chobot & Olsovsky, 1980) and the Japanese coastal industrial areas (Iinuma et al., 1997). In the Karvina district of Czechoslovakia, M. kansasii was isolated from 7% of drinking water samples, and 43.7% industrial water samples and scrapings between 1971 and 1995, during which the annual incidence of disease was 12.03/100000 with a very high incidence of disease in miners (Chobot et al., 1997).

Engel (Engel, et al., 1980a) applied phage typing of M. kansasii isolates to compare strains from clinical and environmental specimens from The Netherlands, Britain and Czechoslovakia. Of the Czech strains, Phage type 13 accounted for 25/28 (89.3%) of the clinical isolates from the Karvina coalmine district. This phage type was also found in 28 of 31 (90.4%) environmental isolates (water from the distribution system of the mines), and 12/18 (66.6%) clinical isolates from the rest of the country. This phage type also dominated clinical isolates from the UK, though only 1 of 7 environmental isolates was of this type, and this was isolated from soil in Uganda. Given developments in the field of strain typing, it has been suggested that this technique did not have sufficient discriminatory power to adequately compare the clinical and environmental strains. Phage typing has not been widely applied because it is labor intensive, and technical problems can result in poor reproducibility and interpretation of results (Picardeau et al., 1997).

M. kansasii Type I was not present among 113 environmental strains isolated in Germany (Alcaide et al 1997), though was found in water in France (Picardeau et al, 1997). Picardeau (1997) characterized five subspecies of M. kansasii using a variety of strain typing methods, including RFLP with the major polymorphic tandem repeat (MPTR) and IS1652 probe, PFGE, AFLP and PRA of the hsp-65 gene. Both clinical (38)

166

and water (24) isolates from France were included in this study. The clinical isolates came from patients from 8 different French towns. The water isolates were recovered from tap water from 6 different hospitals in and around Paris. Whilst there were water isolates and clinical isolates that had the same strain pattern using different methods of typing, only one set of isolates was epidemiologically related. This set consisted of identical isolates from patients from a single family living in the same house, and a water isolate recovered from their home. Five other sets of isolates were considered epidemiologically related but there was no evidence of nosocomial infection or water transmission. The molecular studies also showed that strains isolated from tap water at a single hospital were not all identical, consistent with our finding of a number of diverse strains within the water supply, only some of which may be associated with human infection.

In summary we have demonstrated significant heterogeneity amongst clinical and environmental strains of M. kansasii using automated rep-PCR. The clonality of the dominant clinical strain is in keeping with previous studies, yet rep-PCR provided sufficient discriminatory power to identify strain types within this group. Few water strains were identical to clinical strains, suggesting that if water or aerosol transmission is the mode of infection, then point source contamination likely occurs from an alternative environmental source.

Acknowledgements: This study was funded by grants from the Gallipoli Medical Research Foundation of Greenslopes Private Hospital and by The Prince Charles Hospital Foundation.

167

REFERENCES

Alcaide, F., Richter, I., Bernasconi, C., Springer, B., Hagenau, C., & Telenti, A. (1997) Heterogeneity and clonality among isolates of Mycobacterium kansasii:implications for epidemiological and pathogenicity studies. Journal of Clinical Microbiology, 35(8): 1959-1964. Bloch, K. C., Zwerling, L., Pletcher, M. J., Hahn, J. A., Gerberding, J. L., Ostroff, S. M.,…Reingold, A. L. (1998). Incidence and clinical implications of isolation of Mycobacterium kansasii: Results of a 5-Year, population-based study. Annals of Internal Medicine, 129: 698-704. Chobot, S., Malis, J., Sebakova, H., Pelikan, M., Zatloukal, O., Palicka, P., & Kocurova, D. (1997). Endemic Incidence of Infections caused by Mycobacterium kansasii in the Karvina District in 1968-1995. Central European Journal of Public Health, 5 5(4): 164-173. Dailloux, M., Abalain, M.L., Laurain, C., Lebrun, L., Loos-Ayav, C., Lozniewski, A., Maugein, J and the French Mycobacteria Study Group (2006). Respiratory infections associated with nontuberculous mycobacteria in non-HIV patients. European Respiratory Journal, 28(6): 1211-1215. Engel, H.W.B., Berwald, L.G., Grange, J.M., & Kubin, M. (1980) Phage typing of Mycobacterium kansasii. Tubercle, 61(1): 11-19. Engel, H.W.B., Berwald, L.G., & Havelaar, A.H. (1980) The occurrence of Mycobacterium kansasii in tapwater. Tubercle, 61(1): 21-26. Evans, S. A., Colville, A., Evans, A.J., Crisp,A.J., & Johnston, I.D.A. (1996). Pulmonary Mycobacterium kansasii: comparison of the clinical features, treatment and outcome with pulmonary tuberculosis. Thorax, 51: 1248-1252. Falkinham, J. O. (2010). Impact of human activities on the ecology of nontuberculous mycobacteria. Future Microbiology, 5(6): 951-960. Gaafar, A., Unzaga, M.J., Cisterna, R., Clavo, F.E., Urra, E., Ayarza, R., & Martin, G. (2003). Evaluation of a modified single-enzyme amplified-fragment length polymorphism technique for fingerprinting and differentiating of Mycobacterium kansasii Type I isolates. Journal of Clinical Microbiology, 41(8): 3846-3850.

168

Hoefsloot, W., van Ingen*, J., Andrejak, D., Ängeby, K., Bauriaud, R., Bemer, P., … Wagner, D for NTM-net. (2013). A snapshot of the geographic diversity of nontuberculous mycobacteria isolated from pulmonary samples: An NTM- net collaborative study. European Respiratory Journal In Press. Iinuma, Y., Ichiyama, S., Hasegawa, Y., Shimokata, K., Kawahara, S., & Matsushima, T. (1997). Large-restriction-fragment analysis of Mycobacterium kansasii genomic DNA and its application in molecular typing. Journal of Clinical Microbiology, 35(3): 596-599. Kubin, M., Švanclová, E., Medek, B., Chobot. S., & Olšovský, Z. (1980). Mycobacterium kansasii infection in an endemic area of Czechoslovakia. Tubercle 61(4): 207-212. Maliwan, N., & Zvetina, J.R. (2005) Clinical features and follow up of 302 patients with Mycobacterium kansasii pulmonary infection: a 50 year experience. Postgraduate Medical Journal, 81: 530-533. McSwiggan, D., Collins, C. (1974) The isolation of M.kansasii and M.xenopi from water systems. Tubercle 55: 291-297. Picardeau, M., Prod'hom, G., Raskine, L., LePennec, M.P., & Vincent, V (1997). Genotypic characterization of five subspecies of Mycobacterium kansasii. Journal of Clinical Microbiology, 35(1): 25-32. Psaltis, J. (2005). Mycobacterium kansasii: A Queensland Mycobacterium Reference Laboratory Review 1975-2004. Conference Presentation, Australian Society of Microbiology Annual Scientific Meeting, Canberra, Australia. Santin, M., Alcaide, F., Benitez, M., Salazar, A., Ardanuy, C., Podzamczer, …D., & Gudiol, F., (2004) Incidence and molecular typing of Mycobacterium kansasii in a defined geographical area in Catalonia, Spain. Epidemiology of Infection, 132: 425-432. Steadham, J. E. (1980). High-catalase strains of Mycobacterium kansasii isolated from water in Texas. Journal of Clinical Microbiology, 11(5): 496-498. Taillard, C., Greub, G., Weber, R., Pfyffer, G.E., Bodmer, T., Zimmerli, S.,… & Prod'hom, G, (2003). Clinical Implications of Mycobacterium kansasii Species Heterogeneity: Swiss National Survey. Journal of Clinial Microbiology, 41(3): 1240-1244.

169

Thomson, R. (2010). Changing epidemiology of pulmonary nontuberculous mycobacteria infections. Emerging Infectious Diseases, 16(10): 1576-1582. Thomson, R., Carter, R., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M (2013). Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia. BMC Microbiology, 13;89 Thomson, R., Tolson, C., Carter, R., Coulter, C., Hargreaves, M., Huygens, F (2012). Isolation Of Nontuberculous Mycobacteria (NTM) From Household Water And Shower Aerosols In Patients With NTM Pulmonary Disease. American Journal of Respiratory and Critical Care Medicine, 185: A4030. Wilton, S., & Cousins, D. (1992). Detection and identification of multiple mycobacterial pathogens by DNA amplification in a single tube. PCR Methods and Applications. 4: 269-273.

170

171

Chapter 1:Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1: Objective 1 Aim 3: Objective 5 Aim 2: Objectives 3&4 Chapter 3: Chapter 9: Isolation of Development of Chapter 5: NTM from homes of Mycobacterium method for isolation of patients with NTM lentilavum NTM from water disease

Aim 1:Objective 2 Chapter 4: Isolation of Chapter 6: NTM from Brisbane M. abscessus water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

172

CHAPTER 8: HETEROGENEITY OF CLINICAL AND ENVIRONMENTAL ISOLATES OF M. FORTUITUM USING REPETITIVE UNIT PCR: MUNICIPAL WATER AN UNLIKELY SOURCE OF COMMUNITY ACQUIRED INFECTIONS.

This paper has been submitted to Epidemiology and Infection, HYG-OM-5035-Jun- 13. A third version with minor changes submitted November 2013 -Appendix F. Thomson, R., Tolson, C., Carter, R., Huygens, F., & Hargreaves, M. Heterogeneity of clinical and environmental isolates of M. fortuitum using repetitive unit PCR: municipal water an unlikely source of community acquired infections.

Author contributions RT (60%) designed the study, coordinated the collection of samples, participated in the processing of water samples, collated and analysed the data, and wrote the manuscript. CT (15%) processed water samples, performed sequencing, regrew clinical isolates, performed DNA extraction and rep-PCR experiments RC (15%) performed PFGE and rep-PCR experiments and contributed to the writing of manuscript FH (5%) intellectually contributed to the study design and methodology and the review of the manuscript MH (5%) intellectually contributed to the study design and methodology and the review of the manuscript

ABSTRACT

M. fortuitum is a rapidly growing mycobacteria associated with community acquired infections and nosocomial wound and soft tissue infections. It can cause pulmonary disease and contaminate pulmonary specimens. The environmental niche for this organism is not known, though it has been suspected that water has been the source of infection especially in the hospital setting. The aim of this study was to compare environmental water isolates to those human isolates associated with community acquired and nosocomial infections. Methods: Nineteen strains of M.

173

fortuitum were recovered from municipal water and one strain from a household shower aerosol. Fifty-three patient isolates were available for comparison. PFGE and rep-PCR strain typing were performed on 10 isolates to validate rep-PCR , then the remaining isolates were analysed using rep-PCR (Diversilab). Results: A wide variation in strain types was identified, with thirteen clusters of ≥2 indistinguishable isolates, and 28 patterns consisting of individual isolates. The clusters could be grouped into seven similar groups (>95% similarity). None of the clinical isolates clustered with any of the water isolates. Conclusions: Municipal water and clinical isolates consist of different strain types, making municipal water an unlikely source of human infection. Point source contamination of water from an alternative source such as soil, may account for outbreaks where water is the suspected source of infection.

INTRODUCTION

Mycobacterium fortuitum is a non-pigmented rapidly growing mycobacterium, often associated with skin and soft tissue infections, though pulmonary disease can occur (De Groote & Huitt, 2006). Wound infections (Kuritsky et al., 1983), post- injection abscesses, and post pedicure infections (Vugia, et al., 2005) are well described. M. fortuitum is also a recognised cause of nosocomial infections (Burns et al., 1991) and pseudo-outbreaks (Wallace, Brown, & Griffith, 1998). It is usually considered that municipal and hospital water are the source of these outbreaks, though M. fortuitum is also found in soil and house dust (Dawson, 1971). M. fortuitum has been found in municipal water (Carson et al., 1988; Castillo-Rodal et al., 2012; Le Dantec et al., 2002b) and is relatively resistant to chlorine and other disinfectants (Cortesia, Lopez, de Waard, & Takiff, 2010; Le Dantec et al., 2002a). Whilst there have been many outbreaks reported where water has been the suspected vehicle of transmission, very few have been realised in a timely fashion that would allow full environmental sampling to adequately document the source of the infection. (Kuritsky et al., 1983; Quiñones et al., 2010; Sampaio, Viana-Niero, de Freitas, Hfling-Lima, & Leao, 2006; Wallace et al., 1989; Wallace et al., 1998)

174

In order to determine if strains of M. fortuitum resident in municipal drinking water in Brisbane, Australia were a likely source of infections with this organism in the community, we undertook a prospective study to compare isolates from municipal water and infected patients resident in the area supplied by that water during the same time period (2007-2009).

METHODS

During 2007-2008 citywide water sampling of approximately 220 sites (Trunk Main, Reservoir and Distribution samples) was performed and 19 isolates were identified as M. fortuitum using 16S rDNA and fragment sequencing (Thomson, Carter, Tolson, Huygens, & Hargreaves, 2013). A home sampling study of water, swabs and aerosols from NTM patient houses was also conducted in 2009-10 and 1 isolate of M. fortuitum was recovered from a shower aerosol. (Thomson et al., 2012) Water isolates had been stored in Dubos broth at -20°C and were thawed and sub-cultured onto 7H11 plates as well as Löwenstein-Jensen slopes and incubated at 35°C until sufficient growth was available.

Human samples from patients whose residential address was within the water distribution catchment, were submitted to the Mycobacterium Reference Laboratory during the same time period (2007-2009) and had been digested and decontaminated using 4% NaOH, neutralised with phosphoric acid and centrifuged at 3000g to concentrate the acid-fast bacilli (AFB). Smears were prepared from the sediment and stained by the Ziehl-Niehlsen method. One Löwenstein-Jensen slope (±pyruvate) and 7ml Mycobacterial Growth Indicator Tube (MGIT) were inoculated and incubated at 35°C until growth was detected. ZN staining of colonies confirmed Acid Fast Bacilli. Multiplex PCR (Wilton & Cousins, 1992) was performed to discriminate between M. tuberculosis, M. avium, M. intracellulare, M. abscessus and other Mycobacterium spp. Isolates identified as other Mycobacterium spp were further speciated using Hain Life Sciences GenoType® Mycobacterium AS (additional species) kit (2004-7 only) and/or 16S rRNA sequencing in conjunction

175

with phenotypic characteristics. Patient clinical information was obtained from the NTM database at the QLD Tuberculosis Control unit. Additional isolates from patients living outside the water distribution network were included for comparison.

The clonality of clinical and water M. fortuitum isolates was determined using a rep- PCR based method (Diversilab® system, bioMerieux, Melbourne). DNA was extracted from clinical and water isolates using the Ultraclean Microbial DNA Isolation Kit (Mo Bio Laboratories, CA, USA). The PCR mixture was prepared using AmpliTaq polymerase and PCR buffer (Applied Biosystems, New Jersey, USA) and Mycobacterium Diversilab®- primer mix according to the manufacturer’s instructions. Separation and detection of rep-PCR products was performed by micro-fluidic chips of the Diversilab® System. Fingerprints were analysed with Diversilab® software v.3.4.38 using the Pearson correlation co-efficient and unweighted pair group method with arithmetic means to compare isolates and determine clonal relationship. To validate the method, multiple isolates from the same patient were run in parallel, and Pulsed field gel electrophoresis (PFGE) was performed on 10 clinical isolates and a control strain and results compared to the patterns generated by automated rep-PCR. Based on the Tenover classification of isolates using PFGE, the Diversilab rep-PCR similarity cut offs were determined as >97% (indistinguishable), >95% similar, and <95% different.

PFGE was performed using the method outlined in the BioRad Genpath® Group 6 Kit (BioRad France) with modifications outlined by Mazurek et al (1993) and Burki et al (1995). Organisms were inoculated into 10 ml Middlebrook 7H9 broth (Difco, Becton Dickinson and Company, Sparks, MD USA in-house media) supplemented with 0.2% OADC (Difco, Becton Dickinson and Company, MD USA), 0.1% Tween 80 (MP Biochemicals LLC Ohio USA, cycloserine (1mg/ml Sigma-Aldrich INC St Louis MO USA) and ampicillin (0.1mg/ml Sigma-Aldrich INC St Louis MO USA) and incubated for three days. One ml of broth was centrifuged and the supernatant discarded. Gel plugs were prepared and incubated in 500µl of Lysis Buffer 1 and 20 µl Lysozyme (25mg/ml) at 36°C. Following a wash step, 500µl Proteinase K buffer and

176

20 µl Proteinase K (>600U/ml) were added to each sample. Plugs were incubated for 48 hours at 50°C. The plugs were then washed four times in 1xWash Buffer. After the final wash, the plugs were stored in1xWash Buffer. Digestion was performed using, Xba 1 Enzyme (10U/ml) and the samples were incubated for 18 hours at 36°C.

The plugs were loaded into wells of a 1% PFGE agarose gel (BioRad) electrophoresis gel ensuring there were no air bubbles. Sufficient 0.5xTBE was added to the electrophoresis cell and cooled to 14°C and electrophoresis performed using the following parameters: Initial A time one second, Final A time 40 seconds, Voltage 200 V and time 22 hours. Following electrophoresis, the gel was stained using Ethidium bromide (BioRad), de-stained in running distilled water for 30 minutes and then photographed.

RESULTS

The geographic distribution of water sites and patients residential addresses is demonstrated in Figure 8.1. The clinical isolates were from 53 patients, 27(50.9%) of whom were male. There were 27 pulmonary isolates (50.9%; 24 considered not to be causing invasive disease, 3 significant disease according to ATS/IDSA criteria (Griffiths, D., et al, 1007)). Twelve patients had underlying bronchiectasis and 4 had cavities on chest radiograph. There were 26 soft tissue isolates (5 insulin injection site infections, 3 laparoscopic band infections, 2 isolates from blood associated with line infections, 2 surgical wound infections, and 14 other community acquired soft tissue infections).

177

Figure 8.1. Map of the Brisbane area with water site locations in blue and patient residential locations in red.

Figure 8.2. Rep-PCR of M. fortuitum isolates (numbered 1-10) associated with laparoscopic gastric band infections demonstrating indistinguishable isolates for each patient (same pattern (P) colour) and the degree of difference between patients (lower graphic, percentage similarity in boxes). 178

AM AM AM KG KG RB RB PN ML WT

Figure 8.3. PFGE patterns of 10 M. fortuitum isolates associated with laparoscopic gastric band procedures (same isolates as for Figure 1). Lanes 1-3: isolates from patient AM, lanes 4-5, patient KG; lanes 6-7 patient RB; lane 8 patient PN; lane 9 patient ML and lane 10 wild type control strain.

PFGE and rep-PCR were performed on 10 isolates of M. fortuitum associated with laparoscopic gastric band infections. (Wright et al, 2013: Manuscript submitted- Supplemental material) Isolates from the same patient that were indistinguishable with PFGE, were associated with a >97% similarity cut off on rep-PCR. (Figures 8.2 and 8.3) This concurred with the manufacturers recommendation and was then selected as the cutoff for analysis of the remaining isolates using rep-PCR. Three of the water isolates had patterns generated on rep-PCR that had low intensity bands and were removed from the analysis. There was a wide variation in strain types evident with 41 different patterns generated by 70 isolates. (Figure 8.4) These formed thirteen clusters (P1-13) of 2 or more isolates; the remaining patterns consisted of single isolates only (P14-41). These clusters were further grouped according to similarity of approximately 95% into seven groups. Of the larger clusters, P4 consisted of nine clinical isolates, eight of which were associated with soft tissue infections. P5 (a cluster similar to P4) however, consisted of six isolates, five of which were pulmonary isolates and the other associated with an injection site infection. 179

There was no evidence of geographical clustering of isolates apart from the two isolates making up P6 that came from patients living in adjacent suburbs. A husband and wife both presented with soft tissue infections of the lower limbs and had indistinguishable isolates (P4; #21 and #22). The water isolate groups were distinctly separate from the clinical isolates, best appreciated by viewing the scatterplot in Figure 8.5.

Figure 8.5. Scatterplot of numbered M. fortuitum isolates, colour coded according to site of infection, with water isolates in pink (side legend). Gridline spacing correlates with 5% similarity. (Pulmonary NS=non clinically significant isolates, Sig = clinically significant)

180

Figure 8.4a. Full rep-PCR dendrogram of M. fortuitum water and clinical isolates. Red interrupted line of similarity at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into similar groups (≥95% similarity).

181

Figure 8.4b. Full rep-PCR dendrogram of M. fortuitum water and clinical isolates Red interrupted line of similarity at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into similar groups (≥95% similarity).

182

Figure 8.4c. Full rep-PCR dendrogram of M. fortuitum water and clinical isolates. Red interrupted line of similarity at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into similar groups (≥95% similarity).

DISCUSSION

The aim of this study was to determine if strains of M. fortuitum resident in the municipal water were likely to be the source of sporadic infections with M. fortuitum in the community. The correlation with PFGE, and the wide variety of

183

strain types identified, confirm the utility of rep-PCR as a discriminatory tool for this species. There were no patient isolates that were indistinguishable from water isolates. There did appear to be a dominant group of clinical strains, with a degree of nonsignificant clustering according to infection type (P4 predominantly soft tissue and P5 predominantly pulmonary) but this finding may have occurred by chance as the clusters P4 and P5 were similar. The water strains in contrast formed 3 separate groups that differed from all clinical isolates.

Other studies have demonstrated heterogeneity amongst the species M. fortuitum using a variety of molecular techniques, including PFGE, (Legrand, Radegonde, Goh, & Rastogi, 2002; Sampaio, et al., 2006; Sniezek et al., 2003; Wallace et al., 1989; Winthrop et al., 2002) 16s-23s rDNA ITS genotyping, RAPD and ERIC-PCR (Sampaio, et al., 2006) mainly in the investigation of nosocomial outbreaks. Legrand (2002) examined 51 isolates of M. fortuitum from 47 patients in Guadeloupe, Martinique and French Guiana between 1996-99, suspecting a microepidemic. Only 5 isolates were found to cluster, confirming significant genomic heterogeneity amongst clinical strains and suggesting a “diversity of ecological niches in which this organism may develop”.

The heterogeneity of strains we have demonstrated, suggest that if water is the vehicle for transmission of infection, then point source contamination probably occurs from another environmental reservoir such as soil or dust. Such soil organisms can readily enter water distribution systems via cracks in pipes caused by tree roots etc, or may contaminate surface source water and survive disinfection.

Many surgical outbreaks reported in the literature have occurred in the midst of long periods without NTM infections. If municipal water was the source of organisms for these outbreaks, then it could be postulated that a transient drop in disinfection or an increase in contamination must have occurred. It seems more likely that episodic point source contamination may be responsible.

184

The susceptibility of M. fortuitum to quaternary ammonium disinfection was studied by Cortesia et al (2010), and there were phenotypic and presumably genotypic changes in those strains that persisted after exposure to disinfection. It is possible that the differences we have noted between clinical and water strains has occurred because we have only detected strains that have survived exposure to disinfection in municipal water, and that strains responsible for some clinical infections are protected by survival in amoebae or biofilm inside pipes and hence were not detected by our sampling approach. The changes that occur in strains upon exposure to the chlorine/chloramine disinfection used in drinking water has not been studied adequately to permit comment on whether this theory is plausible.

One of the first reports of nosocomial transmission of M. fortuitum was published by Burns (1991) who investigated a cluster of cases of positive sputum cultures amongst patients in an alcoholic rehabilitation ward. The common factors amongst cases, was usage of 2 ward showers, and M. fortuitum was isolated from the tap water connecting to the showers but not from the showers themselves. Using PFGE the 16 case isolates were found to be identical to the water isolate. However, there were no other parts of the hospital where such cases were identified and no further cases were reported after the showers had been disconnected and decontaminated.

Wallace investigated the sources of M. fortuitum associated with infections following cardiac bypass surgery (Wallace et al., 1989) across 12 states in the USA. In the analysis of a particular outbreak in Texas in 1981, one isolate of M. fortuitum was recovered from an operating room waterbath and had a similar phenotype and enzyme electrophoretic pattern as a cardiac associated isolate. A similar isolate was also recovered from the municipal water coming into the hospital and from an ice machine on one of the hospital floors. Two patients with other surgical wound infections (neck and abdominal) were infected with the same strain. In the 5 years following that outbreak an additional 21 clinical isolates of M. fortuitum were

185

recovered from patients in the same hospital. Six were studied. All differed in phenotypic markers or plasmid profile from the outbreak isolates. In a separate outbreak (Colorado 1976) an isolate of M. fortuitum recovered from a settle plate in an operating room had the same phenotype and electrophoretic pattern as four of the epidemic strains.

Winthrop et al (2002) reported 110 cases of furunculosis due to M. fortuitum acquired following pedicures at a single nail salon in California. All patients had their feet and lower legs soaked in a whirlpool footbath. Large amounts of hair and skin debris were found behind the inlet suction screens of the whirlpool footbath and M. fortuitum was cultured from these areas from all 10 footbaths. The authors concluded that M. fortuitum had entered via the salon tap water, seeded the accumulated organic debris behind the footbath inlet screens then multiplied and circulated within the footbath basin. They felt that because all of the footbaths cultured rapid-growers including multiple strains of M. fortuitum that it was unlikely the baths were contaminated by a client. On the basis of our results and the fact that NTM are so prevalent in soil, one could argue that in the absence identical strains of M. fortuitum being discovered in the salon tap water (M. chelonae/abscessus only isolated), that it is highly likely the footbaths were contaminated by soil or organic matter from the feet of clients. The salon owner reported that these inlet suction areas were never cleaned.

In summary we have compared isolates of M. fortuitum resident in a municipal drinking water distribution system in Brisbane 2007-2009, to those found to cause community and medically acquired infections in patients who receive their ware supply from this system and found significant strain heterogeneity. From these results and appraisal of previous studies, it would seem that most community infections probably originate in soil/dust, and nosocomial outbreaks occur as a result of point contamination of water sources. However further investigation of strain types found in soil and other environmental sources are needed to prove this is the case.

186

REFERENCES

Burki, D.R., Bernasconi, C, Bodmer, T., & Telenti, A. (1995). Evaluation of the relatedness of strains of Mycobacterium avium using pulsed-field gel electrophoresis. European Respiratory Journal, 14(3), 212-217. Burns, D.N., Wallace, R. J. Jr, Schultz, M. E., Zhang, Y. S., Zubairi, S. Q., Pang, Y. J., … Gordin FM. (1991). Nosocomial outbreak of respiratory tract colonization with Mycobacterium fortuitum: demonstration of the usefulness of pulsed- field gel electrophoresis in an epidemiologic investigation. American Review of Respiratory Diseases, 142(2), 468-470. Carson, L. A., Bland, L. A., Cusick, L. B., Favero, M. S., Bolan, G. A., Reingold, A. L., & Good, R. C. (1988). Prevalence of nontuberculous mycobacteria in water supplies of hemodialysis centers. Applied and Environmental Microbiology, 54(12), 3122-3125. Castillo-Rodal, A., Mazari-Hiriart, M., Lloret-Sánchez, L., Sachman-Ruiz, B., Vinuesa, P., & López-Vidal, Y. (2012). Potentially pathogenic nontuberculous mycobacteria found in aquatic systems. Analysis from a reclaimed water and water distribution system in Mexico City. European Journal of Clinical Microbiology & Infectious Diseases, 31(5), 683-694. Cortesia, C., Lopez, G. J., de Waard, J. H., & Takiff, H. E. (2010). The use of quaternary ammonium disinfectants selects for persisters at high frequency from some species of non-tuberculous mycobacteria and may be associated with outbreaks of soft tissue infections. Journal of Antimicrobial Chemotherapy, 65(12), 2574-2581. Dawson, D. J. (1971). Potential pathogens among strains of mycobacteria isolated from house-dusts. Medical Journal of Australia, 1(13), 679-681. De Groote, M-A., & Huitt, G. (2006). Infections due to rapidly growing Mycobacteria. Clinical Infectious Diseases, 42(12), 1756-1763. Girffith, D.E., Aksamit, T., Brown-Elliott, B.A., Catanzaro, A., Daley, D., Gordin, F., et al (2007). An Official ATS/IDSA Statement: Diagnosis, Treatment, and Prevention of Nontuberculous Mycobacterial Diseases. American Journal of Respiratory and Critical Care Medicine, 175 (4), 367-416

187

Kuritsky, J. N., Bullen, M. G., Broome, C. V., Silcox, V. A., Good, R. C., & Wallace, R. J. (1983). Sternal wound infections and endocarditis due to organisms of the Mycbobacterium fortuitum complex. Annals of Internal Medicine, 98(6), 938- 939. Le Dantec, C., Duguet, J-P., Montiel, A., Dumoutier, N., Dubrou, S., & Vincent, V. (2002a). Chlorine Disinfection of Atypical Mycobacteria Isolated from a Water Distribution System. Applied and Environmental Microbiology, 68(3), 1025-1032. Le Dantec, C., Duguet, J-P., Montiel, A., Dumoutier, N., Dubrou, S., & Vincent, V. (2002b). Occurrence of Mycobacteria in Water Treatment Lines and in Water Distribution Systems. Applied and Environmental Microbiology, 68(11), 5318-5325. Legrand, E., Radegonde, N., Goh, K. S., & Rastogi, N. (2002). A pulsed-field gel electrophoresis study of Mycobacterium fortuitum in a Caribbean setting underlines high genetic diversity of the strains and excludes nosocomial outbreaks. International Journal of Medical Microbiology, 292(1), 51-57. Mazurek, G. H., Hartman, S, Zhang, Y., Brown, B. A., Hector, J.S.R., Murphy, D., & Wallace, R. J., Jr. (1993). Large restriction fragment polymorphisms in the Mycobacterium avium-intracellulare Complex: a potential epidemiological tool. Journal of Clinical Microbiology, 31, 390-394. Quiñones, C., Ramalle-Gómara, E., Perucha, M., Lezaun, M. E., Fernández-Vilariño, E., García-Morrás, P., & Simal, G. (2010). An outbreak of Mycobacterium fortuitum cutaneous infection associated with mesotherapy. Journal of the European Academy of Dermatology and Venereology, 24(5), 604-606. Sampaio, J. L. M., Chimara, E., Ferrazoli, L., da Silva Telles, M. A., Del Guercio, V. M. F., Jerico, Z. V. N., . . . & Laeo, S. C. (2006). Application of four molecular typing methods for analysis of Mycobacterium fortuitum group strains causing post-mammaplasty infections. Clinical Microbiology & Infection, 12(2), 142-149.

188

Sampaio, J. L. M., Viana-Niero, C., de Freitas, D., Hfling-Lima, A. L., & Leao, S. C. (2006). Enterobacterial repetitive intergenic consensus PCR is a useful tool for typing Mycobacterium chelonae and Mycobacterium abscessus isolates. Diagnostic Microbiology and Infectious Diseases, 55(2), 107-118. Sniezek, P. J., Graham, B. S., Busch, H. B., Lederman, E. R., Lim, M. L., Poggemyer, K, . . . Winthrop, K. (2003). Rapidly growing mycobacterial infections after pedicures. Archives of Dermatology, 139, 629-634. Thomson, R. M., Carter, R., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2013). Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia. BMC Microbiology, 13;89 Thomson, R. M., Tolson, C., Carter, R., Coulter, C., Hargreaves, M., & Huygens, F. (2012). Isolation Of nontuberculous Mycobacteria (NTM) from household water and shower aerosols in patients with NTM pulmonary disease. American Journal of Respiratory and Critical Care Medicine, 185, A4030. Vugia, D. J., Jang, Y., Zizek, C., Ely, J., Winthrop, K. L., & Desmond E. (2005). Mycobacteria in nail salon whirlpool footbaths, California. Emerging Infectious Diseases, 11(4), 616-618. Wallace, R. J., Musser, J. M., Hull, S. I., Silcox, V. A., Steele, L. D., Forrester, G. D, . . . Selander, R. K. (1989). Diversity and sources of rapidly growing mycobacteria associated with infections following cardiac surgery. Journal of Infectious Diseases, 159(4), 708-716. Wallace, R. J., Brown, B. A., & Griffith, D. E. (1998). Nosocomial outbreaks/pseudooutbreaks caused by nontuberculous mycobacteria. Annual Review of Microbiology, 52(1), 453-490. Wilton, S, & Cousins, Debby. (1992). Detection and identification of multiple mycobacterial pathogens by DNA amplification in a single tube. PCR Methods and Applications, 4, 269-273. Winthrop, K., Abrams, M., Mitchell, Y., Schwartz, I., Ely, J., Gillies, D., & Vugia, D. (2002). An outbreak of mycobacterial furunculosis associated with footbaths at a nail salon. New England Journal of Medicine, 364(18), 1366-1371.

189

Chapter 1:Introduction, Hypothesis, Aims

Chapter 2 Literature Review

Aim 1: Objective 1 Aim 3: Objective 5 Aim 2: Objectives 3&4 Chapter 3: Chapter 9: Isolation of Development of Chapter 5: NTM from homes of Mycobacterium method for isolation patients with NTM lentilavum of NTM from water disease

Aim 1:Objective 2 Chapter 4: Isolation of Chapter 6: NTM from Brisbane M. abscessus water

Chapter 7: M. kansasii

Chapter 8: M. fortuitum

190

CHAPTER 9. ISOLATION OF NTM FROM HOUSEHOLD WATER AND SHOWER AEROSOLS IN PATIENTS WITH NTM PULMONARY DISEASE.

This paper has been presented in abstract form at the TSANZ (Canberra 2012) and ATS (San Francisco 2012). Thomson, R., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2012). Isolation of Nontuberculous mycobacteria from household water and shower aerosols in patients with NTM Pulmonary disease. American Journal of Respiratory and Critical Care Medicine, 185, A4030. (Appendix G1) The manuscript was published in the Journal of Clinical Microbiology July 10 ahead of print. (2013, 51(9):3006)

Author contributions

RT (70%) designed the study, recruited the patients, coordinated the collection of samples, visited the houses and processed the samples, collated and analysed the data, and wrote the manuscript. CT (10%): visited the houses, processed water samples, performed subcultures and sequencing and collated results. CC (2.5%) contributed to organisational matters and intellectual contribution FH (10%) intellectually contributed to the study design and methodology, coordinated and supervised the HRM experiments and the writing of the manuscript. MH (7.5%) intellectually contributed to the study design and methodology, and contributed to the writing of the manuscript.

ABSTRACT

It has been postulated that susceptible individuals may acquire Nontuberculous Mycobacterial (NTM) infection from water and aerosol exposure. This study examined household water and shower aerosols of patients with NTM pulmonary disease. Mycobacteria isolated from clinical samples from 20

191

patients included M. avium (5), M. intracellulare (12), M. abscessus (7), M. gordonae (1), M. lentiflavum (1), M. fortuitum (1), M. peregrinum (1), M. chelonae (1) M. triplex (1) and M. kansasii (1). Methods: One-litre water samples and swabs were collected from all taps, and swimming pools/rainwater tanks. Shower aerosols were sampled using Andersen six-stage cascade impactors. For a subgroup of patients, real time PCR was performed and high resolution melt profiles compared to those of ATCC control strains. Results: Pathogenic mycobacteria were isolated from 19 homes. Species identified in the home matched that found in the patient in seven (35%) cases – M. abscessus (3), M. avium (1), M. gordonae (1), M. lentiflavum (1), and M. kansasii (1). In an additional patient with M. abscessus infection, this species was isolated from potable water supplying her home. NTM grown from aerosols included M. abscessus (3 homes), M. gordonae (2), M. kansasii (1), M. fortuitum complex (4), M. mucogenicum (1) and M. wolinskyi (1). Conclusion: NTM causing human disease can be isolated from household water and aerosols. The evidence appears strongest for M. avium, M. kansasii, M. lentiflavum and M. abscessus. Despite a predominance of disease due to M. intracellulare we found no evidence for acquisition of infection from household water for this species.

INTRODUCTION

Nontuberculous mycobacteria are ubiquitous organisms responsible for progressive pulmonary disease. Water is a documented source of NTM infection in humans, though is not the only source. The incidence of pulmonary disease due to environmental mycobacteria is increasing in many parts of the world including Australia. (Thomson, 2010) Reasons include increased awareness of mycobacteria as pulmonary pathogens, improvements in methods of detection and culture, and an ageing population (as this is often a disease of the elderly). An increase in exposure is also possible as people are exposed to mycobacteria in water through drinking, swimming and bathing. Aerosols generated during these activities can be inhaled, and water can be aspirated when swallowed (Thomson, Armstrong, and Looke, 2007)). Evidence linking human infection due to M. intracellulare and household dust (Reznikov, Leggo and Dawson,

192

1971), soil and potting mix (De Groote, Pace, Fulton and Falkinham III, 2006), and peat moss (Cayer et al, 2007)) exposure also exists.

There is some evidence that susceptible individuals may acquire infection from potable water in their own homes. To date this evidence has been largely reported for Mycobacterium avium (Embil et al, 1997; Falkinham III, 2011; Nischiuchi et al, 2007). We recently documented NTM in the Brisbane potable water distribution system (Thomson et al, 2013) and have shown that strains of M. lentiflavum (Marshall et al, 2011), and M. abscessus (Thomson, Tolson, Carter, Huygens and Hargreaves, 2013)) are highly related/identical to those found in patients with disease. Despite frequent identification of M. kansasii from water samples, pathogenic strains were found infrequently (unpublished data). M. intracellulare and M. avium were found infrequently in potable water. It has been suggested that the low yield of M. intracellulare from potable water samples, is the result of insufficient sample volume, difficulties with culture techniques, and the fact that perhaps this species resides in the biofilm lining pipes and only released intermittently into water. (Falkinham, Horton and Le Chavallier, 2001) Previous investigators have documented the presence of mycobacteria in soil and house dusts in QLD and have shown approximately 50% of environmental strains were also found in human specimens such as sputum (Reznikov et al., 1971; Reznikov and Dawson, 1971; Dawson, 1971).

The aim of this study was to compare NTM found in household water, biofilm and aerosols generated by showering with the species/strains causing infection in patients with known nontuberculous mycobacterial pulmonary disease in Brisbane, Australia.

METHODS

Patients were recruited by the Principal investigator (RT) through the Respiratory Outpatient department at The Prince Charles Hospital, through clinics at the Queensland Tuberculosis Control Centre, and at Greenslopes Private Hospital. Prevalent cases who had been resident in the same dwelling

193

for greater than 5 years prior to the diagnosis of NTM disease, were eligible for home sampling. Following signed informed consent, the PI and assistant visited the homes to collect 1L samples of water into sterile containers directly from kitchen, bathroom and shower taps, rainwater tanks and swimming pools (if applicable). Swabs were taken from inside all taps and shower heads.

Aerosols from showers were sampled using Andersen six-stage cascade impactors for viable air sampling (Grasby-Andersen, Smyrna, GA) that provide validated particle size distribution data (Anderson, 1958) and that have been used previously to collect mycobacterial aerosols (Fennelly et al., 2004; Rautiala et al., 2004; Wendt, George, Parker, Gruft, and Falkinham, 1980). The impactors are calibrated at a flow rate of 28.3 L/min and samples were collected onto petri dishes containing Mycobacteria 7H11 agar, supplemented with OADC enrichment (100mL/L), malachite green (25mg/L) and cycloheximide (500mg/L)(17). The impactor was placed as close as practicable to the shower recess and the shower hot tap was run. Windows were left as the patient would normally have them during showering, and exhaust fans were left off. Collection time was 20 minutes. After collection mycobacterial plates were sealed with parafilm, packed in a plastic bag and incubated at 30°C for 3-6 months. Plates were read weekly and purity plate subcultures performed using 7H11 agar.

Water samples were processed as previously described. (Thomson et al, 2013; Thomson, Carter, Gilpin, Coulter and Hargreaves, 2008) After analysis of water sampling revealed an increased yield for rapidly growing mycobacteria using liquid media (MGIT tubes)(Thomson et al, 2013) – these were included for the last 6 houses that were sampled. Swabs were placed in 2ml SDW at time of collection, sealed and transported to the lab at room temperature. Each swab container was vortexed for 2 minutes and inoculated as per the water samples. Following initial inoculation (M7H11 plates, and MGIT tubes) the samples were decontaminated with 0.005% CPC for 30 minutes and further inoculated for culture.

In a subgroup of seven patients, once samples were processed for culture, the remaining samples for each house were pooled according to source (eg hot

194

water, cold water, swabs) and frozen for later PCR and High Resolution Melt curve analysis.

HRM Methodology: Frozen samples were thawed in a water bath at 65°C. 100 mL of each sample was filtered thro.ugh a 0.45 μm pore-size (47 mm diameter) membrane (Advantec, Tokyo, Japan). Each filter was aseptically transferred into a separate sterile 50 mL tube containing 10 mL sterile distilled water (SDW). The tubes were vortexed for 8-10 min at 2,200 rpm, membranes aseptically removed and tubes centrifuged at 4,750 rpm for 30 min at 4°C in a swinging- bucket rotor. The supernatant was discarded and DNA was extracted using the Power Water DNA Isolation Kit (MO BIO, Carlsbad, California, USA), as follows. The pellet was transferred to the supplied Power Bead tube and processed in a Mini- BeadBeater (BioSpec, Bartlesville, Oklahoma, USA) for 5 min with 60 μL kit- supplied C1 lysis buffer. The kit’s Experienced User Protocol was followed from the second step onwards, with the exception of performing a primary elution with 50 μL DNase RNase free deionized water, and a secondary elution with 100 μL kit-supplied C6 elution buffer. The DNA extracts were stored at -80°C until use. All the samples and NTM ATCC controls (M. peregrinum: ATCC 700686, M. gordonae: ATCC 14470, M. terrae: ATCC 15755, M. scrofulaceum: ATCC 19981, M. chelonae: NCTC 946, M. fortuitum: NCTC 10394, M. abscessus: ATCC 19977, M. avium: 8867/GI/97, M. intracellulare: ATCC 13950, M. smegmatis: ATCC 19420, M. kansasii: TMC 1201, M. malmoense: ATCC 29571) were quantified using Nano Drop 1000 (Thermo Scientific, Australia). The real-time PCR reactions were prepared to a volume of 20μl, whose constituents are shown in Table A (supplemental material). Amplification was performed on the Rotor-gene Q real-time PCR instrument (Qiagen Inc, Germantown, MD). Real-time PCR followed by HRM was performed using the following conditions: Step 1: Hold at 95C for 10 minutes. Step 2: Cycling: 95C for 10 sec, 52C for 15 sec, 72 C for 30 sec and repeated 40 times. Step 3: Hold at 72C for 5 min. Step 4: HRM: Melt from 72C to 95C rising by 0.05C at each step. In all amplification reactions, negative samples (whose constituents are listed in table B, supplemental material) were used to avoid any possible contamination. Each assay included 12 reference strains as positive controls to validate the characterisation of the

195

tested samples. All reactions were performed in duplicates.

Human samples were digested and decontaminated using 4% NaOH, neutralised with phosphoric acid and centrifuged at 3000g to concentrate the acid-fast bacilli (AFB). Smears were prepared from the sediment and stained by the Ziehl-Niehlsen (ZN) method. One Lowenstein-Jensen slope (±pyruvate) and 7ml Mycobacterial Growth Indicator Tube (MGIT) were inoculated and incubated at 35oC until growth was detected. ZN staining of colonies confirmed AFB. Multiplex PCR (7) was performed to discriminate between M. tuberculosis, M. avium, M. intracellulare, M. abscessus and other Mycobacterium spp. Isolates identified as other Mycobacterium spp were further speciated using Hain Life Sciences GenoType® Mycobacterium AS (additional species) kit (2004-7 only) and/or 16S rRNA sequencing in conjunction with phenotypic characteristics.

The clonality of clinical and water isolates was determined using a rep-PCR based method (Diversilab® system, bioMerieux, Melbourne). DNA was extracted from clinical and water isolates using the Ultraclean Microbial DNA Isolation Kit (Mo Bio Laboratories, CA, USA). PCR mixture was prepared using AmpliTaq polymerase and PCR buffer (Applied Biosystems, New Jersey, USA) and Mycobacterium Diversilab®- primer mix according to the manufacturer’s instructions. Separation and detection of Rep-PCR products was performed by micro-fluidic chips of the Diversilab® System. Fingerprints were analysed with Diversilab® software v.3.4.38 using the Pearson correlation co-efficient and unweighted pair group method with arithmetic means to compare isolates and determine clonal relationship. A similarity index of ≥97% was used to define isolates that were indistinguishable, based on the manufacturers recommendations. Isolates that shared 95% similarity were considered similar and those <90% considered ‘different/unrelated’.

The protocol was approved by the Queensland University of Technology Research Ethics unit (Approval No. 0900000085) and the Human Research Ethics Committee of the Prince Charles Hospital (EC 2617). 196

RESULTS

Species of mycobacteria isolated from clinical samples from 20 patients included M. avium (5), M. intracellulare (12), M. abscessus (7), M. gordonae (1), M. lentiflavum (1), M. fortuitum (1), M. peregrinum (1), M. chelonae (1) M. triplex (1) and M. kansasii (1). Patient details are summarised in Table 9.1.

Twenty patient homes were sampled and mycobacteria were cultured from all but one home. Species identified in the home matched that found in the patient in seven (35%) cases – M. abscessus (3), M. avium (1), M. gordonae (1), M. lentiflavum (1), and M. kansasii (1). In an additional patient with M. abscessus infection, this species was isolated from potable water in the reservoir zone supplying the home. (Full details in Table S1, Supplementary material)

197

Table 9.1. Clinical details of patients whose homes were sampled.

Pt ID: gender: Radiology Microbiology + Symptoms Clinical notes age* GP; F; 72yrs BE, S++ Br. Wash 2007 M. abscessus Treated for 6 weeks, clinically Gardener nodules Cough, sputum, haemoptysis stable but remained culture with pots++ positive. Died Ca Pancreas 2011 DP; F; 69 BE, Small Multiple S + samples from 2004, Treated for 4 years cavities progressive radiology, wt loss, (ERClarClof + Amikacin)- Avid gardener haemoptysis converted sputum July ‘09; M. intracellulare Sep, Dec culture negative 2010-June ’04(multiple), Apr ’06, Jun ’07, Jun 2011 ‘08 BE New infiltrate, increased Treated 12 months ERClar – Small cough+haemoptysis converted sputum, in cavities 2 positive sputa June’11 (1/3); Sept remission 2013 ’11 (1/3) M. kansasii HM: M; 72 BE, Multiple S+++ M. intracellulare; Partial clinical response to nodules, sputa from April 2006, sig weight treatment, but remains S++ lge cavities loss, cough C+, end stage disease JI; F; 74 Nodules, S++ M. intracellulare Br wash Feb Treated ERClar, converted >400 orchids; BE, small 2009 sputum Aug ‘09 peat moss cavity PsA, A. fumigatus, Trichosporin sp. Ongoing haemoptysis exposure++ Cough, sputum, haemoptysis Ceased treatment end 2011 Relapse/reinfection 2013- new commenced retreatment Feb infiltrate 2013 1/3 sp S-C+-; Br wash S- C + Jan 2013 M. intracellulare JK; F; 51 BE Multiple sputa pos 2007 M. avium Treated 18mths, converted Rainwater Prev Staph aureus; New radiology, sputum, still in remission Tanks symptoms MB:F; 72 BE Cough, sputum, respiratory failure, Died from disease despite Large fernery, cachexia, progressive radiology treatment 2011 rainwater M. avium (’07);M. intracellulare tanks (’06,’08) MC; F; 85 Nodules, Cough, sputum, wt loss, sweats; Br Treated 2 yrs ERClar, remains Rainwater BE wash April ’08 M. species; OLB Nov in remission tanks ’08 M. lentiflavum PsA MCG; F; 63 BE Recurrent LRTIs, persistent cough; Treated 2 yrs ERClar, in Rainwater Br wash Nov ’07; Jul ’08, May ’10 – remission tanks M. intracellulare MD; F; 63 BE, Haemoptysis Treated 18mths, Gardener nodules Br wash Jan’06, Aug ’07 – M. RMLobectomy 2011 intracellulare Relapse/reinfection 2013 NW; M; 81 BE, Cough, sputum, SOB; Br wash Oct Treated 6 weeks IV amikacin, nodules ’07 M. intracellulare; Sputum M. cefoxitin, clari 2010, 2 weeks abscessus Jan ’08, Apr ’08, Sept ’08, in 2012 2012 RC; F; 75 BE Recurrent infections RML, S. 24 mths ERClarCipro- some aureus, symptomatic improvement, Didn’t use M. intracellulare isolated most but didn’t convert sputum. shower, frequently. Currently stable off treatment gardener, M. intracellulare Mar ’06; Slow with chronic cough and very dusty grower Apr ’08; sputum, lethargy but house M. abscessus Jun ’08; M. reluctant for retreatment 198

intracellulare Jul ’08; M. peregrinum July ’08; M.

gordonae Aug ’08;

M. intracellulare Sept –Dec ’08 (multiple); M. peregrinum Dec ’08; M. intracellulare Feb ’09(mult); M. peregrinum Mar ’09; M. fortuitum Mar ’09; M. fortuitum May ’09; M.

intracellulare Oct ’09; M. fortuitum complex Dec’09; M. intracellulare Jan ‘10 RW; M; 69 BE, M. intracellulare 2003-4; M. triplex Treated over 3-4 yrs, drug nodules, 2006; intolerances, peripheral Cactus House cavities M. avium + M. intracellulare 2007; neuropathy

(>300 plants) M. intracellulare 2008; Nocardia Eventually converted sputum. farcinica 2009; M. intracellulare Remains in remission 2009 LY: F; 60 BE, M. intracellulare repeatedly from RMLobectomy following 4yrs nodules 2000-2007; treatment, drug side effects, M. avium after 15mth remission and short relapse time.

Mar ’05; Now deceased secondary to RGM single isolates whilst on NTM disease treatment for SGM M. chelonae May ’05; M. abscessus Jan ‘07 MO; M; 85 Cavities M. intra Dec ’06, Mar ’07, Apr ‘07 18mths ERClar, converted

sputum. BH;M;25 BE, Post pertussis bronchiolitis, with 6 mths treatment amikacin, nodules, secondary infection M. abscessus cefoxitin, clarithromycin, bronchiolit subs bolletii Jul-Sep ‘11 clofazimine is JB: F; 60 RML BE M. avium Br wash 2007 Treated 6 mths, intolerant of

medication, Stable and culture negative off treatment JD; F; 77 BE M. intracellulare 2000 M. intra treated 2000; 2007- M. abscessus multiple including multiple courses of treatment Sep ’08, Oct ’09, Feb ‘11 for M. abscessus with partial

response, limited by drug toxicity, resp failure sec to disease, RIP JP: F; 53 Lge M. abscessus Sp Oct ’10, Br wash Heavy smoker cigarettes, cavities, Dec ’10 (+S. aureus) marijuana, large cavities. nodules, Dec ’12 and multiple in between Treated with IV therapy-

BE response when abstinent from cigarettes. Died Ca lung and progressive cavitary disease NF; F; 80 BE M. abscessus Mar ’11, Apr ‘11 2 weeks amikacin, cefoxitin, 80 6 weeks clarithromycin,

clofazimine; in remission

* age calculated from date of first isolate; S-Smear; C-Culture; BE- Bronchiectasis; PsA- Pseudomonas aeruginosa;A. fumigatus- Aspergillus fumigatus;E-Ethambutol; R- Rifampicin; Clar-Clarithromycin; Clof- Clofazimine OLB –open lung biopsy

199

Table 9.S1. Home sampling results

Pt Pt species Source Type Te CPC NTM HRM NTM in Town ID m water# p G M. abscessus Kitchen Water Col No M. mucogenicum M. gordonae P d M. mucogenicum Kitchen Swab M. gordonae M. poriforae Bathroom Swab No M. mucogenicum M. fortuitum Bathroom Swab No M. lentiflavum M. gordonae complex Bathroom Water Col No M. mucogenicum M. species NFI d Bathroom Swab M. gordonae Shower Water Col No M. mucogenicum d Shower Swab M. gordonae Shower Aerosol M. mucogenicum Shower Aerosol M. abscessus D M. intracellulare Kitchen Water Col Yes M. gordonae NT M. P M. kansasii d M. kansasii abscessus/chelonae Kitchen Swab Yes M. flavescens M. gordonae Bathroom Water Ho No M. austroafricanum M. szulgai t Bathroom Water Col Yes M. kansasii d Shower Water Ho No M. austroafricanum t Shower Water Ho Yes M. gordonae t M. kansasii Shower Water Col Yes M. gordonae d M. kansasii Shower Swab Yes M. flavescens Shower Aerosol M. gordonae H M. intracellulare Bathroom Swab M. peregrinum M. gordonae M Shower Water Ho Yes M. fortuitum complex M. mucogenicum t M. kansasii M. arupense M. chelonae M. poriforae M. flouroanthenivorans M. gadium M. szulgai M. fortuitum JI M. intracellulare Kitchen Water Ho No M. kansasii NT M. fortuitum t M. mucogenicum complex Kitchen Water Ho Yes M. kansasii M. kansasii t M. lentiflavum Kitchen Water Col Yes M. kansasii M. gordonae d M. flavescens Shower Water Ho No M. kansasii M. mucogenicum t Shower Water Col Yes M. kansasii d Bathroom Water Ho Yes M. kansasii t M. mucogenicum Bathroom Water Col Yes M. kansasii d M. mucogenicum Bathroom Swab Yes M. kansasii M. szulgai Tank Water Yes M. gordonae JK M. avium Kitchen Water Col Yes M. gordonae M. mucogenicum d M. kansasii M. poriferae M. szulgai M. kansasii Kitchen Swab M. gordonae M. gordonae Bathroom Water Col Yes M. neglectum M. cosmeticum d M. kansasii M. diernhoferi Bathroom Swab Yes M. szulgai M. gordonae M. mageritense M. kansasii M. lentiflavum M. gordonae M. poriferae 200

Shower Water Ho Yes M. kansasii t M. szulgai Shower Water Col Yes M. szulgai d M. kansasii Shower Swab No M. intracellulare M. gordonae M. gordonae Tank Water Yes M. abscessus M. intracellulare M M. avium Kitchen Swab M. gordonae M. gordonae B M. intracellulare M. kansasii Bathroom Water Col Yes M. gordonae M. fortuitum d complex Bathroom Swab M. avium M. abscessus M. chelonae Shower Water Col No M. aichiense d M. mucogenicum M. gordonae Shower Swab Yes M. poriforae M. avium M. scrofulaceum Shower Aerosol M. fortuitum complex M. abscessus/chelonae M M. lentiflavum Kitchen Water Ho No M. mucogenicum M. chelonae M. lentiflavum C t M. gordonae M. gordonae Kitchen Water Col Yes M rhodesiae M. intracellulare M. kansasii d M. mucogenicum Kitchen Swab Yes M. gordonae M. interjectum Bathroom Water Col Yes M mucogenicum M. simiae d M. species NFI M. gadium M. lentiflavum Shower Water Ho No M. lentiflavum t Shower Water Ho Yes M. lentiflavum t Shower Water Col Yes M. species NFI d Tank Water Yes M. lentiflavum M. mucogenicum M M. intracellulare Kitchen Water Col No M. kansasii NT M. lentiflavum C d M. tilburgii M. fortuitum G Kitchen Swab No M. osloensis complex Shower Water Ho Yes M. osloensis M. kansasii t Shower Swab Yes M. osloensis Bathroom Water Ho Yes M. species NFI t Bathroom Water Col No M. tilburgii d M. mucogenicum M. kansasii/gastri Pool Water Col No M. abscessus d M. mucogenicum M M. intracellulare Shower Water Col No M. gordonae M. arupense D d M. gordonae Shower Swab M. avium M. fortuitum M. abscessus complex M. chelonae M. kansasii Shower Aerosol M. kansasii N M. abscessus Kitchen Water Ho No M. rhodesiae M. gordonae W M. intracellulare t M. Kitchen Water Col No M. rhodesiae fluoroanthenivorans d M. abscessus Shower Water Ho No M. rhodesiae M. kansasii t Shower Water Ho Yes M. rhodesiae t Shower Water Col Yes M. kansasii d Shower Swab Yes M. rhodesiae M. abscessus M. chlorphenolicum 201

RC M. intracellulare Shower Swab M. gordonae M. gordonae M. fortuitum Bathroom Swab Col No M. gordonae M. gordonae M. mucogenicum M. peregrinum d M. gadium M. lentiflavum M. gordonae Shower Swab M. gordonae M. species NFI Kitchen Aerosol M. gordonae M. mageritense M. arupense R M. intracellulare Kitchen Swab Yes M. poriforae NT M. gordonae W M. avium M. gilvum M. kansasii M. triplex M. gordonae M. fortuitum Shower Water Ho Yes M. gordonae complex t M. mucogenicum Bathroom Water Ho Yes M. gordonae M. sengalense t Bathroom Water Col No M. kansasii d Bathroom Water Col Yes M. mucogenicum d Shower Water Col Yes M. kansasii d M. mucogenicum Shower Aerosol M. wolinskyi LY M. intracellulare Kitchen Water Col M. gordonae NT M. gordonae M. avium d M. gordonae M. mucogenicum M. abscessus Ho M. cosmeticum M. chelonae t M. diernhoferi Shower Water Col M. kansasii M. mageritense d M. lentiflavum Shower Water Ho M. poriferae M. poriferae t Bathroom Water Ho M. gordonae t

Tank Water M. gordonae M M. intracellulare Bathroom Water Ho M. gordonae NT M. gordonae O t M. mucogenicum M. arupense Col M. mucogenicum d M. species Bathroom Water Col M. poriferae M. diernhoferi d M. kansasii Bathroom Water Col M. abscessus/chelonae M. fortuitum d complex Kitchen Water Ho M. mucogenicum t M. gordonae Kitchen Water Col M. abscessus/chelonae d M. kansasii M. hadleri M. gordonae M. hodleri Bath Water Ho M. mucogenicum t Bath Water Col M. rhodesiae d Shower Water Ho M. gordonae t Shower Water Col M. abscessus/chelonae d B M. abscessus Shower Water Col No M. fortuitum NT M. H d fluroanthenivorans Shower Water Col Yes M. fortuitum M. gordonae d M. mucogenicum Shower Water Col No M. mucogenicum M. arupense d M. species Bathroom Water Col No M. mucogenicum M. diernhoferi d M. kansasii Bathroom Water Col Yes M. mucogenicum d Bathroom Water Col Yes M. abscessus d Laundry Water Col Yes M. kansasii d Shower Aerosol Sta M. fortuitum

202

ge 2 Shower Aerosol Sta M. fortuitum ge 3 Shower Aerosol Sta M. fortuitum ge 4 JB M. avium NO NTM Isolated from NT M. gordonae any site M. mucogenicum JD M. abscessus Kitchen Swab Yes M. monalense NT M. fortuitum Shower Water Col No M. mucogenicum complex d M. gordonae Shower Water Col Yes M. mucogenicum M. kansasii d M. szulgai M. arupense M. mucogenicum M. species M. diernhoferi JP M. abscessus Kitchen Water Col No M. mucogenicum NT M. fortuitum d complex Bathroom Water Ho No M. mucogenicum M. mucogenicum t M. kansasii Bathroom Water Col No M. mucogenicum M. flavescens d M. Bathroom Water Col Yes M. mucogenicum fluoroanthenivorans d M. gordonae Bathroom Swab Yes M. mucogenicum Fishtank Water Yes M. fortuitum complex Fishtank Water No M. fortuitum complex N M. abscessus Shower Water Yes M. mucogenicum NT M. lentiflavum F Shower Swab Yes M. mucogenicum M. abscessus Bathroom Water Col No M. mucogenicum M. cosmeticum d M. Bathroom Water Col Yes M. mucogenicum fluroanthenivorans d M. gordonae Bathroom Water Col Yes M. mucogenicum d Shower Aerosol M. fortuitum Shower Aerosol M. fortuitum Shower Aerosol M. fortuitum M M. avium Kitchen Water Col No M. abscessus NT M. species NFI L d M. gordonae Kitchen Water Col Yes M. mucogenicum M. kansasii d M. abscessus Shower Water Ho No M. mucogenicum t Shower Water Col No M. mucogenicum d Shower Water Col Yes M. mucogenicum d Shower Water Col No M. mucogenicum d Bathroom Water Ho Yes M. fortuitum t Bathroom Water Col No M. fortuitum d Bathroom Water Col Yes M. fortuitum d Bathroom Swab No M. fortuitum Bathroom Swab Yes M. fortuitum Shower Aerosol M. fortuitum (Pt-patient; CPC-cetylpiridium chloride decontamination; Temp=tap temperature (hot/cold); HRM-high resolution melt; NT-not tested; NFI – not formally identified; Aerosol stage refers to level of Anderson impactor and reflects particle size)

203

In those instances where the environmental isolate was cultured (as opposed to direct detection using PCR), the strain type matched that of the patient’s infecting strain in the patient with M. kansasii infection (Figure 9.1.A). The clinical isolate of M. lentiflavum was closely related to the M. lentiflavum found in the patient’s house and the local water supply. (Figure 9.1.B) The M. gordonae isolated from patient’s sputum was unrelated to the isolate found in the home water. Despite repeated attempts the M. avium isolated from a shower swab (MB) was unable to be regrown for comparison with the patient’s clinical isolate. In one patient with infection due to M. abscessus subs bolletii, a different strain type of M. abscessus was grown from his shower water. (Figure 9.1.C)

Results from Brisbane wide sampling of water sites (Thomson et al., 2013) revealed a wide variety of NTM. Some of these NTM (esp M. kansasii, M. gordonae, M. fortuitum and M. mucogenicum) that were cultured from 16 homes, matched the species isolated from sampling sites near the patients’ homes. In two instances (NF, NW) M. abscessus was cultured from water close to the homes of these patients with M. abscessus disease and the strains were closely related to those found in their clinical samples (Figure 9.1.D).

Similarly in the patient with M. lentiflavum infection the town water isolates were similar to the home and clinical isolates (Fig 9.1.B). In four other patients where the infecting NTM species was isolated from their home samples (DP, RC, MD, GP) it was also isolated from local water sampling sites.

Full table including Isolates cultured from reservoir zones supplying each patients house, including distance from house to sampling point – Supplementary material Table 9.2.

204

Figure 9.1: Rep-PCR dendrograms of A. M. kansasii isolated from patient sputum (DP) and samples from the patients home; B. M. lentiflavum isolated from patient (#4:MC)), her home water (#1), and surrounding water distribution sites (#2,3,5), demonstrating >90% similarity; C. M. abscessus subs bolletii from one patient patient (#4 & 5) and M. abscessus subs abscessus grown from shower water (# 1 & 2). M. abscessus control #3. D. M. abscessus isolated from 2 patients (#2: NF; #3: NW) and from water supply close to their homes (#1 & 4).

Aerosols were collected in 18 homes. The culture plates were overgrown in 6 patients, negative at 6 months in 3, but grew NTM in the remaining 9 (50%). Species identified included M. abscessus (3 homes including one patient with disease due to M. abscessus), M. gordonae (2), M. kansasii (1), M. fortuitum complex (4), M. mucogenicum (1) and M. wolinskyi (1).

205

Table 9.2. Supplementary Table of patient isolates, home grown isolates and those found in local distribution system samples. ID-patient; Type: Res-reservoir, D-Distribution, TM-Trunk main; HRM-high resolution melt ID Infecting Species identified in Closest Type Species grown from Species home sampling site site (no;Distan ce(km)) H M. intracellulare M. fortuitum 83 (1.70) D M. gordonae M complex M. mucogenicum M. peregrinum 130 (0.96) D Contaminated/NAF 182 (1.98) D M. gordonae M. fortuitum M. kansasii 231 (2.25) R M. gordonae M. mucogenicum R M. intracellulare M. gordonae 80 (1.42) D M. gordonae C 78 (1.7) R M. mucogenicum M. fortuitum M. gadium 264 (1.7) R M. lentiflavum M. peregrinum M. mucogenicum M. species NFI M. gordonae 230,231 R M. gordonae (2.23) M. mucogenicum 316 (0.57) D M. mageritense M. mucogenicum 81 (0.87) D M. mucogenicum 289 R M. gordonae M. lentiflavum 280 R M. arupense M. gordonae M. mucogenicum M. species NFI D M. intracellulare M. gordonae 41 (0.23) D M. abscessus/chelonae P M. kansasii 42 (1.19) D M. abscessus/chelonae M. kansasii M. flavescens 72 (0.88) D M. gordonae M. austroafricanum 132 (1.87) R Contaminated M. gordonae 190 (0.71) TM M. gordonae M. flavescens M. szulgai G M. abscessus M. mucogenicum 210 (1.03) D M. gordonae P M. lentiflavum M. mucogenicum M. abscessus M. poriforae M. gordonae 39 (1.25) 212 (2.59) R M. fortuitum complex M. gordonae M. mucogenicum 231 (2.59) D M. species NFI 226 (5.30) R Contaminated N M. abscessus M. rhodesiae 125 R M. gordonae W M. intracellulare M. kansasii 132 (0.47) R Contaminated M. chlorphenolicum 58 (2.07) R M. gordonae M. abscessus 10 R Contaminated 191 (2.73) R M. fluoroanthenivorans M. kansasii/gastri 57 (1.81) D M. abscessus M. kansasii

206

183,184 R M. gordonae (1.31) M. kansasii M M. intracellulare M. gordonae 40 (0.35) R No sample D 193 (0.90) R No sample M. kansasii 38 (0.92) D M. arupense M. avium 180,181 R M. gordonae M. abscessus (0.35) M. fortuitum complex M. chelonae M. kansasii M M. intracellulare M. abscessus 125 (0.55) D M. species NFI L M. mucogenicum M. gordonae M. fortuitum 58 (0.82) R M. gordonae 183,184 R M. gordonae (1.39) M. kansasii 57 (0.55) D M. abscessus M. kansasii B M. abscessus M. fortuitum 276 (0.46) D M. fluroanthenivorans H M. mucogenicum M. gordonae M. abscessus M. mucogenicum M. kansasii 280, 283, R M. arupense 284 (1.32) M. gordonae M. mucogenicum M. species 285-288 R M. diernhoferi (1.44) M. gordonae M. gordonae M. kansasii M. kansasii JB M. avium Nil 194 (6.85) M. gordonae # M. gordonae M. mucogenicum 220 Contaminated 26 Contam 28 NAF N M. abscessus M. mucogenicum 56 (0.62) D M. lentiflavum F M. fortuitum 274 (2.17) D Contaminated 114(2.6) D M. abscessus M. abscessus M. cosmeticum M. fluroanthenivorans 136(2.78) R No sample 164 (3.16) D M. gordonae J M. abscessus M. monalense 175 (0.82) D M. fortuitum complex D M. mucogenicum M. fortuitum complex M. gordonae M. kansasii M. kansasii M. szulgai 177 (2.12) D Contaminated 219 (1.33) D Contaminated M. kansasii 119 (1.62) D M. kansasii M. kansasii M. kansasii M. szulgai 185,186 R Contaminated (4.56)

207

280, R M. arupense 283,284 M. gordonae (4.79) M. mucogenicum M. species 285-289 R M. diernhoferi (4.73) M. gordonae M. gordonae M. kansasii M. kansasii JP M. abscessus M. mucogenicum 121 (1.55) D M. fortuitum complex M. fortuitum M. mucogenicum complex 219 (3.05) D Contaminated M. kansasii 268 (3.38) D M. flavescens M. fluroanthenivorans M. fortuitum complex M. fortuitum complex M. mucogenicum 63 (4.03) D M. fortuitum complex M. gordonae M. gordonae M. mucogenicum M M. intracellulare M. gordonae 176 (0.12) D M. gordonae O M. mucogenicum 177 (1.53) D Contaminated M. poriferae 280 (1.34) R M. arupense M. M. gordonae abscessus/chelonae M. mucogenicum M. kansasii M. species M. hodleri 283,284 R No sample M. rhodesiae (1.68) 285-289 R M. diernhoferi (1.68) M. gordonae M. gordonae M. kansasii M. kansasii 185,186 R Contaminated (2.17) 277 (1.11) TM M. arupense M. fortuitum complex M. gordonae L M. intracellulare M. gordonae 237 (2.78) D No sample Y M. avium M. kansasii 258 (1.98) D No sample M. abscessus M. poriferae 274 (3.13) D Contaminated M. chelonae 291 (4.57) R Contaminated M. gordonae M. mucogenicum 292 (5.28) R M. cosmeticum M. diernhoferi M. gordonae M. mageritense 293 (3.56) R M. lentiflavum M. mucogenicum M. mucogenicum M. poriferae 290 (3.90) R M. gordonae R M. intracellulare M. poriforae 169 (0.88) D M. gordonae

208

W M. avium M. gilvum M. kansasii M. triplex M. gordonae M. kansasii M. kansasii M. kansasii M. mucogenicum 153 (1.33) TM No sample M. wolinskyi 37 (1.36) D M. gordonae 218 (1.45) D M. fortuitum complex M. gordonae M. kansasii M. mucogenicum 40 (3.44) R No sample 180,181 R M. gordonae (3.44) M. fortuitum complex M. kansasii 193 (3.44) R No sample 154 (2.09) R M. kansasii M. sengalense M M. intracellulare M. kansasii 56 (0.16) D M. lentiflavum C M. tilburgii 281 (1.12) D M. fortuitum complex G M. osloensis M. kansasii M. species NFI 229 (1.47) D M. kansasii M. mucogenicum 241 (2.57) D Contaminated M. 136 (3.07) R No sample abscessus/chelonae

JI M. M. kansasii 281 (0.16) D M. fortuitum complex intracellulare M. kansasii M. mucogenicum 56 (1.0) D M. lentiflavum 164 (1.95) M. gordonae 136 (2.43) R No sample M. szulgai 246 (2.07) D M. flavescens M. gordonae M. mucogenicum

M M. avium M. gordonae 37 (1.52) D M. gordonae B M. abscessus 249 (2.39) D Contaminated M. chelonae M. aichiense 169 (1.96) D M. gordonae M. mucogenicum M. kansasii M. poriforae 40 (3.67) R No sample M. avium 180,181(3. R M. gordonae M. scrofulaceum 6) M. fortuitum complex M. fortuitum M. kansasii complex 193(3.67) R No sample M. abscessus/chelonae M M. lentiflavum M. mucogenicum 149 (1.07) TM No sample C M rhodesiae 183,184 R M. gordonae M. gordonae (1.91) M. kansasii M. species NFI 131 (1.38) TM Contaminated M. lentiflavum M. gordonae M. chelonae M. lentiflavum M. gordonae M. mucogenicum M. intracellulare 128 (1.48) TM M. interjectum M. simiae 125 (2.13) R M. gordonae 52 (1.67) D M. gadium JK M. avium M. gordonae 252 (1.15) D No sample M. kansasii 261(3.73) D M. mucogenicum M. szulgai M. poriferae 209

244 (2.31) D M. mucogenicum M. neglectum 243 (2.89) D M. kansasii M. intracellulare 290-292 R M. gordonae M. abscessus (5.98) M. mucogenicum M. cosmeticum M. diernhoferi M. gordonae M. mageritense 293 (6.4) R M. lentiflavum M. mucogenicum M. poriferae

# House 6.85km from North Pine Water Treatment Plant and possibly outside the distribution network as no testing sites close to house. Sites listed those within North Pine system.

DISCUSSION

In this study we have demonstrated that NTM present in potable water distribution systems can also be found in the houses of patients with NTM disease, and can be grown from aerosols generated by showering. Matching human and environmental strains were found for M. abscessus, M. lentiflavum and M. kansasii. Other investigators have explored the possibility that patients with NTM disease (predominantly M. avium and M. intracellulare) may have acquired their infection from water sources in the home. A Japanese group (Nishiuchi et al., 2007) looked at the isolation of NTM from the homes of 49 patients with MAC pulmonary disease and 43 healthy volunteers without MAC disease. From each home – 3x200ml water samples (kitchen tap, showerhead, and used bathtub water) were collected. In addition, scale on the surface of showerheads, slime from three drains and dust from airconditioners was sampled. Eleven samples contained MAC – 10 were from 9/49 patients’ residences and one from the 43 volunteers residences. Eight isolates were M. avium and three were M. intracellulare – a similar ratio to the species causing pulmonary disease in the Japanese population. The isolates from the patients bathrooms were compared with those from their sputum samples and two had identical PFGE patterns and serotype. For four patients the isolates were the same species but had different RFLP or PFGE profiles. Interestingly only bathroom samples were positive – MAC was not isolated from kitchen tap water, yet the same system was supplying water to the whole house. The authors postulate that favourable temperature and humidity allow MAC to preferentially colonise the

210

bathroom. However the frequency of use of kitchen taps, may allow more flushing out of mycobacteria from pipes and the possibility that expectorating patients may have contaminated bathroom drains should not be discounted. In the single patient who had M. intracellulare infection, where the pathogen was grown only from a drain site, this may certainly have been the case.

Falkinham (Falkinham, 2011) reported the results of home water sampling from 31 patients (with 37 homes) in the USA and Canada with NTM disease. Patients included in the study had infection with M. avium (9), M. intracellulare (6), MAC (11), M. abscessus (4) and M. xenopi (1). 49% households sampled grew NTM including M. avium (10), M. intracellulare (10), M. malmoense (5), M. szulgai (3), M. gordonae (6), M. chelonae (2), M. scrofulaceum (1), M. trivale (1), M. terrae (1). In 17/37 (46%) a species was isolated that matched the patients infecting species, and in 7 of these the strains had matching rep-PCR patterns. Details of these matching species are not reported in the paper – though 2 of the matching strains illustrated were M. avium. In contrast to our findings, there was a greater yield of M. intracellulare and an absence of M. abscessus in home samples.

Aerosol sampling for NTM has been performed previously to investigate hot tubs and indoor pools as a source of NTM (Glazer et al., 2007) It has also been used to document NTM in air samples collected during the remediation of an old building (Rautiala, et al., 2004), and from a peat moss processing plant (Cayer, et al., 2007). De Groote (2006) analysed aerosols generated by the pouring of commercial potting mix using the Andersen impactor. The isolates shown to be acid fast were presumptively called mycobacteria but not identified further.

Ours is the first study to examine the household shower aerosols as a potential source of NTM. We have demonstrated that a variety of species can be aerosolised and cultured from droplets of a respirable size. The most significant of these is M. abscessus, which was found in aerosols of patients with disease. We did not demonstrate the aerosolisation of M. intracellulare, which was the main pathogen associated with disease in the patients studied.

211

The yield of M. intracellulare from both Brisbane water (Thomson, Carter, Tolson, Huygens, & Hargreaves, 2013) and home samples was interesting. Falkinham (Falkinham III, Norton, & Le Chavallier, 2001) has suggested that in previous water studies the low yield was due to insufficient number of specimens or the lack of biofilm specimens. In our study however, a large number of samples and relatively larger volume of water samples were collected, and biofilm samples were collected from every tap. There may be issues with detection, though the use of direct detection using HRM didn’t increase the yield for M. intracellulare (yet did find other NTM). Faezel (Feazel et al., 2009) reported extensive sampling of the biofilm of showerheads in the US and whilst M. avium was found, no M. intracellulare was reported. In both of the other home sampling papers there was no evidence to support the proposal that patients with M. intracellulare disease had acquired infection from water sources.

These findings in combination with the earlier studies done by Dawson (Dawson, 1971; Reznikov, & Dawson, 1971) on house dust and soil, suggest that QLD patients at least may be more likely to acquire M. intracellulare from soil or dust rather than water.

More recently Fujita et al (2012) conducted a large survey of soils from 100 patients with MAC infection. In those with high soil exposure (>2hours/week) 6/16 patients had matching soil and clinical isolates using VNTR profiles (5 M. avium, 1 M. intracellulare). This compared with none of the 19 patients with low or no soil exposure, again suggesting that residential soils may also be a source of exposure for patients with MAC disease.

A major limitation of home water and soil sampling studies relates to the long lag time between exposure and infection that can occur with NTM disease. Quite often a patient has had symptoms for many years prior to diagnosis and hence establishing the environment in which they were exposed is often very difficult. Patients are also potentially exposed to NTM outside of their home. It is also worth

212

considering the close relationship between soil and water in the environment, in activities such as gardening and in the situation of soil entering water distribution networks through cracks in pipes and rainwater tanks via household roofs.

In summary we have shown that pathogenic NTM can be isolated from water, biofilm and aerosols in the households of patients with NTM disease. Matching clinical and patient isolates was demonstrated. From ours and other literature, the evidence supporting water as a source of infection would appear greatest for M. avium, M. abscessus, M. lentiflavum and M. kansasii. However there appears little evidence for M. intracellulare, and soil exposure may be more relevant.

213

REFERENCES.

Andersen, A. A. (1958). New sampler for the collection, sizing, and enumeration of viable airborne particles. Journal of Bacteriology, 76, 471-484. Bernstein, D. I., Lummus, Z. L., Santilli, G., Siskorsky, J., & Berstein, I. L. (1995). Machine operator's lung, a hypersensitivity pneumonitis disorder associated with exposure to metalworking fluid aerosols. Chest, 108, 636-641. Cayer, M-P., Veillette, M., Pageau, P., Hamelin, R., Bergeron, M-J., Mériaux, A.,… Duchaine, Caroline. (2007). Identification of mycobacteria in peat moss processing plants: application of molecular biology approaches. Canadian Journal of Microbiology, 53(1), 92-99. Dawson, DJ. (1971). Potential pathogens among strains of mycobacteria isolated from house-dusts. Medical Journal of Australia, 1(13), 679-681. De Groote, M. A, Pace, N. R., Fulton, K., & Falkinham, Joseph O., III. (2006). Relationships between Mycobacterium Isolates from Patients with Pulmonary Mycobacterial Infection and Potting Soils. Applied and Environmental Microbiology, 72(12), 7602-7606. Embil, J., Warren, P., Yakrus, M., Stark, R., Corne, S., Forrest, D., & Hershfield, E. (1997). Pulmonary illness associated with exposure to Mycobacterium-avium complex in hot tub water. Hypersensitivity pneumonitis or infection? Chest, 111, 813-816. Falkinham III, J. O., Norton, C. D., & Le Chavallier, M. W. (2001). Factors influencing numbers of Mycobacterium avium, Mycobacterium intracellulare and other mycobacteria in drinking water distribution systems. Applied and Environmental Microbiology, 67(3), 1225-1231. Falkinham, J. O.III. (2011). Nontuberculous mycobacteria from household plumbing of patients with nontuberculous mycobacteria disease. Emerging Infectious Diseases, 17(3). Feazel, L. M., Baumgartner, L. K., Peterson, K. L., Frank, D. N., Harris, J. K., & Pace, N. (2009). Opportunistic pathogens enriched in showerhead biofilms. Proceedings of the National Academy of Science, 106(38), 16393-16399.

214

Fennelly, K. P., Martyny, J. W., Fulton, K. E., Orme, I. M., Cave, D. M., & Heifets, L. B. (2004). Cough-generated aerosols of Mycobacterium tuberculosis: A new method to study infectiousness. American Journal of Respiratory and Critical Care Medicine, 169(5), 604-609. Fujita, K., Ito, Y., Hirai, T., Maekawa, K., Imai, S., Tatsumi, S., . . . Mishima, M. (2012). Genetic relatedness of Mycobacterium avium-intracellulare complex isolates from patients with pulmonary MAC disease and their residential soils. Clinical Microbiology and Infection, no-no. doi: 10.1111/j.1469- 0691.2012.03929.x Glazer, C. S., Martyny, J. W., Lee, B., Sanchez, T. L., Sells, T. M., Newman, L. S., . . . Rose, C. S. (2007). Nontuberculous mycobacteria in aerosol droplets and bulk water samples from therapy pools and hot tubs. Journal of Occupational and Environmental Hygiene, 4(11), 831 - 840. Marshall, H., Carter, R., Torbey, M., Minion, S., Tolson, C., Sidjabat, H., . . . Thomson, R.. (2011). Mycobacterium lentiflavum in drinking water supplies, Australia. Emerging Infectious Diseases, 17(3). doi: 10.3201/eid1703.090948 Martin-Casabona, N., Bahrmand, A.R., Bennedsen, J., Thomsen, V. Osltergaard, Curcio, M., Fauville-Dufaux, M., . . . Watt, B. (2004). Non-tuberculous mycobacteria: patterns of isolation. A multi-country retrospective survey. The International Journal of Tuberculosis and Lung Disease, 8, 1186-1193. McSwiggan, D. A., & Collins, C.H. (1974). The isolation of M. kansasii and M. xenopi from water systems. Tubercle, 55, 291-297. Nishiuchi, Y., Maekura, R., Kitada, S., Tamaru, A., Taguri, T., Kira, Y., . . . Ito, M. (2007). The Recovery of Mycobacterium avium-intracellulare Complex (MAC) from the residential bathrooms of patients with pulmonary MAC. Clinical Infectious Diseases, 45(3), 347-351. Rautiala, S., Torvinen, E., Torkko, P., Suomalainen, S., Nevalainen, A., Kalliokoski, P., & Katila, M. L. (2004). Potentially pathogenic, slow-growing mycobacteria released into workplace air during the remediation of buildings. Journal of Occupational and Environmental Hygiene, 1(1), 1-6.

215

Reznikov, M., & Dawson, D. J. (1971a). Serological investigation of strains of Mycobacterium intracellulare ("battey" bacillus) isolated from house-dusts. Medical Journal of Australia, 1(13), 682-683. Reznikov, M., Leggo, J. H., & Dawson, D. J. (1971b). Investigation by seroagglutination of strains of the Mycobacterium intracellulare-M. scrofulaceum group from house dusts and sputum in Southeastern Queensland. American Review of Respiratory Disease, 104(6), 951-953. Rose, C. S., Martyny, J. W., Newman, L. S., Milton, D. K., King, T. E., Beebe, J. L.,… Kreiss, K. (1998). Lifeguard Lung: endemic granulomatous pneumonitis in an indoor swimming pool. American Journal of Public Health, 88(12), 1795- 1800. Shelton, B. G., Flanders, W. D., & Morris, G. K. (1999). Mycobacterium sp. as a possible cause of hypersensitivity pneumonitis in machine workers. Emerging Infectious Diseases, 5(2), 270-273. Slosarek, M., Kubin, M., Jaresova, M. (1993). Water-borne household infections due to Mycobacterium xenopi. Central European Journal of Public Health, 1(2), 78-80. Thomson, R (2010). Changing epidemiology of pulmonary nontuberculous mycobacteria infections. Emerging Infectious Diseases, 16(10), 1576-1582. Thomson, R., Tolson, C., Carter, R., Coulter, C., Huygens, F., & Hargreaves, M. (2013). Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia. BMC Microbiology, 13:89 Thomson, R., Carter, R., Gilpin, C., Coulter, C., & Hargreaves, M. (2008). Comparison of methods for processing drinking water samples for the isolation of Mycobacterium avium and Mycobacterium intracellulare. Applied and Environmental Microbiology, 74(10), 3094-3098. Thomson, R. M., Armstrong, J. G., & Looke, D. F. (2007). Gastroesophageal reflux disease, acid suppression, and Mycobacterium avium complex pulmonary disease. Chest, 131, 1166-1172.

216

Thomson, R. M, Tolson, C., Carter, R., Coulter, C., Huygens, F., & Hargreaves, M. (2012). Mycobacterium abscessus infection and potable water. Am J Respir Crit Care Med, 185, A4022. Wendt, S. L., George, K. L., Parker, B. C., Gruft, H., Falkinham, J. O. 3rd. (1980). Epidemiology of infection by nontuberculous Mycobacteria. III. Isolation of potentially pathogenic mycobacteria from aerosols. American Review of Respiratory Disease, 122(2), 259-263. Wright, E. P., Collins, C. H., & Yates, M. D. (1985). Mycobacterium xenopi and Mycobacterium kansasii in a hospital water supply. Journal of Hospital Infection, 6, 175-178.

217

218

CHAPTER 10. CONCLUSIONS AND FUTURE DIRECTIONS

The hypothesis of this thesis was that environmental mycobacteria, found in potable water and in water and aerosols in patients’ homes, would prove to be a source of infection in at-risk patients.

The first aim was to document NTM causing disease in the Brisbane Distribution system. However, the method for isolation of mycobacteria of interest required refinement. The result of my comparative study, found filtration of water for the isolation of mycobacteria is a more sensitive method of concentration than centrifugation. The addition of Sodium Thiosulphate to water samples (to neutralise residual chlorine) as recommended by the US water authorities is not necessary and may reduce the yield for NTM. CPC decontamination, whilst effective in reducing growth of contaminants also significantly reduced mycobacterial numbers.

Using the results of this methodological study, the first sampling of municipal water in Australia was then performed to ascertain what species of NTM were resident in the distribution system and hence potentially capable of aerosolisation or aspiration by patients. Mycobacteria were isolated from 40.21% sampling sites in summer (76/189) and 82.05% sites in winter (160/195). The winter samples yielded the greatest number and variety of mycobacteria, as there was a high degree of subculture overgrowth and contamination in summer.

NTM disease in Queensland has been increasing over recent decades (Thomson, 2010) and the documentation of pathogenic species within the drinking water distribution system is significant. Previously authorities have been reluctant to accept the significance of NTM, and routine or even periodic monitoring has not been performed. The 2004 Australian Drinking Water Guidelines (NHMRC, 2004) acknowledges “Evidence from some localities suggests that the presence of certain environmental Mycobacterium in drinking water may be associated with opportunistic infection in a minority of susceptible individuals”. This comment predominantly relates to those with HIV infection and AIDS, without considering the

219

burden of pulmonary disease in those not immunosuppressed, nor the soft tissue infections. The guidelines also state there are “insufficient data to establish a guideline value for Mycobacterium in water. When epidemiological associations between the presence of Mycobacterium spp in drinking water and infections in the community can be demonstrated, an investigation into the source of the contamination and its removal, should be carried out.” However if routine or even periodic monitoring is not conducted, such associations will not be detected.

Species that have been documented to cause disease in humans residing in Brisbane that were also found in water include M. gordonae, M. kansasii, M. abscessus, M. chelonae, M. fortuitum complex, M. intracellulare, M. avium complex, M. flavescens, M. interjectum, M. lentiflavum, M. mucogenicum, M. simiae, M. szulgai, M. terrae. M. kansasii was frequently isolated, but M. avium and M. intracellulare (the main pathogens responsible for disease is QLD) were isolated infrequently. Distance of sampling site from treatment plant and elevation above sea level in some cases was associated with isolation of NTM. Pathogenic NTM (defined as those known to cause disease in QLD) were more likely to be identified from sites with narrower diameter pipes, predominantly distribution sample points, and from sites with asbestos cement or modified PVC pipes. It has been suggested previously that narrow diameter pipes and different pipe material may be associated with increased NTM isolation. These factors must be taken into consideration by water authorities to help minimise contamination, and to develop better methods of controlling levels of pathogenic NTM.

As such the revised guidelines from 2011 (NHMRC, 2011), do acknowledge some of the growing evidence that NTM in drinking water distribution systems may be a problem. However there is still no guideline value and the inclusion of NTM in routine monitoring programs is not recommended. “A multiple barrier approach from catchment to tap is recommended to minimise the risk of contamination. Minimising biofilm growth and maintaining cleanliness of distribution systems is a priority. Operation of barriers should be monitored to ensure effectiveness.” Whilst

220

routine monitoring may not be cost effective, periodic surveys to detect changes in NTM populations should be considered.

To further explore the possibility that patients acquire infection from water, the second aim of this thesis was to show that the strains of mycobacteria found in potable water are the same as those found in patients living in the area supplied by that water. The main pathogenic species identified in water included M. lentiflavum, M. abscessus, M. kansasii and M. fortuitum. The strains of these species were then compared to those strains associated with human infection. In the case of M. abscessus and M. lentiflavum, water strains clustered with human strains, strengthening the theory that municipal water may be a source of infection for these patients. In contrast to previous beliefs that these pathogens were water borne, M. kansasii and M. fortuitum strains differed.

M. lentiflavum, a slow-growing non-tuberculous mycobacterium, is a rare cause of human disease. It has been isolated from environmental samples worldwide. The genotypic similarity and geographical relationship between human and potable water isolates in the Brisbane metropolitan area was demonstrated. Automated rep-PCR genotyping revealed a dominant environmental clone closely related to clinical strains, suggesting potable water as a possible source of M. lentiflavum infection in humans.

M. abscessus is a rapidly growing mycobacterium responsible for progressive pulmonary disease, soft tissue and wound infections. Patient isolates formed clusters with all of the water isolates except for WP3. Further patient isolates were unrelated to the water isolates. The high degree of similarity between strains of M. abscessus from potable water and strains causing infection in humans from the same geographical area strengthens the possibility that drinking water may also be the source of infection in these patients. This pathogen is notoriously difficult to treat and carries a high risk of moribidity and mortality and there has been a recent suggestion that infection may be transmitted between patients (Aitken et al, 2012). Transmission through water exposure is therefore of significant concern.

221

M. kansasii is a pulmonary pathogen that has been grown readily from municipal water, but rarely isolated from natural waters. A definitive link between water exposure and disease has not been demonstrated and the environmental niche for this organism is poorly understood. Previous strain typing of clinical isolates has revealed seven subtypes with Type 1 being highly clonal and responsible for most infections worldwide. The prevalence of other subtypes varies geographically. In this study 49 water isolates are compared with 83 patient isolates from the same geographical area (Brisbane, Australia), using automated repetitive unit PCR. The clonality of the dominant clinical strain type is again demonstrated but with this method strain variation within this group is evident comparable with other reported methods. There is significant heterogeneity of water isolates and very few are similar or related to the clinical isolates. This suggests that if water or aerosol transmission is the mode of infection, then point source contamination likely occurs from an alternative environmental source.

The environmental niche for M. fortuitum is not known, though it has been suspected that water has been the source of infection especially in the hospital setting. A wide variation in strain types was identified in Brisbane, with thirteen clusters of ≥2 indistinguishable isolates, and 28 patterns consisting of individual isolates. None of the clinical isolates clustered with any of the water isolates, making municipal water an unlikely source of human infection. Point source contamination of water from an alternative source such as soil, may account for outbreaks where water is the suspected source of infection.

These results further refine the evidence that the risk of acquisition of NTM infection from water is likely to be species specific. Whilst it has been suggested that all NTM patients avoid showering, boil drinking water, and regularly drain and clean hot water systems (NTMinfo.com), this may not be necessary for those whose infections are due to non-water associated species. To this end the next aim of this thesis was to demonstrate that strains of environmental mycobacteria known to

222

cause human pulmonary disease could be aerosolized during normal household activities.

Extensive sampling was performed in the homes of 20 patients with NTM disease. Mycobacteria isolated from clinical samples from 20 patients included M. avium (5), M. intracellulare (12), M. abscessus (7), M. gordonae (1), M. lentiflavum (1), M. fortuitum (1), M. peregrinum (1), M. chelonae (1) M. triplex (1) and M. kansasii (1). Species identified in the home matched that found in the patient in seven (35%) cases – M. abscessus (3), M. avium (1), M. gordonae (1), M. lentiflavum (1), and M. kansasii (1). In an additional patient with M. abscessus infection, this species was isolated from potable water supplying her home. NTM isolated from aerosols included M. abscessus (3 homes), M. gordonae (2), M. kansasii (1), M. fortuitum complex (4), M. mucogenicum (1) and M. wolinskyi (1). The evidence again suggests that a patient’s risk of acquiring NTM from their homes appears species specific for M. avium, M. kansasii, M. lentiflavum and M. abscessus.

Because of the finding of M. avium in potable water in some parts of the world, and the finding of M. intracellulare in water pipe biofilm samples, previous authors have assumed that patients must acquire their infections from water in the home. In addition to ours, other studies of home sampling in patients with NTM have similarly found that not all species are associated with matching human and environmental strains. It has been realised (due to developments in species identification) that most, if not all, of the M. intracellulare strains found in water in these previous studies, is in fact M. chimaera, (personal communication J. Falkinham, R. Wallace). ITS gene fragment sequencing is required to differentiate M. chimaera from M. intracellulare, and this is currently being performed on those stored water isolates identified as M. intracellulare.

Despite a predominance of disease due to M. intracellulare we found no evidence for acquisition of infection from household water for this species and it was isolated infrequently from Brisbane DS water. As infection due to M. intracellulare is one of the most common NTM infections in many parts of the world, including Australia, 223

recommendations regarding water and aerosol exposure may not be relevant to this group of ageing patients, in whom such recommendations carry a significant lifestyle burden. Investigating further the source of the infection for these patients is important however, to enhance treatment success and prevent reinfection. As such, investigation into soil and dust exposure is recommended and identification of environmental variables associated with areas of high bacterial load would be worthwhile.

In order to strengthen the suspicion that M. abscessus strains in water are a cause of human infection, this study needs to be replicated in other centres. There have been criticisms of the Diversilab rep-PCR system of strain typing, particularly with respect to reproducibility. This was an issue more with M. abscessus strains than with other species studied. No strain typing technique published to data has adequate discriminatory power to really compare M. abscessus isolates accurately. As such, whole genome sequencing will be performed on the water isolates obtained, and a collection of clinical isolates from Brisbane. These sequences will also be compared with those of isolates from the UK (in collaboration with the University of Cambridge). Hopefully this will shed some light on the very confusing taxonomy of M. abscessus and whether the environmental and human strains are truly “indistinguishable”. It may also determine whether there truly is a difference in antibiotic susceptibility and outcome associated with the different subspecies.

Whilst M. kansasii isolates found in municipal water differed to those found in patients, we have very little disease due to M. kansasii in Queensland. This may reflect a low level of pathogenic relative to non-pathogenic strains in our water. Areas of high incidence of disease are found around the world, and the prevalence of pathogenic strains in the associated water supplies to these areas has not been shown. One could therefore suspect that if water is a vehicle of transmission, then water contamination occurs from another environmental source. Many high incidence areas are associated with the mining industry. In Queensland we have

224

identified high incidence areas in Roma, Cloncurry, St George and Mt Isa (Figure 10.1)

Figure 10.1. Map of Queensland statistical local areas, showing prevalence of M. kansasii infections 2001-2010.

In conjunction with the University of QLD School of Population Health, climatic and environmental factors will be explored using geospatial-mapping techniques. Further environmental sampling may be undertaken in these areas. Such geospatial analysis is being performed for the main species of NTM causing disease in QLD patients to try and identify environmental factors that contribute to Queenslands relatively higher incidence of disease compared to other states of Australia.

NTM are inherently of low virulence, and being environmental most people would be exposed to them regularly. As such, to develop disease, it is suspected that there must be some defect in protective immunity. As we conduct specialist clinics for management of these infections, a biobank of serum and DNA has been 225

established. A collaborative study with the University of WA examining the role of anti-cytokine antibodies in the development of disease is nearing completion, and an NHMRC application is underway to further investigate the immunological susceptibility to NTM infections.

In summary pathogenic mycobacteria are present in the Brisbane municipal water distribution system and in the water coming from taps in residential homes. These NTM can be aerosolized during showering to a respirable particle size and hence the potential for infection via inhalation is real. The pathogenicity of the water strains of these NTM would appear to be species specific. The evidence from this work, and that from other investigators, that patients acquire infection from water, would appear greatest for M. abscessus, M. avium (Embil et al., 1997; Lumb et al., 2004; Rose et al., 1998), M. lentiflavum, M. porcinum (Brown-Elliot et al., 2011) and possibly M. kansasii. Further work is needed to confirm our findings, and to explore other potential environmental niches (such as soil) for these pathogens.

REFERENCES

Aitken, M. L., Limaye, A., Pottinger, P., Whimbey, E., Goss, C. H., Tonelli, M.R., ... Wallace, R. J. (2012) Respiratory Outbreak of Mycobacterium abscessus subspecies massiliense in a lung transplant and cystic fibrosis center. American Journal of Respiratory and Critical Care Medicine, 185(2), 231-232. Brown-Elliott, B. A., Wallace, R. J., Tichindelean, C., Sarria, J. C., McNulty, S., Vasireddy, R., ... Loeffelholz, M. (2011). Five-year outbreak of community- and hospital-acquired Mycobacterium porcinum infections related to public water supplies. Journal of Clinical Microbiology, 49(12), 4231-4238. Embil, J., Warren, P., Yakrus, M., Stark, R., Corne, S., Forrest, D., & Hershfield, E. (1997). Pulmonary illness associated with exposure to Mycobacterium-avium complex in hot tub water. Hypersensitivity pneumonitis or infection? Chest, 111, 813-816. Lumb, R., Stapledon, R., Scroop, A., Bond, P., Cunliffe, D., Goodwin, A., ... Bastian, I. (2004). Investigation of spa pools associated with lung disorders caused by 226

Mycobacterium avium complex in immunocompetent adults. Applied and Environmental Microbiology, 70(8), 4906-4910. NHMRC (2004) Australian Drinking Water Guidelines Paper 6 National Water Quality Management Strategy. National Health and Medical Research Council, Commonwealth of Australia, Canberra NHMRC, NRMMC (2011) Australian Drinking Water Guidelines Paper 6 National Water Quality Management Strategy. National Health and Medical Research Council, National Resource Management Ministerial Council, Commonwealth of Australia, Canberra Rose, C. S., Martyny, J. W., Newman, L. S., Milton, D. K., King, T. E., Beebe, J. L., ... Kreiss, K. (1998). Lifeguard Lung: endemic granulomatous pneumonitis in an indoor swimming pool. American Journal of Public Health, 88(12), 1795- 1800. Thomson, R. (2010). Changing epidemiology of pulmonary nontuberculous mycobacteria infections. Emerging Infectious Diseases, 16(10), 1576-1582.

227

SUPPLEMENTARY MATERIAL Additional Material 4.1: Characteristics of the Brisbane Water distribution network. (From National Performance Report 2007-2008: urban water utilities. Downloaded 9/1/2012 from www.nwc.gov.au/publications Australian Government National Water Commission (page last updated 25/11/11)

2006-7 2007-8

Volume of water source 11632 387 from surface water (ML) Volume of water sourced 10 2268 from groundwater (ML) Volume of water sourced 6155 5931 from recycling (ML) Volume of water received 140 049 129 226 from bulk supplier (ML)4 Total sourced water 168846 137 812 Total urban water supplied 118 632 101 210 (ML) Average annual residential 153 128 water supplied (kL/property) Number of water treatment 4 4 plants Length of water mains (km) 6340 6369 Properties served per km of 69 69 water main Population receiving water 1 006 1 021 supply services (000s) Total connected properties – 435 441 water supply (000s) Water Quality Compliance Water Quality Guidelines ADWG 2004 ADWG 2004 No. of zones where 3 of 3 3 of 3 microbiological compliance achieved % of population where 100% 100% microbiological compliance achieved No. of zones chemical 3 of 3 3 of 3 compliance achieved

4 Over 75% of the water used in South East Queensland (SEQ) comes from three dams operated by SEQwater (Wivenhoe, Somerset and North Pine) 228

Additional Material 4.2. Pipe materials and characteristics of different sample types.

Figure: Pipe materials of different types of samples.

229

Table: Characteristics of different types of samples. Type of sample N Minimum Maximum Mean Std. Deviation Diameter (mm) 1850 80.00 1670.00 185.46 230.41 Main age (years) 1850 3 103 48.79 21.46 Distribution Elevation (m AHD) 1639 0.00 143.42 32.25 24.45 Distance to nearest 1639 0.02 24.49 5.71 5.76 reservoir (km)

Diameter (mm) 725 100.00 1670.00 691.83 466.14 Main age (years) 725 16 88 44.86 14.92 Reservoir Elevation (m AHD) 666 17.58 283.68 80.75 53.65 Distance to nearest 666 0.01 21.81 0.66 3.34 reservoir (km)

Diameter (mm) 484 100.00 1670.00 984.92 392.93 Main age (years) 484 34 119 50.80 15.67 Trunk Main Elevation (m AHD) 382 0.00 94.27 27.42 21.42 Distance to nearest 382 0.08 20.46 9.72 7.08 reservoir (km)

230

Additional material 4.3: Species of NTM isolated from different sample types.

Species Trunk main Reservoir Distribution n= 484 n= 725 n=1850 M. abscessus 1 (1.4) 1 (0.8) 12 (3.6) M. angelicum/szulgai 1 (0.3)

M. arupense 3 (4.1) 2 (1.5) 4 (1.2) M. austroafricanum 1 (0.3) M. bolletii/M. massiliense 1 (0.3) M. chelonae 1 (1.4) 1 (0.3)

M. cookii 2 (0.6) M. cosmeticum 1 (1.4) 1 (0.3) M. diernhoferi 1 (1.4) 3 (2.3) M. farcinogenes 1 (0.8) 2(0.6)

M. flavescens 1 (0.8) 2 (0.6) M. fluoranthenivorans 2 (2.7) 4 (3.1) 9 (2.7) M. fortuitum complex 1 (1.4) 4 (3.1) 22 (6.6) M. gadium 1 (1.4) 1 (0.8) 3 (0.9)

M. gilvum 1 (0.3) M. gordonae 25 (33.8) 51 (38.9) 68 (20.8) M. interjectum 2 (2.7) 6 (1.8) M. intracellulare 1 (0.8) 1 (0.3)

M. kansasii 19 (25.7) 30 (22.9) 89 (26.8) M. lentiflavum 1 (1.4) 3 (2.3) 15 (4.5) M. mageritense 1 (1.4) 4 (1.2) M. moriokaense 1 (1.4)

M. mucogenicum 9 (12.2) 15 (11.5) 49 (14.8) M. poriforae 8 (6.1) 15 (14.5) M. rhodesiae 1 (0.8) M. sengalense 1 (1.4) 1 (0.3)

M. simiae 1 (1.4) 1 (0.3) M. species NFI 1 (1.4) 2 (1.5) 6 (1.8) M. szulgai 1 (1.4) 1 (0.8) 8 (2.4) M. terrae 1 (0.8) 1 (0.3)

M. tilburgii 1 (1.4) 1 (0.8) 1 (0.3) M. triplex 1 (0.3) M. wolinsky 1 (0.3) MAC 3 (0.9)

Total 74 129 332

231

Additional material 6.1. Full Diversilab report of M. abscessus analysis

Name: Report #696: MAB Add mat 1 UQCCR + PATH QLD RBH Date: 03/16/2013 User: Hanna Sidjabat Type: Typing Kit: Mycobacterium Library: None

Group Pattern Key Sample ID Species Source Location Note 1 1 1 483134518R M. abscessus Pulmonary Brisbane Sou 2 483132823R M. abscessus Pulmonary Brisbane Sou 3 29457824R M. abscessus Control Wild type 2 4 450617067R M. abscessus Pulmonary Brisbane Nor 5 379715752R M. abscessus Cutaneous/so Brisbane Wes 6 Tank BWCPCR M. abscessus Rainwater ta Brisbane Sou 7 483138005R M. abscessus Pulmonary Brisbane Nor 8 274808657R M. abscessus Lap band inf Far North Qu 9 454458408R M. abscessus Pulmonary CF South Coast 3 10 Myco_BH Bath M. abscessus Bathroom tap Brisbane Nor 11 Myco_BH Bath M. abscessus Bathroom tap Brisbane Nor 4 12 360024143R M. abscessus Pulmonary Brisbane Eas 13 419312442R M. abscessus Pulmonary South Coast 14 M.abscessus M. abscessus Laboratory Control 5 15 379714981R M. abscessus Pulmonary Brisbane Nor 6 16 MCG Pool Wat M. abscessus Swimming poo Brisbane Sou 17 492604092 M. abscessus Pulmonary CF Wide Bay 7 18 324036586 M. abscessus Pulmonary Brisbane Sou 8 19 483127533R M. abscessus Pulmonary Central QLD 9 20 505126835 M. abscessus Pulmonary CF Brisbane Eas 10 21 505090218 M. abscessus Pulmonary CF Brisbane Eas 11 22 Sp 320_CPC 1 M. abscessus Municipal Wa Brisbane Wes 2 12 23 255815033 M. abscessus Pulmonary Brisbane Nor 24 318566653 M. abscessus Bloodstream Brisbane Cen 13 25 275134670 M. abscessus Pulmonary Brisbane Nor 14 26 393818813 M. abscessus Pulmonary Brisbane Nor 27 275134456 M. abscessus Device/Line Central QLD 15 28 328446647 M. abscessus Pulmonary Wide Bay 29 328445363 M. abscessus Pulmonary Brisbane Nor 30 417748364rep M. abscessus Pulmonary Brisbane Sou 31 29386674 M. abscessus Soft Tissue Brisbane Sou 32 291938261 M. abscessus Pulmonary Brisbane Nor 33 275135839 M. abscessus Pulmonary Brisbane Cen 34 321221479 M. abscessus Pulmonary Brisbane Nor 35 318543716 M. abscessus Pulmonary West Moreton 36 404271453 M. abscessus Pulmonary Brisbane Sou 37 328449605 M. abscessus Pulmonary CF Brisbane Eas 38 328447721 M. abscessus Pulmonary Brisbane Nor 39 275136432rep M. abscessus Pulmonary Brisbane Eas 40 275135807rep M. abscessus Cutaneous/so Far North Qu 41 394473438rep M. abscessus Pulmonary South Coast 42 SP 101 WCPC_ M. abscessus Municipal Wa Brisbane Sou 43 275133084 M. abscessus Pulmonary Brisbane Sou

Analysis Report #696 MAB Add mat 1 Page 1 of 8

232

44 377711337 M. abscessus Cutaneous/so Brisbane Nor 45 351754090 M. abscessus Pulmonary Brisbane Nor 46 321981832 M. abscessus Injection Si Brisbane Eas 47 328446068 M. abscessus Injection Si Brisbane Nor 48 379755630rep M. abscessus Pulmonary CF West Moreton 3 16 49 409899621rep M. abscessus Pulmonary Brisbane Wes 17 50 381048028rep M. abscessus Cutaneous/so Brisbane Sou 51 SP142 WCPC R M. abscessus Municipal Wa Brisbane Wes 52 328448789 M. abscessus Pulmonary Brisbane Sou 53 328440882rep M. abscessus Pulmonary Brisbane Nor 54 271995461 M. abscessus Pulmonary South Coast 55 29389421rep M. abscesss Injection Si Brisbane Sou 56 275139959 M. abscessus Pulmonary Brisbane Sou 57 SP114 R M. abscessus Municipal Wa Brisbane Sou 58 29385807rep M. abscessus Pulmonary Brisbane Sou 59 SP101 R M. abscessus Municipal Wa Brisbane Sou 60 328558972rep M. abscessus Pulmonary Brisbane Eas 61 SP1100207200 M. abscessus Municipal Wa Brisbane Wes 62 SP110 white M. abscessus Municipal Wa Brisbane Wes 63 275137963 M. abscessus Pulmonary Brisbane Eas 64 311911744rep M. abscessus Pulmonary Wide Bay 65 296555363 M. abscessus Pulmonary Brisbane Sou 66 SP101 OWPWC M. abscessus Municipal Wa Brisbane Sou 67 SP42 WCPC R M. abscessus Municipal Wa Brisbane Sou 68 275134359rep M. abscessus Cutaneous/so Wide Bay 69 341211044rep M. abscessus Cutaneous/so South Coast 70 328444147rep M. abscessus Surgical wou Brisbane Sou 71 336395997rep M. abscessus Pulmonary CF Brisbane Eas 18 72 318540128 M. abscessus Pulmonary Brisbane Nor 19 73 328447600rep M. abscessus Cutaneous/So Brisbane Nor 4 20 74 387640523rep M.a bscessus Surgical wou Brisbane Cen 75 297683281 M. abscessus Pulmonary Brisbane Wes 76 SP110 R M. abscessus Municipal wa Brisbane Wes 77 366921511 M. abscessus Pulmonary Brisbane Eas 21 78 379714112rep M. abscessus Pulmonary Sunshine Coa 22 79 SP114 WCPC R M. abscessus Municipal wa Brisbane Sou 80 280885764rep M. abscessus Pulmonary Brisbane Wes 23 81 275138510 M. abscessus Cutaneous/So South Coast 5 24 82 274804151 M. abscessus Surgical wou Brisbane Cen 25 83 274802998 M. abscessus Cutaneous/So South Coast 6 26 84 275136552 M. abscessus Pulmonary Brisbane Nor 85 29387102 M. abscessus Pulmonary Brisbane Wes 86 275133697rep M. abscessus Pulmonary Brisbane Sou 27 87 328447776rep M. abscessus Pulmonary Brisbane Sou 28 88 390900825rep M. abscessus Pulmonary Brisbane Sou 89 328447798rep M. abscessus Pulmonary Brisbane Sou 90 427760051rep M. abscessus Pulmonary Brisbane Nor 91 275134478 M. abscessus Pulmonary Brisbane Nor

Reviewed by Date

Analysis Report #696 MAB Add mat 1 Page 2 of 8

233

Analysis Report #696 MAB Add mat 1 Page 4 of 8

234

Analysis Report #696 MAB Add mat 1 Page 5 of 8

235

Analysis Report #696 MAB Add mat 1 Page 6 of 8

236

Analysis Report #696 MAB Add mat 1 Page 7 of 8

237

Analysis Report #696 MAB Add mat 1 Page 8 of 8

238

APPENDIX A

Thomson, R., Carter, R., Gilpin, C., Coulter, C., & Hargreaves, M. (2008). Comparison of methods for processing drinking water samples for the isolation of Mycobacterium avium and Mycobacterium intracellulare. Applied and Environmental Microbiology, 74, 3094-3098

239

VOL. 74, 2008 ISOLATION OF MYCOBACTERIA FROM WATER 3095

FIG. 1. Flow chart for the processing of sterile water samples.

of viable mycobacteria; 1% NaOH and 5% oxalic acid resulted azlocillin, nalidixic acid, trimethoprim, and amphotericin B in reductions of 64 and 59%, respectively. [PANTA]), and incubation temperature (32°C, 35°C, and 35°C

Le Dantec et al. (18) used membrane filtration followed by with CO2). decontamination with sodium dodecyl sulfate and NaOH, adjusting the pH with 40% phosphoric acid. Using M. gordonae- MATERIALS AND METHODS spiked sterile tap water, these authors showed that this decon- M. avium ATCC 35765 and M. intracellulare ATCC 13950 were inoculated into tamination method reduced the number of mycobacteria to 7H9 broth (0.5 McFarland standard, correlating to a concentration of 1.5 ϫ 108 1% of the original number. CFU/ml) and diluted to a concentration of 100 CFU/500 ml water. Falkinham et al. (7, 8) suggested that decontamination of Control samples. Organisms (M. avium and M. intracellulare separately) were drinking water may not be required. In a study published in added to eight 500-ml samples of sterile water (sterilized by filtration to preserve chlorination using a 0.2-␮m MediaKap-2 hollow-fiber medium filter [Spectrum 2001 (8) these authors initially processed samples without de- Laboratories Inc.]) to a final concentration of 100 CFU/500 ml. Sodium thiosul- contamination, but if plates were overgrown, they reprocessed fate (0.5 ml of a 10% solution) was added to four of the samples (two M. avium them using CPC. Unfortunately, it is not stated in the paper samples and two M. intracellulare samples). One half of the samples were pro- how often decontamination was necessary. Only 15% of the cessed using filtration, and the other half were processed using centrifugation samples contained slowly growing mycobacteria (3% M. avium (Fig. 1). Filtration. Filtration was performed using 0.45-␮m cellulose nitrate filters and 1% M. intracellulare), and there were 2% rapid growers. (Sartorius AG, Go¨ttingen, Germany). The filters were then rinsed with 2 ml Other variables that may affect the yield of mycobacteria sterile distilled water (SDW) and macerated in 3 ml SDW. From the 3 ml of from environmental water samples include the sample volume, SDW, 0.1-ml aliquots were then transferred in triplicate to LJ slants and M7H10 the use of sodium thiosulfate to neutralize chlorine-based dis- and M7H11 plates and sealed in gas-permeable plastic bags for incubation at 32°C, at 35°C, and at 35°C with CO2. Aliquots (0.5 ml) were transferred to two infectants, the method of concentration (e.g., filtration versus MGIT tubes, one of which contained PANTA. centrifugation), the culture medium, and the incubation tem- Centrifugation. Four 500-ml samples (two samples containing sodium thiosul- perature. fate, one M. avium sample, and one M. intracellulare sample) were centrifuged in In Queensland the main mycobacterial pathogen associated 250-ml sterile bottles at 5,000 ϫ g for 20 min at 25°C. The pelleted cells were with pulmonary disease is M. intracellulare, followed by M. rinsed twice with phosphate-buffered saline (21). The resulting suspension was added to sterile diluent to obtain a 3-ml sample, and 0.1-ml aliquots were used avium, M. abscessus, and M. kansasii. It has been postulated to inoculate in triplicate each of the following: LJ slants and M7H10 and M7H11 that patients acquire disease by inhaling aerosols containing plates. The plates were sealed in gas-permeable plastic bags and incubated as mycobacteria from environmental water sources and water indicated above. Additional 0.5-ml aliquots were used to inoculate two MGIT outlets in their homes (20). Patients may also aspirate contam- tubes, one with PANTA and one without PANTA. Tap water. Tap water samples (four 500-ml samples) were collected after 2 inated water as a result of swallowing disorders or severe min of flushing from a single tap in the laboratory. These tap water samples were gastroesophageal reflux disease (31). spiked with M. avium (two samples) and M. intracellulare (two samples) to obtain This pilot study was undertaken to try to identify the best a final concentration of 100 CFU/500 ml. The samples were then decontaminated method of processing water samples for the isolation of myco- with 0.005% CPC and incubated at room temperature for 30 min. Two samples bacteria prior to a larger environmental survey. The aim of this (one M. avium sample and one M. intracellulare sample) were then processed using filtration, and two samples were processed using centrifugation, as de- study was to compare different methods of processing drinking scribed above for sterile samples. water samples for the isolation of species of mycobacteria All plates were read weekly. At 3 weeks all plates were photographed digitally pathogenic to humans, particularly M. intracellulare and M. and colonies were counted. Colonies from plates demonstrating growth were avium, with regard to the concentration method (centrifuga- stained to confirm the presence of acid-fast bacilli, and morphologically different colonies were subcultured on M7H10 agar and incubated at 35°C. Subcultured tion versus filtration), culture medium (Lowenstein-Jensen organisms were then identified to the species level using multiplex PCR as [LJ] slants, Middlebrook 7H10 and 7H11 plates, mycobacterial described by Wilton and Cousins (36). All colonies grown from the tap water growth indicator tubes [MGIT], and MGIT with polymyxin, samples were treated similarly.

240

3096 THOMSON ET AL. APPL.ENVIRON.MICROBIOL.

TABLE 1. Cultures positive for mycobacteria after concentration by ﬁltration using different culture media after 1 and 6 weeks (n ϭ 66)

Week 1 Final culture results Culture n medium No. No. No. No. No. No. contaminated negative positive overgrown negative positive

7H10 18 1 5 12 0 2 16 7H11 18 0 5 13 1 0 17 LJ 18 2 16 0 0 1 17 MGIT 6 0 6 0 0 5 1 MGIT ϩ 60 3 3 0 2 4 PANTA

reading, compared to 81.8% of the untreated samples. How- ever, the colony counts were significantly reduced for spiked tap water samples (7.4 Ϯ 8.5 CFU/liter [mean Ϯ standard deviation]) compared to sterile samples (205.4 Ϯ 262.4 CFU/ liter; P ϭ 0.0001). This equates to a mean reduction to 3.6% of the original numbers. At 3 weeks, three samples not treated with CPC were overgrown, compared to none of the treated samples. In 9 of 88 FIG. 2. Box plot showing the median concentrations (CFU/liter) (10.2%) spiked sterile samples contaminants grew in addition (solid bars), middle two quartiles (boxes), and ranges (error bars) for to mycobacteria, compared to 13/44 (29.5%) tap water samples mycobacteria in spiked sterile water concentrated by filtration and processed with and without addition of sodium thiosulfate. (P ϭ 0.012). The contaminants did not affect the ability to isolate mycobacteria. For the spiked sterile samples, on two of the plates there was fungal overgrowth at week 4, a week after they were photographed, and this was likely due to aerial Data were analyzed using SPSS v12.0 for Windows 2003 (Apache Software Foundation). Tests of association were performed using Fisher’s exact test for contamination when the plates were inspected for photogra- chi-square 2 ϫ 2 tables. Statistical significance was defined as a two-sided P value phy. Of the remaining seven plates, four had single nonbuff of Ͻ0.05. Colony counts were also compared using the Mann-Whitney U test as colonies, two had two colonies, and one had three colonies. the values were not normally distributed. While the colonies were not formally identified, it is presumed that they entered the system during the processing of samples. RESULTS For a number of samples morphologically different colonies grew on Middlebrook medium plates. These colonies were There were 88 spiked sterile water cultures and 44 spiked tap water samples. Mycobacteria grew in 83.3% of all filtered subcultured and then identified to the species level using mul- samples, compared to 12.1% for all centrifuged samples (P Ͻ tiplex PCR, and they were found to be the same organism. 0.0001). Pulsed-field gel electrophoresis was not performed for these Mycobacteria grew in 52.3% of the spiked sterile samples isolates. All colonies grown from the tap water samples were not treated with sodium thiosulfate, compared to 43.2% for similarly processed. No mycobacteria other than the spiked samples that were treated (P ϭ 0.223). For filtered samples the organisms were identified in the tap water samples. addition of sodium thiosulfate did not affect recovery. How- ever, for centrifuged samples, 4.5% of the treated samples were positive, compared to 22.7% of the untreated samples (P ϭ 0.058). The colony counts were lower for filtered sterile TABLE 2. Comparison of incubation temperatures for culture of samples to which sodium thiosulfate was added (151.7 Ϯ 169.8 mycobacteria in both spiked sterile and tap water samples CFU/liter versus 259.0 Ϯ 352.8 CFU/liter [means Ϯ standard processed by using centrifugation or filtration deviations]); however, the difference was not statistically sig- Processing Incubation No. No. No. Total nificant (P ϭ 0.178; Mann-Whitney U test, P ϭ 0.709) (Fig. 2). method conditions contaminated negative positive no. There was no overall difference between Middlebrook 7H10 Centrifugation 32°C 2 14 2 18 and 7H11; 12 and 13 of 18 filtered samples, respectively, 35°C 2 14 2 18 showed positive growth after 1 week. The LJ slants initially 35°C ϩ CO2 3 12 3 18 appeared to be less sensitive, but there was no difference be- BACTEC 0 11 1 12 tween them and the Middlebrook media at 3 weeks (Table 1). Total 7 51 8 66 There was no difference overall between the different incuba- Filtration 32°C 0 0 18 18 tion temperatures (Table 2). 35°C 0 0 18 18 For filtered samples CPC decontamination did not appear to 35°C ϩ CO2 1 0 17 18 affect the number of positive cultures at 3 weeks; 86.4% of the BACTEC 0 7 5 12 Total 1 7 58 66 filtered samples treated with CPC were positive at the final

241

VOL. 74, 2008 ISOLATION OF MYCOBACTERIA FROM WATER 3097

DISCUSSION samples (i.e., no decontamination) and the MGIT system (with PANTA to control contamination) is currently under way. The In this study we demonstrated that filtration is a more ef- MGIT system without PANTA used in this study did not in- fective method than centrifugation for isolating mycobacteria clude oleic acid-albumin-dextrose-catalase enrichment, which from water samples. Apart from providing a far greater yield, may explain the lower yield using this system. filtration was also simpler and more time efficient. To our There have been a number of studies using different meth- knowledge, there have been no direct comparative studies, but ods to isolate mycobacteria from water samples, and there is previous authors have been able to isolate mycobacteria from no established standard. We demonstrated that sodium thio- water samples processed using centrifugation. Perhaps it was sulfate may not be necessary and may interfere with growth. our centrifugation technique, or alternatively, the success of We confirmed the findings of previous authors that CPC con- previous authors may have been related to much higher controls contamination but also significantly reduces mycobacte- centrations of mycobacteria in the water sampled. In this study rial growth. While it would be appealing to process samples low concentrations of target organisms were used, like those without decontamination, the utility of the method would de- that may be expected to exist in suburban, treated water sup- pend on the origin of the samples. plies (3, 8, 9, 12, 18). This study added refinement to concentration and culture In the majority of reported studies the workers have pro- techniques for the isolation of mycobacteria from water; how- cessed samples with sodium thiosulfate to neutralize residual ever, the major challenge remains the need for decontamina- chlorine (2). It is not known whether neutralizing residual tion to reduce bacterial and fungal overgrowth. We and other chlorine interferes with the ability to isolate mycobacteria by workers have demonstrated that addition of CPC is effective increasing bacterial overgrowth or whether the presence of for this purpose; however, we quantified the reduction in yield residual chlorine reduces the yield and diversity of the species of M. intracellulare and M. avium, two of the main pathogens of mycobacteria subsequently isolated. As most opportunistic associated with lung disease, and found that it is significant. pathogenic NTM are relatively resistant to chlorine (18, 24, 25, Given that the major environmental niche for M. intracellulare 28, 30), the addition of sodium thiosulfate may not be neces- is biofilms (8) and only small numbers of this bacterium are sary and may increase contamination rates. found in water samples, the detection of low concentrations of The thiosulfate anion characteristically reacts with dilute acids organisms is important. Perhaps a metagenomic study may 2Ϫ to produce sulfur, sulfur dioxide, and water: S2O3 (aq) ϩ obviate the need for any decontamination and culture method, ϩ 2H (aq) 3 S(s) ϩ SO2(g) ϩ H2O(l), where aq, g, s, and l are and developments in this area are awaited with interest. aqueous, gas, solid, and liquid phases, respectively. Thiosulfate reduces the hypochlorite and in so doing is oxidized to sulfate. REFERENCES

The complete reaction is 4NaClO ϩ Na2S2O3 ϩ 2NaOH 3 1. Chang, C.-T., L.-Y. Wang, C.-Y. Liao, and S.-P. Huang. 2002. Identification 4NaCl ϩ 2Na SO ϩ H O(13). of nontuberculous mycobacteria existing in tap water by PCR-restriction 2 4 2 fragment length polymorphism. Appl. Environ. Microbiol. 68:3159–3161. From our results it appears that sodium thiosulfate may have 2. Clesceri, L., A. Greenberg, and A. Eaton (ed.). 1998. Standard methods for some antibacterial properties in water, perhaps due to gener- the examination of water and wastewater, 20th ed. American Public Health Association, Washington, DC. ation of sulfur, as the contamination rates and mycobacterial 3. Covert, T. C., M. R. Rodgers, A. L. Reyes, and G. N. Stelma, Jr. 1999. colony counts were less in the treated samples. Although not Occurrence of nontuberculous mycobacteria in environmental samples. statistically significant, this is an interesting observation. Im- Appl. Environ. Microbiol. 65:2492–2496. 4. du Moulin, G., K. Stottmeier, P. Pelletier, A. Tsang, and J. Hedley-Whyte. portantly, it seems that for the purpose of isolating mycobac- 1988. Concentration of Mycobacterium avium by hospital hot water systems. teria from water, the addition of sodium thiosulfate is unnec- JAMA 260:1599–1601. essary. 5. du Moulin, G. C., and K. D. Stottmeier. 1978. Use of cetylpyridinium chloride in the decontamination of water for culture of mycobacteria. Appl. The addition of CPC to tap water samples spiked with M. Environ. Microbiol. 36:771–773. avium and M. intracellulare resulted in survival of 3.6% of the 6. Embil, J., P. Warren, M. Yakrus, R. Stark, S. Corne, D. Forrest, and E. Hershfield. 1997. Pulmonary illness associated with exposure to Mycobacte- organisms, but it did not affect the number of positive samples rium avium complex in hot tub water. Hypersensitivity pneumonitis or in- with the concentration of organisms used. The organisms used fection? Chest 111:813–816. in our study were grown in 7H9 broth. It has been shown that 7. Falkinham, J. 2004. Environmental sources of Mycobacterium avium linked to routes of exposure, p. 26–38. In S. Pedley, J. Bartram, G. Rees, A. Dufour, antecedent growth conditions may affect susceptibility to chlo- and J. A. Cotruvo (ed.), Pathogenic mycobacteria in water: a guide to public rine-based disinfectants. Water-grown strains of M. avium health consequences, monitoring and management. IWA Publishing, Lon- were shown to be significantly more chlorine resistant than don, United Kingdom. 8. Falkinham, J., III, C. Norton, and M. Le Chavallier. 2001. Factors influenc- strains grown in medium (30). The magnitude of growth re- ing numbers of Mycobacterium avium, Mycobacterium intracellulare, and duction that we observed may not necessarily apply to water- other mycobacteria in drinking water distribution systems. Appl. Environ. Microbiol. 67:1225–1231. grown organisms from environmental or distribution system 9. Galassi, L., R. Donato, E. Tortoli, D. Burrini, D. Santianni, and R. Dei. 2003. samples. Nontuberculous mycobacteria in hospital water systems: application of There were no differences between the temperatures tested HPLC for identification of environmental mycobacteria. J. Water Health 1:133–139. or between the different solid media overall. However, the 10. Glover, N., A. Holtzman, T. Aronson, S. Froman, O. Berlin, P. Dominguez, Middlebrook media were more sensitive at 1 week and had the K. Kunkel, G. Overturf, G. Stelma, Jr., C. Smith, and M. Yakrus. 1994. The advantage of quantitation of growth over LJ slants. The MGIT isolation and identification of Mycobacterium avium complex (MAC) recovered from Los Angeles potable water, a possible source of infection in AIDS system has recently been introduced for the culture of clinical patients. Int. J. Environ. Health Res. 4:63–72. specimens and has not been used widely for the processing of 11. Goslee, S., and E. Wolinsky. 1976. Water as a source of potentially pathogenic mycobacteria. Am. Rev. Respir. Dis. 113:287–292. water samples. Supplementation with PANTA is used to re- 12. Hilborn, E. D., T. C. Covert, M. A. Yakrus, S. I. Harris, S. F. Donnelly, E. W. duce contamination. A further study utilizing raw tap water Rice, S. Toney, S. A. Bailey, and G. N. Stelma, Jr. 2006. Persistence of

242

3098 THOMSON ET AL. APPL.ENVIRON.MICROBIOL.

nontuberculous mycobacteria in a drinking water system after addition of 26. Pryor, M., S. Springthorpe, S. Riffard, T. Brooks, Y. Huo, G. Davis, and S. A. filtration treatment. Appl. Environ. Microbiol. 72:5864–5869. Sattar. 2004. Investigation of opportunistic pathogens in municipal drinking 13. Holleman, A., and E. Wiberg. 2001. Inorganic chemistry. Academic Press, water under different supply and treatment regimes. Water Sci. Technol. San Diego, CA. 50:83–90. 14. Iivanainen, E., J. Northrup, R. D. Arbeit, M. Ristola, M.-L. Katila, and C. F. 27. Schulze-Robbecke, R., A. Weber, and R. Fischeder. 1991. Comparison of Von Reyn. 1999. Isolation of mycobacteria from indoor swimming pools in decontamination methods for the isolation of mycobacteria from drinking Finland. APMIS 107:193–200. water samples. J. Microbiol. Methods 14:177–183. 15. Kubalek, I., and S. Komenda. 1995. Seasonal variations in the occurrence of 28. Steed, K. A., and J. O. Falkinham III. 2006. Effect of growth in biofilms on environmental mycobacteria in potable water. APMIS 103:327–330. chlorine susceptibility of Mycobacterium avium and Mycobacterium intracel- 16. Kubalek, I., and J. Mysak. 1996. The prevalence of environmental myco- lulare. Appl. Environ. Microbiol. 72:4007–4011. bacteria in drinking water supply systems in a demarcated region in Czech 29. Stinear, T., T. Ford, and V. Vincent. 2004. Analytical methods for the de- Republic, in the period 1984–1989. Eur. J. Epidemiol. V 12:471–474. tection of waterborne and environmental pathogenic mycobacteria, p. 55–73. 17. Le Dantec, C., J.-P. Duguet, A. Montiel, N. Dumoutier, S. Dubrou, and V. In S. Pedley, J. Bartram, G. Rees, A. Dufour, and J. A. Cotruvo (ed.), Vincent. 2002. Chlorine disinfection of atypical mycobacteria isolated from a Pathogenic mycobacteria in water: a guide to public health consequences, water distribution system. Appl. Environ. Microbiol. 68:1025–1032. monitoring and management. IWA Publishing, London, United Kingdom. 18. Le Dantec, C., J.-P. Duguet, A. Montiel, N. Dumoutier, S. Dubrou, and V. 30. Taylor, R. H., J. O. Falkinham III, C. D. Norton, and M. W. LeChevallier. Vincent. 2002. Occurrence of mycobacteria in water treatment lines and in 2000. Chlorine, chloramine, chlorine dioxide, and ozone susceptibility of 68: water distribution systems. Appl. Environ. Microbiol. 5318–5325. Mycobacterium avium. Appl. Environ. Microbiol. 66:1702–1705. 19. Leoni, E., P. Legnani, M. T. Mucci, and R. Pirani. 1999. Prevalence of 31. Thomson, R. M., J. G. Armstrong, and D. F. Looke. 2007. Gastroesophageal mycobacteria in a swimming pool environment. J. Appl. Microbiol. 87:683– reflux disease, acid suppression, and Mycobacterium avium complex pulmo- 688. nary disease. Chest 131:1166–1172. 20. Marras, T. K., R. J. Wallace, Jr., L. L. Koth, M. S. Stulbarg, C. T. Cowl, and C. L. Daley. 2005. Hypersensitivity pneumonitis reaction to Mycobacterium 32. Tobin-D’Angelo, M. J., M. A. Blass, C. del Rio, J. S. Halvosa, H. M. Blum- avium in household water. Chest 127:664–671. berg, and C. R. Horsburgh, Jr. 2004. Hospital water as a source of Myco- 21. Murray, P. R., E. J. Baron, M. A. Pfaller, F. C. Tenover, and R. H. Yolken bacterium avium complex isolates in respiratory specimens. J. Infect. Dis. (ed.). 1999. Manual of clinical microbiology, 7th ed. American Society for 189:98–104. Microbiology, Washington, DC. 33. Torvinen, E., S. Suomalainen, M. J. Lehtola, I. T. Miettinen, O. Zacheus, L. 22. Neumann, M., R. Schulze-Robbecke, C. Hagenau, and K. Behringer. 1997. Paulin, M.-L. Katila, and P. J. Martikainen. 2004. Mycobacteria in water Comparison of methods for isolation of mycobacteria from water. Appl. and loose deposits of drinking water distribution systems in Finland. Appl. Environ. Microbiol. 63:547–552. Environ. Microbiol. 70:1973–1981. 23. Norton, C. D., M. W. LeChevallier, and J. O. Falkinham III. 2004. Survival 34. Vaerewijck, M. J., G. Huys, J. C. Palomino, J. Swings, and F. Portaels. 2005. of Mycobacterium avium in a model distribution system. Water Res. 38:1457– Mycobacteria in drinking water distribution systems: ecology and signifi- 1466. cance for human health. FEMS Microbiol. Rev. 29:911–934. 24. Payment, P. 1999. Poor efficacy of residual chlorine disinfectant in drinking 35. von Reyn, C. F., J. N. Marlow, R. D. Arbeit, T. W. Barber, and J. O. water to inactivate water borne pathogens in distribution systems. Can. J. Falkinham. 1994. Persistent colonisation of potable water as a source of Microbiol. 45:709. Mycobacterium avium infection in AIDS. Lancet 343:1137–1141. 25. Pelletier, P. A., G. C. du Moulin, and K. D. Stottmeier. 1988. Mycobacteria 36. Wilton, S., and D. Cousins. 1992. Detection and identification of multiple in public water supplies: comparative resistance to chlorine. Microbiol. Sci. mycobacterial pathogens by DNA amplification in a single tube. PCR Meth- 5:147–148. ods Applic. 4:269–273.

243

APPENDIX B1

244

APPENDIX B2 Thomson, R. M., Carter, R., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2013) Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia. BMC Microbiology, 13:89

Thomson et al. BMC Microbiology 2013, 13:89 http://www.biomedcentral.com/1471-2180/13/89

RESEARCH ARTICLE Open Access Factors associated with the isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in Brisbane, Australia Rachel M Thomson1*, Robyn Carter2, Carla Tolson2, Chris Coulter2, Flavia Huygens3 and Megan Hargreaves4

Abstract Background: Nontuberculous mycobacteria (NTM) are normal inhabitants of a variety of environmental reservoirs including natural and municipal water. The aim of this study was to document the variety of species of NTM in potable water in Brisbane, QLD, with a specific interest in the main pathogens responsible for disease in this region and to explore factors associated with the isolation of NTM. One-litre water samples were collected from 189 routine collection sites in summer and 195 sites in winter. Samples were split, with half decontaminated with CPC 0.005%, then concentrated by filtration and cultured on 7H11 plates in MGIT tubes (winter only). Results: Mycobacteria were grown from 40.21% sites in Summer (76/189) and 82.05% sites in winter (160/195). The winter samples yielded the greatest number and variety of mycobacteria as there was a high degree of subculture overgrowth and contamination in summer. Of those samples that did yield mycobacteria in summer, the variety of species differed from those isolated in winter. The inclusion of liquid media increased the yield for some species of NTM. Species that have been documented to cause disease in humans residing in Brisbane that were also found in water include M. gordonae, M. kansasii, M. abscessus, M. chelonae, M. fortuitum complex, M. intracellulare, M. avium complex, M. flavescens, M. interjectum, M. lentiflavum, M. mucogenicum, M. simiae, M. szulgai, M. terrae. M. kansasii was frequently isolated, but M. avium and M. intracellulare (the main pathogens responsible for disease is QLD) were isolated infrequently. Distance of sampling site from treatment plant in summer was associated with isolation of NTM. Pathogenic NTM (defined as those known to cause disease in QLD) were more likely to be identified from sites with narrower diameter pipes, predominantly distribution sample points, and from sites with asbestos cement or modified PVC pipes. Conclusions: NTM responsible for human disease can be found in large urban water distribution systems in Australia. Based on our findings, additional point chlorination, maintenance of more constant pressure gradients in the system, and the utilisation of particular pipe materials should be considered.

Background many parts of the world including Queensland [2]. Clinic- The environmental or nontuberculous mycobacteria ally significant cases representapproximatelyone-thirdof (NTM) are a group of human and animal pathogens that all NTM pulmonary patient-isolates processed by laborator- have significant impacts on the morbidity and mortality of ies in the state. Postulated reasons for this increase include humans and important economic impacts on agriculture increased awareness of mycobacteria as pulmonary patho- [1]. They are normal inhabitants of a variety of environ- gens, improvements in methods of detection and culture, mental reservoirs including natural and municipal water, and an ageing population (as this is often a disease of soil, aerosols, and protozoans. The incidence of pulmonary the elderly). disease due to environmental mycobacteria is increasing in It has been shown that mycobacteria are resident in drinking water distribution systems [3-8]. They have also * Correspondence: [email protected] been found in hospital water distribution systems [9-11] 1Gallipoli Medical Research Centre, Greenslopes Private Hospital, Brisbane, and domestic tap water [12-16]. In 2007, Brisbane Water QLD, Australia managed the supply of potable water to the population Full list of author information is available at the end of the article

© 2013 Thomson et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

245

Thomson et al. BMC Microbiology 2013, 13:89 Page 2 of 8 http://www.biomedcentral.com/1471-2180/13/89

of Brisbane. Details of the system supplying water to ap- Because of the widely varying growth rates of different proximately 1 million residents are tabled in Additional NTM species, and the presence of multiple different file 1: Table S1. Water is treated by chloramination at colony types and species in samples from each site, deter- the treatment plants and by point chlorination at points mination of concentrations of individual NTM species in of entry into the system. There are 45 Reservoirs within CFU/ml was not determined. The number of different the network. species/strains from each site was determined and There have been no published studies examining the expressed per site (1L sample) for each season. presence of mycobacteria in water distribution systems Information regarding the different sampling sites was in Australia, and routine sampling/monitoring is not obtained and included mains age, pipe material, distance mandated as part of public health practice. from nearest reservoir, and elevation above sea level. The aim of this study was to determine if mycobacteria Distances between main treatment plants and sampling that currently cause disease in Queensland residents and sites were calculated from latitude and longitude values those that are found as contaminants in human samples, provided for each site. are present in Brisbane drinking water and to explore Statistics: Statistical analysis was performed using IBM system factors that may be associated with the presence or SPSS v 20. absence of these NTM. Sampling site variables were analysed against individual site culture results using a one-way ANOVA with post Methods hoc Bonferroni correction. Culture method variables were Water authorities routinely sample approximately 220 analysed against results of individual replicates. Where sites across Brisbane as part of water quality maintenance. variables were not normally distributed, the non paramet- These sites are a mixture of Trunk Main (TM) samples, ric Kruskal Wallis test was used. Elevation above sea level Reservoir (R) samples, and Distribution point (D) samples. was transformed (square root) and analysed using one way For this study an extra litre (sample) was collected from ANOVA. Categorical variables were analysed using Chi each site to allow filtration and culture for mycobacteria. Square rxc tables. Values are represented as mean +/− SD Samples were collected in 1Litre sterile bottles and or median (where non normal distribution); significance transported at 4°C to the QLD Mycobacterial Reference level p = 0.05. Laboratory. Samples were collected over a 3-week period in both Winter (July-August 2007) and Summer (December 2007- Results January 2008). Each sample was halved, with 500 ml Sampling site analysis treated with 0.005% Cetylpyridinium chloride (CPC) for Of a total of 217 sites, 1L-samples from 189 sites in sum- 30 minutes. Filtration was performed through 0.45 μm mer and 195 sites in winter were received. Because of the cellulose nitrate filters (Sartorius AG 37070 Goettingen, drought conditions experienced in QLD at the time of the Germany). Filters were then rinsed with 2ml sterile distilled study and subsequent water restrictions, 17 of the sampling water (SDW) and macerated and then 0.1ml aliquots were sites were dry during summer and not able to be sampled. then transferred in triplicate to Middlebrook 7H11 plates, An additional 11 sites were therefore recruited that had which were sealed in gas permeable plastic bags for incuba- not been part of the sampling routine during the preceding tion at 32°C. For the winter samples 0.5 ml aliquots were winter. Overall mycobacteria were identified in 61.5% also transferred to two Mycobacterial Growth Indicator samples. Mycobacteria were grown from 40.2% sites in Tubes (MGITs), one containing PANTA (polymixin, summer (76/189) and 82.1% sites in winter (160/195). The azlocillin, nalidixic acid, trimethoprim, amphotericin B) and lower yield in summer was due to higher rates of contam- incubated using the Bactec 960 system (Becton Dickinson, ination, including that of subculture plates. Of the colonies North Ryde, NSW). As a result, each sample collected in subcultured and sequenced, 236 colonies were subse- winter resulted in 10 processed cultures and each sample in quently identified as NTM. Winter yields were greater summer resulted in six processed cultures. (Mean 2.59 ± 1.62 colonies per site sample; range 1–10) Plates were inspected weekly and a representative selec- compared with summer (1.70 ± 0.84; 1–4). tion of each morphological type of Ziehl Nielsen (ZN) For those sites that were supplied water from Mt Crosby positive colonies from each site sample were subcultured (152 sites in summer, 158 sites in winter), the distance of onto 7H11 plates. Multiplex PCR [17] was performed the sampling site from the treatment plant was associated followed by 16S-rRNA sequencing of mycobacterial iso- with culture result particularly in summer; the mean dis- lates and compared using RIDOM and GenBank database tance from plant to site was 81.75 ± 6.99 km for negative [18,19]. Sequence homology of ≥97% was accepted. Those sites, 82.50 ± 6.17 km for contaminated/overgrown sites identified as M. abscessus/M. chelonae underwent hsp65 and 85.40 ± 6.46 km for positive sites (p = 0.015). In winter and rpoB sequencing for more definitive identification. the distances were similar (negative 84.95 ± 6.77km;

246

Thomson et al. BMC Microbiology 2013, 13:89 Page 3 of 8 http://www.biomedcentral.com/1471-2180/13/89

contaminated/overgrown 82.49 ± 6.77 km; positive 83.34 ± (4.3%), Asbestos Cement (2.2%) or Ductile Iron Cement 6.65 km; p = 0.581). Lined (2.2%) In contrast the majority of distribution For those 17 sites receiving water from the Pine treat- samples came from Asbestos cement or Cast Iron Spun ment plant or from both treatment plants (19 summer, lined pipes. 17 winter), the distance of sampling site from the treat- There was no statistically significant difference between ment plant didn’t correlate with culture result. culture results overall and sample type (ie Distribution/ Reservoir/Trunk Main), mains age, pipe material or dis- Type of sample tance of sampling site from nearest reservoir (Table 1). Samples came from distribution points (D), reservoirs (R) This held true when winter and summer samples were or trunk mains (TM). By their nature, the samples differed analysed separately, though there was a trend towards significantly according to differences in pipe diameter, and more positive sites that were distribution samples pipe material. The characteristics of the different type of (p = 0.074) with narrower diameter pipes in winter samples are shown in Additional file 2: Figure S1 and (p = 0.114). Whilst there were differences in the culture Table S2. The majority of Trunk Main samples (also larger results from different pipe materials the numbers in some diameter) were of Mild Steel Cement Lined (88%), the categories were too small to be statistically meaningful. remainder were Cast iron spun lined (6.6%), cast iron Trunk Main samples grew M. kansasii, M. gordonae, M. cement lined (3.3%) or Mild steel unlined black piping mucogenicum, M. abscessus, M. chelonae, M. lentiflavum, (2.1%). Reservoir samples similarly came mostly from Mild M. simiae, M. szulgai, M. fortuitum complex, and hence Steel Cement lined pipes (75.5%), with the remainder from these species are also potentially present in more Cast Iron spun lined (13%), Cast Iron Cement Lined distal sites.

Table 1 Summary of NTM positive and negative sampling site variables NTM Negative NTM Positive Significance (p value) Sampling Site Factor (Mean ± SD) Site elevation (meters above sea level)* 44.75 ± 40.12 43.78 ± 39.99 0.977 S 44.94 ± 41.92 44.88 ± 38.86 0.680 W 43.51 ± 26.54 43.26 ± 40.63 0.751 Pipe Diameter (cm) 438.01 ± 459.91 435.21 ± 461.92 0.954 S 403.23 ± 417.56 489.15 ± 513.25 0.211 W 553.94 ± 571.58 409.59 ± 434.81 0.103 Mains Age (years) 46.56 ± 19.53 48.94 ± 19.15 0.246 S 46.15 ± 19.83 50.97 ± 17.74 0.091 W 47.91 ± 18.71 47.97 ± 19.77 0.987 Pipe material Asbestos cement 28 (30.8% 63 (69.2) Cement lined† 77 (41.8) 107 (58.2) PVC 6 (42.9) 8 (57.1) 0.166 Cast iron spun lined 30 (35.7) 54 (64.3) Other‡ 7 (63.3) 4 (36.4) Sample type N (%) Distribution 86 (37.1) 146 (62.9) Reservoir 36 (39.1) 56 (60.9) 0.668 Trunk Main 26 (43.3) 34 (56.7) Surface water source N (%) Mt Crosby 120 (38.6) 191 (61.4) Pine 14 (37.8) 23 (62.2) 0.995 Mixed 14 (38.9) 22 (61.1) *Elevation non normally distributed, square root transformation to analyse. †Cast iron, ductile iron or mild steel cement lined. ‡Steel unlined/ polyethylene/unknown.

247

Thomson et al. BMC Microbiology 2013, 13:89 Page 4 of 8 http://www.biomedcentral.com/1471-2180/13/89

Some species relevant to humans, namely M. intra- Table 2 Presence of pathogenic NTM against different cellulare, and M. flavescens were grown from reservoir variables samples though may not have been detected more dis- Variable Pathogenic Non pathogenic P value tally in distribution point samples because of the limita- NTM NTM tions of culture techniques (overgrowth, contamination Sample type etc.). (Additional file 3: Species of NTM isolated from Distribution 203 129 0.001 different sample types) Reservoir 56 75 All variables were examined between different species Treatment Plant 33 41 of NTM. Pathogenic NTM (defined as those that had Surface water source been found in human samples in QLD and known to Crosby 231 195 cause disease) were more likely to be identified from 0.695 sites with narrower diameter pipes, predominantly dis- Mixed 25 25 tribution sample points, and from sites with asbestos Pine 36 26 cement or modified PVC pipes. No other variables were Distance to nearest 4.46 (5.01) 4.85 (6.18) 0.423 found to be significant (Table 2). reservoir (km) Mean (±SD) Age of water mains (yrs) 49.45 (19.50) 49.03 (19.11) 0.646 Mean (±SD) Location The reservoir zones that cluster around the Central Pipe diameter (mm) 360.82 (414.90) 509.74 (503.47) <0.0001 Mean (±SD) Brisbane District (CBD) appeared to contribute more positive sites than those in more peripheral zones (ie had more Site elevation 43.26 (45.50) 44.97 (37.17) 0.638 positive sites relative to the proportion of sites sampled) Pipe material however this did not meet statistical significance. Of the Asbestos cement 91 (62.3) 55 (37.7) sites within an approximate 5-kilometre radius of the Cast iron cement lined 26 (56.5) 20 (43.5) CBD, 64.8% grew NTM, compared to 59.9% of sites out- Cast iron spun lined 68 (59.1) 47 (40.9) side this area (p = 0.431 Fisher’s exact test). Ductile Iron cement lined 14 (50) 14 (50) Mild steel cement lined 75 (44.4) 95 (55.9) 0.046 Methodological factors associated with positive culture Mild steel unlined black 3 (42.9) 4 (57.1) results To assess the effect of decontamination and the relative Modified PVC 5 (88.3) 1 (16.7) contribution of the different media to positive results Polyethylene 0 1 (100) and species variety, the individual results of each culture Unplasticized PVC 8 (61.5) 5 (38.5) taken per site was analysed. The results were analysed for summer and winter separately as contamination is- though there were some specific contributions of the liquid sues in summer would have confounded the result. media discussed below. In winter, there were 10 cultures per site, and in summer 6 cultures per site. Hence, there were 1176 plates and 784 MGITs processed in winter (with PANTA added to half of Summer these) and 1140 7H11 plates were processed in summer. Of 1140 cultures processed in summer, only 1.6% were For funding reasons, MGITs were not used in summer. negative, 30.1% were overgrown, 50.2% were positive and Overall 65.3% of cultures were positive for mycobacterial 18.2% were positive with contaminants. Unfortunately, of growth, though there were statistically significant differ- the positive plates that were subcultured, a large percent- ences between summer and winter (p < 0.0001). age became contaminated and the mycobacterial yield was disappointing. Winter There was a wide variety of species identified using 16s Of 1960 cultures processed during winter, 528 (26.9%) rRNA sequencing (Table 3). Those isolates identified as failed to grow any colonies and 188 (9.6%) were overgrown M. abscessus/M. chelonae underwent subsequent hsp65 to the extent that mycobacteria could not be detected, if and rpoB gene fragment sequencing for further differenti- they were present; 847 (43.2%) of cultures had positive ation. Exhaustive speciation was not performed as only growth and 397 (20.3%) were positive but with contami- potentially pathogenic mycobacteria were of interest. nants (presumed fungal on the basis of plate morphology, Overall there were more species identified in winter. All but not formally identified). The winter cultures yielded of the M. intracellulare,MAC,M. lentiflavum, M. simiae, the greatest number and variety of mycobacteria (Table 3). M. chelonae isolates were found in winter, along with the This held true even if MGIT samples were excluded, majority of other pathogenic species such as M. abscessus,

248

Thomson et al. BMC Microbiology 2013, 13:89 Page 5 of 8 http://www.biomedcentral.com/1471-2180/13/89

Table 3 Species of NTM identified in water samples overgrowth and contamination of positive plates, and in- collected in winter and summer creased the yield from positive plates (Table 4). Summer Winter The use of the MGIT tubes in winter increased the yield M. abscessus 2 11 for certain species of mycobacteria. There were 13 isolates M. abscessus/chelonae 1 of M. abscessus – 10 of these were only grown using liquid media (7 MGIT + PANTA, 3MGIT). However the three M. angelicum/szulgai 1 isolates that grew on solid media were from sites that were M. arupense 45not picked up by liquid media. M. lentiflavum was only M. austroafricanum 1 identified in winter samples. Eight sites grew M. lentiflavum M. bolletii/M. massiliense 1 from MGIT only (1 MGIT, 7 MGIT + PANTA). It was M. chelonae 2 grown on solid media from 6 sites – 4ofthesesitesalso M. cookii 11had positive MGITs (2) and MGIT + PANTA (3). For the majority of sites from which M. gordonae was identified, it M. cosmeticum 11 was detected using M7H11. However, from ten sites it was M. diernhoferi 11only grown from MGIT tubes (4 MGIT + PANTA). M. farcinogenes 12Twenty-four sites grew M. kansasii only from MGIT (11 M. flavescens 21MGIT + PANTA only). Several of the pathogenic species M. fluoranthenivorans 11 4 shown in Table 5, were not grown in liquid media and only M. fortuitum complex 13 14 grew on solid media (eg M. fortuitum, M. intracellulare, M. avium). M. gadium 14 M. gilvum 1 Discussion M. gordonae 24 120 This is the first study to document the presence of po- M. interjectum 17tentially pathogenic NTM in an Australian drinking M. intracellulare 2 water distribution system (DS). The incidence of disease M. kansasii 5 133 is increasing [2] and water as a potential source of infection needs to be addressed. NTM have been reported in M. lentiflavum 19 potable water studies from other countries. Mycobac- M. mageritense 14teria were isolated from 38% (16/42) of drinking water M. moriokaense 1 DS in the USA [7], from 21.3% (42/197) in Greece [20] M. mucogenicum 31 42 and 72% (104/144) in Paris [8]. Mycobacteria were found M. poriforae 18 6 in Finnish DS samples – from 35%, up to 80% at sites M. rhodesiae 1 more distal in the network [21]. Mycobacterial numbers reported are similar in DS that used groundwater com- M. sengalense 2 pared to surface water [7]. In our study we identified M. simiae 2 NTM in samples from 82.1% of sites tested in winter M. species NFI 54and 40.2% sites in summer. M. terrae 2 Kubalek demonstrated seasonal variations in the occur- M. tilburgii 21rence of environmental mycobacteria in potable water in M. triplex 1 the Czech Republic between 1984 and 1989 [22]. Forty two percent of samples were positive for mycobacteria, with sig- M. wolinsky 1 nificantly more positive in Spring than Autumn. We have MAC 3 similarly shown differences in seasonal isolation of NTM, TOTAL 130 408 and differences in the species isolated between seasons. Factors associated with the isolation of pathogenic M. kansasii, and M. mucogenicum. M. poriforae and NTM included distance of sampling points from the main M. fluoranthenivorans were predominantly found in sum- treatment plant, diameter of the pipes at point of sam- mer samples and M. fortuitum and M. mucogenicum were pling, and certain pipe materials. Pelletier found that free found equally in winter and summer. chlorine concentrations gradually decrease as water travels down the distribution system [23]. From previous studies Decontamination [21,24] one would expect that mycobacterial growth Decontamination made a statistically significant differ- would be greater the further from disinfection. Du Moulin ence to culture results for all media used (p < 0.0001 found that communities in Massachusetts were more for all). Overall decontamination did decrease the likely to have patients with MAC isolates if they lived

249

Thomson et al. BMC Microbiology 2013, 13:89 Page 6 of 8 http://www.biomedcentral.com/1471-2180/13/89

Table 4 Effect of decontamination on culture results for different media Media Decontamination Culture result n (%) Significance Negative Positive Positive + contaminants Overgrown MGIT Winter No 43 (21.9) 35 (17.9) 115 (58.7) 3 (1.5) p < 0.0001* Yes 99 (50.8) 49 (25.1) 46 (23.6) 1 (0.5) MGIT + PANTA Winter No 4 (0.7) 48 (24.7) 82 (42.3) 3 (1.5) p < 0.0001* Yes 14 (2.5) 64 (32.8) 19 (9.7) 1 (0.5) 7H11 Summer No 4 (0.7) 234 (41.1) 145 (25.4) 187 (32.8) p < 0.0001 Yes 14 (2.5) 338 (59.3) 62 (10.9) 156 (27.4) 7H11 Winter No 46 (7.9) 313 (53.5) 115 (19.7) 111 (19) Yes 161 (27.5) 335 (57.3) 20 (3.4) 69 (11.8)

further away from water treatment plants, and if they lived (sloughing). Falkinham [24] demonstrated significantly in more densely populated areas [25]. This can be higher mycobacterial numbers in distribution samples explained by more complex water distribution systems in (average 25000 fold) than those collected immediately urban areas, with increased numbers of smaller diameter downstream from treatment plants, indicating that pipes, coupled with greater transit time of water in the mycobacteria actively grow within the distribution sys- system allowing for degradation of disinfection products. tem. Whilst we didn’t find that smaller diameter pipes By their hydrophobic nature, mycobacteria have the were more likely to yield NTM, pathogenic species more ability to form biofilms in pipes of distribution networks, certainly more likely to come from sites with smaller contributing to their proliferation and survival [1]. Sma- diameter pipes. ller pipes have a greater surface area of biofilms to con- Some pipe materials have been shown to contribute to tribute NTM into the water, and are usually present at biofilm formation particularly Iron pipes (compared to more distal points of the system where complex bends chlorinated PVC) [26]. However the survival of mycobac- occur. During pressure transients at point of turbulence teria in DS is dependent upon a complex interaction be- such as the bends in pipes, release of biofilms occurs tween pipe surface, nutrient levels and disinfectants. In one study [27], when biofilms were grown on non-corroded surfaces (copper or PVC) free chlorine was more effective Table 5 Number of NTM isolates cultured using different for controlling HPC and M. avium,butmonochloramine media controlled bacterial levels better on corroded iron pipe Winter Summer surfaces. M. avium biofilm levels were higher on iron and media media galvanized pipe surfaces than on copper or cPVC surfaces. Species MGIT MGIT + PANTA 7H11 7H11 In this study we were unable to assess the relative contri- M. abscessus 373 bution of disinfectant concentrations, and nutrient levels, M. abscessus/chelonae 12however there did seem to be some pipe surfaces (such as M. bolletii/M. massiliense 1 asbestos cement or modified PVC) associated with a greater yield of pathogenic mycobacteria at point of M. fortuitum complex 14 13 sampling. These results were consistent for both summer M. gordonae 14 12 94 24 and winter, when chlorine concentrations may have been M. kansasii 34 45 52 5 different (due to heat inactivation). M. mucogenicum 6 4 32 31 There was a wide variety of species isolated from water, M. terrae 2 many of which have been documented to cause disease in M. intracellulare 21QLD patients [2]. M. intracellulare is the main pathogen associated with pulmonary disease in many parts of the M. lentiflavum 3 10 6 world (including Australia andtheUnitedStates)[28].In MAC 3 our study, the isolation of M. intracellulare from water dis- M. flavescens 12tribution samples was disappointing and similar to previous M. interjectum 161investigators. This has been attributed to the difficulties as- M. simiae 2 sociated with culturing this organism from environmental M. szulgai 19samples as high concentrations have been found in biofilm MGIT: Mycobacterial Growth Indicator Tube; PANTA: polymixin, azlocillin, samples from water meters or pipes [24]. However as dis- naladixic acid, trimethroprim, amphotericin B. ease associated serotypes of M. intracellulare have been

250

Thomson et al. BMC Microbiology 2013, 13:89 Page 7 of 8 http://www.biomedcentral.com/1471-2180/13/89

found in soil and house dust, [29,30] and rainwater tanks, Additional files [31] the environmental niche for M. intracellulare may not necessarily be potable water, rather soil and dust contami- Additional file 1: Table S1. Characteristics of the Brisbane Water nates water supplies through breaches in distribution sys- distribution network. (From National Performance Report 2007–2008: urban water utilities. Downloaded 9/1/2012 from www.nwc.gov.au/ tems (e.g. cracked underground pipes). publications Australian Government National Water Commission It has long been recognised that M. kansasii can be (page last updated 25/11/11). found in potable water [4,32,33]. Disease due to this or- Additional file 2: Figure S1. Culture results according to pipe material ganism is not common in Queensland (approximately 20 at sampling site (complements Figure 2). Table S2. Site factors (Pipe diameter, mains age, elevation and distance from treatment plants) cases of significant pulmonary disease per year), yet this associated with culture result. species was readily isolated from potable water. Additional file 3: Species of NTM isolated from different sample types. M. abscessus has become an increasing clinical problem in the last 5–10 years, and its presence in potable water has Authors’ contributions not previously been emphasized. In the majority of pub- RT designed the study, coordinated the collection of samples, participated in lished studies looking for NTM in water, no M. abscessus the processing of water samples, collated and analysed the data, and wrote was documented. There have been taxonomical changes, the manuscript. RC coordinated, received and processed the water samples (including subculturing and sequencing), collated the results and reviewed which led to M. abscessus being recognised as independent the manuscript. CT processed water samples, performed sequencing and from M. chelonae,soolderstudiesreportingM. chelonae collated results.CC contributed to the study design, provided institutional may have included M. abscessus.Butinstudiesdonesince support and reviewed the manuscript. FH intellectually contributed to the study design and methodology and the writing of the manuscript. MH 2000 M. abscessus has been rarely reported [24,33-35]. The intellectually contributed to the study design and methodology, liaised with inclusion of liquid media in our study may have increased Brisbane Water, and contributed to the writing of the manuscript. All authors the yield for M. abscessus. read and approved the final manuscript. The universal problem with studies of environmental Acknowledgements samples has been the difficulty in culturing these slow The authors would like to acknowledge the contribution from Brisbane growing organisms in the presence of fungal and other Water in providing water samples. Urban Utilitie provided a map of the bacterial contaminants [1,36]. Direct detection using PCR distribution network and data on the individual site points. We are grateful also to the staff of the QLD Mycobacterial Reference Laboratory for probes or a metagenomic approach is appealing however assistance and accommodation of this work. positive results may indicate the presence of mycobacterial DNA, but not necessarily viable organisms. This is espe- Funding cially relevant in the presence of disinfection, such as with The study was funded by grants from The Prince Charles Hospital potable water. Foundation and the Gallipoli Medical Research Foundation of Greenslopes Private Hospital. AmajorstudyexaminingshowerheadsintheUSAusing such an approach [37], did find M. avium and M. gordonae Author details 1 in multiple samples. M. abscessus was not reported. Gallipoli Medical Research Centre, Greenslopes Private Hospital, Brisbane, QLD, Australia. 2QLD Mycobacterial Reference Laboratory, Pathology Queensland, RBWH Campus, Herston Rd, Herston, QLD 4006, Australia. Conclusion 3Queensland University of Technology, Institute of Health and Biomedical 4 We have documented pathogenic NTM in the municipal Innovation, Kelvin Grove Campus, Brisbane, QLD 4059, Australia. Queensland University of Technology, Faculty of Science and Technology, George Street, drinking water distribution system of a major Australian Brisbane, QLD 4001, Australia. city. Distance of sampling sites from treatment plants, narrower diameter pipes (predominantly distribution Received: 5 July 2012 Accepted: 8 April 2013 Published: 22 April 2013 point sites) and sites with asbestos cement or modified PVC pipes were more likely to harbor pathogenic NTM. It References is predicted that the interaction between humans and 1. Falkingham J III: Nontuberculous mycobacteria in the environment. mycobacteria will increase, resulting in more cases of dis- Clin Chest Med 2002, 23:529–551. 2. Thomson R: Changing epidemiology of pulmonary nontuberculous ease. Factors driving this increase include disinfection of mycobacteria infections. EID 2010, 16:1576–1582. drinking water with chlorine, selecting mycobacteria by 3. Martín-Casabona N, Bahrmand AR, Bennedsen J, Osltergaard Thomsen V, reducing competition and the increasing percentage of Curcio M, Fauville-Dufaux M, Feldman K, Havelkova M, Katila M-L, Koksalan K, Pereira MF, Rodrigues F, Pfyffer GE, Portaels F, Rossello Urgell J, Rusch- our population with predisposing conditions, especially Gerdes S, Spanish Group for Non-Tuberculosis Mycobacteria, Tortoli E, age and immunosuppression. Public and environmental Vincent V, Watt B: Non-tuberculous mycobacteria: patterns of isolation. A health efforts must therefore focus on actions that will multi-country retrospective survey. Int J Tuberc Lung Dis 2004, 8:1186–1193. 4. Engel HWB, Berwald LG, Havelaar AH: The occurrence of Mycobacterium specifically remove mycobacteria from habitats where sus- kansasii in Tapwater. Tubercle 1980, 61:21–26. ceptible humans are exposed. Based on our findings, add- 5. Mankiewicz EM, Majdaniw O: Atypical mycobacteria in tapwater. Can J itional point chlorination, maintenance of more constant Public Health 1982, 73:358–360. 6. Carson LA, Bland LA, Cusick LB, Favero MS, Bolan GA, Reingold AL, Good RC: pressure gradients in the system, and the utilisation of par- Prevalence of nontuberculous mycobacteria in water supplies of ticular pipe materials should be considered. hemodialysis centers. Appl Environ Microbiol 1988, 54:3122–3125.

251

Thomson et al. BMC Microbiology 2013, 13:89 Page 8 of 8 http://www.biomedcentral.com/1471-2180/13/89

7. Covert TC, Rodgers MR, Reyes AL, Stelma GN Jr: Occurrence of Thomson R, Tórtola Fernandez MT, Tortoli E, Totten SE, Tyrrell G, Vasankari T, nontuberculous mycobacteria in environmental samples. Appl Environ Villar M, Walkiewicz R, Winthrop K, Wagner D, for NTM-NET: The geographic Microbiol 1999, 65:2492–2496. diversity of nontuberculous mycobacteria isolated from pulmonary 8. Le Dantec C, Duguet J-P, Montiel A, Dumoutier N, Dubrou S, Vincent V: samples: A NTM-NET collaborative study. Eur Respir J 2013. doi:10.1183/ Occurrence of mycobacteria in water treatment lines and in water 09031936.00149212. erj01492-2012; published ahead of print. distribution systems. Appl Environ Microbiol 2002, 68:5318–5325. 29. Dawson D: Potential pathogens among strains of mycobacteria isolated 9. du Moulin G, Stottmeier K, Pelletier P, Tsang A, Hedley-Whyte J: from house-dusts. Med J Aust 1971, 1:679–681. Concentration of Mycobacterium avium by hospital hot water systems. 30. Reznikov M, Leggo JH, Dawson DJ: Investigation by seroagglutination of JAMA 1988, 260:1599–1601. strains of the Mycobacterium intracellulare-M. scrofulaceum group from 10. Tobin-D’Angelo MJ, Blass MA, del Rio C, Halvosa JS, Blumberg HM, house dusts and sputum in Southeastern Queensland. Am Rev Respir Dis Horsburgh CR Jr: Hospital water as a source of Mycobacterium avium 1971, 104:951–953. complex isolates in respiratory specimens. J Inf Dis 2004, 189:98–104. 31. Tuffley RJ, Hollbeche D: Isolation of the mycobacterium avium-M. 11. Fox C, Smith B, Brogan O, Rayner A, Harris G, Watt B: Non-tuberculous intracellulare-M. scrofulaceum complex from tank water in Queensland, mycobacteria in a hospital’s piped water supply. J Hosp Infect 1992, Australia. Appl Environ Microbiol 1980, 39:48–53. 21:152–154. 32. McSwiggan DA, Collins CH: The isolation of M. kansasii and M. xenopi 12. Gangadharam PLJ, Awe RJ, Jenkins DE: Mycobacterial contamination from water systems. Tubercle 1974, 55:291–297. through tap water. Am Rev Respir Dis 1976, 113:894. 33. September S, Brozel V, Venter S: Diversity of nontuberculoid 13. Peters MMC, Rusch-Gerdes S, Seidel C, Gobel U, Pohle HD, Ruf B: Isolation mycobacterium species in biofilms of urban and semiurban drinking of atypical mycobacteria from tap water in hospitals and homes: Is this a water distribution systems. Appl Environ Microbiol 2004, 70:7571–7573. possible source of disseminated MAC infection in AIDS patients? 34. Van Ingen J, Boeree M, Dekhuijzen P, Van Soolingen D: Environmental J Infection 1995, 31:39–44. sources of rapid growing nontuberculous mycobacteria causing disease 14. von Reyn CF, Marlow JN, Arbeit RD, Barber TW, Falkinham JO: Persistent in humans. Clin Micro Inf 2009, 15:888–892. colonisation of potable water as a source of Mycobacterium avium 35. Huang W-C, Chiou C-S, Chen J-H, Shen G-H: Molecular epidemiology of infection in AIDS. The Lancet 1994, 343:1137–1141. Mycobacterium abscessus in a subtropcal chronic ventilatory setting. 15. Chang C, Wang L, Liao C, Huang S: Identification of nontuberculous J Med Micro 2010, 59:1203–1211. mycobacteria existing in tap water by PCR-restriction fragment length 36. Pedley SBJ, Rees G, Dufour A, Cotruvo J: Pathogenic Mycobacteria in Water. polymorphism. Appl Environ Microbiol 2002, 68:3159–3161. London: IWA Publishing; 2004. 16. Goslee S, Wolinsky E: Water as a source of potentially pathogenic 37. Feazel L, Baumgartner L, Peterson K, Frank D, Harris J, Pace N: Opportunistic mycobacteria. Am Rev Respir Dis 1976, 113:287–292. pathogens enriched in showerhead biofilms. PNAS 2009, 106:16393–16399. 17. Wilton S, Cousins D: Detection and identification of multiple mycobacterial pathogens by DNA amplification in a single tube. doi:10.1186/1471-2180-13-89 PCR Methods Appl 1992, 4:269–273. Cite this article as: Thomson et al.: Factors associated with the isolation 18. Harmsen D, Rothgänger J, Frosch M, Albert J: RIDOM: Ribosomal of Nontuberculous mycobacteria (NTM) from a large municipal water differentiation of medical microorganisms database. Nucleic Acids Res system in Brisbane, Australia. BMC Microbiology 2013 13:89. 2002, 30:416–417. 19. Benson D, Karsch-Mizrachi I, Lipman D, Ostell J, Sayers E: Genbank. Nucleic Acids Res 2008, 37(databse issue):D26–31. 20. Tsintzou A, Vantarakis A, Pagonopoulou O, Athanassuadou A, Papapetropoulou M: Environmental mycobacteria in drinking water before and after replacement of the water distribution network. Water, Air and Soil Pollut 2000, 120:273–282. 21. Torvinen E, Suomalainen S, Lehtola MJ, Miettinen IT, Zacheus O, Paulin L, Katila M-L, Martikainen PJ: Mycobacteria in water and loose deposits of drinking water distribution systems in Finland. Appl Environ Microbiol 2004, 70:1973–1981. 22. Kubalek I, Komenda S: Seasonal variations in the occurrence of environmental mycobacteria in potable water. APMIS 1995, 103:327–330. 23. Pelletier P, Du Moulin G, Stottmeier KD: Mycobacteria in public water supplies: comparative resistance to chlorine. Microbiological sciences 1988, 5:147–148. 24. Falkinham J III, Norton C, Le Chavallier M: Factors influencing numbers of Mycobacterium avium, Mycobacterium intracellulare and other mycobacteria in drinking water distribution systems. Appl Environ Microbiol 2001, 67:1225–1231. 25. du Moulin GC, Sherman IH, Hoaglin DC, Stottmeier KD: Mycobacterium avium complex, an emerging pathogen in Massachusetts. Journal of Clinical Microbiology 1985, 22:9–12. 26. Norton CD, LeChevallier MW: A pilot study of bacteriological population Submit your next manuscript to BioMed Central changes through potable water treatment and distribution. Appl Environ and take full advantage of: Microbiol 2000, 66:268–276. 27. Norton CD, LeChevallier MW, Falkinham JO III: Survival of Mycobacterium • Convenient online submission avium in a model distribution system. Water Research 2004, 38:1457–1466. 28. Hoefsloot W, van Ingen J, Andrejak C, Angeby K, Bauriaud R, Bemer P, Beylis • Thorough peer review N, Boeree MJ, Cacho J, Chihota V, Chimara E, Churchyard G, Cias R, Dasa R, • No space constraints or color ﬁgure charges Daley CL, Dekhuijzen PNR, Domingo D, Drobniewski F, Esteban J, Fauville- Dufaux M, Folkvardsen DB, Gibbons N, Gómez-Mampaso E, Gonzalez R, • Immediate publication on acceptance Hoffmann H, Hsueh PR, Indra A, Jagielski T, Jamieson F, Jankovic M, Jong E, • Inclusion in PubMed, CAS, Scopus and Google Scholar Keane J, Koh WJ, Lange B, Leao S, Macedo R, Mannsåker T, Marras TK, • Research which is freely available for redistribution Maugein J, Milburn HJ, Mlinkó T, Morcillo N, Morimoto K, Papaventsis D, Palenque E, Paez-Peña M, Piersimoni C, Polanová M, Rastogi N, Richter E, Ruiz-Serrano MJ, Silva A, da Silva MP, Simsek H, van Soolingen D, Szabó N, Submit your manuscript at www.biomedcentral.com/submit

252

APPENDIX C

Marshall, H., Carter, R., Torbey, M. J., Minion, S., Tolson, C., Sidjabat, H.E., Hargreaves, M., & Thomson, R. M. (2011). Mycobacterium lentiflavum in drinking water supplies, Australia. Emerging Infectious Diseases, 17(3), 395-402

Mycobacterium lentiﬂ avum in Drinking Water Supplies, Australia Henry M. Marshall, Robyn Carter, Matthew J. Torbey, Sharri Minion, Carla Tolson, Hanna E. Sidjabat, Flavia Huygens, Megan Hargreaves, and Rachel M. Thomson

Mycobacterium lentifl avum!"#"$%&'()*&'+,)",&,-./0*( (')'0%#'+'*(! +"! M. avium! /")D0*S! 25M?6E! 7'GG*#*&+'%+'"&! 1.%&.$"231&/#1-0*+.2!"+$"#"*#*0"1#.$0"&4"5.2#,"6+$0#$07"8-" /%&! -*! 7'G!!/,0+! <'+C",+! )"0*/,0%#! '7*&+'!!/%+'"&E! C*&/*E! 5#$"/00,"+$&%#-06"4*&2"0,9+*&,20,-#%"$#2:%0$"'&*%6'+607" )'(/0%(('!!/%+'"&!'&!+C*!D%(+!'(!D"(('-0*!226; ;&"#$$0$$"-50"1%+,+1#%"$+),+!"1#,10"&4"M. lentifl avum"+$&%#-0$" M(!<'+C!"+C*#!345E!M. lentifl avum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lentifl avum 345!7'(*%(*!'(!&"+'!!%-0*;!M!/*&+#%0!#*G*#*&/*!0%-"#%+"#1! '#$" 1.%-.*06" 4*&2" AH" &4" ?@I" 6*+,K+,)" '#-0*" $+-0$7" ;50$0" D*#G"#)(!(D*/'%+'"&!"G!%00!D"('+'F*!'("0%+*(;!Q&!=IIHE!9II! $+-0$"&90*%#::06")0&)*#:5+1#%%3"'+-5"5&20"#66*0$$0$"&4"-50" '("0%+*(!"G!345!<*#*!#*D"#+*7; :#-+0,-$" '5&" 5#6" 1%+,+1#%%3" $+),+!"1#,-" 6+$0#$07">.-&2#-06" K+#%'&! F%#'%+'"&! <'+C'&! )1/"-%/+*#'%0! (D*/'*(! '(! <*00! *0:0-+-+90" $0L.0,10B/#$06" M=N" )0,&-3:+,)" $5&'06" #" O&"<&;! M0+C",$C! *D'7*)'"0"$'/! (+,7'*(! D#"F'7*! ,(*G,0! 6&2+,#,-" 0,9+*&,20,-#%" 1%&,0" 1%&$0%3" *0%#-06" -&" 1%+,+1#%" '&G"#)%+'"&E! )"0*/,0%#! (+#%'&! +1D'&$! /%&! -*! '&F%0,%-0*E! $-*#+,$7";5+$"!",6+,)"$.))0$-$""":&-#/%0"'#-0*"#$"#":&$$+/%0" *(D*/'%001! 'G! %! ('&$0*! /0"&*! /%&! -*! 0'&O*7! +"! %&! ",+-#*%O! $&.*10"&4"M. lentifl avum"+,401-+&,"+,"5.2#,$7 (",#/*;!W,0(*7.!!*07!$*0!*0*/+#"DC"#*('(!2WXYZ6!C%(!-**&! /"&('7*#*7! +C*! (+%&7%#7! G"#! )1/"-%/+*#'%0! (+#%'&! +1D'&$! ycobacterium lentifl avum! "#$%&'()(! %#*! &"&+,-*#. -,+! '(! +')*.! %&7! 0%-"#.! '&+*&('F*! %&7! #*J,'#*(! *SD*&('F*! M /,0",(! )1/"-%/+*#'%! 23456! !!#(+! '7*&+'!!*7! '&! 7*7'/%+*7! *J,'D)*&+;! M0("E! P3M! 7*$#%7%+'"&! /%&! 899:! 216;! M. lentifl avum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`7" lentifl avum!'&G*/+'"&!'&!U,**&(0%&7;!K*/"&7E!<*!'&+*&7*7! Y.3)0,$!"U7"Y#*)*0#90$[J"#,6"<.00,$%#,6";./0*1.%&$+$"=&,-*&%" +"! *SD0"#*! +C*! $*&"+1D'/! %&7! $*"$#%DC'/! #*0%+'"&(C'D! =0,-*0!"E*+$/#,0"ZN7U7";5&2$&,[ -*+<**&!D%+'*&+!'("0%+*(!%&7!D"+%-0*!<%+*#!'("0%+*(!'&!+C*! Pa8X"A@7H?@AR0+6AG@H7@V@VFC R#'(-%&*!%#*%;

" O20*)+,)"8,401-+&.$"P+$0#$0$"Q"'''71617)&9R0+6"Q"S&%7"AG!"T&7"H!"U#*15"?@AA" HVW

253

NO]O>N=Y

Methods Water Sampling _*!#*F'*<*7!+C*!#*/"#7(!"G!%00!D%+'*&+(!G#")!?! %&7! /0'&'/%0! ('$&'!!/%&/*! %//"#7'&$! +"! M)*#'/%&! 4C"#%/'/! D#"/*((*7! <'+C'&! =V! C",#(;! T%0G! "G! *%/C! (%)D0*! <%(! K"/'*+1! 2M4K6aQ&G*/+'",(! P'(*%(*(! K"/'*+1! "G! M)*#'/%! 7*/"&+%)'&%+*7!-1!,('&$!I;IINc!/*+10D'#'7'&',)!/C0"#'7*E! 2QPKM6!/#'+*#'%!226!24%-0*!86;!P,#'&$!=IIB@=IIHE!D"+%-0*! %&7!*%/C!NII.)[!%0'J,"+!<%(!!!0+*#*7!(*D%#%+*01!-1!,('&$! <%+*#!<%(!/"00*/+*7!G#")!=I:!('+*(!'&!R#'(-%&*b(!7#'&O'&$! VN.)! /*00,0"(*! &'+#%+*! !!0+*#(! 2K%#+"#',(! MYE! Y"++'&$*&E! <%+*#!(1(+*); Y*#)%&16;!4C*!!!0+*#(!<*#*!#'&(*7!%&7!)%/*#%+*7!'&!A!)[! (+*#'0*!7'(+'00*7!<%+*#;!M0'J,"+(!2I;8!)[6!<*#*!+#%&(G*##*7! Laboratory Identiﬁ cation '&! +#'D0'/%+*! +"! 5BT88! D0%+*(E! (*%0*7! '&! $%(.D*#)*%-0*! T,)%&! (%)D0*(! <*#*! 7'$*(+*7! %&7! 7*/"&+%)'&%+*7! D0%(+'/! -%$(E! %&7! '&/,-%+*7! %+! A=>?;! M0'J,"+(! 2I;N! )[6! -1! ,('&$! Vc! 3%dTE! &*,+#%0'e*7! <'+C! DC"(DC"#'/! %/'7E! <*#*!+#%&(G*##*7!+"!=!)1/"-%/+*#'%0!$#"<+C!'&7'/%+"#!+,-*(E! %&7!/*&+#'G,$*7!+"!/"&/*&+#%+*!+C*!%/'7.G%(+!-%/'00'!2MXR6;! 8!"G!?! ,&+'0! $#"<+C! <%(! 7*+*/+*7;! f3! (+%'&'&$! /")D%#*7!-1!,('&$!L'-"(")%0!P'GG*#*&+'%+'"&!"G!5*7'/%0! "G! /"0"&'*(! /"&!!#)*7! MXR;! 5,0+'D0*S! W?L! 276! <%(! 5'/#""#$%&'()(!%&7!Y*&R%&O!7%+%-%(*(!29E106; D*#G"#)*7! +"! 7'(/#')'&%+*! -*+<**&! M. tuberculosisE! M. aviumE! M. intracellulareE! M. abscessusE! %&7! "+C*#! Automated Rep-PCR Strain Typing Mycobacterium!(DD;!d+C*#!Mycobacterium! (DD;! <*#*! 4C*!(')'0%#'+1!"G!8:!/0'&'/%0!%&7!B!<%+*#!'("0%+*(!<%(! G,#+C*#!(D*/'%+*7!-1!,('&$!T%'&!['G*!K/'*&/*(!Y*&"41D*! 7*+*#)'&*7! -1! ,('&$! %! #*D.W?L! )*+C"7! 2P'F*#('[%-6;! 51/"-%/+*#',)!MK!2%77'+'"&%0!(D*/'*(6!O'+!2=IIV@=IIB! P3M! <%(! *S+#%/+*7! G#")! /0'&'/%0! %&7! <%+*#! '("0%+*(! "&01\! T%'&! ['G*(/'*&/*E! 3*C#*&E! Y*#)%&16! %&7a"#! -1! ,('&$! +C*! i0+#%/0*%&! 5'/#"-'%0! P3M! Q("0%+'"&! j'+! 8:K! #L3M! (*J,*&/'&$! '&! /"&g,&/+'"&! <'+C! DC*&"+1D'/! 25"-'"![%-"#%+"#'*(E!?%#0(-%7E!?ME!iKM6;!W?L!)'S+,#*! /C%#%/+*#'(+'/(; <%(! D#*D%#*7! -1! ,('&$! M)D0'4%J! D"01)*#%(*! %&7! W?L! -,GG*#! 2MDD0'*7! R'"(1(+*)(E! X"(+*#! ?'+1E! ?ME! iKM6! %&7! ;#/%0"A7">20*+1#,";5&*#1+1"]&1+0-3R8,401-+&.$"P+$0#$0$"]&1+0-3" 51/"-%/+*#',)! P'F*#('[%-! D#')*#! )'S! %//"#7'&$! +"! +C*! &4">20*+1#"6+#),&$-+1"1*+-0*+#"4&*"T;U"%.,)"6+$0#$0b" )%&,G%/+,#*#b(! '&(+#,/+'"&(;! L*D.W?L! D#"7,/+(! <*#*! =%+,+1#%"1*+-0*+#" " (*D%#%+*7! %&7! 7*+*/+*7! -1! ,('&$! )'/#""!,'7'/! /C'D(! "G! "M.%2&,#*3"$32:-&2$">TP" "T&6.%#*"&*"1#9+-#*3"&:#1+-+0$"&,"150$-"*#6+&)*#:5"aN" +C*! P'F*#('[%-! (1(+*);! X'&$*#D#'&+(! <*#*! %&%01e*7! <'+C! "U.%-+4&1#%"/*&,15+01-#$+$"'+-5"2.%-+:%0"$2#%%",&6.%0$"&,"" P'F*#('[%-!("G+<%#*!F*#('"&!A;V;AH!-1!,('&$!+C*!W*%#("&! "5+)5(*0$&%.-+&,"1&2:.-0*+c06"-&2&)*#:53">TP" /"##*0%+'"&!/"*G!!/'*&+!%&7!,&<*'$C+*7!D%'#.$#",D!)*+C"7! ">::*&:*+#-0"0D1%.$+&,"&4"&-50*"6+#),&$0$" <'+C!%#'+C)*+'/!)*%&(!+"!/")D%#*!'("0%+*(!%&7!7*+*#)'&*! U+1*&/+&%&)+1 /0"&%0!#*0%+'"&(C'D(; "M&$+-+90"1.%-.*0"*0$.%-$"4*&2"#-"%0#$-"?"$0:#*#-0"0D:01-&*#-06" "$:.-.2"$#2:%0$"aN" "M&$+-+90"1.%-.*0"*0$.%-$"4*&2"#-"%0#$-"A"/*&,15+#%"'#$5"&*"" Results "%#9#)0"aN" " E+&:$3d"$5&'+,)")*#,.%&2#-&.$"+,4%#22#-+&,"&*"#1+6(4#$-"" Clinical Isolates "/#1+%%+"and":&$+-+90"1.%-.*0"aN" X"#+1.(*F*&! '("0%+*(! "G! M. lentiﬂ avum! <*#*! #*D"#+*7! "E+&:$3d"$5&'+,)")*#,.%&2#-&.$"+,4%#22#-+&,"&*"#1+6(4#$-" "/#1+%%+"and"&,0"&*"2&*0"1.%-.*0(:&$+-+90"$:.-.2"&*"/*&,15+#%"" G#")!A:!D%+'*&+(!2X'$,#*!8\!4%-0*!!=6;!X,00!/0'&'/%0!'&G"#)%+'"&! "'#$5+,)$" <%(! %F%'0%-0*! G"#! A=! 2H9c6! D%+'*&+(;! X",#! D%+'*&+(! 2H! =&220,-$" '("0%+*(6!C%7!/0'&'/%001!('$&'!!/%&+!7'(*%(*;!K*F*&!D%+'*&+(! x"N+$K(/0,04+-"&4"-50*#:3"$5&.%6"/0"1&,$+60*06"4&*"0#15":#-+0,-" <*#*!+%O'&$!+#*%+)*&+!"#!<*#*!,&7*#!(,#F*'00%&/*!G"#!5M?! "/04&*0"+,$-+-.-+&,"&4"-50*#:3" 28!"#!=!'("0%+*(!*%/C6\!&"!+#*%+)*&+!/C%&$*7!'&!#*(D"&(*!+"! x"OD:0*-"1&,$.%-#-+&,"$5&.%6"/0"&/-#+,06"'50,"T;U"#*0"" +C*!&*6#:-06"4*&2"Z2[7"T;U!",&,-./0*1.%&.$"231&/#1-0*+#7" d;*#,$/*&,15+#%"&*"&-50*"%.,)"/+&:$37" D"('+'F*! MXR! ()*%#(! -1! f3! (+%'&'&$;! dG! +C*! A=! D%+'*&+(!

HVI" O20*)+,)"8,401-+&.$"P+$0#$0$"Q"'''71617)&9R0+6"Q"S&%7"AG!"T&7"H!"U#*15"?@AA

254

M. lentiﬂ avum"+,"P*+,K+,)"e#-0*"].::%+0$

(C%#*7!+C*!(%)*!$#",&7<%+*#!(",#/*!-,+!<*#*!7'(+#'-,+*7! %)"&$!8I!7'GG*#*&+!#*(*#F"'#!e"&*(;!X"#!8=!D%+'*&+(!0'F'&$! <'+C'&!=I!O)!"G!R#'(-%&*!/*&+#%0!-,('&*((!7'(+#'/+E!+C*!)*%&! 7'(+%&/*! -*+<**&! +C*'#! #*('7*&+'%0! %77#*((*(! %&7! &*%#*(+! D"('+'F*!<%+*#!('+*!<%(!A;V9!O)!2#%&$*!I;9@9;H!O)6;!4C*! V!D*#("&(!<'+C!/0'&'/%001!('$&'!!/%&+!'00&*((!0'F*7!%!)*%&!"G! =;B!O)!G#")!%!D"('+'F*!<%+*#!('+*!2X'$,#*!=6;

Case Descriptions for Signiﬁ cant Isolates 4C*! V! D%+'*&+(!

Case 1: Disseminated Infection M!VA.1*%#."07!<")%&!.$-*#%+#!" %//*((! 7*F'/*! <%(! D0%/*7! '&! X*-#,%#1! =II:;! Q&! MD#'0! ?@@AB?@@C7 =IIBE! (C*! (",$C+! /%#*! G"#! C*D%+"(D0*&")*$%01! %&7! )'07! D%&/1+"D*&'%;! ['F*#! %&7! $%(+#'/! 01)DC! &"7*! -'"D('*(! (C"<*7! $#%&,0")%+%;! 4<"! -"&*! )%##"

;#/%0"?7"=5#*#1-0*+$-+1$"&4":#-+0,-$"4*&2"'5&2"Mycobacterium lentiflavum"'#$"+$&%#-06"#,6"$&.*10"&4"+$&%#-0!"<.00,$%#,6!">.$-*#%+#!" ?@@AB?@@Cb" ]&.*10!",&7"+$&%#-0$" T&7 U06+#,"#)0!"3"Z*#,)0[J" E*&,15+#% e&.,6 =5#*#1-0*+$-+1 :#-+0,-$ $0D!"UR`" '#$5+,)" ]:.-.2 $'#/R#$:+*#-0" a-50*" >6.%-$ "]+),+4+1#,-"1%+,+1#%"+%%,0$$" H FV"ZF?BCW[J"@RH" ? @ @ E%&&6!"A" "T&,$+),+4+1#,-"1%+,+1#%"+%%,0$$" AC IG"Z??BCC[J"A?RI" V F F E%&&6!"A" "M*&/#/%0",&,$+),+4+1#,-"1%+,+1#%"+%%,0$$ F GF"ZWVBCA[J"HRA" A ? @ >$1+-0$!"A" "T&,$+),+4+1#,-"1%+,+1#%"+%%,0$$"'+-5"U>=" G II"ZFVBGW[J"FRH" @ G @ @ =5+%6*0," "]+),+4+1#,-"1%+,+1#%"+%%,0$$" A A7IJ"@RA" @ @ A @ "T&,$+),+4+1#,-"1%+,+1#%"+%%,0$$ H A?"ZA7IBAG[J"AR?" @ ? A @ bU>=!"Mycobacterium avium"1&2:%0D7"

" O20*)+,)"8,401-+&.$"P+$0#$0$"Q"'''71617)&9R0+6"Q"S&%7"AG!"T&7"H!"U#*15"?@AA" HVG

255

NO]O>N=Y

M. lentiﬂ avum! G"#! +C*! !!#(+! +')*;! Q&! X*-#,%#1! =II9E! (C*! -*$%&!*+C%)-,+"0!2HII!)$6E!#'G%)D'&!2VNI!)$!8la76E!%&7! /0%#'+C#")1/'&! 2NII! )$! =la76;! T*#! (1)D+")(! ')D#"F*7E! %&7!(C*!/")D0*+*7!8H!)"&+C(!"G!+#*%+)*&+;!R#"&/C"(/"D'/! <%(C'&$(! D"(++#*%+)*&+! <*#*! f3! (+%'&! %&7! MXR! /,0+,#*! &*$%+'F*;

Case 3: Bronchiectasis M! V9.1*%#."07! 4%'<%&*(*! <")%&! .$-*#%+#!" ?@@AB?@@C7" f*#3" $5#6+,)!" <*00E!<'+C!&"!*S%/*#-%+'"&(!'&!+C*!D#*F'",(!1*%#!%&7!(+%-0*! #::*&D+2#-0" .*/#," 0D-0,-J" 1+*1%0!" ?@(K2" *#6+.$" 4*&2" 10,-*#%" /.$+,0$$"6+$-*+1-7 #%7'"$#%DC'/!%DD*%#%&/*;

Case 4: Cervical Lymphadenitis 2?46! (C"<*7! D*#'-#"&/C'%0! +C'/O*&'&$! %&7! -#"&/C'"0'+'(! M!=I.)"&+C."07!$'#0!<%(!*S%)'&*7!G"#!%!V.<**O!C'(+"#1! -,+! &"! 01)DC%7*&"D%+C1;! 4C*! D%+'*&+! <%(! *)D'#'/%001! "G! -'0%+*#%0! /*#F'/%0! 01)DC%7*&"D%+C1;! 4C*! 0%#$*(+! &"7*! $'F*&!'("&'%e'7E!#'G%)D'/'&E!D1#%e'&%)'7*E!/0%#'+C#")1/'&E! 2=I!l!=V!))6!<%(!*S/'(*7;!3*/#"+'e'&$!$#%&,0")%+%!<*#*! %&7!*+C%)-,+"0;!d#%0!D#*7&'("&*!2=N!)$!8la76!')D#"F*7! (**&;! M. lentiﬂ avum! <%(! /,0+,#*7;! 3"! %&+')1/"-%/+*#'%0! (1)D+")(! %&7! 0'F*#! -'"/C*)'(+#1! %&7! 7*/#*%(*7! (D0*&'/! +C*#%D1!<%(!%7)'&'(+*#*7\!(C*!#*/"F*#*7!G,001; ('e*;! KC*! <%(! 7'(/C%#$*7! "&! D#*7&'("&*! 28N! )$! 8la76E! '("&'%e'7!2AII!)$!8la76E!*+C%)-,+"0!2VII!)$!=la76E!%&7! Nonsigniﬁ cant Isolates /0%#'+C#")1/'&!2NII!)$!=la76;!T*#!"#$%&")*$%01!')D#"F*7! M! =9.1*%#."07! <")%&! ,&7*#<*&+! -'0%+*#%0! 0,&$! "F*#!+C*!&*S+!:!)"&+C(;!4C*!'&+#%F*&",(!D"#+!<%(!#*)"F*7;! +#%&(D0%&+%+'"&;!L",+'&*!D"(++#%&(D0%&+!-#"&/C'%0!<%(C'&$(! 4*&!)"&+C(!0%+*#E!(C*!#*)%'&*7!<*00!%&7!/")D0'%&+!<'+C! $#*

HVC" O20*)+,)"8,401-+&.$"P+$0#$0$"Q"'''71617)&9R0+6"Q"S&%7"AG!"T&7"H!"U#*15"?@AA

256

M. lentiﬂ avum"+,"P*+,K+,)"e#-0*"].::%+0$

*S%)'&%+'"&E!-,+!/1+"0"$'/!*S%)'&%+'"&!"G!%!01)DC!&"7*! Z0*F*&!<*#*!#*D"#+*701!(+%-0*!"#!')D#"F*7!%+!G"00"<.,DE!:! %(D'#%+*!G#")!+C*!/C'07!(C"<*7!01)DC"/1+*(E!)%/#"DC%$*(E! 7'*7E!%&7!A!C%7!,&/*#+%'&!",+/")*(; &*,+#"DC'0(E!%&7!G#%$)*&+(!"G!*D'+C*0'"'7!C'(+'"/1+*(!-,+!&"! <*00.G"#)*7!$#%&,0")%(;!X#")!=!"+C*#!D%+'*&+(!2AN.1*%#. Discussion "07! <")%&E! /C#"&'/! 0*$! ,0/*#\! N9.1*%#."07! <")%&E! D"(+! M. lentiﬂ avum!7'(*%(*!/%&!-*!7'G!!/,0+! +"! 7'%$&"(*E! +C1#"'7*/+")1!<",&7!%-(/*((6E!M. lentiﬂ avum!<'+C",+!S. %(! +C*! /%(*(! '&! +C'(! #*D"#+! *S*)D0'G1;! 4C*! /0'&'/%0! aureus!<*#*!/,0+,#*7\!+C*!D%+'*&+(!<*#*!+#*%+*7!<'+C!<",&7! '&G"#)%+'"&!<*!$%+C*#*7!<%(!0%#$*01!#*+#"(D*/+'F*E!

Global Case Reports Q&!AI!/%(*(!"G!/0'&'/%001!('$&'!!/%&+!7'(*%(*!D,-0'(C*7! '&! Z&$0'(C! 2"&0'&*! MDD*&7'S! 4%-0*E! <<<;/7/;$"FaZQPa /"&+*&+a8BaAaA9N.%DD4;C+)6E! 7'(*%(*! (D*/+#,)! F%#'*7! G#")!/*#F'/%0!01)DC%7*&'+'(!2H!"G!9!/%(*(!'&!/C'07#*&6!+"! %/,+*!"#!/C#"&'/!7'(*%(*!,(,%001!%GG*/+'&$!0,&$(!%&7!D0*,#%! 2'&!!0+#%+*(E!/%F'+'*(E!&"7,0*(E!*GG,('"&(6!-,+!%0("!%#+C#'+'(a 7'(/'+'(E!-"&*!0*('"&(E!(O'&!,0/*#(E!%&7!C*D%+"(D0*&")*$%01;! 4C*!#%D'7!"&(*+!"G!/*#F'/%0!01)DC%7*&'+'(!C%(!-**&!&"+*7! '&! )%&1! #*D"#+(E! ,(,%001! <'+C! %&! *S/*00*&+! ",+/")*! G#")! *S/'('"&! %0"&*;! 4C*! )*%&! %$*! "G! %7,0+(! <'+C! `+).*0" H7" P0,6*&)*#2" #,6" 9+*-.#%" )0%" +2#)0$" *0:*0$0,-+,)" *0:( M=N" !",)0*:*+,-" :#--0*,$" &4" AI" 5.2#," #,6" G" '#-0*" +$&%#-0$" &4" &"&01)DC%7*&'+'(!7'(*%(*!2=I!/%(*(6!<%(!N:!1*%#(!2#%&$*! Mycobacterium lentiﬂ avum!" E*+$/#,0!" <.00,$%#,6!" >.$-*#%+#!" =A@HB!1*%#(6E!%&7!+C*1!<*#*!*F*&01!(D0'+!-*+<**&!+C*!(*S*(;! ?@@AB?@@C7" =8!" 1%+,+1#%" +$&%#-0J" e!" :&-#/%0" '#-0*" +$&%#-0J" E>g!" Z0*F*&!/%(*.D%+'*&+(!C%7!%(("/'%+*7!')),&"/")D#")'(*;! /*&,15&#%90&%#*"%#9#)07"b=%+,+1#%%3"$+),+!"1#,-"+$&%#-07

" O20*)+,)"8,401-+&.$"P+$0#$0$"Q"'''71617)&9R0+6"Q"S&%7"AG!"T&7"H!"U#*15"?@AA" HVV

257

NO]O>N=Y

,('&$! "!,/0"S%/'00'&!<'+C!"#!<'+C",+!7#%'&%$*!%GG*/+*7!%! D"01)"#DC'()! %&7! #*D.W?LE! #*D"#+'&$! 9Ac! (')'0%#'+1! /")D0*+*!/,#*!'&!%00!/%(*(; %(! +C*! /,+"GG! F%0,*! G"#! /0,(+*#*7! M. tuberculosis! '("0%+*(! M. lentiﬂ avum!'(!%!#%#*!'("0%+*!%&7!%&!,&,(,%0!/%,(*!"G! %&7!9=c!G"#!M. avium!266;!4C*!%&%01('(!"G!M. abscessus 7'(*%(*!'&!C,)%&(;!M(!<'+C!"+C*#!345E!'+!/%&!-*!'("0%+*7! -1!f*0%e&1!*+!%0;E!+C*!0%#$*(+!(+,71!"G!#*D.W?L!'&!345E! G#")! /"&+%)'&%+*7! (%)D0*(n! /0'&'/%0! ('$&'!!/%&/*! (C",07! ,(*7!#*D.W?L!+"!(,//*((G,001!/0,(+*#!M. abscessus!(+#%'&(! -*! %((*((*7! -*G"#*! %&1! +#*%+)*&+! '(! /"&('7*#*7! 226;! Q&! +C%+!<*#*!/0"&%001!#*0%+*7!-1!WXYZ!%&%01('(!2356;!X",#!"G! =IINE! "G! VHH! D%+'*&+(! <'+C! D,0)"&%#1! 345! '("0%+*(! '&! +C*!<%+*#!(%)D0*(!/"&(+'+,+'&$!D#"!!0*!MA!%&7!8!,&#*0%+*7! U,**&(0%&7E!"&01!=:;:c!<*#*!/"&('7*#*7!+"!C%F*!/0'&'/%001! (+#%'&!2D#"!!0*!Y6!/%)*!G#")!('+*(!+C%+!(C%#*7!%!$#",&7<%+*#! ('$&'!!/%&+!7'(*%(*!2316;!4C*!D#"D"#+'"&!<%(!C'$C*#!G"#!M. (",#/*;!4C*(*!!!&7'&$(!(,$$*(+!%!7")'&%&+!*&F'#"&)*&+%0! intracellulare!2A9;Vc6E!M. avium!2AA;Ac6E!%&7!M. kansasii! (+#%'&! /0"(*01! #*0%+*7! 29Ic6E! -,+! &"+! '7*&+'/%0E! +"! (+#%'&(! 2N=;:c6!%&7!),/C!0"<*#!G"#!(D*/'*(!+#%7'+'"&%001!+C",$C+! G",&7!'&!C,)%&!(D*/')*&(!%&7!%(!%!/%,(*!"G!C,)%&!7'(*%(*;! +"!-*!)"#*!0'O*01!/"&+%)'&%&+(E!*;$;E!M. gordonae!288;8c6;! 4C*! +C*"#1! "G! 7")'&%&+! 0"/%0! *&F'#"&)*&+%0! (+#%'&(! '(! Q&! ",#! (+,71E! '("0%+*(! G"#! V! 288c6! "G! A:! <*#*! /0'&'/%001! (,DD"#+*7!-1!+C*!!!&7'&$!"G!%!7'GG*#*&+!(+#%'&!+1D*!G#")!=! ('$&'!!/%&+E! (')'0%#! +"! D,-0'(C*7! *(+')%+*(! "G! 8Ic@=8c! D%+'*&+(!0'F'&$!&*%#!*%/C!"+C*#!-,+!VNI!O)!G#")!R#'(-%&*! 216E226;! 4C'(! D#"D"#+'"&! )%1! -*! %&! ,&7*#*(+')%+*! $'F*&! 2D#"!!0*!P6; +C%+! <*! /",07! &"+! 7*+*#)'&*! /0'&'/%0! ('$&'!!/%&/*! '&! V! W#"!!0*! M! /"&+%'&*7! /0'&'/%001! ('$&'!!/%&+! %&7! D%+'*&+(;!_"#07<'7*E!M. lentiﬂ avum!)%1!-*!,&7*##*D"#+*7! &"&('$&'!!/%&+!'("0%+*(;!W#"!!0*!R!%0("!/"&+%'&*7!%!D%'#!"G! %&7!'&/"##*/+01!'7*&+'!!*7!%(!"+C*#E!)"#*!G%)'0'%#!(D*/'*(E! C'$C01! (')'0%#! '("0%+*(! 29Vc6! "G!

F@@" O20*)+,)"8,401-+&.$"P+$0#$0$"Q"'''71617)&9R0+6"Q"S&%7"AG!"T&7"H!"U#*15"?@AA

258

M. lentiﬂ avum"+,"P*+,K+,)"e#-0*"].::%+0$

Acknowledgments 8I;!! R*&("&!PME!j%#(/C.5'e#%/C'!QE!['D)%&!PÈ!d(+*00!È!K%1*#(!Z_;! _*!+C%&O!?C#'(!5%/7*#)"++!G"#!345!7%+%-%(*!%(('(+%&/*E! Y*&R%&O;! 3,/0*'/! M/'7(! L*(;! =II9\AB! 27%+%-%(*! '((,*6nP=:@A8;! ZD,-!=IIH!d/+!=8; ?C#'(! ?",0+*#! G"#! %7F'/*E! P%F'7! [;! W%+*#("&! G"#! D#"F'7'&$! +C*! 88;!! Q-tu*e!LE!K*##%&".T*#%&e!LE!`')]&*e.W%0"D!5E!L")t&!?E!?"#+*$. Mycobacterium!O'+E!%&7![%-?C'D!2-'"5*#'*,SE!5*0-",#&*E!^Q?E! ,*#%!5E!`')]&*e!K;!P'((*)'&%+*7!'&G*/+'"&!/%,(*7!-1!(0"<.$#"<. M,(+#%0'%6!G"#!,(*!"G!+C*!M$'0*&+!=8II!M&%01e*#!%&7!!!&$*#D#'&+! '&$! Mycobacterium lentifl avum;! Z,#! `! ?0'&! 5'/#"-'"0! Q&G*/+! P'(;! %&%01('(! ("G+<%#*;! _*! %0("! +C%&O! +C*! G"00"<'&$! /0'&'/'%&(! =II=\=8n:98@=;!PdQn!8I;8IIBa(8II9:.II=.IHIV.A 8=;!! 5"&+*g"!5E!Y"'/"*+S*%!È!M$*(+%!3E!Y'0!ME!i##%!ZE!`')*&*e!5K;! G"#! D#"F'7'&$! /0'&'/%0! 7%+%n! `"C&! M,O*(E! 5'/C%*0! R'&+E! M00%&! ?,+%&*",(!'&G*/+'"&!-1!Mycobacterium lentifl avum!'&!%!D%+'*&+!<'+C! X'&&')"#*E! W%,0! Y#'G!!&E! W*+*#! T"DO'&(E! `"(*DC! 5/?"#)%/OE! TQ^;!P*#)%+"0"$1;!=II:\=8An8BA@V;!PdQn!8I;88N9aIIII9AH:A _'00'%)!d0'F*#E!Z#'/!W*#+&'O"F(E!K%&7#%!L'ee"E!P%F'7!K"<7*&E! 8A;!! T%%(*!YE!j*&+#,D!TE!KO"D&'O!TE!KD#'&$*#!RE!R"++$*#!Z?;!Myco- %&7!KC1%)!K,&7*#; bacterium lentifl avumn!%&!*+'"0"$'/!%$*&+!"G!/*#F'/%0!01)DC%7*&'+'(;! ?0'&!Q&G*/+!P'(;!899B\=Nn8=VN@:;!PdQn!8I;8IH:aN8:9NH 4C*! R#'(-%&*! <%+*#! (%)D0'&$! <%(! G,&7*7! -1! 4C*! W#'&/*! 8V;!! 4"#+"0'!ZE!W'*#(')"&'!?E!j'#(/C&*#!WE!R%#+"0"&'!ME!R,##'&'!?E![%/. ?C%#0*(!T"(D'+%0!L*(*%#/C!X",&7%+'"&!%&7!+C*!Y%00'D"0'!L*(*%#/C! /C'&'!?E!*+!%0;!?C%#%/+*#'e%+'"&!"G!)1/"-%/+*#'%0!'("0%+*(!DC10"$*. &*+'/%001!#*0%+*7!+"E!-,+!7'GG*#*&+!G#")!Mycobacterium simiae.!`!?0'&! X",&7%+'"&!"G!Y#**&(0"D*(!W#'F%+*!T"(D'+%0; 5'/#"-'"0;!899B\ANn:9B@BI=; 8N;!! ?%-#'%! XE! 4"##*(! 5Ê! Y%#/'%.?'%! `QE! P")'&$,*e.Y%##'7"! 53E! P#! 5%#(C%00! '(! %! +C"#%/'/! )*7'/'&*! G*00"

" O20*)+,)"8,401-+&.$"P+$0#$0$"Q"'''71617)&9R0+6"Q"S&%7"AG!"T&7"H!"U#*15"?@AA" F@A

259

NO]O>N=Y

=H;!! KC'&!KE!o""&!`TE!K"&$!KTE!j')!Z?;!Q("0%+'"&!"G!Mycobacterium AA;!! 4('+O"!QE!L%CO'0%!LE!W#'C%!dE!M0'.^*C)%(!4E!4*#*G*<"#O!fE!K"'&'!TE! lentifl avum!G#")!%!D%+'*&+!<'+C!%!0,&$!7*(+#"1*7!-1!+,-*#/,0"('(;!j". *+!%0;!Q("0%+'"&!%&7!%,+")%+*7!#'-"+1D'&$!"G!Mycobacterium lentifl a- #*%&!`![%-!5*7;!=IIB\=Bn8=V@B;!PdQn!8I;AAVAaOg0);=IIB;=B;=;8=V vum!G#")!7#'&O'&$!<%+*#!7'(+#'-,+'"&!(1(+*)!%&7!/0'&'/%0!(D*/')*&(;! =9;!! [**! QjE! [',! `_;! d(+*")1*0'+'(! /"&/,##*&+01! /%,(*7! -1! Staphy- XZ5K! 5'/#"-'"0! [*++;! =II:\=N:n=A:@VA;! PdQn! 8I;8888ag;8NBV. lococcus aureus! %&7! Mycobacterium tuberculosis.! K",+C! 5*7! `;! :9:H;=II:;II88:;S =IIB\8IIn9IA@N; AV;!! [**!ZKE![**!5oE!T%&!KTE!j%!`d;!d//,##*&/*!%&7!)"0*/,0%#!7'G. AI;!! X#%&/".W%#*7*(! ?E! R0,)-*#$! T5;! W("%(! ),(/0*! %-(/*((! /%,(*7! G*#*&+'%+'"&!"G!*&F'#"&)*&+%0!)1/"-%/+*#'%!'&!(,#G%/*!<%+*#(;!`!5'. -1! Mycobacterium tuberculosis! %&7! Staphylococcus aureusn! /#"-'"0!R'"+*/C&"0;!=IIH\8Hn8=IB@8N; /%(*! #*D"#+! %&7! #*F'*<;! M)! `! 5*7! K/';! =II8\A=8nV8N@B;! PdQn! AN;!! f*0%e&1!M5E!L""+!`5E!KC*%!oLE!?"0")-"!LZE!KC%)D,+%!Q?E!K+"/O! 8I;8I9BaIIIIIVV8.=II8I:III.IIIIH XE!*+!%0;!?"C"#+!(+,71!"G!)"0*/,0%#!'7*&+'!!/%+'"&!%&7!+1D'&$!"G!My- A8;!! 4C")("&!L5;!?C%&$'&$!*D'7*)'"0"$1!"G!D,0)"&%#1!&"&+,-*#/,. cobacterium abscessus, Mycobacterium massilienseE!%&7!Mycobac- 0",(!)1/"-%/+*#'%!'&G*/+'"&(;!Z)*#$!Q&G*/+!P'(;!=I8I\8:n8NB:@HA; terium bolletii.!`!?0'&!5'/#"-'"0;!=II9\VBn89HN@9N;!PdQn!8I;88=Ha A=;!! 4"#F'&*&!ZE!K,")%0%'&*&!KE![*C+"0%!5È!5'*++'&*&!Q4E!f%/C*,(!dE! `?5;I8:HH.IH W%,0'&![E!*+!%0;!51/"-%/+*#'%!'&!<%+*#!%&7!0""(*!7*D"('+(!"G!7#'&O. '&$!<%+*#!7'(+#'-,+'"&!(1(+*)(!'&!X'&0%&7;!MDD0!Z&F'#"&!5'/#"-'"0;! M77#*((! G"#! /"##*(D"&7*&/*n! T*! 5;! 5%#(C%00E! 4C*! W#'&/*! ?C%#0*(! =IIV\BIn89BA@H8;!PdQn!8I;88=HaMZ5;BI;V;89BA.89H8;=IIV T"(D'+%0@4C"#%/'/!5*7'/'&*E!L"7*!L7E!?C*#)('7*E!R#'(-%&*E!U[P!VIA=E! M,(+#%0'%\!*)%'0n!C*z)%#(C%00{C*%0+C;J07;$"F;%,

F@?" O20*)+,)"8,401-+&.$"P+$0#$0$"Q"'''71617)&9R0+6"Q"S&%7"AG!"T&7"H!"U#*15"?@AA

260

APPENDIX D1

Mycobacterium abscessus in Potable Water

Rachel Thomson123, Carla Tolson4, Chris Coulter4, Flavia Huygens5, Megan Hargreaves5 1Gallipoli Medical Research Centre, Greenslopes Private Hospital, 2Thoracic Medicine, The Prince Charles Hospital, 3QLD TB control centre, 4QLD Mycobacterial Reference Laboratory, 5Queensland University of Technology, QLD, Australia

Introduction Results Mycobacterium abscessus is a rapidly growing The drinking water isolates formed two distinct mycobacteria responsible for progressive pulmonary strain patterns (A and B) that shared >90% disease, soft tissue and wound infections, and can similarity. (Figure 2) The tankwater isolate Figure 8: Strain type of M.Abscessus according to origin of isolate contaminate clinical specimens. The incidence of (pattern C) shared >85% similarity with the disease due to M. abscessus has been increasing potable water isolates, but the pool isolate in Queensland and these infections are notoriously (D) was distinctly different. difﬁcult to treat. Nontuberculous mycobacteria Fifty-three clinical isolates clustered tightly (NTM) are generally considered environmental 40 organisms though M. abscessus has not featured (>95% similarity) with the Group A potable water isolates (Figure 3), 4 patients with Group B (Figure Diversilab frequently in environmental studies, particularly 30 Strain 4). Thirteen patient isolates clustered with the Figure 2: M. abscessus water and pool samples Pattern those examining potable water. In a study of A Rainwater tank isolate (Figure 5). One patient B 20 C Brisbane potable water, M. abscessus was isolated COUNT matched the pool isolate. There were a further D from ten different locations. The aim of this study E 3 patient isolates that were unrelated to the water F was to compare the strains of M. abscessus isolated 10 G from water with those found in clinical samples isolates. (Figure 6) No differences were found between strain types in terms of geographic origin, from patients living in the same catchment area. 0 gender, age, or site/type of infection. LUNG

Methods There were 6 different clinical strain types (A–F) Figure 4: M. abscessus Dendrogam Clone B with Pearson Correlation Analysis INJECTION SITE ASIVE DEVICE/LINE

From a study of Brisbane potable water from various locations around Brisbane (Figure 7) INV

between 2007 and 2009, ten isolates conﬁrmed and from a variety of clinical sources (Figure 8) . SURGICAL WOUND INFECTION Figure 7: Map of Brisbane area showing residential addresses of patients with

as M. abscessus were recovered. In addition, M. abscessus isolates labelled according to Diversilab strain pattern (A,B,C etc) CUT ANEOUS/SOFT TISSUE INFECTION one strain was isolated from a rainwater tank and site of isolate (PS-Pulmonary signiﬁcant, PNS – Pulmonary not signiﬁcant, ST- BLOODSTREAM INFECTION (LINE ASSOC) Soft Tissue, SW- surgical wound, BL- Blood (line related), IS- injection site). Sites SITE OF INFECTION of a patient with disease due to M. avium, and of isolation of M. abscessus isolated from PW (Potable water), TW (Tank water) another from the swimming pool of a patient with M. and SPW (Swimming pool water) similarly shown with Diversilab strain type. Figure 8: Strain type of M. abscessus according to origin of isolate Age* Gender n (%) Species intracellulare disease. A random sample of 74 clinical Mean±SD M:F

isolates referred to the QLD Mycobacterial reference M. abscessus 32 (59.3%) LUNG 54 (73) 58.24±24.53 27:27 M. bolletii 2 (3.7%) laboratory during the same time period was available M. massiliense 20 (37%) for comparison using repPCR strain typing (Diversilab). Conclusion M. abscessus 7 (63.6%) CUTANEOUS/SOFT 11 (14.9) 49.18±14.49 5:6 M. bolletii 2 (18.2%) Figure 5: M. abscessus Dendrogam Clone C with Pearson Correlation Analysis TISSUE INFECTION M. massiliense 2 (18.2%) s M. abscessus has become an increasingly important clinical problem in the last 10 years, and its presence Repetitive DNA elements distributed more or Separation and detection of products by in potable water has not previously been emphasized. INVASIVE DEVICE/LINE 3 (4.1) 42.67±19.63 2:1 M. abscessus 3 (100%) less randomly over the genome electrophoresis using micro-ﬂuidic chips

SURGICAL WOUND M. abscessus 1 (50%) s In the majority of published studies looking for NTM in water, no M. abscessus was documented. 2 (2.7) 46.00±7.07 1:1 INFECTION M. massiliense 1 (50%) s In our study we have more definitively identified members of the M. abscessus complex in potable DNA Fingerprints BLOODSTREAM INFECTION 1 (1.4) 47.00±19.86 1:0 M. abscessus 1 (100%) analysed with (LINE ASSOC) Diversilab® software- water, as strains that are highly related or indistinguishable from those found in samples from patients Pearson correlation Primers that anneal INJECTION SITE 3 (4.1) 41.33±22.75 2:1 M. abscessus 3 (100%) to these repetitive co-efficient and known to have disease. elements unweighted pair group method with arithmetic M. abscessus 47 (63.5%) means to compare Total 74 (100) 55.09±22.88 38:36 M. bolletii 4 (5.4%) s Given the difficulty in treating these infections and the devastating effects they have for patients, isolates and determine M. massiliense 23 (31.1%) clonal relationship efforts to identify effective disinfection methods for M. abscessus should be a high priority. Table 2: Clinical details of patients with M. abscessus isolates used for Figure 6: Swimming pool isolate and single related clinical isolate, Figure 1: Strain typing using rep PCR (Diversilab) strain typing comparison with other unrelated clinical strains of M. abscessus. Figure 3: M. abscessus Dendrogam Clone A with Pearson Correlation Analysis

261

APPENDIX D2

Thomson et al. BMC Infectious Diseases 2013, 13:241 http://www.biomedcentral.com/1471-2334/13/241

RESEARCH ARTICLE Open Access Mycobacterium abscessus isolated from municipal water - a potential source of human infection Rachel Thomson1*, Carla Tolson2, Hanna Sidjabat3, Flavia Huygens4 and Megan Hargreaves5

Abstract Background: Mycobacterium abscessus is a rapidly growing mycobacterium responsible for progressive pulmonary disease, soft tissue and wound infections. The incidence of disease due to M. abscessus has been increasing in Queensland. In a study of Brisbane drinking water, M. abscessus was isolated from ten different locations. The aim of this study was to compare genotypically the M. abscessus isolates obtained from water to those obtained from human clinical specimens. Methods: Between 2007 and 2009, eleven isolates confirmed as M. abscessus were recovered from potable water, one strain was isolated from a rainwater tank and another from a swimming pool and two from domestic taps. Seventy-four clinical isolates referred during the same time period were available for comparison using rep-PCR strain typing (Diversilab). Results: The drinking water isolates formed two clusters with ≥97% genetic similarity (Water patterns 1 and 2). The tankwater isolate (WP4), one municipal water isolate (WP3) and the pool isolate (WP5) were distinctly different. Patient isolates formed clusters with all of the water isolates except for WP3. Further patient isolates were unrelated to the water isolates. Conclusion: The high degree of similarity between strains of M. abscessus from potable water and strains causing infection in humans from the same geographical area, strengthens the possibility that drinking water may be the source of infection in these patients.

Background swimming and bathing. Also aerosols generated during Nontuberculous mycobacteria are environmental organ- these activities may be inhaled, [8] potentially resulting isms that can cause progressive lung disease in suscep- in disease. There is a recognized association between tible patients. M. abscessus is a significant problem as it pulmonary infection with rapid growing mycobacteria is highly resistant to antimicrobial agents and usually [9,10] (M. abscessus and M. fortuitum) and esophageal requires prolonged treatment (>six months) with intra- disorders. It is possible that patients acquire infection by venous and oral antibiotics in combination [1-3]. The in- aspirating contaminated water [11]. Outbreak investiga- fection is often relentless despite treatment, and even if tions have identified M. abscessus in hospital water, dia- treatment is apparently effective, relapse is common. lysis [12,13] and surgical equipment and endoscopy Rapidly growing mycobacteria such as M. abscessus are cleaning equipment [14,15]. In many of these instances also well documented causes of skin and soft tissue in- it is not clear whether the source of these outbreaks was fections, especially complicating surgical procedures and an infected patient or water. injection sites [4]. The taxonomy of Mycobacterium abscessus has been a Water is an important environmental reservoir of topic of debate in the literature. M. massiliense and M. mycobacteria causing human disease [5-7]. Humans are bolletii cannot be separated from M. abscessus on the exposed to waterborne mycobacteria through drinking, basis of extensive phenotypic analysis and genotypic studies have lead to the proposition that the three taxa * Correspondence: [email protected] represent a single species with internal variability [16]. 1Gallipoli Medical Research Centre, Greenslopes Private Hospital, Brisbane, Strain variation amongst species of mycobacteria is well QLD, Australia known and Pulsed Field Gel Electrophoresis (PFGE) has Full list of author information is available at the end of the article

262

Thomson et al. BMC Infectious Diseases 2013, 13:241 Page 2 of 7 http://www.biomedcentral.com/1471-2334/13/241

been considered the “Gold standard” for strain typing between M. tuberculosis, M. avium, M. intracellulare, M. of many mycobacteria. ERIC (enterobacterial repetitive abscessus and other Mycobacterium spp. Isolates identified intergenic consensus) PCR has been used successfully as other Mycobacterium spp were further speciated using to differentiate strains of mycobacteria associated with Hain Life Sciences GenoType® Mycobacterium CM kit outbreaks of disease in mesotherapy clinics due to M. (2004–7 only) and/or 16S rDNA sequencing in conjunc- abscessus and M. chelonae, and in post mammoplasty tion with phenotypic characteristics. All clinical isolates patients with M. fortuitum infections [17,18]. A high underwent gene fragment sequencing for hsp65 and rpoB. concordance with PFGE was shown. More recently rep- PCR (Diversilab) was used in comparison to PFGE in a Strain typing using automated Rep-PCR suspected outbreak of M. abscessus in a cystic fibrosis The clonality of clinical and water M. abscessus isolates clinic [19]. Isolates that were identical on PFGE shared and a control strain (ATCC 19977) was determined using only 90% similarity with rep-PCR, suggesting that the a rep-PCR based method (Diversilab® system, bioMerieux, latter method may be more discriminatory. Rep-PCR Melbourne). DNA was extracted from clinical and water was similarly shown to concur with strain typing using isolates using the Ultraclean Microbial DNA Isolation Kit VNTR (variable number tandem repeats) [20]. This (Mo Bio Laboratories, CA, USA). PCR mixture was pre- method has significant advantages in cost and time sa- pared using AmpliTaq polymerase and PCR buffer (Ap- ving over other methods and has demonstrated high plied Biosystems, New Jersey, USA) and Mycobacterium discriminatory power. Diversilab® primer mix according to the manufacturer’s in- In Queensland NTM disease remains notifiable and a structions. Separation and detection of Rep-PCR products central reference laboratory performs speciation of all was performed by micro-fluidic chips of the Diversilab® culture positive isolates, providing a unique opportunity System. A laboratory control strain of M. abscessus was to capture isolates from around the state. The incidence included, and two chips were repeated because of low in- of disease due to NTM has been increasing [21] and be- tensity banding. In most cases when the intensity im- tween 2007–2008 water sampling was conducted to in- proved the pattern didn’t change from that obtained with vestigate the presence of pathogenic NTM in potable low intensity. An additional chip was run twice to confirm water [22]. This paper reports the comparison of these reproducibility of strain patterns. Any aberrances were ex- water isolates with human isolates collected from pa- cluded from the analysis. Fingerprints were analysed with tients in South east Queensland during the same time Diversilab® software v.3.4.38 using the Pearson correlation period. co-efficient and unweighted pair group method with arithmetic means to compare isolates and determine clonal re- Methods lationship. A similarity index of ≥97% was used to define Water isolates obtained in previous studies [22] had been isolates that were indistinguishable, based on the manu- stored in Dubos broth at −20°C and were thawed and facturers recommendations despite previously published subbed onto 7H11 plates as well as Lowenstein-Jensen comparisons of rep-PCR with both PFGE and Variable slopes and incubated at 35°C until sufficient growth was Number Tandem Repeats, [20] where a cut off of 90% was available. used. Isolates that shared 95% similarity were considered Those identified using 16S rDNA sequencing as M. similar (and formed groups), and those <90% considered abscessus/M. chelonae underwent hsp65 and rpoBgene ‘different/unrelated’. fragment sequencing for more definitive identification. The study protocol was approved by the Human Re- According to the proposed changes by Leao et al. [16] search Ethics Committee of The Prince Charles Hospital in line with the Bacteriological code, for the purposes (EC-2617). of this study all isolates identified as M. abscessus, M. bolletii, M. massiliense were included as M. abscessus species. Results Human samples that were collected as part of routine During the 10 years from 2001–2010, there were 486 pa- patient care, were digested and decontaminated using 4% tient notifications: 68.1% were pulmonary, 22.2% cutane- NaOH, neutralized with phosphoric acid and centrifuged ous or soft tissue (Table 1). at 3000g to concentrate the acid-fast bacilli (AFB). Smears Seventy-four clinical isolates collected during 2007–8 were prepared from the sediment and stained by the were available for comparison with fifteen water isolates. Ziehl-Niehlsen (ZN) method. One Lowenstein-Jensen The mean age of patients was 55.94 yrs (SD 22.25; me- slope (±pyruvate) and 7ml Mycobacterial Growth Indica- dian 58.5; range 2-90yrs). Fifty percent were male. Forty- tor Tube (MGIT) were inoculated and incubated at 35°C four patients had clinically significant pulmonary disease until growth was detected. ZN staining of colonies con- (according to clinician notification and the ATS/IDSA firmed AFB. Multiplex PCR was performed to discriminate criteria), 11 patients had pulmonary isolates that were

263

Thomson et al. BMC Infectious Diseases 2013, 13:241 Page 3 of 7 http://www.biomedcentral.com/1471-2334/13/241

Table 1 Sites of isolation of M. abscessus from human Of 19 water isolates identified as M. abscessus/ samples 2001-2010 M. chelonae using 16s rDNA sequencing, 17 underwent Site of isolate Frequency (%) successful hsp65 and rpoB sequencing. This identified Blood 14 (2.9) 14 isolates as M. abscessus and one as M. bolletii – all Bones and joints 5 (1) considered M. abscessus subs abscessus according to the proposed taxonomy by Leao [16]. There was insuffi- Cutaneous/Soft tissue 108 (22.2) cient DNA in two samples. Fifteen water isolates of Eye 1 (0.2) M. abscessus were strain typed using rep-PCR. Eleven Lymph node + Other 1 (0.2) isolates came from six different potable water distribu- Lymph nodes 3 (0.6) tion system sampling sites; two came from domestic Peritoneal 2 (0.4) taps in the same dwelling, one came from an unrelated Post surgical/Medical access device 19 (3.9) rainwater tank and the other from a suburban domestic swimming pool. Pulmonary 331 (68.1) The 15 water isolates formed 6 different patterns Unknown 2 (0.4) (Figure 1). Municipal water isolates formed two clus- Total 486 (100) ters of indistinguishable isolates (≥97% similarity) and were classified as Group WP1 (eight isolates) and Group WP2 (two isolates) and a single isolate WP3. The rain- considered contaminants or not clinically significant and water tank isolate (Group WP4) and the swimming pool 19 patients had isolates from extra-pulmonary sites. isolate (Group WP5) were distinctly different. The two Eleven patients (14.9%) were included from other parts isolates from a patient’s domestic taps were indistin- of Australia who lived >500 km outside the Brisbane guishable (WP 6) (Figure 1). water distribution network, and would have received When the clinical isolates were analysed with the potable water supply to their homes from different net- water isolates there were 28 patterns/clusters that shared works. Of the remaining patients, 31/63 (49.2%) patients >97% similarity (indistinguishable). Thirteen of these clus- lived in reservoir zones that contained sample points ters contained more than 1 isolate, and the remaining that grew M. abscessus. patterns were single isolates only. The indistinguishable

Figure 1 rep-PCR dendrogram of M. abscessus water strains with Pearson Correlation Analysis.

264

Thomson et al. BMC Infectious Diseases 2013, 13:241 Page 4 of 7 http://www.biomedcentral.com/1471-2334/13/241

clusters that were similar (>95% similarity) were then The isolate from water site 320 (WP3) differed from grouped into seven main groups. Full report and graphics all the clinical isolates. in Additional file 1. There were no significant differences between the There were two large similar clusters of 21 (P15) and characteristics of groups of patients in each strain cluster 22 (P17) isolates respectively. Examples demonstrated (age, gender, type of disease, site of infection, ground in Figure 2. P15 included one of the municipal water water source or reservoir zone) (Table 2). Some of the isolates from WP1 and P17 contained the remaining isolates from more remote areas did have unique strain seven isolates from WP1. A further three clusters (P12- patterns, but equally so did some of the Brisbane area 14) containing 5 clinical isolates were similar to cluster isolates. Some of the remote area isolates were similar to P15 and a single isolate P16 was similar to P17. those of patients from Brisbane. However as is often the The two water isolates from WP2 also clustered with case with NTM infections, the precise location of where six clinical isolates (P20-22;Gp 4: Additional file 1). Five infection is acquired can be difficult to pinpoint. clinical isolates were indistinguishable from the tank water isolate (P2). A further three clinical isolates and Discussion the two domestic tap water isolates were similar to this M. abscessus has become an increasingly important cli- cluster (Group 1). One clinical isolate was indistinguish- nical problem in the last 10 years, and its presence in able from the swimming pool isolate Figure 3. potable water has not previously been emphasized.

Figure 2 rep-PCR dendrogram of examples of main clinical strain patterns 17 (1, green) and 15 (2, orange) demonstrating similarities between water and clinical isolates. (Full dendrogram in Additional file 1).

265

Thomson et al. BMC Infectious Diseases 2013, 13:241 Page 5 of 7 http://www.biomedcentral.com/1471-2334/13/241

Figure 3 rep-PCR dendrogram showing examples of different strain patterns where there were similarities between clinical and water strains.

In the majority of published studies looking for NTM in Taiwan, M. abscessus was grown from one water sample in water, no M. abscessus was documented [23]. How- from one of the hospitals involved however this differed ever there have been taxonomical changes that have led on strain typing from the isolates found in patients [25]. to M. abscessus being recognized as independent from Falkinham [26] in 2001 sampled 8 different water treat- M. chelonae, so older studies reporting M. chelonae may ment systems (raw, treated and Distribution System (DS) not necessarily have differentiated M. chelonae subsp samples). M. abscessus was only found in one raw water abscessus, particularly as both species have identical 16s sample from one system. There were no rapid growers in rDNA sequences. But in studies of potable water done treated water, two rapid growers were found in the DS since 2000 M. abscessus has been rarely reported. samples but the paper doesn’t clarify the species. In a more A study done in Pretoria, South Africa in 2004 [24] recent study (2009), van Ingen reported no M. abscessus in reported an analysis of 78 biofilm and water samples – 14 shower or tap water in the Netherlands [27]. We have had NTM – no MAC, one had M. abscessus (16s rDNA found that the inclusion of liquid media increased the yield sequencing only). In a multi-hospital outbreak investigation for M. abscessus from potable water samples [22]. However

Table 2 Clinical details of patients with M. abscessus isolates used for strain typing comparison n (%) Age* Mean±SD Gender M:F Species Lung 54 (73) 58.24±24.53 27:27 M. abscessus subs abscessus 34 (63%) M. abscessus subs bolletii 20 (37%) Cutaneous/Soft Tissue Infection 11 (14.9) 49.18±14.49 5:6 M. abscessus subs abscessus 9 (81.8%) M. abscessus subs bolletii 2 (18.2%) Invasive Device/Line 3 (4.1) 42.67±19.63 2:1 M. abscessus subs abscessus 3 (100%) Surgical Wound Infection 2 (2.7) 46.00±7.07 1:1 M. abscessus subs abscessus 1 (50%) M. abscessus subs bolletii 1 (50%) Bloodstream Infection (Line Assoc) 1 (1.4) 47.00±19.86 1:0 M. abscessus subs abscessus 1 (100%) Injection Site 3 (4.1) 41.33±22.75 2:1 M. abscessus subs abscessus 3 (100%) Total 74 (100) 55.09±22.88 38:36 M. abscessus subs abscessus 51 (68.9%) M. abscessus subs bolletii 23 (31.1%)

266

Thomson et al. BMC Infectious Diseases 2013, 13:241 Page 6 of 7 http://www.biomedcentral.com/1471-2334/13/241

three of the 13 isolates were only grown on solid media – Harris [20] describes a novel VNTR (variable number suggesting that for future studies using culture-based tandem repeats) method to type 41 M. abscessus com- techniques, both media forms should be included. plex strains from 17 pediatric cystic fibrosis patients and It has long been recognized that outbreaks of disease compares this with rep-PCR (Diversilab). As with the due to rapid growers may have originated in hospital and Aitken paper, isolates with 90% similarity were regarded tap water, particularly for M. chelonae and M. fortuitum as indistinguishable. The finding of six distinct profiles [28] though in some studies the authors have acknowl- among 12 patients with M. abscessus complex suggests edged that speciation separating M. abscessus has not oc- that the typing methods have sufficient discriminatory curred. Outbreak investigations have linked M. abscessus power. infections and pseudo infections to hospital tap water, and endoscope cleaning fluids [29,30]. Conclusion In all of these hospital outbreaks, the potential for con- We have analysed both patient and water isolates from tamination of hospital sources and equipment by infected quite diverse and epidemiologically unrelated settings patients or other environmental sources (such as dust or across a geographical area of >1367km2 and found 28 dirt) exists; hence the origin of the culprit strain remains clusters. We have more definitively identified M. abscessus unclear. strains in potable water, as strains that are indistinguish- Zelazny [31] compared partial sequencing of rpoB, able from those found in samples from patients known to hsp65 and secA gene fragments and rep-PCR (Diversilab) have disease. Whilst point source investigations of out- in order to differentiate M. abscessus from subspecies breaks have linked M. abscessus infection and hospital formerly named M. massiliense and M. bolletii. They water and equipment, the present study tightens the link were able to group the majority of M. abscessus complex between potable water, to which patients are regularly ex- clinical isolates into a group similar to the ATCC strain posed, and the acquisition of infection. This has important 19977, with a degree of diversity as low as 80% similarity implications for those patients at risk. Given the difficulty to the reference strain. There were two other strain in treating these infections and the devastating effects they groups neither of which resembled the strain patterns have for patients, efforts to identify effective disinfection obtained in our study. The geographic origin of the methods for M. abscessus should be a high priority. patients included in this study was not reported nor examined, but as the study was performed by a tertiary Additional file referral centre many came from patients around the United States (A. Zelazny, personal communication). In Additional file 1: Full Diversilab rep-PCR analysis of water and our study the strains obtained both from patients and human isolates. from water were closely related, and differed from the strain patterns reported in the US, suggesting geogra- Competing interests phical differences in strain variation. The diversity of the All authors declare that they have no competing interests. US strains, may be due to the diverse origins of the Authors’ contributions isolates reported, reflecting differences in the dominant RT conceived and designed the study, processed the water samples, environmental strains. There was a degree of geographic performed the Diversilab analysis and wrote the manuscript. CT processed water and human samples, performed DNA extraction and rep-PCR. HS diversity of our strains with unique strain types iden- performed rep-PCR experiments with CT. FH and MH contributed to the tified from central and far north Queensland (>500km design and conduct of the study, the analysis of the results, and the writing and > 1000km from Brisbane respectively) and two unique of the manuscript. All authors reviewed and approved the final manuscript. strains from the bayside of Brisbane (> 30km from the Acknowledgements CBD). This work was supported by research grants from The Gallipoli Medical Whilst rep-PCR using the Diversilab has only recently Research Foundation and The Prince Charles Hospital Foundation, Brisbane, Australia. The authors acknowledge the contribution of Dr Chris Coulter and been used for strain typing mycobacteria, published studies staff at the QLD Mycobacterial Reference Laboratory for their support. support its performance as equivalent or perhaps even more discriminatory than PFGE. Aitken et al [19] reported Author details 1Gallipoli Medical Research Centre, Greenslopes Private Hospital, Brisbane, an outbreak of M. abscessus infections in a lung transplant- QLD, Australia. 2QLD Mycobacterial Reference Laboratory, Pathology ation and cystic fibrosis center. Isolates involved were Queensland, RBWH Campus, Herston Rd, Herston, QLD 4006, Australia. 3 strain typed using both PFGE and rep-PCR (Diversilab). University of Queensland Centre for Clinical Research, Herston Rd, Herston, QLD 4006, Australia. 4Institute of Health and Biomedical Innovation, Kelvin The isolate of the index case and outbreak cases were Grove Campus, Queensland University of Technology, Brisbane, QLD 4059, reported to be “genetically identical” using PFGE. The rep- Australia. 5Queensland University of Technology, Faculty of Science and PCR patterns of the same isolates showed only 90% genetic Technology, George Street, Brisbane, QLD 4001, Australia. similarity, suggesting in fact that these strains are different, Received: 29 January 2013 Accepted: 10 May 2013 and that rep-PCR may be more discriminatory than PFGE. Published: 25 May 2013

267

Thomson et al. BMC Infectious Diseases 2013, 13:241 Page 7 of 7 http://www.biomedcentral.com/1471-2334/13/241

References 22. Thomson RM, Carter R, Tolson C, Coulter C, Huygens F, Hargreaves M: 1. Jeon K, Kwon OJ, Lee NY, Kim B-J, Kook Y-H, Lee S-H, Park YK, Kim CK, Koh Factors associated with the isolation of Nontuberculous Mycobacteria W-J: Antibiotic treatment of Mycobacterium abscessus lung disease: (NTM) from a large municipal water system in Brisbane, Australia. BMC a retrospective analysis of 65 patients. Am J Respir Crit Care Med 2009, Microbiology, 13:89. 180(9):896–902. 23. Van Ingen J, Boeree M, Dekhuijzen P, Van Soolingen D: Environmental 2. Thomson R, Yew W-W: When and how to treat pulmonary non-tuberculous sources of rapid growing nontuberculous mycobacteria causing disease mycobacterial diseases. Respirology 2009, 14(1):12–26. in humans. Clin Micro Inf 2009, 15:888–892. 3. Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, 24. September SM, Brozel VS, Venter SN: Diversity of nontuberculoid Holland SM, Horsburgh R, Huitt G, Iademarco MF, Iseman M, Olivier K, Mycobacterium species in biofilms of urban and semiurban drinking Ruoss S, von Reyn CF, Wallace RJ Jr, Winthrop K, on behalf of the ATS water distribution systems. Appl Environ Microbiol 2004, 70(12):7571–3. Mycobacterial Diseases Subcommitte: An Official ATS/IDSA statement: 25. Huang W-C, Chiou C-S, Chen J-H, Shen G-H: Molecular epidemiology of diagnosis, treatment, and prevention of nontuberculous mycobacterial Mycobacterium abscessus infections in a subtropical chronic ventilatory diseases. Am J Respir Crit Care Med 2007, 175:367–416. setting. J Med Microbiol 2010, 59(10):1203–11. 4. De Groote MA, Huitt G: Infections due to rapidly growing Mycobacteria. 26. Falkinham J III, Norton C, Le Chavallier M: Factors influencing numbers of Clin Infect Dis 2006, 42(12):1756–63. Mycobacterium avium, Mycobacterium intracellulare and other 5. Covert TC, Rodgers MR, Reyes AL, Stelma GN Jr: Occurrence of mycobacteria in drinking water distribution systems. Appl Environ nontuberculous mycobacteria in environmental samples. Appl Environ Microbiol 2001, 67(3):1225–31. Microbiol 1999, 65(6):2492–6. 27. Van Ingen J, Blaak H, de Beer J, de Roda Husman AM, Van Soolingen D: 6. Fox C, Smith B, Brogan O, Rayner A, Harris G, Watt B: Non-tuberculous Rapidly growing nontuberculous mycobacteria cultured from home tap mycobacteria in a hospital's piped water supply. J Hosp Infect 1992, and shower water. App Environ Microbiol 2010, 76(17):6017–6019. 21(2):152–4. 28. Wallace RJ, Brown BA, Griffith DE: Nosocomial Outbreaks/pseudooutbreaks 7. Le Dantec C, Duguet J-P, Montiel A, Dumoutier N, Dubrou S, Vincent V: caused by nontuberculous mycobacteria. Annu Rev Microbiol 1998, Occurrence of mycobacteria in water treatment lines and in water 52(1):453–90. distribution systems. Appl Environ Microbiol 2002, 68(11):5318–25. 29. Maloney S, Welbel S, Daves B, Adams K, Becker S, Bland L, Arduino M, 8. Falkinham J III: Mycobacterial aerosols and respiratory disease. Emerg Wallace R Jr, Zhang Y, Buck G, Risch P, Jarvis WR: Mycobacterium abscessus Infect Dis. 2003, 9:763–7. pseudoinfection traced to an automated endoscope washer: utility of 9. Hadjiliadis D, Adlakha A, Prakash UB: Rapidly growing mycobacterial lung epidemiologic and laboratory investigation. J Infect Dis 1994, 169:1166–9. infection in association with esophageal disorders. Mayo Clinic Proc. 1999, 30. Zhang Y, Rajagopalan M, Brown BA, Wallace RJ Jr: Randomly amplified 74:45–51. polymorphic DNA PCR for comparison of Mycobacterium abscessus 10. Hutchins GM, Boitnott JK: Atypical mycobacterial infection complicating strains from nosocomial outbreaks. J Clin Microbiol 1997, 35(12):3132–9. mineral oil pneumonia. JAMA. 1978, 240:539–41. 31. Zelazny AM, Root JM, Shea YR, Colombo RE, Shamputa IC, Stock F, Conlan 11. Thomson R, Armstrong J, Looke D: Gastroesophageal reflux disease, acid S, McNulty S, Brown-Elliott BA, Wallace RJ Jr, Olivier KN, Holland SM, suppression, and Mycobacterium avium complex pulmonary disease. Sampaio EP: Cohort study of molecular identification and typing of Chest 2007, 131:1166–72. Mycobacterium abscessus, Mycobacterium massiliense, and Mycobacterium 12. Carson LA, Bland LA, Cusick LB, Favero MS, Bolan GA, Reingold AL, Good RC: bolletii. J Clin Microbiol 2009, 47(7):1985–95. Prevalence of nontuberculous mycobacteria in water supplies of hemodialysis centers. Appl Environ Microbiol 1988, 54(12):3122–5. doi:10.1186/1471-2334-13-241 13. Gomila M, Ramirez A, Lalucat J: Diversity of environmental mycobacterium Cite this article as: Thomson et al.: Mycobacterium abscessus isolated isolates from hemodialysis water as shown by a multigene sequencing from municipal water - a potential source of human infection. BMC Infectious Diseases 2013 13:241. approach. Appl Environ Microbiol 2007, 73(12):3787–97. 14. Gubler J, Salfinger M, von Graevenitz A: Pseudoepidemic of nontuberculous mycobacteria due to a contaminated bronchoscope cleaning machine. Report of an outbreak and review of the literature. Chest 1992, 101(5):1245–9. 15. Dawson DJ, Armstrong J, Blacklock ZM: Mycobacterial cross-contamination of bronchoscopy specimens. Am Rev Respir Dis 1982, 126(6):1095–7. 16. Leao SC, Tortoli E, Euzeby JP, Garcia MJ: Proposal that Mycobacterium massiliense and Mycobacterium bolletii be united and reclassified as Mycobacterium abscessus subsp. bolletii comb. nov., designation of Mycobacterium abscessus subsp. abscessus subsp. nov. and emended description of Mycobacterium abscessus. Int J Syst Evol Microbiol. 2011, 61:2311–3. 17. Sampaio JLM, Viana-Niero C, de Freitas D, Hfling-Lima AL, Leao SC: Enterobacterial repetitive intergenic consensus PCR is a useful tool for typing Mycobacterium chelonae and Mycobacterium abscessus isolates. Diagn Micr Infec Dis 2006, 55(2):107–18. 18. Sampaio JLM, Chimara E, Ferrazoli L, da Silva Telles MA, del Guercio VMF, Jerico ZVN, Miyashiro K, Fortaleza CMCB, Padoveze MC, Leao SC: Submit your next manuscript to BioMed Central Application of four molecular typing methods for analysis of and take full advantage of: Mycobacterium fortuitum group strains causing post-mammaplasty infections. Clin Microbiol Inf 2006, 12(2):142–9. • Convenient online submission 19. Aitken ML, Limaye A, Pottinger P, Whimbey E, Goss CH, Tonelli MR, Cangelosi GA, Dirac MA, Olivier KN, Brown-Elliott BA, McNulty S, Wallace RJ: • Thorough peer review Respiratory outbreak of Mycobacterium abscessus subspecies massiliense • No space constraints or color ﬁgure charges in a lung transplant and cystic fibrosis center. Am J Respir Crit Care Med 2012, 185(2):231–2. • Immediate publication on acceptance 20. Harris KA, Kenna DTD, Blauwendraat C, Hartley JC, Turton JF, Aurora P, Dixon • Inclusion in PubMed, CAS, Scopus and Google Scholar GLJ: Molecular fingerprinting of Mycobacterium abscessus strains in a cohort • Research which is freely available for redistribution of pediatric cystic fibrosis patients. J Clin Microbiol, 50(5):1758–61. 21. Thomson R: Changing epidemiology of pulmonary nontuberculous mycobacteria infections. Emerg Inf Diseases 2010, 16(10):1576–82. Submit your manuscript at www.biomedcentral.com/submit

268

APPENDIX E PDF of Manuscript submitted to the International Journal of Medical Microbiology February 14th, 2013

Elsevier Editorial System(tm) for International Journal of Medical Microbiology Manuscript Draft

Manuscript Number:

Title: Strain variation amongst clinical and potable water isolates of M. kansasii using automated repetitive unit PCR.

Article Type: Research Paper

Keywords: Mycobacterium kansasii; Nontuberculous mycobacteria; molecular epidemiology; genotyping; microbial transmission

Corresponding Author: Dr. Rachel Thomson, MBBS

Corresponding Author's Institution: Greenslopes Private Hospital

First Author: Rachel Thomson, MBBS

Order of Authors: Rachel Thomson, MBBS; Tolson Carla, B App Sci; Hanna Sidjabat, PhD; Flavia Huygens, PhD; Megan Hargreaves, PhD

Abstract: Mycobacterium kansasii is a pulmonary pathogen that has been grown readily from municipal water, bur rarely isolated from natural waters. A definitive link between water exposure and disease has not been demonstrated and the environmental niche for this organism is poorly understood. Strain typing of clinical isolates has revealed seven subtypes with Type 1 being highly clonal and responsible for most infections worldwide. The prevalence of other subtypes varies geographically. In this study 49 water isolates are compared with 83 patient isolates from the same geographical area (Brisbane, Australia), using automated repetitive unit PCR. The clonality of the dominant clinical strain type is again demonstrated but with this method strain variation within this group is evident comparable with other reported methods. There is significant heterogeneity of water isolates and very few are similar or related to the clinical isolates. This suggests that if water or aerosol transmission is the mode of infection, then point source contamination likely occurs from an alternative environmental source.

269

*Manuscript

1 Strain variation amongst clinical and potable water isolates of M. kansasii using 2 automated repetitive unit PCR. 3 4 Rachel Thomson MBBS * 5 Gallipoli Medical Research Centre 6 Greenslopes Private Hospital, Brisbane, QLD, Australia 7 [email protected] 8 Ph: 61 7 38478890 Fax: 61 7 38478891 9 10 Carla Tolson B App Sci 11 QLD Mycobacterial Reference Laboratory, Pathology Queensland 12 RBWH Campus 13 Herston Rd, Herston QLD Australia 4006 14 15 Hanna Sidjabat PhD 16 University of Queensland Centre for Clinical Research 17 Herston Rd, Herston QLD Australia 4006 18 [email protected] 19 20 Flavia Huygens PhD 21 Queensland University of Technology, Institute of Health and Biomedical Innovation 22 Kelvin Grove Campus, Brisbane, QLD Australia 4059 23 [email protected] 24 25 Megan Hargreaves PhD 26 Queensland University of Technology, Faculty of Science and Technology 27 George Street, Brisbane QLD Australia 4001 28 [email protected] 29 30 *Corresponding author 31 32 Running title: Strain variation of M. kansasii 33

270

34 Summary: Mycobacterium kansasii is a pulmonary pathogen that has been grown readily 35 from municipal water, bur rarely isolated from natural waters. A definitive link between water 36 exposure and disease has not been demonstrated and the environmental niche for this 37 organism is poorly understood. Strain typing of clinical isolates has revealed seven subtypes 38 with Type 1 being highly clonal and responsible for most infections worldwide. The 39 prevalence of other subtypes varies geographically. In this study 49 water isolates are 40 compared with 83 patient isolates from the same geographical area (Brisbane, Australia), 41 using automated repetitive unit PCR. The clonality of the dominant clinical strain type is 42 again demonstrated but with this method strain variation within this group is evident 43 comparable with other reported methods. There is significant heterogeneity of water isolates 44 and very few are similar or related to the clinical isolates. This suggests that if water or 45 aerosol transmission is the mode of infection, then point source contamination likely occurs 46 from an alternative environmental source. 47 48 49 50 51 52 53 54 55 56 57 58 59 60

61 2

271

62 Introduction 63

64 Mycobacterium kansasii is a pulmonary pathogen found in tap water (Engel et al, 1980). It is 65 the second most common cause of Nontuberculous Mycobacterial (NTM) disease in the 66 United States, South America and areas of high incidence occur around the world – 67 particularly Poland, Slovakia, Paris, (Hoefsloot, van Ingen* et al 2013) South-east England 68 and Wales. In Queensland, Australia, it is the fourth most common cause of NTM pulmonary

69 disease (behind M. intracellulare, M. avium and M. abscessus) (Thomson 2010). 70 71 The typical presentation of disease is a cavitary upper lobe pulmonary disease, affecting 72 middle-aged men, often with underlying COPD and heavy alcohol intake. Underlying 73 malignancy, HIV co-infection and silicosis are known risk factors for disease. The nodular

74 and bronchiectatic form of pulmonary disease seen with M. avium complex has also been

75 described with M. kansasii. Soft tissue infection can also occur, though is rare. (Evans et al 76 1996, Bloch et al 1998, Maliwan and Zvetina 2005) 77

78 The reservoir of M. kansasii is still not well understood, though it has been postulated that

79 water is the natural habitat. Whilst M. kansasii has been readily isolated from tap water, 80 (McSwiggan and Collins 1974, Engel et al 1980b, Steadham 1980) it has only been rarely 81 isolated from natural water sources such as rivers or lakes. It has been rarely recovered from 82 soil samples or animals. A definitive link between an environmental source of infection and 83 human disease has not been shown, nor has human-to-human transmission. 84

85 Seven subtypes of M. kansasii have been described using amplification of the 16s-23s r RNA 86 spacer region, and PCR-restriction fragment length polymorphism (RFLP) analysis of the 87 hsp65 gene (Iinuma et al 1997, Picardeau et al 1997) Type I is the most prevalent clinical 88 isolate worldwide (Alcaide et al 1997, Taillard et al 2003, Santin et al 2004). In Queensland

89 typing was performed on 140 stored clinical isolates of M. kansasii. Type I was again the 3

272

90 most dominant, with other types only evident in the last 2 decades. (Psaltis 2005, Conference

91 presentation) Figure 1. 92 93 In Queensland, the incidence of disease due to NTM has been increasing (Thomson 2010) and 94 between 2007-2008 water sampling was conducted to investigate the presence of pathogenic 95 NTM in potable water (Thomson et al 2013, manuscript under review). A study of NTM 96 patients homes was also conducted (Thomson et al 2013, manuscript under review). The aim

97 of this study was to compare M. kansasii strains found in potable water and patient houses to 98 clinical isolates from QLD patients. In Queensland NTM disease remains notifiable and a 99 central reference laboratory performs speciation of all culture positive isolates, providing a 100 unique opportunity to capture patient isolates from the whole state. This paper reports the 101 comparison of these water isolates with human isolates collected from patients in Queensland 102 during the same time period. 103

104 Methods 105 106 During 2007-2008 citywide water sampling of approximately 220 sites (Trunk Main,

107 Reservoir and Distribution samples) was performed and 49 isolates were identified as M.

108 kansasii using 16s rRNA and hsp 65 gene fragment sequencing. A home sampling study of 109 water, swabs and aerosols from NTM patients houses was also conducted in 2009-10 and 16

110 isolates of M. kansasii were recovered. M. kansasii ATCC strain 12478 was included for 111 comparison. Water isolates had been stored in Dubos broth at -20°C and were thawed and 112 sub-cultured onto 7H11 plates as well as Lowenstein-Jensen slopes and incubated at 35°C 113 until sufficient growth was available. 114 115 We recovered 83 patient isolates from 71 patients– 10 patients had 2 isolates, 1 patient had 3 116 isolates. Human samples submitted to the MRL had been digested and decontaminated using 117 4% NaOH, neutralised with phosphoric acid and centrifuged at 3000g to concentrate the acid- 4

273

118 fast bacilli (AFB). Smears were prepared from the sediment and stained by the Ziehl-Niehlsen

119 method. One Lowenstein-Jensen slope (pyruvate) and 7ml Mycobacterial Growth Indicator

120 Tube (MGIT) were inoculated and incubated at 35oC until growth was detected. ZN staining 121 of colonies confirmed Acid Fast Bacilli (AFB). Multiplex PCR (Wilton and Cousins, 1992)

122 was performed to discriminate between M. tuberculosis, M. avium, M. intracellulare, M.

123 abscessus and other Mycobacterium spp. Isolates identified as other Mycobacterium spp were

124 further speciated using Hain Life Sciences GenoType Mycobacterium AS (additional

125 species) kit (2004-7 only) and/or 16S rRNA sequencing in conjunction with phenotypic 126 characteristics. Clinical information on the patients was obtained from the NTM database at 127 the QLD Tuberculosis Control unit. 128

129 The clonality of clinical and water M. kansasii isolates was determined using a rep-PCR 130 based method (Diversilab® system, bioMerieux, Melbourne, Australia). DNA was extracted 131 from clinical and water isolates using the Ultraclean Microbial DNA Isolation Kit (Mo Bio 132 Laboratories, CA, USA). PCR mixture was prepared using AmpliTaq polymerase and PCR 133 buffer (Applied Biosystems, New Jersey, USA) and Mycobacterium Diversilab®- primer mix 134 according to the manufacturer’s instructions. Separation and detection of Rep-PCR products 135 was performed by micro-fluidic chips of the Diversilab® System. Fingerprints were analysed 136 with Diversilab® software v.3.4.38 using the Pearson correlation co-efficient and unweighted 137 pair group method with arithmetic means to compare isolates and determine clonal 138 relationship. This method emphasizes peak intensities more so than peak presence/absence 139 and was selected based on the appearance of the virtual gel images generated and as 140 recommended by the manufacturer (bioMerioux, Australia). A cutoff of 97% strain similarity 141 was used to define indistinguishable strains, those with 95% similarity were considered 142 related, and those <95% different. A large cluster of indistinguishable isolates determined 143 using the Pearson analysis, was further analysed using the Extended Jaccard method that

144 weighs the presence or absence of peaks more so than intensities.

274

145

146 Results 147 The mean age of the 71 patients was 64.40 years (SD 18.47; Range 5-95 years). Forty-one 148 (57.7%) were male. The majority (90.14%) of patients had pulmonary infection, 4 had soft 149 tissue isolates and 1 isolate came from a child’s lymph node. 150 151 Using a similarity cut off of 97%, there were 14 strain types identified amongst all clinical

152 isolates. Figure 2a The majority of clinical strains were of type CP (Clinical pattern) 11 – a 153 highly clonal strain (presumed Type I on ITS_RFLP). Four isolates were labeled strain type 154 CP4, 3 CP12, 2 CP1and 2 CP3 with the remaining strain types consisting of single isolates 155 only (including the ATCC strain, CP6). The strains belonging to clusters CP1-4 were related 156 (95% similarity), as were the strains of CP 11 and 12. The remaining clusters differed. The

157 relationship between clusters is demonstrated on the scatterplot in Figure 2b. 158 Of the 4 soft tissue isolates – 2 were strain type CP11, one CP12 and one CP13. The lymph 159 node isolate was CP11 and the remaining less common strain types were pulmonary isolates.

160 There was no evidence of geographical clustering of strains across the state of Queensland.

161 162 The dominant clinical cluster CP11 was further analysed using the Extended Jaccard method. 163 (Figure 3a and 3b) Using a similarity index of 97%, 13 strain patterns were observed. The 164 dominant patterns were P1 containing 26 isolates, P3 (n=17), P4 (n=13), P2 (n=2). The 165 remaining patterns consisted of single isolates only. The geographic distribution of these 166 isolates is shown in the scatterplot. All three isolates from West Moreton (Ipswich) were P1 167 but isolates from all other areas were distributed across different groups. 168 169 There was significant heterogeneity amongst the water isolates, forming 28 patterns (WP1-

170 28) Figure 4. The majority of water isolates were different to the clinical isolates, best 171 appreciated on the scatterplot Figure 5. Of the water isolates there were 2 isolates (Cluster 172 WP5) that matched the dominant clinical strain type CP11. A dominant cluster of water

275

173 isolates (WP2) matched a single patient isolate (CP9). The remaining water and clinical 174 isolates were unrelated. 175

176 Discussion 177

178 Several typing methods have revealed significant heterogeneity of M. kansasii strains, yet a 179 high degree of clonality of Type I strains exists; this is the most common clinical strain in 180 most parts of the world. (Alcaide et al 1997, Iinuma et al. 1997) In this study we have shown 181 that automated rep PCR has sufficient discriminatory power to similarly demonstrate 182 heterogeneity of clinical and environmental isolates. We have also shown the majority of

183 clinical isolates of M. kansasii found in Queensland are highly clonal using rep PCR, and 184 most likely correlate to the Type 1 strain identified using ITS_RFLP in a previous study

185 (Psaltis 2005, conference presentation). Whilst the study conducted on M. kansasii clinical 186 isolates from QLD using ITS_RFLP was from an earlier time period, it would appear that 187 either rep-PCR is more discriminatory as a typing tool, or that there has been further 188 evolution/acquisition of clinical strain types over time. 189 190 Using the Extended Jaccard analysis tool it is possible to identify strain variation within this

191 dominant highly related group of clinical isolates. PFGE of M. kansasii DNA digested with 192 DraI revealed limited polymorphism within Type I isolates from Spain and Switzerland, 193 though did reveal significant polymorphism amongst Type II isolates. Large restriction

194 fragment PFGE (Iinuma, et al. 1997) digested with VspI revealed 21restriction patterns when 195 applied to clinical isolates from Japan. Eighteen of the 21 types formed a cluster showing 196 genetic relatedness with a similarity coefficient of ≥0.87. The other three were divergent from 197 the cluster and from each other. Amplified-fragment length polymorphism (AFLP)(Gaafar et 198 al. 2003) of 253 Type 1 clinical strains, using the combination of patterns generated by three 199 primers, identified 12 clusters. The majority of isolates however were members of 5 clusters 200 (C122, n=121; C111,n=59, C222; n=51; C221; n=12; C112,n=3). The remaining seven 7

276

201 clusters contained only single isolates. In this study automated rep-PCR was comparable in 202 discriminating strain patterns within the clinical strains assumed to be Type I. 203 204 In comparison to clinical isolates, potable water isolates showed greater diversity. Relative to

205 the number of strains of M. kansasii that were isolated from potable water, there were very

206 few that matched clinical isolates. M. kansasii is not a common infection in Queensland (186 207 patient notifications between 2001-2010, Population 4 million; 0.465/100 000/year) and this 208 may be a reflection of the low burden of pathogenic strains within the water supply. However 209 clinical isolates were included from a variety of locations across the state, and water was only 210 sampled from the distribution system of the capital city. The possibility that the water supply 211 in other locations contains pathogenic strains is still possible. It is also possible that at certain 212 points the water becomes contaminated by another environmental source and hence there may 213 be another environmental niche for this pathogen. 214 215 This possibility is supported by reports of high incidence areas of M. kansasii disease 216 associated with the mining industry in South Africa, (Engel et al 1980a, Falkinham 2010), 217 industrial zones of Czechoslovakia (Chobot et al 1997) (particularly mining and iron 218 manufacture) (Kubin et al 1980) and the Japanese coastal industrial areas (Iinuma et al. 1997).

219 In the Karvina district of Czechoslovakia, M. kansasii was isolated from 7% of drinking water 220 samples, and 43.7% industrial water samples and scrapings between 1971 and 1995, during 221 which the annual incidence of disease was 12.03/100000 with a very high incidence of 222 disease in miners (Chobot et al 1997). 223

224 Engel (Engel 1980a) applied phage typing of M. kansasii isolates to compare strains from 225 clinical and environmental specimens from The Netherlands, Britain and Czechoslovakia. Of 226 the Czech strains, Phage type 13 accounted for 25/28 (89.3%) of the clinical isolates from the 227 Karvina coalmine district. This phage type was also found in 28 of 31 (90.4%) environmental 228 isolates (water from the distribution system of the mines), and 12/18 (66.6%) clinical isolates 8

277

229 from the rest of the country. This phage type also dominated clinical isolates from the UK, 230 though only 1 of 7 environmental isolates was of this type, and this was isolated from soil in

231 Uganda. Given developments in the field of strain typing, it has been suggested that this 232 technique did not have sufficient discriminatory power to adequately compare the clinical and 233 environmental strains. Phage typing has not been widely applied because it is labor intensive, 234 and technical problems can result in poor reproducibility and interpretation of results

235 (Picardeau et al 1997). 236

237 M. kansasii Type I was not present among 113 environmental strains isolated in Germany 238 (Alcaide et al 1997), though was found in water in France (Picardeau 1997). Picardeau

239 characterized five subspecies of M. kansasii (Picardeau et al 1997) using a variety of strain 240 typing methods, including RFLP with the major polymorphic tandem repeat (MPTR) and

241 IS1652 probe, PFGE, AFLP and PRA of the hsp-65 gene. Both clinical (38) and water (24) 242 isolates from France were included in this study. The clinical isolates came from patients 243 from 8 different French towns. The water isolates were recovered from tap water from 6 244 different hospitals in and around Paris. Whilst there were water isolates and clinical isolates 245 that had the same strain pattern using different methods of typing, only 1 set of isolates was 246 epidemiologically related. This set consisted of identical isolates from patients from a single 247 family living in the same house, and a water isolate recovered from their home. Five other 248 sets of isolates were considered epidemiologically related but there was no evidence of 249 nosocomial infection or water transmission. The molecular studies also showed that strains 250 isolated from tap water at a single hospital were not all identical, consistent with our finding 251 of a number of diverse strains within the water supply, some of which may be associated with

252 human infection.

253 254 In summary we have demonstrated significant heterogeneity amongst clinical and

255 environmental strains of M. kansasii using automated rep-PCR. The clonality of the dominant 256 clinical strain is in keeping with previous studies, yet rep-PCR provided sufficient

278

257 discriminatory power to identify strain types within this group. Few water strains were 258 identical to clinical strains, suggesting that if water or aerosol transmission is the mode of 259 infection, then point source contamination likely occurs from an alternative environmental 260 source. 261 262 263 264 265 266 267

268 Acknowledgements: This study was funded by grants from the Gallipoli Medical Research 269 Foundation of Greenslopes Private Hospital and by The Prince Charles Hospital Foundation. 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 10

279

285

286 References 287

288 Alcaide, F., Richter,I., Bernasconi, C., Springer, B., Hagenau, C., Telenti, A. (1997) Heterogeneity 289 and clonality among isolates of Mycobacterium kansasii:implications for epidemiological and 290 pathogenicity studies. J Clin Micro 35(8): 1959-1964. 291 292 Bloch, K. C., Zwerling, L., Pletcher,M.J., Hahn,J.A., Gerberding, J.L., Ostroff, S.M., Vugia,D.J., 293 Reingold,A.L. (1998). Incidence and Clinical Implications of Isolation of Mycobacterium kansasii: 294 Results of a 5-Year, Population-Based Study. Ann Intern Med 129: 698-704. 295 296 Chobot, S., Malis, J., Sebakova, H., Pelikan, M., Zatloukal, O., Palicka, P., Kocurova, D., (1997) 297 Endemic Incidence of Infections caused by Mycobacterium kansasii in the Karvina District in 298 1968-1995." Centr Eur J Publ Health 5 5(4): 164-173. 299 300 Dailloux, M., Abalain, M.L., Laurain, C., Lebrun, L., Loos-Ayav, C., Lozniewski, A., Maugein, J and the 301 French Mycobacteria Study Group (2006). Respiratory infections associated with nontuberculous 302 mycobacteria in non-HIV patients. Eur Respir J 28(6): 1211-1215. 303 304 Engel, H.W.B., Berwald, L.G., Grange, J.M., Kubin, M. (1980) Phage typing of Mycobacterium 305 kansasii. Tubercle 61(1): 11-19. 306 307 Engel, H.W.B., Berwald, L.G., Havelaar, A.H. (1980) The occurrence of Mycobacterium kansasii in 308 tapwater. Tubercle 61(1): 21-26. 309 310 Evans, S. A., Colville, A., Evans, A.J., Crisp,A.J., Johnston, I.D.A. (1996) Pulmonary Mycobacterium 311 kansasii: comparison of the clinical features, treatment and outcome with pulmonary 312 tuberculosis. Thorax 51: 1248-1252. 313

280

314 Falkinham, J. O. (2010). Impact of human activities on the ecology of nontuberculous 315 mycobacteria. Future Microbiology 5(6): 951-960. 316 317 Gaafar, A., Unzaga, M.J., Cisterna, R., Clavo, F.E., Urra, E., Ayarza, R., Martin, G. (2003) Evaluation of 318 a Modified Single-Enzyme Amplified-Fragment Length Polymorphism Technique for 319 Fingerprinting and Differentiating of Mycobacterium kansasii Type I Isolates. J. Clin. Microbiol. 320 41(8): 3846-3850. 321 322 Hoefsloot, W., van Ingen*, J., Andrejak, D., Ängeby, K., Bauriaud, R., Bemer, P., Boeree, M.J., Cacho, 323 J., Chimara, E., Cias, R., Dasa, R., Dekhuijzen, P., Domingo, D., Drobniewski, F., Esteban, J., Fauville- 324 Dufaux, M., Folkvardsen, D., Gibbons, N., Gómez-Mampaso, E., Gonzalez, R., Hoffmann, H., Hsueh, 325 P., Indra, A., Jagielski, T., Jamieson, F., Jankovic, M., Keane, J., Koh, W., Leao, S., Macedo, R., 326 Mannsåker, T., Marras, T., Maugein, J., Milburn, H., Mlinkó, T., Morcillo, N., Morimoto, N., 327 Papaventis, D., Palenque, E., Paez-Peña, M., Piersimoni, C., Polanová, M., Rastogi, N., Richter, E., 328 Ruiz-Serrano, M., Silva, A., Simsek, H., van Soolingen, D., Szabó, N., Thomson, R.M.,Tórtola 329 Fernandez, M.T., Tortoli, Totten, S., Tyrrell, G., Vasankari, T., Villar, M., Walkiewicz, R., Winthrop, 330 K., Wagner, D for NTM-net. (2013). A snapshot of the geographic diversity of nontuberculous 331 mycobacteria isolated from pulmonary samples: An NTM-net collaborative study. Eur Respir J In 332 Press. 333 334 Iinuma, Y., Ichiyama, S., Hasegawa, Y., Shimokata, K., Kawahara, S., Matsushima, T., (1997) Large- 335 restriction-fragment analysis of Mycobacterium kansasii genomic DNA and its application in 336 molecular typing. J Clin Micro 35(3): 596-599. 337 338 Kubin, M., Švanclová, E., Medek, B., Chobot. S., Olšovský, Z. (1980). Mycobacterium kansasii 339 infection in an endemic area of Czechoslovakia. Tubercle 61(4): 207-212. 340 341 Maliwan, N., Zvetina, J.R. (2005) Clinical features and follow up of 302 patients with 342 Mycobacterium kansasii pulmonary infection: a 50 year experience. Postgrad Med J 81: 530-533. 343

281

344 McSwiggan, D., Collins, C. (1974) The isolation of M.kansasii and M.xenopi from water systems. 345 Tubercle 55: 291-297. 346 347 Picardeau, M., Prod'hom, G., Raskine, L., LePennec, M.P., Vincent, V (1997). Genotypic 348 characterization of five subspecies of Mycobacterium kansasii. J Clin Micro 35(1): 25-32. 349 350 Psaltis, J. (2005). Mycobacterium kansasii: A Queensland Mycobacterium Reference Laboratory 351 Review 1975-2004. Conference Presentation, Australian Society of Microbiology Annual 352 Scientific Meeting, Canberra, Australia. 353 354 Santin, M., Alcaide, F., Benitez, M., Salazar, A., Ardanuy, C., Podzamczer, D., Rufi, G., Dorca, J., 355 Martin, R., Gudiol, F., (2004) Incidence and molecular typing of Mycobacterium kansasii in a 356 defined geographical area in Catalonia, Spain. Epidemiol Infect 132: 425-432. 357 358 Steadham, J. E. (1980). High-catalase strains of Mycobacterium kansasii isolated from water in 359 Texas. J Clin Micro 11(5): 496-498. 360 361 Taillard, C., Greub, G., Weber, R., Pfyffer, G.E., Bodmer, T., Zimmerli, S., Frei, R., Bassetti, S., 362 Rohner, P., Piffaretti, J-C., Bernasconi, E., Bille, J., Telenti, A., Prod'hom, G, (2003). Clinical 363 Implications of Mycobacterium kansasii Species Heterogeneity: Swiss National Survey. J Clin 364 Micro 41(3): 1240-1244. 365 366 Thomson, R. (2010). Changing epidemiology of pulmonary nontuberculous mycobacteria 367 infections. Emerg Inf Dis 16(10): 1576-1582. 368 369 Thomson, R., Carter, R., Tolson, C., Huygens, F., Hargreaves, M (2013). Factors associated with the 370 isolation of Nontuberculous mycobacteria (NTM) from a large municipal water system in 371 Brisbane, Australia. BMC Microbiology Manuscript revision under review. 372

282

373 Thomson, R., Tolson, C., Carter, R., Coulter, C., Hargreaves, M., Huygens, F (2012). Isolation Of 374 Nontuberculous Mycobacteria (NTM) From Household Water And Shower Aerosols In Patients 375 With NTM Pulmonary Disease. Am J Respir Crit Care Med 185: A4030. 376 377 Wilton, S., Cousins, D. (1992) Detection and identification of multiple mycobacterial pathogens by 378 DNA amplification in a single tube. PCR Methods Appl. 4: 269-273. 379 380 381

283

Figure 1

Figure 1: ITS-RFLP Strain typing of M. kansasii isolates from Queensland patients between 1975 and 2004.

Genotypes 1975-2004

15 Type VI Type V 10 Type IV Number 5 Type I

1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Year

284

Figure 2a Click here to download high resolution image

Figure 2b Click here to download high resolution image

285

Figures 3a & 3b Click here to download high resolution image

Figures 4a & 4b Click here to download high resolution image

286

Figures 5 Click here to download high resolution image

287

APPENDIX F PDF of manuscript (3rd revision) submitted to Epidemiology and Infection, November 2013.

Epidemiology and Infection

Heterogeneity of clinical and environmental isolates of Mycobacterium fortuitum using repetitive unit-PCR: municipal water an unlikely source of community acquired infections.

For Review Only Journal: Epidemiology and Infection

Manuscript ID: HYG-OM-5035-Jun-13.R3

Manuscript Type: Original Manuscript

Date Submitted by the Author: n/a

Complete List of Authors: Thomson, Rachel; Greenslopes Private Hospital, Gallipoli Medical Research Centre; The Prince Charles Hospital, Thoracic Medicine Tolson, Carla; Pathology Queensland, Mycobacterial Reference Laboratory Carter, Robyn; Pathology Queensland, Mycobacterial Reference Laboratory Huygens, Flavia; Queensland University of Technology, Institute of Health and Biomedical Innovation Hargreaves, Megan; Queensland University of Technology, Faculty of Science and Technology

Mycobacteria, Infectious disease, Clinical microbiology, Water-borne Keyword: infections

London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK

288

Page 1 of 21 Epidemiology and Infection

1 2 3 4 Heterogeneity of clinical and environmental isolates of Mycobacterium 5 6 fortuitum using repetitive unit-PCR: municipal water an unlikely source of 7 8 9 community acquired infections. 10 11 R. M. Thomson,1 C. E. Tolson,2 R. Carter,2 F. Huygens,3 M. Hargreaves.4 12 13 14 15 1. Gallipoli Medical Research Centre, Greenslopes Private Hospital, Brisbane, 16 17 18 QLD, Australia.For Review Only 19 20 2. QLD Mycobacterial Reference Laboratory, Pathology Queensland 21 22 RBWH Campus, Herston Rd, Herston QLD Australia 4006 23 24 25 3. Queensland University of Technology, Institute of Health and Biomedical 26 27 Innovation, Kelvin Grove Campus, Brisbane, QLD Australia 4059 28 29 4. Queensland University of Technology, Faculty of Science and Technology 30 31 George Street, Brisbane QLD Australia 4001 32 33 34 35 36 37 38 39 40 Running title: Heterogeneity of Mycobacterium fortuitum strains. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine,1 Keppel Street, London WC1E 7HT, UK

289

Epidemiology and Infection Page 2 of 21

1 2 3 Summary 4 5 M. fortuitum is a rapidly growing mycobacterium associated with community 6 7 8 acquired and nosocomial wound, soft tissue and pulmonary infections. It has 9 10 been postulated that water has been the source of infection especially in the 11 12 hospital setting. The aim of this study was to determine if municipal water may 13 14 be the source of community acquired or nosocomial infections in the Brisbane 15 16 17 area. Between 2007-2009, 20 strains of M. fortuitum were recovered from 18 For Review Only 19 municipal water and 53 patient isolates were submitted to the reference 20 21 laboratory. A wide variation in strain types was identified using rep-PCR, with 22 23 thirteen clusters of ≥2 indistinguishable isolates, and 28 patterns consisting of 24 25 26 individual isolates. The clusters could be grouped into seven similar groups 27 28 (>95% similarity). Municipal water and clinical isolates collected during the 29 30 same time period and from the same geographical area, consisted of different 31 32 strain types, making municipal water an unlikely source of sporadic human 33 34 35 infection. 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine,2 Keppel Street, London WC1E 7HT, UK

290

Page 3 of 21 Epidemiology and Infection

1 2 3 INTRODUCTION 4 5 6 7 8 Mycobacterium fortuitum is a non-pigmented rapidly growing mycobacterium, 9 10 often associated with skin and soft tissue infections, though pulmonary disease 11 12 can occur. [1] Wound infections [2], post-injection abscesses, and post pedicure 13 14 infections [3] are well described. M. fortuitum is also a recognised cause of 15 16 17 nosocomial infections [4] and pseudo-outbreaks.[5] It is usually considered that 18 For Review Only 19 municipal and hospital water are the source of these outbreaks, though M. 20 21 fortuitum is also found in soil and house dust.[6] M. fortuitum has been found in 22 23 municipal water [7-9] and is relatively resistant to chlorine and other 24 25 26 disinfectants. [10, 11] Whilst there have been many outbreaks reported where 27 28 water has been the suspected vehicle of transmission, very few have been 29 30 realised in a timely fashion that would allow full environmental sampling to 31 32 adequately document the source of the infection. [2, 5, 12-14] 33 34 35 In order to determine if strains of M. fortuitum resident in municipal drinking 36 37 water in Brisbane, Australia were a likely source of infections with this organism 38 39 in the community, we undertook a prospective study (2007-2009) to compare 40 41 42 isolates from municipal water and infected patients resident in the area supplied 43 44 by that water. 45 46 47 48 METHODS 49 50 51 52 During 2007-2008 citywide water sampling of approximately 220 sites (Trunk 53 54 Main, Reservoir and Distribution water samples) was performed and 19 isolates 55 56 were identified as M. fortuitum using 16S rDNA gene fragment sequencing. [15] A 57 58 59 60 London School of Hygiene and Tropical Medicine,3 Keppel Street, London WC1E 7HT, UK

291

Epidemiology and Infection Page 4 of 21

1 2 3 home sampling study of water, swabs and aerosols from NTM patient houses 4 5 was also conducted in 2009-10 and 1 isolate of M. fortuitum was recovered from 6 7 8 a shower aerosol. [16] Water isolates had been stored in Dubos broth at -20°C 9 10 and were thawed and sub-cultured onto 7H11 plates as well as Lowenstein- 11 12 Jensen slopes and incubated at 35°C until sufficient growth was available. 13 14 15 16 17 Human samples from patients whose residential address was within the water 18 For Review Only 19 distribution catchment, were submitted to the Mycobacterium Reference 20 21 Laboratory during the same time period (2007-2009) and had been digested and 22 23 decontaminated using 4% NaOH, neutralised with phosphoric acid and 24 25 26 centrifuged at 3000g to concentrate the acid-fast bacilli (AFB). Smears were 27 28 prepared from the sediment and stained by the Ziehl-Niehlsen method. One 29 30 Lowenstein-Jensen slope (±pyruvate) and 7ml Mycobacterial Growth Indicator 31 32 33 Tube (MGIT) were inoculated and incubated at 35°C until growth was detected. 34 35 ZN staining of colonies confirmed acid-fast bacilli. Multiplex PCR [17] was 36 37 performed to discriminate between M. tuberculosis, M. avium, M. intracellulare, 38 39 M. abscessus and other Mycobacterium spp. Isolates identified as other 40 41 42 Mycobacterium spp were further speciated using Hain Life Sciences GenoType 43 44 Mycobacterium AS (additional species) kit (2004-7 only) and/or 16S rRNA 45 46 sequencing in conjunction with phenotypic characteristics. Patient clinical 47 48 49 information was obtained from the NTM database at the QLD Tuberculosis 50 51 Control unit. Additional isolates from patients living outside the water 52 53 distribution network were included for comparison. 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine,4 Keppel Street, London WC1E 7HT, UK

292

Page 5 of 21 Epidemiology and Infection

1 2 3 The clonality of clinical and water M. fortuitum isolates was determined using a 4 5 rep-PCR based method (Diversilab® system, bioMerieux, Melbourne). DNA was 6 7 8 extracted from clinical and water isolates using the Ultraclean Microbial DNA 9 10 Isolation Kit (Mo Bio Laboratories, CA, USA). The PCR mixture was prepared 11 12 using AmpliTaq polymerase and PCR buffer (Applied Biosystems, New Jersey, 13 14 USA) and Mycobacterium Diversilab®- primer mix according to the 15 16 17 manufacturer’s instructions. Separation and detection of rep-PCR products was 18 For Review Only 19 performed by micro-fluidic chips of the Diversilab® System. Fingerprints were 20 21 analysed with Diversilab® software v.3.4.38 using the Pearson correlation co- 22 23 efficient and unweighted pair group method with arithmetic means to compare 24 25 26 isolates and determine clonal relationship. Pulsed field gel electrophoresis 27 28 (PFGE) was performed on 10 clinical isolates and a control strain and results 29 30 compared to the patterns generated by automated rep-PCR. 31 32

33 34 35 PFGE was performed using the method outlined in the BioRad Genpath® Group 36 37 6 Kit (BioRad France) with modifications outlined by Mazurek [18] and Burki 38 39 [19]. Organisms were inoculated into 10 ml Middlebrook 7H9 broth (Difco, 40 41 42 Becton Dickinson and Company, Sparks, MD USA in-house media) supplemented 43 44 with 0.2% OADC (Difco, Becton Dickinson and Company, MD USA), 0.1% Tween 45 46 80 (MP Biochemicals LLC Ohio USA, cycloserine (1mg/ml Sigma-Aldrich INC St 47 48 Louis MO USA) and ampicillin (0.1mg/ml Sigma-Aldrich INC St Louis MO USA) 49 50 51 and incubated for three days. One ml of broth was centrifuged and the 52 53 supernatant discarded. 54 55 Gel plugs were prepared and incubated in 500µl of Lysis Buffer 1 and 20 µl 56 57 Lysozyme (25mg/ml) at 36°C. Following a wash step, 500µl Proteinase K buffer 58 59 60 London School of Hygiene and Tropical Medicine,5 Keppel Street, London WC1E 7HT, UK

293

Epidemiology and Infection Page 6 of 21

1 2 3 and 20 µl Proteinase K (>600U/ml) were added to each sample. Plugs were 4 5 incubated for 48 hours at 50°C. The plugs were then washed four times in 6 7 8 1xWash Buffer. After the final wash, the plugs were stored in1x Wash Buffer. 9 10 Digestion was performed using, Xba 1 Enzyme (10U/ml) and the samples were 11 12 incubated for 18 hours at 36°C. 13 14 The plugs were loaded into wells of a 1% PFGE agarose (BioRad) electrophoresis 15 16 17 gel ensuring there were no air bubbles. Sufficient 0.5xTBE was added to the 18 For Review Only 19 electrophoresis cell and cooled to 14°C and electrophoresis performed using the 20 21 following parameters: Initial A time one second, Final A time 40 seconds, Voltage 22 23 200 V and time 22 hours. The gel was stained using Ethidium bromide (BioRad), 24 25 26 de-stained in running distilled water for 30 minutes and then photographed. 27 28 Based on the Tenover classification of isolates using PFGE, the Diversilab rep- 29 30 PCR similarity cut offs were determined as >97% (indistinguishable), >95% 31 32 similar, and <95% different. 33 34 35 36 37 RESULTS 38 39 The geographic distribution of water sites and patients residential addresses is 40 41 42 demonstrated in Figure 1. The clinical isolates came from 53 patients, 43 44 27(50.9%) of whom were male. There were 27 pulmonary isolates (50.9%; 24 45 46 considered not to be causing invasive disease, and three associated with 47 48 significant disease according to ATS/IDSA criteria [20]). Twelve patients had 49 50 51 underlying bronchiectasis and four had cavities on chest radiograph. There were 52 53 26 soft tissue isolates (five insulin injection site infections, three laparoscopic 54 55 band infections, two isolates from blood associated with line infections, two 56 57 58 59 60 London School of Hygiene and Tropical Medicine,6 Keppel Street, London WC1E 7HT, UK

294

Page 7 of 21 Epidemiology and Infection

1 2 3 surgical wound infections, and 14 other community acquired soft tissue 4 5 infections). 6 7 8 PFGE and rep-PCR were performed on 10 isolates of M. fortuitum associated with 9 10 laparoscopic gastric band infections. Isolates from the same patient that were 11 12 indistinguishable with PFGE, were associated with a >97% similarity cut off on 13 14 rep-PCR. (Figures shown in supplemental material) 15 16 17 Three of the water isolates had low intensity band patterns generated by rep- 18 For Review Only 19 PCR and were removed from the analysis. There was a wide variation in strain 20 21 types evident with 41 different patterns generated by 70 isolates. (Figure 2) 22 23 These formed thirteen clusters (P1-13) of two or more isolates; the remaining 24 25 26 patterns consisted of single isolates only (P14-41). These clusters were further 27 28 grouped according to similarity of approximately 95% into seven groups. Of the 29 30 larger clusters, P4 consisted of nine clinical isolates, eight of which were 31 32 associated with soft tissue infections. P5 (a cluster similar to P4) however, 33 34 35 consisted of six isolates, five of which were pulmonary isolates and the other 36 37 associated with an injection site infection. 38 39 There was no evidence of geographical clustering of strains apart from the two 40 41 42 isolates making up P6 that came from patients living in adjacent suburbs. A 43 44 husband and wife both presented with soft tissue infections of the lower limbs 45 46 and had indistinguishable isolates (P4; #21 and #22). The water isolate groups 47 48 were distinctly separate from the clinical isolates, best appreciated by viewing 49 50 51 the scatterplot in Figure 3. 52 53 54 55 DISCUSSION 56 57 58 59 60 London School of Hygiene and Tropical Medicine,7 Keppel Street, London WC1E 7HT, UK

295

Epidemiology and Infection Page 8 of 21

1 2 3 The aim of this study was to determine if strains of M. fortuitum resident in the 4 5 municipal water were likely to be the source of sporadic infections with M. 6 7 8 fortuitum in the community. The correlation with PFGE and the wide variety of 9 10 strain types identified confirms the utility of rep-PCR as a discriminatory tool for 11 12 this species. No patient isolates were indistinguishable from water isolates. 13 14 There did appear to be a dominant group of clinical strains, with a degree of 15 16 17 nonsignificant clustering according to infection type (P4 predominantly soft 18 For Review Only 19 tissue and P5 predominantly pulmonary) but this finding may have occurred by 20 21 chance as the clusters P4 and P5 were similar. The water strains in contrast 22 23 formed 3 separate groups that differed from all clinical isolates. 24 25 26 27 28 Other studies have demonstrated heterogeneity amongst strains of M. fortuitum 29 30 using a variety of molecular techniques, including PFGE, [13, 21-24] 16s-23s 31 32 rRNA ITS genotyping, RAPD and ERIC-PCR [21] mainly in the investigation of 33 34 35 nosocomial outbreaks. Legrand [22] examined 51 isolates of M. fortuitum from 36 37 47 patients in Guadeloupe, Martinique and French Guiana between 1996-99, 38 39 suspecting a microepidemic. Only five isolates were found to cluster, confirming 40 41 42 significant genomic heterogeneity amongst clinical strains and suggesting a 43 44 “diversity of ecological niches in which this organism may develop”. 45 46 47 48 One of the first reports of nosocomial transmission of M. fortuitum through water 49 50 51 was published by Burns [4] who investigated a cluster of cases of positive 52 53 sputum cultures amongst patients in an alcoholic rehabilitation ward. The 54 55 common factors amongst cases, was usage of two ward showers, and M. 56 57 fortuitum was isolated from the tap water connecting to the showers but not 58 59 60 London School of Hygiene and Tropical Medicine,8 Keppel Street, London WC1E 7HT, UK

296

Page 9 of 21 Epidemiology and Infection

1 2 3 from the showers themselves. Using PFGE, the 16 case isolates were found to be 4 5 identical to the water isolate. However, there were no other parts of the hospital 6 7 8 where such cases were identified and no further cases were reported after the 9 10 showers had been disconnected and decontaminated. 11 12 13 14 Wallace investigated the sources of M. fortuitum associated with infections 15 16 17 following cardiac bypass surgery [13] across 12 states in the USA. In the analysis 18 For Review Only 19 of a particular outbreak in Texas in 1981, one isolate of M. fortuitum was 20 21 recovered from an operating room waterbath and had a similar phenotype and 22 23 enzyme electrophoretic pattern as a cardiac associated isolate. A similar isolate 24 25 26 was also recovered from the municipal water coming into the hospital and from 27 28 an ice machine on one of the hospital floors. Two patients with other surgical 29 30 wound infections (neck and abdominal) were infected with the same strain. In 31 32 the 5 years following that outbreak an additional 21 clinical isolates of M. 33 34 35 fortuitum were recovered from patients in the same hospital. Six were studied. 36 37 All differed in phenotypic markers or plasmid profile from the outbreak isolates. 38 39 In a separate outbreak (Colorado 1976), an isolate of M. fortuitum recovered 40 41 42 from a settle plate in an operating room had the same phenotype and 43 44 electrophoretic pattern as four of the epidemic strains. 45 46 47 48 Winthrop et al [23] reported 110 cases of furunculosis acquired following 49 50 51 pedicures at a single nail salon in California (M. fortuitum cultured from 32 52 53 cases). All patients had their feet and lower legs soaked in a whirlpool footbath. 54 55 Large amounts of hair and skin debris were found behind the inlet suction 56 57 screens of the whirlpool footbath and M. fortuitum was cultured from these areas 58 59 60 London School of Hygiene and Tropical Medicine,9 Keppel Street, London WC1E 7HT, UK

297

Epidemiology and Infection Page 10 of 21

1 2 3 from all 10 footbaths. The authors concluded that M. fortuitum had entered via 4 5 the salon tap water, seeded the accumulated organic debris behind the footbath 6 7 8 inlet screens then multiplied and circulated within the footbath basin. They felt 9 10 that because all of the footbaths yielded rapid-growers including multiple strains 11 12 of M. fortuitum that it was unlikely the baths were contaminated by a client. 13 14 Based on our results and the fact that NTM are so prevalent in soil, one could 15 16 17 argue that in the absence of identical strains of M. fortuitum being discovered in 18 For Review Only 19 the salon tap water (only M. chelonae/abscessus were isolated), it is highly likely 20 21 that the footbaths were contaminated by soil or organic matter from the feet of 22 23 clients. The salon owner reported that these inlet suction areas were never 24 25 26 cleaned. 27 28 29 30 The heterogeneity of strains we have demonstrated, suggest that if water is the 31 32 vehicle for transmission of infection, then point source contamination probably 33 34 35 occurs from another environmental reservoir such as soil or dust. Such soil 36 37 organisms can readily enter water distribution systems via cracks in pipes 38 39 caused by tree roots or other trauma, or may contaminate surface source water 40 41 42 and survive disinfection. 43 44 45 46 Many nosocomial outbreaks in the literature have been reported in the midst of 47 48 long periods without NTM surgical infections. If municipal water was the source 49 50 51 of organisms, then it could be postulated that a transient drop in disinfection or 52 53 an increase in contamination must have occurred. It seems more likely that 54 55 episodic point source contamination may be responsible. 56 57 58 59 60 London School of Hygiene and Tropical Medicine,10 Keppel Street, London WC1E 7HT, UK

298

Page 11 of 21 Epidemiology and Infection

1 2 3 The susceptibility of M. fortuitum to quaternary ammonium disinfection was 4 5 studied by Cortesia et al [10] and there were phenotypic and presumably 6 7 8 genotypic changes in those strains that persisted after exposure to disinfection. 9 10 It is possible that the differences we have noted between clinical and water 11 12 strains have occurred because we have only detected strains that have survived 13 14 exposure to disinfection in municipal water and that strains responsible for 15 16 17 some clinical infections are protected by survival in amoebae or biofilm inside 18 For Review Only 19 pipes and hence were not detected by our sampling approach. The changes that 20 21 occur in strains upon exposure to the chlorine/chloramine disinfection used in 22 23 drinking water have not been studied adequately to permit comment on whether 24 25 26 this theory is plausible. [10] 27 28 29 30 In summary, we have compared isolates of M. fortuitum resident in a municipal 31 32 drinking water distribution system, to those found to cause community and 33 34 35 medically acquired infections. We have confirmed significant heterogeneity 36 37 amongst strains of M. fortuitum from patients and municipal water. These 38 39 results and appraisal of previous studies, would suggest that most community 40 41 42 infections probably originate in soil/dust, and nosocomial outbreaks occur as a 43 44 result of point contamination of water sources. However further comparison of 45 46 strain types found in soil and other environmental sources (such as amoebae and 47 48 biofilm) are needed to prove this is the case. 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine,11 Keppel Street, London WC1E 7HT, UK

299

Epidemiology and Infection Page 12 of 21

1 2 3 4 5 6 7 8 9 10 REFERENCES 11 12 13 1. De Groote MA, Huitt G. Infections due to rapidly growing Mycobacteria. 14 15 Clinical Infectious Diseases 2006;42:1756-1763. 16 17 2. Kuritsky J, et al. Sternal wound infections and endocarditis due to 18 organisms of theFor Mycbobacterium Review fortuitum complex. Only Annals of Internal Medicine 19 20 1983;98:938-939. 21 22 3. Vugia D, et al Mycobacteria in nail salon whirlpool footbaths, California. 23 Emerging Infectious Diseases. 2005;11:616-618. 24 25 4. Burns DN et al. Nosocomial outbreak of respiratory tract colonization 26 27 with Mycobacterium fortuitum: demonstration of the usefulness of pulsed-field 28 29 gel electrophoresis in an epidemiologic investigation. American Review of 30 Respiratory Disease 1991;142:468-470. 31 32 5. Wallace RJ, Brown BA, Griffith DE. Nosocomial 33 34 outbreaks/pseudooutbreaks caused by nontuberculous mycobacteria. Annual 35 Review of Microbiology 1998;52:453-490. 36 37 6. Dawson D. Potential pathogens among strains of mycobacteria isolated 38 39 from house-dusts. Medical Journal of Australia 1971;1(13):679-681. 40 7. Castillo-Rodal A, et al. Potentially pathogenic nontuberculous 41 42 mycobacteria found in aquatic systems. Analysis from a reclaimed water and 43 44 water distribution system in Mexico City. European Journal of Clinical 45 46 Microbiology & Infectious Diseases. 2012;31:683-694. 47 8. Carson LA, et al. Prevalence of nontuberculous mycobacteria in water 48 49 supplies of hemodialysis centers. Applied and Environmental Microbiology 50 51 1988;54:3122-3125. 52 9. Le Dantec C, et al. Occurrence of mycobacteria in water treatment lines 53 54 and in water distribution systems. Applied and Environmental Microbiology. 55 56 2002;68:5318-5325. 57 58 59 60 London School of Hygiene and Tropical Medicine,12 Keppel Street, London WC1E 7HT, UK

300

Page 13 of 21 Epidemiology and Infection

1 2 3 10. Cortesia C, et al. The use of quaternary ammonium disinfectants selects 4 5 for persisters at high frequency from some species of non-tuberculous 6 mycobacteria and may be associated with outbreaks of soft tissue infections. 7 8 Journal of Antimicrobial Chemotherapy 2010;65:2574-81. 9 10 11. Le Dantec C, et al. Chlorine disinfection of atypical mycobacteria isolated 11 12 from a water distribution system. Applied and Environmental Microbiology. 13 2002;68:1025-1032. 14 15 12. Quiñones C, et al. An outbreak of Mycobacterium fortuitum cutaneous 16 17 infection associated with mesotherapy. Journal of the European Academy of 18 Dermatology andFor Venereology Review 2010;24:604-606. Only 19 20 13. Wallace RJ, et al. Diversity and sources of rapidly growing mycobacteria 21 22 associated with infections following cardiac surgery. Journal of Infectious 23 Diseases. 1989;159:708-716. 24 25 14. Sampaio J, et al. Enterobacterial repetitive intergenic consensus PCR is a 26 27 useful tool for typing Mycobacterium chelonae and Mycobacterium abscessus 28 29 isolates. Diagnostic Microbiology and Infectious Disease. 2006;55:107-118. 30 15. Thomson R, et al. Factors associated with the isolation of 31 32 Nontuberculous mycobacteria (NTM) from a large municipal water system in 33 34 Brisbane, Australia. BMC Microbiology 2013;13:89 35 16. Thomson R, et al. Isolation Of nontuberculous mycobacteria (NTM) from 36 37 household water and shower aerosols in patients with NTM pulmonary disease. 38 39 American Journal of Respiratory and Critical Care Medicine. 2012;185:A4030. 40 17. Wilton S, Cousins D. Detection and identification of multiple 41 42 mycobacterial pathogens by DNA amplification in a single tube. PCR Methods and 43 44 Applications. 1992;4:269-273. 45 46 18. Mazurek GH, et al. Large restriction fragment polymorphisms in the 47 Mycobacterium avium-intracellulare complex: a potential epidemiological tool. 48 49 Journal of Clinical Microbiology. 1993;31:390-394. 50 51 19. Burki DR, et al. Evaluation of the relatedness of strains of Mycobacterium 52 avium using pulsed-field gel electrophoresis. European Respiratory Journal. 53 54 1995;14:212-217. 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine,13 Keppel Street, London WC1E 7HT, UK

301

Epidemiology and Infection Page 14 of 21

1 2 3 20. Griffith DE, et al. An Official ATS/IDSA statement: diagnosis, treatment, 4 5 and prevention of nontuberculous mycobacterial diseases. American Journal of 6 Respiratoty and Critical Care Medicine. 2007;175:367-416. 7 8 21. Sampaio J, et al. Application of four molecular typing methods for 9 10 analysis of Mycobacterium fortuitum group strains causing post-mammaplasty 11 12 infections. Clinical Microbiology & Infection. 2006;12:142-149. 13 22. Legrand E, et al. A pulsed-field gel electrophoresis study of 14 15 Mycobacterium fortuitum in a Caribbean setting underlines high genetic diversity 16 17 of the strains and excludes nosocomial outbreaks. International Journal of 18 Medical MicrobiologyFor. 2002; Review292:51-57. Only 19 20 23. Winthrop K, et al. An outbreak of mycobacterial furunculosis associated 21 22 with footbaths at a nail salon. New England Journal of Medicine. 2002;364:1366- 23 1371. 24 25 24. Sniezek P, et al. Rapidly growing mycobacterial infections after 26 27 pedicures. Archives of Dermatology. 2003;139:629-634. 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine,14 Keppel Street, London WC1E 7HT, UK

302

Page 15 of 21 Epidemiology and Infection

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Review Only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Figure 1. Map of Brisbane area with water sites marked in blue, and patient locations in red. 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK

303

Epidemiology and Infection Page 16 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Review Only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 2a. Rep-PCR dendrogram of M. fortuitum water and clinical isolates. Red interrupted line of similarity 33 at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into 34 similar groups (≥95% similarity) 35 254x190mm (72 x 72 DPI) 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK

304

Page 17 of 21 Epidemiology and Infection

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Review Only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 2b. Rep-PCR dendrogram of M. fortuitum water and clinical isolates. Red interrupted line of similarity 33 at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into 34 similar groups (≥95% similarity) 35 254x190mm (72 x 72 DPI) 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK

305

Epidemiology and Infection Page 18 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Review Only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 2c. Rep-PCR dendrogram of M. fortuitum water and clinical isolates. Red interrupted line of similarity 33 at 96.7%. Coloured bars under P= indistinguishable patterns (≥97% similarity), grouped under G into 34 similar groups (≥95% similarity) 35 254x190mm (72 x 72 DPI) 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK

306

Page 19 of 21 Epidemiology and Infection

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Review Only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 3. Scatterplot of numbered M. fortuitum isolates, colour coded according to site of infection, with 33 water isolates in pink (side legend). Gridline spacing correlates with 5% similarity. (Pulmonary NS=non 34 clinically significant isolates; Sig=clinically significant) 35 254x190mm (72 x 72 DPI) 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK

307

Epidemiology and Infection Page 20 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Review Only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure S1. Rep-PCR of M. fortuitum isolates (numbered 1-10) associated with laparoscopic gastric band 33 infections demonstrating indistinguishable isolates for each patient (same pattern (P) colour) and the degree 34 of difference between patients (lower graphic, percentage similarity in boxes) 35 254x190mm (72 x 72 DPI) 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK

308

Page 21 of 21 Epidemiology and Infection

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Review Only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure S2. PFGE patterns of 10 M. fortuitum isolates associated with laparoscopic gastric band procedures 33 (same isolates as for Figure S1). Lanes 1-3, isolates from patient AM; lanes 4-5, patient KG; lanes 6-7, 34 patient RB; lane 8, patient PN; lane 9, patient ML and lane 10, wild type control strain. 35 254x190mm (72 x 72 DPI) 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK

309

APPENDIX G1 Thomson, R. M., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2012) Isolation of NTM from household water and shower aerosols in patients with NTM pulmonary disease. American Journal of Respiratory and Critical Care Medicine, 185, A4030. (conference poster)

310

APPENDIX G2 Thomson, R. M., Tolson, C., Coulter, C., Huygens, F., & Hargreaves, M. (2013) Isolation of NTM from household water and shower aerosols in patients with NTM pulmonary disease. Journal of Clinical Microbiology 2013, 51(9):3006. Published ahead of print July 10th, 2013.

Isolation of Nontuberculous Mycobacteria (NTM) from Household Water and Shower Aerosols in Patients with Pulmonary Disease Caused by NTM

Rachel Thomson, Carla Tolson, Robyn Carter, Chris Coulter, Flavia Huygens and Megan Hargreaves J. Clin. Microbiol. 2013, 51(9):3006. DOI: 10.1128/JCM.00899-13. Published Ahead of Print 10 July 2013. Downloaded from

Updated information and services can be found at: http://jcm.asm.org/content/51/9/3006

These include: SUPPLEMENTAL MATERIAL Supplemental material http://jcm.asm.org/

REFERENCES This article cites 22 articles, 6 of which can be accessed free at: http://jcm.asm.org/content/51/9/3006#ref-list-1

CONTENT ALERTS Receive: RSS Feeds, eTOCs, free email alerts (when new articles cite this article), more» on August 22, 2013 by guest

Information about commercial reprint orders: http://journals.asm.org/site/misc/reprints.xhtml To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/

311

Isolation of Nontuberculous Mycobacteria (NTM) from Household Water and Shower Aerosols in Patients with Pulmonary Disease Caused by NTM

Rachel Thomson,a Carla Tolson,b Robyn Carter,b Chris Coulter,b Flavia Huygens,c Megan Hargreavesd Gallipoli Medical Research Centre, Greenslopes Private Hospital, Brisbane, Queensland, Australiaa; Queensland Mycobacterium Reference Laboratory, Pathology Queensland, RBWH Campus, Herston, Queensland, Australiab; Queensland University of Technology, Institute of Health and Biomedical Innovation, Kelvin Grove Campus, Brisbane, Queensland, Australiac; Queensland University of Technology, Faculty of Science and Technology, Brisbane, Queensland, Australiad

It has been postulated that susceptible individuals may acquire infection with nontuberculous mycobacteria (NTM) from water and aerosol exposure. This study examined household water and shower aerosols of patients with NTM pulmonary disease. The Downloaded from mycobacteria isolated from clinical samples from 20 patients included M. avium (5 patients), M. intracellulare (12 patients), M. abscessus (7 patients), M. gordonae (1 patient), M. lentiflavum (1 patient), M. fortuitum (1 patient), M. peregrinum (1 patient), M. chelonae (1 patient), M. triplex (1 patient), and M. kansasii (1 patient). One-liter water samples and swabs were collected from all taps, and swimming pools or rainwater tanks. Shower aerosols were sampled using Andersen six-stage cascade impactors. For a subgroup of patients, real-time PCR was performed and high-resolution melt profiles were compared to those of ATCC control strains. Pathogenic mycobacteria were isolated from 19 homes. Species identified in the home matched that found in the patient in seven (35%) cases: M. abscessus (3 cases), M. avium (1 case), M. gordonae (1 case), M. lentiflavum (1 case), and

M. kansasii (1 case). In an additional patient with M. abscessus infection, this species was isolated from potable water supplying http://jcm.asm.org/ her home. NTM grown from aerosols included M. abscessus (3 homes), M. gordonae (2 homes), M. kansasii (1 home), M. fortuitum complex (4 homes), M. mucogenicum (1 home), and M. wolinskyi (1 home). NTM causing human disease can be isolated from household water and aerosols. The evidence appears strongest for M. avium, M. kansasii, M. lentiﬂavum, and M. abscessus. Despite a predominance of disease due to M. intracellulare, we found no evidence for acquisition of infection from household water for this species.

ontuberculous mycobacteria (NTM) are ubiquitous organ- only intermittently into water (12). Previous investigators have on August 22, 2013 by guest Nisms responsible for progressive pulmonary disease. Water is documented the presence of mycobacteria in soil and house dusts a documented source of NTM infection in humans, although it is in Queensland (QLD) and have shown that ϳ50% of environ- not the only source. The incidence of pulmonary disease due to mental strains were also found in human specimens, such as spu- environmental mycobacteria is increasing in many parts of the tum (3, 13, 14). world, including Australia (1). Reasons include increased aware- The aim of this study was to compare NTM found in household ness of mycobacteria as pulmonary pathogens, improvements in water, biofilm, and aerosols generated by showering with the species methods of detection and culture, and an aging population (as this or strains causing infection in patients with known nontuberculous is often a disease of the elderly). An increase in exposure is also mycobacterial pulmonary disease in Brisbane, Australia. possible as people are exposed to mycobacteria in water through drinking, swimming, and bathing. Aerosols generated during MATERIALS AND METHODS these activities can be inhaled, and water can be aspirated when Patients were recruited by the principal investigator (R.T.) through the swallowed (2). Evidence linking human infection due to exposure Respiratory Outpatient Department at The Prince Charles Hospital, to Mycobacterium intracellulare and household dust (3), soil and through clinics at the Queensland Tuberculosis Control Centre, and at potting mix (4), and peat moss (5) also exists. Greenslopes Private Hospital. Prevalent patients who had been resident in There is some evidence that susceptible individuals may ac- the same dwelling for greater than 5 years prior to the diagnosis of NTM disease were eligible for home sampling. Following signed informed con- quire infection from potable water in their own homes. To date sent, the PI and assistant visited the homes to collect 1-liter samples of this evidence has been largely reported for Mycobacterium avium water into sterile containers directly from kitchen, bathroom, and shower (6–8). We recently documented NTM in the Brisbane, Australia, potable water distribution system (9) and have shown that strains of M. lentiflavum (10) and M. abscessus (11) are highly related or Received 5 April 2013 Returned for modification 28 May 2013 identical to those found in patients with disease. Despite frequent Accepted 9 July 2013 identification of Mycobacterium kansasii from water samples, Published ahead of print 10 July 2013 pathogenic strains were found infrequently (unpublished data). Address correspondence to Rachel Thomson, [email protected]. M. intracellulare and M. avium were found infrequently in potable Supplemental material for this article may be found at http://dx.doi.org/10.1128 water. It has been suggested that the low yield of M. intracellulare /JCM.00899-13. from potable water samples is the result of insufficient sample Copyright © 2013, American Society for Microbiology. All Rights Reserved. volume, difficulties with culture techniques, and the fact that per- doi:10.1128/JCM.00899-13 haps this species resides in the biofilm lining pipes and is released

3006 jcm.asm.org Journal of Clinical Microbiology p. 3006–3011 September 2013 Volume 51 Number 9

312

NTM Patient Home Sampling

taps, rainwater tanks, and swimming pools (if applicable). Swabs were cluded 12 reference strains as positive controls to validate the character- taken from inside all taps and showerheads. ization of the tested samples. All reactions were performed in duplicate. Aerosols from showers were sampled using Andersen six-stage cascade Human samples were digested and decontaminated using 4% NaOH, impactors for viable air sampling (Grasby-Andersen, Smyrna, GA), which neutralized with phosphoric acid, and centrifuged at 3,000 ϫ g to concen- provide validated particle size distribution data (15) and which have been trate the acid-fast bacilli (AFB). Smears were prepared from the sediment used previously to collect mycobacterial aerosols (16–18). The impactors and stained by the Ziehl-Neelsen (ZN) method. One Lowenstein-Jensen are calibrated at a flow rate of 28.3 liters/min, and samples were collected slope (Ϯpyruvate) and a 7-ml MGIT tube were inoculated and incubated onto petri dishes containing Mycobacteria 7H11 agar, supplemented with at 35°C until growth was detected. ZN staining of colonies confirmed OADC (oleic acid-albumin-dextrose-catalase) enrichment (100 ml/liter), AFB. Multiplex PCR (7) was performed to discriminate between M. tu- malachite green (25 mg/liter), and cycloheximide (500 mg/liter) (17). The berculosis, M. avium, M. intracellulare, M. abscessus, and other Mycobac- impactor was placed as close as practicable to the shower recess, and the terium spp. Isolates identified as other Mycobacterium spp. were further shower hot tap was run. Windows were left as the patient would normally determined to the species level using Hain Life Sciences GenoType Myco- have them during showering, and exhaust fans were left off. The collection bacterium AS (additional species) kit (2004 to 2007 only) and/or 16S time was 20 min. After collection, mycobacterial plates were sealed with rRNA sequencing in conjunction with phenotypic characteristics. The clonality of clinical and water isolates was determined using a parafilm, packed in a plastic bag, and incubated at 30°C for 3 to 6 months. method based on repetitive element palindromic PCR (Rep-PCR) (Diver-

Plates were read weekly, and purity plate subcultures were performed Downloaded from silab system; bioMérieux, Melbourne, Australia). DNA was extracted using 7H11 agar. from clinical and water isolates using the Ultraclean microbial DNA iso- Water samples were processed as previously described (9, 19). After lation kit (Mo Bio Laboratories). The PCR mixture was prepared using analysis of water sampling revealed an increased yield for rapidly growing AmpliTaq polymerase and PCR buffer (Applied Biosystems, Hamman- mycobacteria using liquid medium (MGIT tubes) (9), these were in- ton, NJ) and Mycobacterium Diversilab primer mix according to the cluded for the last 6 houses that were sampled. Swabs were placed in 2 ml manufacturer’s instructions. Separation and detection of Rep-PCR prod- sterile distilled water (SDW) at time of collection, sealed, and transported ucts were performed by microfluidic chips of the Diversilab System. Fin- to the lab at room temperature. Each swab container was vortexed for 2 gerprints were analyzed with Diversilab software v.3.4.38 using the Pear- min and inoculated as per the water samples. Following initial inoculation son correlation coefficient and unweighted-pair group method with (M7H11 plates and MGIT tubes), the samples were decontaminated with arithmetic means to compare isolates and determine clonal relationships. http://jcm.asm.org/ 0.005% cetylpyridium chloride (CPC) for 30 min and further inoculated A similarity index of Ն97% was used to define isolates that were indistin- for culture. guishable, based on the manufacturer’s recommendations. Isolates that In a subgroup of seven patients, once samples were processed for cul- shared 95% similarity were considered similar, and those with Ͻ90% ture, the remaining samples for each house were pooled according to similarity were considered “different/unrelated.” source (e.g., hot water, cold water, or swabs) and frozen for later PCR and The protocol was approved by the Queensland University of Technol- high-resolution melt curve (HRM) analysis. ogy Research Ethics Unit (approval no. 0900000085) and the Human HRM methodology. Frozen samples were thawed in a water bath at Research Ethics Committee of the Prince Charles Hospital (EC 2617). 65°C. One hundred milliliters of each sample was filtered through a 0.45- ␮m-pore-size (47-mm-diameter) membrane (Advantec, Tokyo, Japan). RESULTS on August 22, 2013 by guest Each filter was aseptically transferred into a separate sterile 50-ml tube The species of mycobacteria isolated from clinical samples from containing 10 ml sterile distilled water (SDW). The tubes were vortexed 20 patients included M. avium (5 patients), M. intracellulare (12 for 8 to 10 min at 2,200 rpm, membranes were aseptically removed, and patients), M. abscessus (7 patients), M. gordonae (1 patient), M. the tubes were centrifuged at 4,750 rpm for 30 min at 4°C in a swinging- lentiflavum (1 patient), M. fortuitum (1 patient), M. peregrinum (1 bucket rotor. patient), M. chelonae (1 patient), M. triplex (1 patient), and M. The supernatant was discarded, and DNA was extracted using the kansasii (1 patient). Patient details are summarized in Table 1. Power Water DNA isolation kit (Mo Bio Laboratories, Carlsbad, CA), as Twenty patients’ homes were sampled, and mycobacteria were follows. The pellet was transferred to the supplied Power Bead tube and processed in a Mini-BeadBeater (BioSpec, Bartlesville, OK) for 5 min with cultured from all but one home. (Full details are provided in Table 60 ␮l kit-supplied C1 lysis buffer. The kit’s “Experienced user protocol” S1 in the supplemental material.) Species identified in the home was followed from the second step onwards, with the exception of a pri- matched that found in the patient in seven (35%) cases: M. absces- mary elution performed with 50 ␮l DNase- and RNase-free deionized sus (3 cases), M. avium (1 case), M. gordonae (1 case), M. lentifla- water and a secondary elution performed with 100 ␮l kit-supplied C6 vum (1 case), and M. kansasii (1 case). In an additional patient elution buffer. The DNA extracts were stored at Ϫ80°C until use. All with M. abscessus infection, this species was isolated from potable samples and NTM ATCC controls (M. peregrinum ATCC 700686, M. water in the reservoir zone supplying the home. gordonae ATCC 14470, M. terrae ATCC 15755, M. scrofulaceum ATCC In those instances where the environmental isolate was cul- 19981, M. chelonae NCTC 946, M. fortuitum NCTC 10394, M. abscessus tured (as opposed to direct detection using PCR), the strain type ATCC 19977, M. avium 8867/GI/97, M. intracellulare ATCC 13950, M. matched that of the patient’s infecting strain in the patient with M. smegmatis ATCC 19420, M. kansasii TMC 1201, and M. malmoense ATCC kansasii infection (Fig. 1A). The clinical isolate of M. lentiflavum 29571) were quantified using Nano Drop 1000 (Thermo Scientific, Aus- was closely related to the M. lentiflavum isolate found in the pa- tralia). The real-time PCRs were prepared to a volume of 20 ␮l; their tient’s house and the local water supply. (Fig. 1B) The M. gordonae constituents are shown in Table S1 in the supplemental material. Ampli- isolate from patient’s sputum was unrelated to the isolate found in fication was performed on the Rotor-gene Q real-time PCR instrument (Qiagen Inc., Germantown, MD). Real-time PCR followed by HRM was the home water. Despite repeated attempts, the M. avium isolate performed using the following conditions: step 1, hold at 95°C for 10 min; from the shower swab (patient P6) was unable to be regrown for step 2, cycling at 95°C for 10 s, 52°C for 15 s, and 72°C for 30 s and repeated comparison with the patient’s clinical isolate. In one patient with 40 times; step 3, hold at 72°C for 5 min; and step 4, HRM, melt from 72 to infection due to M. abscessus subsp. bolleti, a different strain type 95°C rising by 0.05°C at each step. In all amplification reactions, negative of M. abscessus was grown from his shower water (Fig. 1C). samples (whose constituents are listed in Table S2 in the supplemental Results from Brisbane-wide sampling of water sites (distribu- material) were used to avoid any possible contamination. Each assay in- tion, reservoir, and trunk main samples) revealed a wide variety of

September 2013 Volume 51 Number 9 jcm.asm.org 3007

313

Thomson et al.

TABLE 1 Clinical details of patients whose homes were sampleda Patient ID (gender; Other patient/sampling age [yr])b environment characteristic(s) Radiology Microbiology and symptoms Clinical notes P1 (F; 72) Gardener with ϩϩ pots BE Sϩϩ bronchial wash in 2007, M. abscessus; cough, Treated for 6 wk, clinically stable but sputum, hemoptysis remained culture positive, died from cancer of the pancreas in 2011 P2 (F; 69) Avid gardener BE Multiple Sϩ samples from 2004, progressive Treated for 4 yr (ethambutol, radiology, wt loss, hemoptysis; M. intracellulare rifampin, clarithromycin, in Sept and Dec 2004 (multiple), April 2006, clofazimine, and amikacin), June 2007, and June 2008 converted sputum in July 2009; culture negative from 2010 to June 2011 BE, small cavities New infiltrate, increased cough and hemoptysis; 2 Treated for 12 mo with ethambutol, positive sputa in June 2011 (1/3); in Sept 2011 rifampin, and clarithromycin, (1/3), M. kansasii converted sputum, in remission in 2013 P3 (M; 72) BE, nodules, large Multiple Sϩϩϩ M. intracellulare, sputa from Partial clinical response to cavities April 2006, significant wt loss, cough treatment, but remains Sϩϩ Cϩ, end-stage disease P4 (F; 74) Ͼ400 orchids, exposure to Nodules, BE, small Sϩϩ M. intracellulare bronchial wash in Feb Treated with ethambutol, rifampin, Downloaded from ϩϩ peat moss cavity 2009; Pseudomonas aeruginosa, Aspergillus and clarithromycin, converted fumigatus, Trichosporin sp.; cough, sputum, sputum in Aug 2009; ongoing hemoptysis hemoptysis; ceased treatment at end of 2011 Relapse/reinfection in 2013, new infiltrate; 1/3 Commenced retreatment in Feb sputa SϪ Cϩ/Ϫ; bronchial wash SϪ Cϩ in Jan 2013 2013, M. intracellulare P5 (F; 51) Rainwater tanks BE Multiple sputa positive in 2007 for M. avium; Treated for 18 mo, converted previous S. aureus, new radiology, symptoms sputum, still in remission P6 (F; 72) Large fernery, rainwater tanks BE Cough, sputum, respiratory failure, cachexia, Died from disease despite treatment

progressive radiology; M. avium in 2007; M. in 2011 http://jcm.asm.org/ intracellulare in 2006 and 2008 P7 (F; 85) Rainwater tanks Nodules, BE Cough, sputum, wt loss, sweats; bronchial wash in Treated for 2 yr with ethambutol, April 2008, Mycobacterium species; OLB in Nov rifampin, and clarithromycin, 2008, M. lentiﬂavum remains in remission; P. aeruginosa P8 (F; 63) Rainwater tanks BE Recurrent LRTIs, persistent cough; bronchial Treated for 2 yr with ethambutol, wash in Nov 2007; in July 2008 and May 2010, rifampin, and clarithromycin, in M. intracellulare remission P9 (F; 63) Gardener BE, nodules Hemoptysis; bronchial wash in Jan 2006, in Aug Treated for 18 mo, RML lobectomy 2007, M. intracellulare in 2011; relapse/reinfection in 2013

P10 (M; 81) BE, nodules Cough, sputum, SOB; bronchial wash in Oct 2007, Treated for 6 wk with i.v. amikacin, on August 22, 2013 by guest M. intracellulare; sputum, M. abscessus, in Jan cefoxitin, and clarithromycin in 2008, Apr 2008, and Sept 2008 and in 2012 2010 and for 2 wk in 2012 P11 (F; 75) Didn’t use shower, gardener, BE Recurrent infections in RML, S. aureus; M. Treated for 24 mo with ethambutol, very dusty house intracellulare isolated most frequently; M. rifampin, clarithromycin, and intracellulare in March 2006; slow grower in ciproﬂoxacin, showed some April 2008; M. abscessus in June 2008; M. symptomatic improvement, but intracellulare in July 2008; M. peregrinum in didn’t convert sputum; currently July 2008; M. gordonae in Aug 2008; M. stable off treatment with chronic intracellulare in Sept to Dec 2008 (multiple); cough, sputum, and lethargy but M. peregrinum in Dec 2008; M. intracellulare in reluctant for retreatment Feb 2009 (multiple); M. peregrinum in March 2009; M. fortuitum in March 2009; M. fortuitum in May 2009; M. intracellulare in Oct 2009; M. fortuitum complex in Dec 2009; M. intracellulare in Jan 2010 P12 (M; 69) Cactus house (Ͼ300 plants) BE, nodules, cavities M. intracellulare in 2003–2004; M. triplex in 2006; Treated over 3–4 yr, drug M. avium and M. intracellulare in 2007; M. intolerances, peripheral intracellulare in 2008; Nocardia farcinica in neuropathy; eventually converted 2009; M. intracellulare in 2009 sputum; remains in remission P13 (F; 60) BE, nodules M. intracellulare repeatedly from 2000–2007; M. RML lobectomy following 4 yr of avium after 15-mo remission in March 2005; treatment, drug side effects, and RGM single isolates while on treatment for short relapse time; now deceased SGM; M. chelonae in May 2005; M. abscessus in secondary to NTM disease Jan 2007 P14 (M; 85) Cavities M. intracellulare in Dec 2006, March 2007, and Treated for 18 mo with ethambutol, April 2007 rifampin, and clarithromycin, converted sputum P15 (M; 25) BE, nodules, Post-pertussis bronchiolitis with secondary Treated for 6 mo with amikacin, bronchiolitis infection, M. abscessus subsp. bolletii in July- cefoxitin, clarithromycin, and Sept 2011 clofazimine P16 (F; 60) RML BE M. avium bronchial wash in 2007 Treated for 6 mo, intolerant of medication; stable and culture negative off treatment P17 (F; 77) BE M. intracellulare in 2000; M. abscessus (multiple), M. intracellulare treated in 2000; in including Sept 2008, Oct 2009, and Feb 2011 2007, multiple courses of treatment for M. abscessus with partial response, limited by drug toxicity; respiratory failure secondary to disease, death (Continued on following page)

3008 jcm.asm.org Journal of Clinical Microbiology

314

NTM Patient Home Sampling

TABLE 1 (Continued) Patient ID (gender; Other patient/sampling age [yr])b environment characteristic(s) Radiology Microbiology and symptoms Clinical notes P18 (F; 53) Large cavities, M. abscessus sputum in Oct 2010, bronchial wash Heavy smoker of cigarettes, nodules, BE in Dec 2010 (with S. aureus) and Dec 2012 and marijuana, large cavities; treated multiple in between with i.v. therapy, response when abstinent from cigarettes; died from cancer of the lung and progressive cavitary disease P19 (F; 80) BE M. abscessus in March and April 2011 Treated for 2 wk with amikacin and cefoxitin and 6 wk with clarithromycin and clofazimine; in remission P20 (F; 65) Nodular BE M. intracellulare in 2007, bronchial wash, Progressive radiology and granulomata on TBB symptoms; treated with ethambutol, rifabutin, and clarithromycin in 2010; rash and hepatitis; treated with azithromycin, ethambutol, clofazimine, and amikacin, in Downloaded from remission a Abbreviations: F, female; M, male; BE, bronchiectasis; S, smear; C, culture; LRTIs, lower respiratory tract infections; SOB, shortness of breath; RGM, rapidly growing mycobacteria; SGM, slowly growing mycobacteria; OLB, open lung biopsy; TBB, transbronchial lung biopsy. For smear and culture results, ϩ and Ϫ indicate positive and negative to the general degree shown (where ϩ is positive and ϩϩϩ is highly positive). b Age calculated from date of ﬁrst isolate.

NTM (18). Some of these NTM (especially M. kansasii, M. gor- were from 9/49 patients’ residences, and 1 was from the 43 volun- donae, M. fortuitum, and M. mucogenicum), which were cultured teers’ residences. Eight isolates were M. avium, and 3 were http://jcm.asm.org/ from 16 homes, matched the species isolated from sampling sites M. intracellulare—a similar ratio to the species causing pulmonary near the patients’ homes. In two instances (patients P10 and P19), disease in the Japanese population. The isolates from the patients’ M. abscessus was cultured from water close to the homes of these bathrooms were compared with those from their sputum samples, patients with M. abscessus disease and the strains were closely re- and 2 had identical pulsed-field gel electrophoresis (PFGE) pat- lated to those found in their clinical samples (Fig. 1D). Similarly in terns and serotypes. For 4 patients the isolates were the same spe- the patient with M. lentiflavum infection, the municipal water cies but had different restriction fragment length polymorphism isolates were similar to the home and clinical isolates (Fig. 1B). Of (RFLP) or PFGE profiles. Interestingly only bathroom samples four other patients for which the infecting NTM species was iso- were positive: MAC was not isolated from kitchen tap water, yet on August 22, 2013 by guest lated from their home samples (patients P1, P2, P9, and P11), the the same system was supplying water to the whole house. The species was also isolated from local water sampling sites. (A full authors postulate that favorable temperature and humidity allow table that includes isolates cultured from reservoir zones supply- MAC to preferentially colonize the bathroom. However, the fre- ing each patient’s house and the distance from the house to the quency of use of kitchen taps may allow more flushing out of sampling point is given in Table S2 in the supplemental material.) mycobacteria from pipes, and the possibility that expectorating Aerosols were collected in 18 homes. The culture plates were patients may have contaminated bathroom drains should not be overgrown in 6 patients, negative at 6 months in 3, but grew NTM discounted. In the single patient who had M. intracellulare infec- in the remaining 9 (50%). The species identified included M. ab- tion, where the pathogen was grown only from a drain site, this scessus (3 homes, including one patient with disease due to M. may certainly have been the case. abscessus), M. gordonae (2 patients), M. kansasii (1 patient), M. Falkinham (7) reported the results of home water sampling fortuitum complex (4 patients), M. mucogenicum (1 patient), and from 31 patients (with 37 homes) in the United States and Canada M. wolinskyi (1 patient). with NTM disease. Patients included in the study had infection DISCUSSION with M. avium (9 patients), M. intracellulare (6 patients), MAC In this study, we have demonstrated that NTM present in potable (11 patients), M. abscessus (4 patients), and M. xenopi (1 patient). water distribution systems can also be found in the houses of pa- Forty-nine percent of households sampled grew NTM, including tients with NTM disease and can be grown from aerosols gener- M. avium (10 homes), M. intracellulare (10 homes), M. malmoense ated by showering. Matching human and environmental strains (5 homes), M. szulgai (3 homes), M. gordonae (6 homes), M. che- were found for M. abscessus, M. lentiflavum, and M. kansasii. lonae (2 homes), M. scrofulaceum (1 home), M. triviale (1 home), Other investigators have explored the possibility that patients with and M. terrae (1 home). In 17/37 (46%) homes, a species was NTM disease (predominantly M. avium and M. intracellulare) isolated that matched the patient’s infecting species, and in seven may have acquired their infection from water sources in the home. of these the strains had matching Rep-PCR patterns. Details of A Japanese group (8) looked at the isolation of NTM from the these matching species are not reported in the paper—although homes of 49 patients with M. avium complex (MAC) pulmonary two of the matching strains illustrated were M. avium. In contrast disease and 43 healthy volunteers without MAC disease. From to our findings, there were a greater yield of M. intracellulare and each home, three 200-ml water samples (kitchen tap, showerhead, an absence of M. abscessus in home samples. and used bathtub water) were collected. In addition, scale on the Aerosol sampling for NTM has been performed previously to surface of showerheads, slime from 3 drains, and dust from air investigate hot tubs and indoor pools as a source of NTM (20). It conditioners was sampled. Eleven samples contained MAC: 10 has also been used to document NTM in air samples collected

September 2013 Volume 51 Number 9 jcm.asm.org 3009

315

Thomson et al. Downloaded from http://jcm.asm.org/ on August 22, 2013 by guest

FIG 1 Rep-PCR dendrograms of (A) M. kansasii isolated from patient sputum (patient P2) and samples from the patient’s home, (B) M. lentiﬂavum isolated from patient P7, her home water, and surrounding water distribution sites, (C) M. abscessus subsp. bolletii from one patient and M. abscessus subsp. abscessus grown from shower water, and (D) M. abscessus isolated from 2 patients and from the water supply close to their homes.

during the remediation of an old building (17) and from a peat number of specimens or the lack of biofilm specimens. In our moss processing plant (5). De Groote (4) analyzed aerosols gen- study, however, a large number of samples and relatively larger erated by the pouring of commercial potting mix using the An- volume of water samples were collected, and biofilm samples were dersen impactor. The isolates shown to be acid fast were presump- collected from every tap. There may be issues with detection, al- tively called mycobacteria but not identified further. though the use of direct detection using HRM didn’t increase the Ours is the first study to examine the household shower aero- yield for M. intracellulare (yet did find other NTM). Faezel (21) sols as a potential source of NTM. We have demonstrated that a reported extensive sampling of the biofilm of showerheads in the variety of species can be aerosolized and cultured from droplets of United States, and while M. avium was found, no M. intracellulare a respirable size. The most significant of these is M. abscessus, was reported. In both of the other home sampling papers, there which was found in aerosols of patients with disease. We did not was no evidence to support the proposal that patients with M. demonstrate the aerosolization of M. intracellulare, which was the intracellulare disease had acquired infection from water sources. main pathogen associated with disease in the patients studied. These findings in combination with the earlier studies done by The yield of M. intracellulare from both Brisbane water (9) and Dawson (3, 13, 14) on house dust and soil suggest that Queensland home samples was interesting. Falkinham (12) has suggested that patients at least may be more likely to acquire M. intracellulare in previous water studies the low yield was due to an insufficient from soil or dust than from water.

3010 jcm.asm.org Journal of Clinical Microbiology

316

NTM Patient Home Sampling

More recently Fujita (22) conducted a large survey of soils plumbing of patients with nontuberculous mycobacteria disease. Emerg. from 100 patients with MAC infection. In those with high soil Infect. Dis. 17:419–424. exposure (Ͼ2 h/week), 6/16 patients had matching soil and clin- 8. Nishiuchi Y, Maekura R, Kitada S, Tamaru A, Taguri T, Kira Y, Hiraga T, Hirotani A, Yoshimura K, Miki M, Ito M. 2007. The recovery of ical isolates using variable-number tandem repeat (VNTR) pro- Mycobacterium avium-intracellulare complex (MAC) from the residential files (5 M. avium and 1 M. intracellulare). This compared with bathrooms of patients with pulmonary MAC. Clin. Infect. Dis. 45:347– none of the 19 patients with low or no soil exposure, again sug- 351. gesting that residential soils may also be a source of exposure for 9. Thomson R, Carter R, Tolson C, Coulter C, Huygens F, Hargreaves M. 2013. Factors associated with isolation of nontuberculous mycobacteria patients with MAC disease. (NTM) from a large municipal water system in Brisbane, Australia. BMC A major limitation of home water and soil sampling studies Microbiol. 13:89. doi:10.1186/1471-2180-13-89. relates to the long lag time between exposure and infection that 10. Marshall H, Carter R, Torbey M, Minion S, Tolson C, Sidjabat H, can occur with NTM disease. Quite often a patient has had symp- Huygens F, Hargreaves M, Thomson R. 2011. Mycobacterium lentiflavum in drinking water supplies, Australia. Emerg. Infect. Dis. 17:395–402. toms for many years prior to diagnosis, and hence, establishing the 11. Thomson R, Tolson C, Carter R, Huygens F, Hargreaves M. 2013. environment in which they were exposed is often very difficult. Mycobacterium abscessus isolated from potable water—a potential source Patients are also potentially exposed to NTM outside their home. of human infection. BMC Infect. Dis. 13:241. doi:10.1186/1471-2334-13 -241.

It is also worth considering the close relationship between soil and Downloaded from water in the environment, in activities such as gardening and in 12. Falkinham J, III, Norton C, Le Chavallier M. 2001. Factors influencing numbers of Mycobacterium avium, Mycobacterium intracellulare,and the situation of soil entering water distribution networks through other mycobacteria in drinking water distribution systems. Appl. Environ. cracks in pipes and rainwater tanks via household roofs. Microbiol. 67:1225–1231. In summary, we have shown that pathogenic NTM can be iso- 13. Reznikov M, Dawson D. 1971. Serological investigation of strains of lated from water, biofilm, and aerosols in the households of pa- Mycobacterium intracellulare (“Battey” bacillus) isolated from house- dusts. Med. J. Aust. 1:682–683. tients with NTM disease. The presence of matching clinical and 14. Dawson D. 1971. Potential pathogens among strains of mycobacteria patient isolates was demonstrated. From our work and other lit- isolated from house-dusts. Med. J. Aust. 1:679–681. erature, the evidence supporting water as a source of infection 15. Andersen A. 1958. New sampler for the collection, sizing, and enumera- http://jcm.asm.org/ would appear greatest for M. avium, M. abscessus, M. lentiflavum, tion of viable airborne particles. J. Bacteriol. 76:471–484. 16. Fennelly KP, Martyny JW, Fulton KE, Orme IM, Cave DM, Heifets LB. and M. kansasii. However, there appears little evidence for M. 2004. Cough-generated aerosols of Mycobacterium tuberculosis:anew intracellulare, and soil exposure may be more relevant. method to study infectiousness. Am. J. Respir. Crit. Care Med. 169:604– 609. 17. Rautiala S, Torvinen E, Torkko P, Suomalainen S, Nevalainen A, REFERENCES Kalliokoski P, Katila ML. 2004. Potentially pathogenic, slow-growing 1. Thomson R. 2010. Changing epidemiology of pulmonary nontubercu- mycobacteria released into workplace air during the remediation of build- lous mycobacteria infections. Emerg. Infect. Dis. 16:1576–1582. ings. J. Occup. Environ. Hyg. 1:1–6. 2. Thomson R, Armstrong J, Looke D. 2007. Gastroesophageal reflux 18. Wendt SL, George K, Parker BC, Gruft H, Falkinham JO, III. 1980. disease, acid suppression, and Mycobacterium avium complex pulmonary Epidemiology of infection by nontuberculous mycobacteria. III. Isolation on August 22, 2013 by guest disease. Chest 131:1166–1172. of potentially pathogenic mycobacteria from aerosols. Am. Rev. Respir. 3. Reznikov M, Leggo J, Dawson DJ. 1971. Investigation by seroagglutina- Dis. 122:259–263. tion of strains of the Mycobacterium intracellulare-M. scrofulaceum group 19. Thomson R, Carter R, Gilpin C, Coulter C, Hargreaves M. 2008. from house dusts and sputum in Southeastern Queensland. Am. Rev. Comparison of methods for processing drinking water samples for the Respir. Dis. 104:951–953. isolation of Mycobacterium avium and Mycobacterium intracellulare. Appl. 4. De Groote MA, Pace NR, Fulton K, Falkinham JO, III. 2006. Relation- Environ. Microbiol. 74:3094–3098. ships between Mycobacterium isolates from patients with pulmonary my- 20. Glazer CS, Martyny JW, Lee B, Sanchez TL, Sells TM, Newman LS, cobacterial infection and potting soils. Appl. Environ. Microbiol. 72: Murphy J, Heifets L, Rose C. 2007. Nontuberculous mycobacteria in 7602–7606. aerosol droplets and bulk water samples from therapy pools and hot tubs. 5. Cayer M-P, Veillette M, Pageau P, Hamelin R, Bergeron M-J, Mériaux J. Occup. Environ. Hyg. 4:831–840. A, Cormier Y, Duchaine C. 2007. Identification of mycobacteria in peat 21. Feazel L, Baumgartner L, Peterson K, Frank D, Harris J, Pace N. 2009. moss processing plants: application of molecular biology approaches. Opportunistic pathogens enriched in showerhead biofilms. Proc. Natl. Can. J. Microbiol. 53:92–99. Acad. Sci. USA 106:16393–16399. 6. Embil J, Warren P, Yakrus M, Stark R, Corne S, Forrest D, Hershfield 22. Fujita K, Ito Y, Hirai T, Maekawa K, Imai S, Tatsumi S, Niimi A, E. 1997. Pulmonary illness associated with exposure to Mycobacterium- Iinuma Y, Ichiyama S, Mishima M. 2013. Genetic relatedness of Myco- avium complex in hot tub water. Hypersensitivity pneumonitis or infec- bacterium avium-intracellulare complex isolates from patients with pul- tion? Chest 111:813–816. monary MAC disease and their residential soils. Clin. Microbiol. Infect. 7. Falkinham JI. 2011. Nontuberculous mycobacteria from household 19:537–541.

September 2013 Volume 51 Number 9 jcm.asm.org 3011

317

Supplementary,material,for,online,posting, , Table,S1:,Home%sampling%results, ! Pt% Source% Type% Temp% CPC% NTM% HRM% Pt%species% ID% P1, Kitchen% Water% Cold% No% M.#mucogenicum# % M.#abscessus# % Kitchen% Swab% % NA% # M.#gordonae# # % Bathroom% Swab% % No% M.#mucogenicum# M.#gordonae# # % Bathroom% Swab% % No% M.#lentiflavum# M.#gordonae# # % Bathroom% Water% Cold% No% M.#mucogenicum# # # % Bathroom% Swab% % NA% # M.#gordonae# # % Shower% Water% Cold% No% M.#mucogenicum# # # % Shower% Swab% % % # M#gordonae# # % Shower%% Aerosol% % % M.#mucogenicum# # # % Shower% Aerosol% % % M.#abscessus# # # # P2, Kitchen% Water% Cold% Yes% M.#gordonae# NT% M.#intracellulare# M.#kansasii# M.#kansasii# % Kitchen% Swab% % Yes% M.#flavescens# # # % Bathroom% Water% Hot% No% M.#austroafricanum# # # % Bathroom% Water% Cold% Yes% M.#kansasii# # # %% Shower% Water% Hot% No% M.#austroafricanum# # # % Shower% Water% Hot% Yes% M.#gordonae# # # M.#kansasii# % Shower% Water% Cold% Yes% M.#gordonae# # # M.#kansasii# % Shower% Swab% % Yes% M.#flavescens# # # % Shower% Aerosol% % % M.#gordonae# # # # P3% Bathroom% Swab% % % # M.#peregrinum# M.#intracellulare# % Shower% Water% Hot% Yes% M.#fortuitum#complex# # # # P4% Kitchen% Water% Hot% No% M.#kansasii# NT% M.##intracellulare# M.#mucogenicum# % Kitchen%% Water%% Hot% Yes% M.#kansasii# # # % Kitchen% Water% Cold% Yes% M.#kansasii# # # % Shower% Water% Hot% No% M.#kansasii# # # % Shower% Water% Cold% Yes% M.#kansasii# # # % Bathroom% Water% Hot% Yes% M.#kansasii# # # M.#mucogenicum# % Bathroom% Water% Cold% Yes% M.#kansasii# # # M.#mucogenicum# % Bathroom% Swab% % Yes% M.#kansasii# # # M.#szulgai# % Tank% Water% % Yes% M.#gordonae# # # # P5% Kitchen% Water% Cold% Yes% M.#gordonae# # M.#avium# M.#kansasii# M.#szulgai# % Kitchen% Swab% % % # M.#gordonae# # % Bathroom% Water% Cold% Yes% M.#neglectum# # # M.#kansasii# % Bathroom% Swab% % Yes% M.#szulgai# M.#gordonae# # M.#kansasii# M.#gordonae#

318

% Shower% Water% Hot% Yes% M.#kansasii# # # M.#szulgai# % Shower% Water% Cold% Yes% M.#szulgai# # # M.#kansasii# % Shower% Swab% % No% M.##intracellulare# M.#gordonae# # M.#gordonae# % %Tank% Water% % Yes% M.#abscessus# % # M.#intracellulare# # P6, Kitchen% Swab% % % # M.gordonae# M.#avium# % Bathroom% Water% Cold% Yes% M.#gordonae# % M.#intracellulare# % Bathroom% Swab% % % # M.#avium# M.#abscessus# M.#chelonae% % Shower% Water% Cold% No% M.#aichiense# % M.#mucogenicum# M.#gordonae# % Shower% Swab% % Yes% M.#poriforae# % M.#avium# M.#scrofulaceum# % Shower% Aerosol% % % M.#fortuitum#complex# % M.# abscessus/chelonae# # P7, Kitchen% Water% Hot% No% M.#mucogenicum# M.#chelonae# M.#lentiflavum# M.#gordonae# % Kitchen% Water% Cold% Yes% M#rhodesiae# M.# intracellulare# % Kitchen% Swab% % Yes% M.#gordonae# % % Bathroom% Water% Cold% Yes% M#mucogenicum# % M.#species#NFI# M.#lentiflavum# % Shower% Water% Hot% No% M.#lentiflavum# % % Shower% Water% Hot% Yes% M.#lentiflavum# % % Shower% Water% Cold% Yes% M.#species#NFI# % % Tank% Water% % Yes% M.#lentiflavum# % M.#mucogenicum# % % % % % # % # P8% Kitchen% Water% Cold% No% M.#kansasii# NT% M.#intracellulare# M.#tilburgii# % Kitchen% Swab% % No% M.#osloensis# % # % Shower% Water% Hot% Yes% M.#osloensis# % # % Shower% Swab% % Yes% M.#osloensis# % # % Bathroom% Water% Hot% Yes% M.#species#NFI# % # % Bathroom% Water% Cold% No% M.#tilburgii# % # M.#mucogenicum# M.#kansasii# % Pool% Water% Cold% No% M.#abscessus# % # M.#mucogenicum# # P9% Shower% Water% Cold% No% M.#gordonae# # M.#intracellulare# % Shower% Swab% % % # M.#avium# M.#abscessus# M.#chelonae# # % Shower% Aerosol% % % M.#kansasii# %

319

# P10, Kitchen% Water% Hot% No% M.#rhodesiae# % M.#abscessus# % Kitchen% Water% Cold% No% M.#rhodesiae# % M.#intracellulare# % Shower% Water% Hot% No% M.#rhodesiae# % % Shower% Water% Hot% Yes% M.#rhodesiae# % % Shower% Water% Cold% Yes% M.#kansasii# % % Shower% Swab% % Yes% M.#rhodesiae# M.#abscessus# M.#chlorphenolicum# # P11, Shower% Swab% % % # M.#gordonae# M.#intracellulare# % Bathroom% Swab% Cold% No% M.#gordonae# M.#gordonae# M.#gordonae# M.#gadium# M.#fortuitum# % Shower% Swab% % % # M.#gordonae# M.#peregrinum# % Kitchen% Aerosol% % % M.#gordonae# % M.#gordonae# # P12% Kitchen% Swab% % Yes% M.#poriforae# NT% M.#intracellulare# M.#gilvum# M.#avium# M.#gordonae# M.#triplex# % Shower% Water% Hot% Yes% M.#gordonae# % % Bathroom% Water% Hot% Yes% M.#gordonae# % % Bathroom% Water% Cold% No% M.#kansasii# % % Bathroom% Water% Cold% Yes% M.#mucogenicum# % % Shower% Water% Cold% Yes% M.#kansasii# % M.#mucogenicum# % Shower% Aerosol% % % M.#wolinskyi# % # P13% Kitchen% Water% Cold% % M.#gordonae# NT% M.#intracellulare# % Shower% Water% % % M.#kansasii# % M.#avium# M.#abscessus# M.#chelonae# # P14% Bathroom% Hot% % % M.#gordonae# NT% M.#intracellulare# M.#mucogenicum# % Kitchen% Hot% % % M.#mucogenicum# % % Kitchen% Cold% % % M.# % abscessus/chelonae# M.#kansasii# % Bath% Hot% % % M.#mucogencium# % % Bath% Cold% % % M.#rhodesiae# % % Shower% Hot% % % M.#gordonae# % % P15% Shower% Water% Cold% No% M.#fortuitum# NT% M.#abscessus# % Shower% Water% Cold% Yes% M.#fortuitum# % % Shower% Water% Cold% No% M.#mucogenicum# % % Bathroom% Water% Cold% No% M.#mucogenicum# % % Bathroom% Water% Cold% Yes% M.#mucogenicum# % % Bathroom% Water% Cold% Yes% M.#abscessus# % % Laundry% Water% Cold% Yes% M.#kansasii# % % Shower% Aerosol% Stage% % M.#fortuitum# % 2% % Shower% Aerosol% Stage% % M.#fortuitum# % 3% % Shower% Aerosol% Stage% % M.#fortuitum# % 4% P16% % % % % No#NTM#isolated# NT% M.#avium# from#any#site#

320

# P17% Kitchen% Swab% % Yes% M.#monalense# NT% M.#abscessus# % Shower% Water% Cold% No% M.#mucogenicum# % Shower% Water% Cold% Yes% M.#mucogenicum# % P18% Kitchen% Water% Cold% No% M.#mucogenicum# NT% M.#abscessus# % Bathroom% Water% Hot% No% M.#mucogenicum# % % Bathroom% Water% Cold% No% M.#mucogenicum# % Bathroom% Water% Cold% Yes% M.#mucogenicum# % Bathroom% Swab% % Yes% M.#mucogenicum# % Fishtank%% Water% % Yes% M.#fortuitum#complex# % Fishtank% Water% % No% M.#fortuitum#complex# % P19% Shower%% Water% % Yes% M.#mucogenicum# NT% M.#abscessus# % Shower% Swab% % Yes% M.#mucogenicum# # % Bathroom% Water% Cold% No% M.#mucogenicum# % % Bathroom% Water% Cold% Yes% M.#mucogenicum# % Bathroom% Water% Cold% Yes% M.#mucogenicum# % Shower% Aerosol% % % M.#fortuitum# % Shower% Aerosol% % % M.#fortuitum# % Shower% Aerosol% % % M.#fortuitum# % % % % % # % % P20% Kitchen% Water% Cold% No% M.#abscessus# NT% M.#intracellulare# % Kitchen% Water% Cold% Yes% M.#mucogenicum# % Shower% Water% Hot% No% M.#mucogenicum# % Shower%% Water% Cold% No% M.#mucogenicum# % Shower% Water% Cold% Yes% M.#mucogenicum# % Shower% Water% Cold% No% M.#mucogenicum# % Bathroom% Water% Hot% Yes% M.#fortuitum# % Bathroom% Water% Cold% No% M.#fortuitum# % Bathroom% Water% Cold% Yes% M.#fortuitum# % Bathroom% Swab% % No% M.#fortuitum# % Bathroom% Swab% % Yes% M.#fortuitum# % Shower% Aerosol% % % M.#fortuitum# % % % % % # % % ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

321

Table,S2:%Home%sampling%results,%with%NTM%grown%from%local%municipal%water%sampling%sites% and%their%proximity%to%houses%sampled.%

ID% Infecting% Species%grown% Species% Closest% Type*% Species%grown%from% Species% from%home% from%HRM% sampling%site% site% (no;Distance)% P3% M.#intracellulare# M.#fortuitum# M.# 83%(1.70)% D% M.#gordonae# complex# peregrinum# M.#mucogenicum# 130%(0.96)% D% Contaminated/NAF% 182%(1.98)% D% M.#gordonae# M.#fortuitum# M.#kansasii# 231%(2.25)% R%% M.#gordonae# M.#mucogenicum# # P11% M.#intracellulare# M.#gordonae# M.# 80%(1.42)% D% M.#gordonae# gordonae# 78%(1.7)% R% M.#mucogenicum# M.#fortuitum# M.#gadium% 264%(1.7)% R% M.#lentiflavum# M.#peregrinum# M.#mucogenicum# M.#species#NFI# M.#gordonae, 230,231%(2.23)% R% M.#gordonae# M.#mucogenicum# 316%(0.57)% D% M.##mageritense# M.#mucogenicum# 81%(0.87)% D% M.#mucogenicum# 289% R% M.#gordonae# M.#lentiflavum# 280% R% M.#arupense# M.#gordonae# M.#mucogenicum# M.#species#NFI# # P2% M.#intracellulare# M.#gordonae# NT% 41%(0.23)% D% M.#abscessus/chelonae# M.#kansasii# 42%(1.19)% D% M.#abscessus/chelonae# M.#kansasii# M.#flavescens# 72%(0.88)% D% M.#gordonae# M.# 132%(1.87)% R% Contaminated# austroafricanum# 190%(0.71)% TM% M.#gordonae# M.#gordonae# M.##szulgai# M.#flavescens# # P1% M.#abscessus# M.# M.# 210%(1.03)% D% M.#gordonae# mucogenicum# gordonae# M.#mucogenicum# # M.#poriforae# M.#lentiflavum# 39%(1.25)% % # # 212%(2.59)% R% M.#fortuitum#complex# M.#abscessus# M.#gordonae# M.#mucogenicum# 231%(2.59)% D% M.#species#NFI# 226%(5.30)% R% Contaminated# # P10% M.#abscessus# M.#rhodesiae# M.# 125% R% M.#gordonae# M.#intracellulare# M.#kansasii# abscessus# 132%(0.47)% R% Contaminated# M.# 58%(2.07)% R% M.##gordonae# chlorphenolicum# 10% R% Contaminated# 191%(2.73)% R% M.#fluoroanthenivorans# M.#kansasii/gastri#

322

57%(1.81)% D% M.#abscessus# M.#kansasii# 183,184%(1.31)% R% M.#gordonae# M.#kansasii# # P9% M.#intracellulare# M.#gordonae# M.#avium# 40%(0.35)% R% No#sample# # M.# 193%(0.90)% R% No#sample# M.#kansasii# abscessus# 38%(0.92)% D% M.#arupense# M.#chelonae# 180,181%(0.35)% R% M.#gordonae# M.#fortuitum#complex# M.#kansasii# # P20% M.#intracellulare# M.#abscessus# NT% 125%(0.55)% D% M.#species#NFI# M.# M.#gordonae# mucogenicum# 58%(0.82)% R% M.#gordonae# M.#fortuitum# 183,184%(1.39)% R% M.#gordonae# # M.#kansasii# 57%(0.55)% D% M.#abscessus# M.#kansasii# # P15% M.#abscessus# M.#fortuitum# NT% 276%(0.46)% D% M.#fluroanthenivorans# M.# M.#gordonae# mucogenicum# M.#mucogenicum# M.#abscessus# 280,%283,%284% R% M.#arupense# M.#kansasii# (1.32)% M.#gordonae# # M.#mucogenicum# # M.#species# 285Z288%(1.44)% R% M.#diernhoferi# M.#gordonae# M.#gordonae# M.#kansasii# M.#kansasii# # P16 M.#avium# Nil% NT% 194%(6.85)%% % M.#gordonae# #% M.#gordonae# M.#mucogenicum# 220% % Contaminated# 26% % Contam# 28% % NAF# # P19% M.#abscessus# M.# NT% 56%(0.62)% D% M.#lentiflavum# mucogenicum# 274%(2.17)% D% Contaminated## M.#fortuitum# 114(2.6)% D% M.#abscessus# # M.#abscessus# M.#cosmeticum# M.#fluroanthenivorans# 136(2.78)% R% No#sample# 164%(3.16)% D% M.#gordonae# # P17% M.#abscessus# M.#monalense# NT% 175%(0.82)% D% M.#fortuitum#complex# M.# M.#fortuitum#complex# mucogenicum# M.#gordonae# # M.#kansasii# M.#kansasii# M.#szulgai# 177%(2.12)% D% Contaminated# 219%(1.33)% D% Contaminated#

323

M.#kansasii# 119%(1.62)% D% M.#kansasii# M.#kansasii# M.#kansasii# M.#szulgai# 185,186%(4.56)% R% Contaminated# 280,%283,284% R% M.#arupense# (4.79)% M.#gordonae# M.#mucogenicum# M.#species# 285Z289%(4.73)% R% M.#diernhoferi# M.#gordonae# M.#gordonae# M.#kansasii# M.#kansasii# # P18% M.#abscessus# M.# NT% 121%(1.55)% D% M.#fortuitum#complex# mucogenicum# M.#mucogenicum# M.#fortuitum# 219%(3.05)% D% Contaminated# complex# M.#kansasii# # 268%(3.38)% D% M.#flavescens# M.#fluroanthenivorans# M.#fortuitum#complex# M.#fortuitum#complex# M.#mucogenicum# 63%(4.03)% D% M.#fortuitum#complex# M.#gordonae# M.#gordonae# M.#mucogenicum# # P14% M.#intracellulare# M.#gordonae# NT% 176%(0.12)% D% M.#gordonae# M.# 177%(1.53)% D% Contaminated# mucogenicum# 280%(1.34)% R% M.#arupense# M.#poriferae# M.#gordonae# M.# M.#mucogenicum# abscessus/chelo M.#species# nae# 283,284%(1.68)% R% No#sample# M.#kansasii# 285Z289%(1.68)% R% M.#diernhoferi# M.#hodleri# M.#gordonae# M.#rhodesiae# M.#gordonae# # M.#kansasii# M.#kansasii# 185,186%(2.17)% R% Contaminated# 277%(1.11)% TM% M.#arupense# M.#fortuitum#complex# M.#gordonae# # P13% M.#intracellulare# M.#gordonae# NT% 237%(2.78)% D% No#sample# M.#avium# M.#kansasii# 258%(1.98)% D% No#sample# M.#abscessus# M.#poriferae# 274%(3.13)% D% Contaminated# M.#chelonae# # 291%(4.57)% R% Contaminated# M.#gordonae# M.#mucogenicum# 292%(5.28)% R% M.#cosmeticum# M.#diernhoferi# M.#gordonae# M.#mageritense#

324

293%(3.56)% R% M.#lentiflavum# M.#mucogenicum# M.#mucogenicum# M.#poriferae# 290%(3.90)% R% M.#gordonae# # P12% M.#intracellulare# M.#poriforae# NT% 169%(0.88)% D% M.#gordonae# M.#avium# M.#gilvum# M.#kansasii# M.#triplex# M.#gordonae# M.#kansasii# M.#kansasii# M.#kansasii# M.# 153%(1.33)% TM% No#sample# mucogenicum# 37%(1.36)% D% M.#gordonae# M.#wolinskyi# 218%(1.45)% D% M.#fortuitum#complex# M.#gordonae# M.#kansasii# M.#mucogenicum# 40%(3.44)% R% No#sample# 180,181%(3.44)% R% M.#gordonae# M.#fortuitum#complex# M.#kansasii# 193%(3.44)% R% No#sample# 154%(2.09)% R% M.#kansasii# M.#sengalense# # P8% M.#intracellulare# M.#kansasii# NT% 56%(0.16)% D% M.#lentiflavum# M.#tilburgii# 281%(1.12)% D% M.#fortuitum#complex# M.#osloensis# M.#kansasii# M.#species#NFI# 229%(1.47)% D% M.#kansasii# M.# 241%(2.57)% D% Contaminated# mucogenicum# 136%(3.07)% R% No#sample# M.# abscessus/chelo nae# # # P4% M.# M.#kansasii# NT% 281%(0.16)% D% M.#fortuitum#complex# intracellulare# M.#kansasii# M.# 56%(1.0)% D% M.#lentiflavum# mucogenicum# 164%(1.95)% % M.#gordonae# M.#szulgai# 136%(2.43)% R% No#sample# M.#gordonae# 246%(2.07)% D% M.#flavescens# M.#mucogenicum# # P6% M.#avium# M.#gordonae# NT% 37%(1.52)% D% M.#gordonae# M.#abscessus# # 249%(2.39)% D% Contaminated# M.#chelonae# M.#aichiense# 169%(1.96)% D% M.#gordonae# M.# M.#kansasii# mucogenicum# 40%(3.67)% R% No#sample# M.#poriforae# 180,181(3.6)% R% M.#gordonae# M.#avium# M.#fortuitum#complex# M.#scrofulaceum# M.#kansasii# M.#fortuitum# 193(3.67)% R% No#sample# complex# M.# abscessus/chelo nae# #

325

P7% M.#lentiflavum# M.# M.#chelonae# 149%(1.07)% TM% No#sample# mucogenicum# M.# 183,184%(1.91)% R% M.#gordonae# M#rhodesiae# gordonae# M.#kansasii# M.#gordonae# M.# 131%(1.38)% TM% Contaminated# M.#species#NFI# intracellular M.#gordonae# M.#lentiflavum# e# M.#lentiflavum# # M.#mucogenicum# 128%(1.48)% TM% M.#interjectum# M.#simiae# 125%(2.13)% R% M.#gordonae# 52%(1.67)% D% M.#gadium# # P5% M.#avium# M.#gordonae# M.# 252%(1.15)% D% No#sample## M.#kansasii# gordonae# 261(3.73)% D% M.#mucogenicum# M.#szulgai# # M.#poriferae# # M.# 244%(2.31)% D% M.#mucogenicum# M.#neglectum# gordonae# 243%(2.89)% D% M.#kansasii# M.## # 290Z292%(5.98)% R% M.#gordonae# intracellulare# # M.#mucogenicum# M.#abscessus# M.# M.#cosmeticum# gordonae# M.#diernhoferi# M.#gordonae# M.#mageritense# 293%(6.4)% R% M.#lentiflavum# M.#mucogenicum# M.#poriferae# %

#%House%6.85km%from%North%Pine%Water%Treatment%Plant%and%possibly%outside%the%distribution%

network%as%no%testing%sites%close%to%house.%Sites%listed%those%within%North%Pine%system.%%

*%DZDistribution;%RZ%Reservoir;%TMZTrunk%Main%

% %

! ! ! ! ! !

326

APPENDIX H

Thomson,R. (2010). Changing epidemiology of NTM infections in QLD, Australia. Emerging Infectious Diseases, 16(10), 1576-83

@LML9@NO

Changing Epidemiology of Pulmonary Nontuberculous Mycobacteria Infections Rachel M. Thomson, on behalf of the NTM working group at the Queensland TB Control Centre and Queensland Mycobacterial Reference Laboratory

MedscapeCME ACTIVITY

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ategory 1 Credits\:,];.+0)0/#+,+;"%*5,"#*.,)*/0-,)('50$, )"--'#+%(/$',?0$;,$;','[$'#$,">,$;'0(,=/($0)0=/$0"#,0#,$;',/)$0B0$.:,9**,"$;'(,)*0#0)0/#+,)"-=*'$0#C,$;0+,/)$0B0$.,?0**,&',0++%'5,/,)'($0!,6 )/$',">,=/($0)0=/$0"#:,2",=/($0)0=/$',0#,$;0+,Z"%(#/*,N3L,/)$0B0$.X,1E4,('B0'?,$;',*'/(#0#C,"&Z')$0B'+,/#5,/%$;"(,50+)*"+%('+^,1G4,+$%5., $;','5%)/$0"#,)"#$'#$^,1J4,$/K',$;',="+$6$'+$,/#5U"(,)"-=*'$',$;','B/*%/$0"#,/$,www.medscapecme.com/journal/eid^,1_4,B0'?U=(0#$, )'($0!,)/$':

Learning Objectives `="#,)"-=*'$0"#,">,$;0+,/)$0B0$.8,=/($0)0=/#$+,?0**,&',/&*',$"X R5'#$0>.,$;',=('B/*'#)',/#5,'=05'-0"*"C.,">,#"#$%&'()%*"%+,-.)"&/)$'(0/,1!234,0#>')$0"#,/#5,$(/50$0"#/*,(0+K,>/)$"(+, /#5,=('+'#$/$0"#,">,=%*-"#/(.,!23,0#>')$0"#:

N"#+$(%)$,/#,/==("=(0/$',50/C#"+$0),+$(/$'C.,>"(,=/$0'#$+,?0$;,+%+=')$'5,!23:

Editor Beverly D. Merritt, 2');#0)/*,D(0$'(UL50$"(8 Emerging Infectious Diseases. Disclosure: Beverly D. Merritt has disclosed no relevant ﬁ nancial relationships.

CME Author Desiree Lie, MD, MSED, N*0#0)/*,](">'++"(,">,a/-0*.,3'50)0#'8,S0(')$"(,">,@'+'/();,/#5,a/)%*$.,S'B'*"=-'#$8,`#0B'(+0$.,">,N/*06 >"(#0/8,R(B0#',/$,W(/#C'8,N/*0>"(#0/: Désirée Lie, MD, MSEd, has disclosed the following relevant ﬁ nancial relationship: served as a nonproduct speaker for: “Topics in Health” for Merck Speaker Services.

Authors Disclosure: Rachel M. Thomson, MD, has disclosed no relevant ﬁ nancial relationships.

!"#$%&'()%*"%+,-.)"&/)$'(0/,1!234,50+'/+',0+,/,#"$06 ?;",+-"K',$",!,&("#"5%*/(,50+'/+',0#,'*5'(*.,?"-'#:,My- !,/&*',)"#50$0"#,0#,7%''#+*/#58,9%+$(/*0/:,3.)"&/)$'(0/*,0+"6 cobacterium intracellulare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–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

EIPQ, L-'(C0#C,R#>')$0"%+,S0+'/+'+,T,???:)5):C"BU'05,T,V"*:,EQ8,!":,EH8,W)$"&'(,GHEH

327

N;/#C0#C,L=05'-0"*"C.,">,!23,R#>')$0"#+

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ycobacterium intracellulare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`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a'1(/1(*H"2'(40*"=*B-(#'2/*bB8HJLaHBc:*4"*/11'14*'.*4<(* (+'%(-'"$";'2*%/4/*".*4<'1*2".%'4'".*1<",$%*3(*"=*'.4(#(14* %'/;."1'1*"=*1';.'!*2/.4*%'1(/1(>* 4"*+,3$'2*<(/$4<*(D+(#41> * 9"#(* %(4/'$(%* 2$'.'2/$* '.="#-/4'".* '1* #(G,(14(%* ="#* H4,%'(1* "=* /@'/.* @(#1,1* <,-/.* 9/.4",D* 4(14'.;* '.* +/4'(.41*&<"** 8<(#(="#()* 1'.2(* 4#(/4-(.4*#(2('@(%>*?#"@'1'".*"=*4<'1*'.="#-/4'".*'1*@"$,.5 4<(*'.4#"%,24'".*"=*8M*2".4#"$*1(#@'2(1)*%'1(/1(*2/,1(%*30* 4/#0>*d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d"#*+/4'(.41* "=*+,$-"./#0*%'1(/1(*2/,1(%*30*M. avium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e+(5 "=*-(/.'.;=,$*2$'.'2/$*%/4/*/.%*="$$"&5,+*"=*2$'.'2/$$0*1';5 .`+'f* e+(.* H",#2(* `+'%(-'"$";'2* H4/4'14'21* ="#* ?,3$'2* .'!*2/.4*789*2/1(1> F(/$4<)* @(#1'".* X>V>* 6&&&>"+(.(+'>2"-:>* ?"+,$/4'".* 14/5 4'14'21* &(#(* "34/'.(%* =#"-* 4<(* B,14#/$'/.* M,#(/,* "=* H4/5 Methods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g*X* -/'.*$/3"#/4"#'(1*/.%*/*=(&*1-/$$(#*$/3"#/4"#'(1>*8<(*X*+#'5 4/3$(1*'.*e+(.`+'>*

, L-'(C0#C,R#>')$0"%+,S0+'/+'+,T,???:)5):C"BU'05,T,V"*:,EQ8,!":,EH8,W)$"&'(,GHEH, EIPP,

328

@LML9@NO

Results B*-"#(*/;;#(11'@(*2/+4,#(*"=*1'.;$()*+"1'4'@()*,.1+(2'5 /4(%*'1"$/4(1*"22,##(%*'.*PQQQE*4<(*4"4/$*.,-3(#*"=*789* '1"$/4(1*&/1*PW>RJPTT)TTT*+"+,$/4'".>*M(2/,1(*4<(*.,-3(#* "=* 1+(2'/4(%* +,$-"./#0* '1"$/4(1* '.* (/2<* 0(/#* 6PQQQ* /.%* XTTS:*1<",$%*."4*PhPV>UJPTT)TTT* +"+,$/4'".:>* 8<(* 4"4/$* .,-3(#*"=*'1"$/4(1*'.*XTTS*&/1*PS>RJPTT)TTT*+"+,$/4'".>* B11,-'.;*4<(*+#"+"#4'".*"=*,.1+(2'/4(%*'1"$/4(1*#(-/'.(%* 2".14/.4)*/.%*".*4<(*3/1'1*"=*4<(*'.2#(/1(*'.*1+(2'/4(%*'1"5 $/4(1)*&(*(14'-/4(%*4<(*4"4/$*.,-3(#*"=*'1"$/4(1*'.*XTTS*4"* 3(*XX>PJPTT)TTT*+"+,$/4'".> 8<(* %'==(#(.2(* 3(4&((.* 4<(* X* 0(/#1* &/1* %,(* 4"* 4<(* <';<(#* .,-3(#* "=* +,$-"./#0* '1"$/4(1* /1* %(-".14#/4(%* '.* 8/3$(* P>* H+(2'/4(%* +,$-"./#0* '1"$/4(1* #"1(* 1';.'!*2/.4$0)* a0C%(',E:,N;/#C'+,0#,>('<%'#).,">,0+"*/$0"#,>"(,+*"?6C("?0#C,1Mi4, #"#$%&'()%*"%+, -.)"&/)$'(0/, 1/&+"*%$', #%-&'(+48, 7%''#+*/#58, =#"-*S>S*4"*PT>XJPTT)TTT*+"+,$/4'".>* 9%+$(/*0/8,EFFF,/#5,GHHI:,39N8,Mycobacterium avium,)"-=*'[: 8<(#(*&(#(*%'==(#(.2(1*3(4&((.*4<(*1+(2'(1*"=*789* '1"$/4(%*'.*PQQQ*/.%*XTTS*6d';,#(1*P)*X:>*8<(*1+(2'(1*/25 2",.4'.;*="#*-"14*"=*4<(*2*9"#(*"=*4<(*9BO*'1"$/4(1* 2"--".*1+(2'(1*&/1*M. fortuitum*6VVJPWV:)*="$$"&(%*30* /#(* ."&* '%(.4'!*(%* 30* -,$4'+$(D* ?OY* /1* ('4<(#* M. avium* M. intracellulare*6PX:)*M. abscessus*6PP:)*/.%*M. chelonae* "#* M. intracellulare>* a,#'.;* PQQQhXTTS)* 4<(* .,-3(#* "=* 6PP:>* e@(#/$$)* UR>S\* "=* .".+,$-"./#0* '1"$/4(1* &(#(* =($4* M. avium* '1"$/4(1* '.2#(/1(%* =#"-* VS* 6PJPTT)TTT:* 4"* UX* 4"* 3(* /11"2'/4(%* &'4<* 2$'.'2/$$0* 1';.'!*2/.4* 789* %'1(/1(>* 6P>SSJPTTTTT:*/.%*M. intracellulare*=#"-*ii*6X>XJPTT)TTT:* 8/3$(*W*1<"&1*4<(*14/4'14'2/$*1';.'!*2/.2(*"=*4<(1(*'1"$/4(1* 4"*XPX*6S>VJPTT)TTT:)*&<'$(*9BO*%(2#(/1(%*=#"-*VS*4"*V>* /22"#%'.;*4"*1+(2'(1>* M. abscessus*'1"$/4(1*'.2#(/1(%*=#"-*XV*6T>TUSJPTT)*TTT:* '.*PQQQ*4"*WT*6PJPTT)TTT:*'.*XTTS>*M. fortuitum*'1"$/4(1*'.5 Pulmonary Isolates 2#(/1(%*=#"-*WP6P>PiJPTT)TTT:*'.*PQQQ*4"*ST*6P>XSJPTT)TTT:* e=*4<(*PPP*.(&*."4'!*2/4'".1*'.*XTTS*&'4<*1';.'!*2/.4* '.*XTTS>* +,$-"./#0*%'1(/1(*4*e=*WRR*+/4'(.41*&'4<* ="#*SU>R\*"=*2/1(1>*M. avium*6Pi:*&/1*4<(*.(D4*-"14*2"-5 +,$-"./#0*789*'1"$/4(1)*".$0*XU>U\*V\:)*="$$"&(%*30*M. kansasii*6Q*bR>P\c:>*B*4"4/$* (/1(*"@(#/$$>*?(#2(.4/;(1*&(#(*1$';<4$0*<';<(#*="#*+/4'(.41* "=*PQ*2/1(1*&(#(*#(+(/4*."4'!*2/4'".1*6'>(>)*#($/+1(J#('.=(25 &'4<*M. intracellulare*'1"$/4(1*6VQ>W\:)*M. avium*6VV>V\:)* 4'".E*M. intracellulare*6PS:)*M. avium*6P:)*M. kansasii*6P:)* M. kansasii*6SX>U\:)*/.%*-,2<*$"&(#*="#*+/4'(.41*'.=(24(%* M. fortuitum*6P:)*/.%*M. abscessus*6P:>* &'4<* 1+(2'(1* 4#/%'4'"./$$0* 4<",;<4* -"#(* $'C($0* 4"* 3(* 2".5 8<(*/;(*/.%*1(D*%'14#'3,4'".*"=*+/4'(.41*&'4<*1';.'!*5 4/-'./.41)*1,2<*/1*M. gordonae*6PP>P\:>*e4<(#*#/#(#*1+(5 2/.4*+,$-"./#0*%'1(/1(*%'==(#(%*'.*PQQQ*/.%*XTTS>*L.*PQQQ)* 2'(1*&(#(*."4*4<",;<4*4"*3(*/11"2'/4(%*&'4<*+,$-"./#0*%'15 4<(* /31"$,4(* .,-3(#* "=* -(.* &'4<* %'1(/1(* ",4.,-3(#(%* (/1(*68/3$(*V:>* &"-(.*'.*/$$*/;(*;#",+1*&'4<*4<(*(D2(+4'".*"=*4<(*[iS50(/#*

2/&*',E:,N"-=/(0+"#,">,#%-&'(+,">,#"#$%&'()%*"%+,-.)"&/)$'(0/,0+"*/$'+,0#,EFFF,/#5,GHHI8,7%''#+*/#58,9%+$(/*0/c, !":,0+"*/$'+,1(/$'UEHH8HHH,="=%*/$0"#^,FId,NR4, R+"*/$'+ EFFF, GHHI =,B/*%', 2"$/*,=/$0'#$,0+"*/$'+ IEPe 1E_:b^,EJ:_fE_:F4, QJEg 1EI:b^,E_:JfEQ:F4, !M , ]%*-"#/(., JbG,1EH:F^,F:FfEG:E4, _bb,1EG:G^,EE:EfEJ:J4, !M ,!"#=%*-"#/(., EJI,1J:F^,J:_f_:G4, E_J,1J:Q^,J:Ef_:H4, !M M=')0/$'5,0+"*/$'+, JEb,1F:E^,b:GfEH:G4, I__,1EJ:Q^,EG:JfE_:P4, hH:HHHHHHE, ,]%*-"#/(., EFG,1I:I^,_:IfQ:Q4, _EJ,1EH:G^,F:_fEE:I4, hH:HHHHHHE, ,!"#=%*-"#/(., EEF,1J:_^,J:HfJ:b4, EJE,1J:J^,G:FfJ:P4, !M cNR8,)"#>05'#)',0#$'(B/*^,!M8,#"$,+0C#0>0)/#$:,]"=%*/$0"#,">,7%''#+*/#5,0#,EFFF8,J:I,-0**0"#^,0#,GHHI8,_,-0**0"#:, eR#,EFFF8,-"(',+-'/(6#'C/$0B'8,+0#C*'6="+0$0B',%#+=')0/$'5,0+"*/$'+,?'(',)/=$%('5,$;/#,%+%/*8,0#>*/$0#C,$;'+',#%-&'(+:, g2;',)/*)%*/$'5,#%-&'(,">,="+0$0B',0+"*/$'+,0#,GHHI,?/+,GG:EUEHH8HHH,="=%*/$0"#8,&/+'5,"#,$;',"&+'(B'5,0#)('/+',0#,+=')0/$'5,0+"*/$'+8,/++%-0#C,$;', =("="($0"#,">,0+"*/$'+,#"$,('>'(('5,>"(,+=')0/$0"#,('-/0#'5,)"#+$/#$:,,

EIPb, L-'(C0#C,R#>')$0"%+,S0+'/+'+,T,???:)5):C"BU'05,T,V"*:,EQ8,!":,EH8,W)$"&'(,GHEH

329

N;/#C0#C,L=05'-0"*"C.,">,!23,R#>')$0"#+

'.2#(/1(* =#"-* PQQQ>* Z'44$(* 2*F"&(@(#)* 4<(*'.2'%(.2(*"=*'1"$/4'".*"=*789*=#"-*<,-/.*+,$-"./#0* 1+(2'-(.1*PJPTT)TTT*+"+,$/4'".>*8<(* '.2'%(.2(*"=*."4'!*(%*1';.'!*2/.4*+,$-"./#0*%'1(/1(*X*4"*V>XJPTT)TTT*+"+,$/4'".*+(#*0(/#>*8<'1* '.2#(/1(*'1*$/#;($0*%,(*4"*/.*'.2#(/1(*'.*4<(*.,-3(#*"=*($5 %(#$0*&"-(.*&'4<*.".2/@'4/#0*%'1(/1(>* 8<(* "31(#@(%* 2*M. intracellulare*'1*4<(*-/'.*+/4<";(.* /11"2'/4(%*&'4<*4<(*'.2#(/1(*'.*%'1(/1()*="$$"&(%*30*M. abscessus,*/.%*M. kansasii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hiW*0(/#1*"=*/;(*6+*j*T>TTTS:)* B8HJLaHB*14/.%/#%1*"=*XTTi*615:*TVT:*8<(#(* 2#"3'"$";'2/$* 2#'4(#'/)* /.%* '4* '1* $'C($0* -"#(* 2/1(5+/4'(.41* &/1*/*2"-3'.(%*"@(#/$$*'.2#(/1(*'.*+/4'(.41*UThiW*0(/#1*"=* &",$%*."&*3(*2".1'%(#(%*4"**M(2/,1(*789* /;(*6+*j*T>TTX:*/.%*/*%(2#(/1(*'.*4<"1(*VThWW*0(/#1*"=*/;(* %'1(/1(* '1* /* 2<#".'2* %'1(/1()* "=4(.* "@(#* -/.0* 0(/#1)* 4<(* 6+*j*T>TPV:>* +#(@/$(.2(*"=*%'1(/1(*&'$$*3(*-,2<*<';<(#*/.%*-"#(*$'C($0* 8<(* #/%'"$";'2* +/44(#.* "=* %'1(/1(* *`@(.*&'4<* 6d';,#(*W:>*9"14*"=*4<(*'.2#(/1(*&/1*'.*+/4'(.41*&'4<*.".5 4<(*2,##(.4*."4'!*2/4'".*+#"2(11*'.*N,((.1$/.%)*&(*=/'$(%*4"* 2/@'4/#0*%'1(/1()*=#"-*XS*6T>iJPTT)TTT:*4"*RW*6X>PJPTT)TTT:* 2/+4,#(*/$$*1'.;$()*+"1'4'@()*,.1+(2'/4(%*'1"$/4(1>* 2/1(1*6+kT>TTTP:>*B*1$';<4*%#"+*=#"-*VV*6T>QJPTT)TTT:*2/15 B* #'1'.;* 3"%0* "=* $'4(#/4,#(* 1,++"#41* 4<(* "31(#@/4'".* (1*"=*2/@'4/#0*%'1(/1(*'.*PQQQ*4"*XR*6T>iJPTT)TTT:*"22,##(%* 30*2$'.'2'/.1*4XS:>* '1*'.2#(/1'.;>*e4<(#*14,%'(1*'.*B,14#/$'/** 8<(* -"14* Discussion #(2(.4* #(+"#4* '1* =#"-* 4<(* 7"#4<(#.* 8(##'4"#0>* B$$* '1"$/4(1* L.* XTTS)* 789* &(#(* '1"$/4(%* =#"-* PS>U* +(#1".1* +(#* =#"-* PQRQhPQQi* &(#(* (D/-'.(%* 616:>* a'1(/1(* '.2'%(.2(* PTT)TTT*+"+,$/4'".E*<"&(@(#)*4<(*4#,(*.,-3(#*"=*'1"$/4(1* '.2#(/1(%*=#"-*X>i*4"*W>i*+(#*PTT)TTT*+"+,$/4'".*3(4&((.* =#"-*4PJPTT)TTT)*/*1';.'!*2/.4* 4<(* !*#14* /.%* 1(2".%* * ?/.;* #(5 2/&*',G:,M0C#0>0)/#)',">,#"#$%&'()%*"%+,-.)"&/)$'(0/,0+"*/$'+,/+,#"$0>0'58,7%''#+*/#58,9%+$(/*0/8,EFFF,/#5,GHHIc, !"#$%&'()%*"%+,-.)"&/)$'(0/, EFFF, GHHI, 0+"*/$'+ !":,1d4,0+"*/$'+, @/$'UEHH8HHH,="=%*/$0"#, !":,1d4,0+"*/$'+, @/$'UEHH8HHH,="=%*/$0"#, ]%*-"#/(., ,M0C#0>0)/#$,, PP,1GH:G4, ,G:G, EJH,1GQ:Q4, J:GI ,`#5'(,'B/*%/$0"#,, E,1H:GQ4, H:HJ IG,1EH:Q4, E:J ,!"$,+0C#0>0)/#$, JH_,1PF:Q4, ,b:P, JHQ,1QG:P4, P:QI L[$(/=%*-"#/(., ,M0C#0>0)/#$,, EHQ,1Pb:I4, G:QI Fb,1Qb:I4, G:_I ,`#5'(,'B/*%/$0"#, H f _,1G:b4, ,f, ,!"$,+0C#0>0)/#$, GF,1GE:I4, H:bJ,, _E,1Gb:P4, E:HJ cf8,#"$,/==*0)/&*',

, L-'(C0#C,R#>')$0"%+,S0+'/+'+,T,???:)5):C"BU'05,T,V"*:,EQ8,!":,EH8,W)$"&'(,GHEH, EIPF,

330

@LML9@NO

2/&*',J:,N*0#0)/*,+0C#0>0)/#)',">,#"#$%&'()%*"%+,-.)"&/)$'(0/,0+"*/$'+,>("-,=%*-"#/(.,+0$'+8,7%''#+*/#58,9%+$(/*0/8,GHHIc, M=')0'+ M0C#0>0)/#$, `#5'( 'B/*%/$0"#, !"$,+0C#0>0)/#$, 2"$/* M*"?,C("?'(+, Mycobacterium intracellulare PF,1JF:P4, GG,1EE:E4, Fb,1_F:G4, EFF M. avium Eb,1JJ:J4, b,1E_:b4, Gb,1_b:J4, I_ 39N f f J 1EHH4, J M. kansasii EH,1IG:Q4, J,1EI:b4, Q,1JE:Q4, EF M. scrofulaceum J1JH4, f P1PH4, EH M. gordonae G,1EE:E4, f EQ,1bb:F4, Eb M. lentiflavum f f I 1EHH4, I M. asiaticum f f J 1EHH4, J M. haemophilum f f E 1EHH4, E M. simiae f f G 1EHH4, G ,`#+=')0/$'5, f f G_,1EHH4, G_ @/=05 C("?'(+, f f f f M. abscessus P,1G_:E4, _,1EJ:b4, Eb,1QG:E4, GF M. chelonae J,1GE:_4, G,1E_:J4, F,1Q_:J4, E_ M. fortuitum E,1I:J4, E,1I:J4, EP,1bF:_4, EF M. peregrinum f E,1JJ:J4, G,1QQ:P4, J M. mucogenicum f f G 1EHH4, G M. nonchromogenicum fE,1IH4,E,1IH4,G ,`#+=')0/$'5, G,1Gb:Q4, f I,1PE:_4, P 9$.=0)/*,-.)"&/)$'(0/,!aR, I1P:H4, EH,1E_:E4, IQ,1Pb:F4, PE W$;'(, f f J 1EHH4, J 2"$/* EJH IG JHQ _bb cV/*%'+,/(',C0B'#,/+,#":,0+"*/$'+,1d,">,$"$/*,>"(,+=')0'+4:,39N8,Mycobacterium avium,)"-=*'[^,f8,#"$,/==*0)/&*'^,!aR8,#"$,>%($;'(,05'#$0>0'5:,, +"#4(%*/*V5="$%*'.2#(/1(*'.*M. kansasii*%'1(/1(*'.*!(14(#.* 4';(.1*4"*/11(11*'.=(24'".*"#*/4*$(/14*(D+"1,#(*4"*789*&/1* B,14#/$'/*%,#'.;*PQUXhPQRX*/.%*PQRVhPQRi*619:>*9"#(*#(5 '.'4'/$$0*,1(%*4"*%(-".14#/4(*4<(*2".1'%(#/3$(*;(";#/+<'2* 2(.4$0*9/##/1*(4*/$>*63:*#(+"#4(%*/.*'.2#(/1(*'.*+"+,$/4'".* %'14#'3,4'".*"=*789*'.=(24'".*620,21:>*9"#(*#(2(.4$0*l*%(-".14#/4(%*/.*'.2#(/1(*'.*1C'.*4(14*#(/24'@'40*4"*M. 6Q>PJPTT)TTT:*4"*XTTi*6PW>PJPTT)TTT:>*B$4<",;<*/%(G,/4(* intracellulare*'.*PQiP*/.%*PQiX*6PP>X\:*/.%*PQQQhXTTT* 2$'.'2/$*%/4/*&(#(*."4*/@/'$/3$(*="#*/$$*+/4'(.41)*'4*'1*$'C($0* 6PU>U\:*/-".;*#(+#(1(.4/4'@(*mH*+"+,$/4'".*2"<"#41*622:>* 4*HC'.*4(14'.;*30*,1'.;*9BO*/.5 4<(.*4<(*"31(#@(%*'.2#(/1(1*2/.."4*3(*/44#'3,4(%*4"*4<(*'.5

2/&*',_:,N*0#0)/*,+0C#0>0)/#)',">,#"#$%&'()%*"%+,-.)"&/)$'(0/,0+"*/$'+,>("-,#"#=%*-"#/(.,+0$'+8,7%''#+*/#58,9%+$(/*0/8,GHHIc, M=')0'+ M0C#0>0)/#$, `#5'( 'B/*%/$0"#, !"$,+0C#0>0)/#$, 2"$/* Mycobacterium intracellulare P,1Ib:J4, f I,1_E:P4, EG M. avium Q,1QQ:P4, G,1GG:G4, E,1EE:E4, F M. scrofulaceum J,1QH4, E,1GH4, E,1GH4, I M. gordonae E,1JJ:J4, f G,1QQ:P4, J M. abscessus F,1bE:b4, f G,1Eb:G4, EE M. chelonae b,1PG:P4, f J,1GP:J4, EE M. fortuitum GJ,1QF:P4, J,1F:E4, P,1GE:G4, JJ M. peregrinum P,1EHH4, f f P M. asiaticum f f E 1EHH4, E M. haemophilum E,1JJ:J4, f G,1QQ:P4, J M. lentiflavum f f E 1EHH4, E M. mucogenicum f f E 1EHH4, E M. marinum f f E 1EHH4, E M. smegmatis E1IH4, f E1IH4, G M. szulgai E,1EHH4, f f E M. ulcerans _,1EHH4, f f _ 9$.=0)/*,-.)"&/)$'(0/,!aR, J,1GP:J4, f b,1PG:P4, EE M*"?,C("?'(,%#+=')0/$'5, E,1IH4, f E,1IH4, G W$;'(, f f E 1EHH4, E 2"$/* FP,1QP:b4, Q,1_:G4, _H,1Gb4, E_J cV/*%'+,/(',C0B'#,/+,#":,0+"*/$'+,1d,">,$"$/*,>"(,+=')0'+4:,f8,#"$,/==*0)/&*'^,!aR8,#"$,>%($;'(,05'#$0>0'5:,

EIbH, L-'(C0#C,R#>')$0"%+,S0+'/+'+,T,???:)5):C"BU'05,T,V"*:,EQ8,!":,EH8,W)$"&'(,GHEH

331

N;/#C0#C,L=05'-0"*"C.,">,!23,R#>')$0"#+

8<"-1".)*,.+,3>*%/4/:>*a,#'.;*4<(*14,%0*+(#'"%)*N,((.5 1$/.%** 8<(* #(%,24'".*'.*"*"&*&'4<'.*%'14#'3,4'".*1014(-1*-/0**?#(@'",1$0)*M. scrofulaceum*&/1*4<(*-/'.*+/4<";(.*/11"2'/4(%*&'4<*2<'$%<""%* $0-+*B*2*8<(*#(1,#;(.2(*"=*4<'1*1+(2'(1*'.*XTTS*1,++"#41*4<(* ."4'".*"=*'.2#(/1(%*2<$"#'.(*%(;#/%/4'".)*3#",;<4*/3",4*30* "*"&*#(%,24'".1*3,4*+"4(.4'/$$0*/$1"*30*4(-+(#/4,#(*'.2#(/15 (1*/11"2'/4(%*&'4<*2$'-/4(*2* O* 8<(*2*F"&(@(#)*1"-(*/,4<"#1* *Y(2(.4$0)*d(/A($*(4* /$>* 625:* =",.%* -02"3/24(#'/$* a7B* 2".2(.4#/4(%* '.* 4<(* 3'"!*$-1*"=*1<"&(#<(/%1*'.*4<(*m.'4(%*H4/4(1>*N,((.1$/.%* '-+$(-(.4(%* /* 14#".;* +,3$'2* /&/#(.(11* 2/-+/';.* +#"5 -"4'.;*W5-'.,4(*1<"&(#1*4"*#(%,2(*&/4(#*,1/;()*&<'2<*'.* 4<("#0*1<",$%*#(%,2(*(D+"1,#(*4"*-02"3/24(#'/*4<#",;<* 4<'1*#",4(>* H(@(#/$*2/1(*1(#'(1**`/#$'(#*14,%'(1**9"#(*#(5 2(.4*14,%'(1** 8<(*%'==(#(.2(1*3(4&((.*4<(1(*;#",+1*"=*14,%'(1*-/0*3(* /44#'3,4(%* 4"* 1/-+$'.;* 3'/1* /.%* +"+,$/4'".* %'==(#(.2(1>* !(*

a0C%(', J:, 9C'6/5Z%+$'5, (/$'+, >"(, )/+'+, ">, #"#$%&'()%*"%+, -.)"&/)$'(0/, 50+'/+', /-"#C, ?"-'#, 1948, -'#, 1Y48, /#5, /**, ('+05'#$+,1N48,7%''#+*/#58,9%+$(/*0/8,EFFF,/#5,GHHI:,c=,j,H:HHHI^, e=,j,H:HJH^,g=hH:HHI^,k=,j,H:HIP:

2#(/1(%*,1(*"=*2"-+,4(%*4"-";#/+<0*12/.1)*3(44(#*$/3"#/5 4"#0*4(2<.'G,(1)*"#*;#(/4(#*2$'.'2'/.*/&/#(.(11>*L-+#"@(%* %(4(24'".*-(4<"%1)*1,2<*/1*4<(*'.4#"%,24'".*"=*3#"4<*2,$4,#(* 69IL8*MBO8`O*QUTE*M(24".*a'2C'.1".)*d#/.C$'.*Z/C(1)* 7n)*mHB:*3(="#(*4<(*%/4/*2"$$(24'".*"=*PQQQ)*/$1"*2/.."4* (D+$/'.*",#*"31(#@/4'".1>*B.*'.2#(/1(*'.*(.@'#".-(.4/$*(D5 +"1,#(* -/0* *8<(*-/'.*(.@'#".-(.4/$*.'2<(*="#*M. intracellulare*/++(/#1*4"*3(*'.*3'"!*$-1*$'.'.;*%#'.C'.;*&/4(#*+'+(1* a0C%(', _:, N;/#C'+, 0#, $;', (/50"*"C0), /=='/(/#)', ">, )/+'+, ">, 67:> M. intracellulare** EFFF,/#5,GHHI:

, L-'(C0#C,R#>')$0"%+,S0+'/+'+,T,???:)5):C"BU'05,T,V"*:,EQ8,!":,EH8,W)$"&'(,GHEH, EIbE,

332

@LML9@NO

/;(%*-(.*&<"*1-"C(*4"*4<(*."%,$/#*3#".2<'(24/4'2*="#-* **Q>** B3#/*`+'%(-'"$";'2/$*(@'%(.2(*"=*4<(*+#(15 "=*%'1(/1(*'.*($%(#$0*&"-(.>*N,((.1$/.%*;"@(#.-(.4*14/5 (.2(*"=*.".54,3(#2,$",1*1(.1'4'@'40*4"*4,3(#2,$'.*'.*N,((.1$/.%>*8,5 3(#2$(>*PQUPEWXfWRihQQ>*aeLf*PT>PTPUJHTTWP5VRiQ6UP:RTPVR5W 4'14'21*1,;;(14*4** B3#/*H(.1'4'@'40*4"*/@'/.*/.%*<,-/.*??a*'.* +#(%'24(%*4"*#'1(*=#"-*VU>V*0(/#1*'.*XTTU*4"*WS*0(/#1*'.* M#'13/.(*12<""$*2<'$%#(.>*8,3(#2$(>*PQUiEWRfiQhQW>*aeLf*PT>PTPUJ XTSP)*/.%*4<(*.,-3(#*"=*+(#1".1*[US*0(/#1*"=*/;(*&'$$* HTTWP5VRiQ6Ui:RTTTX5V '.2#(/1(*W5="$%*30*XTSP>*B1*4<(*@,$.(#/3$(*+"+,$/4'".*'.5 PP>** B3#/*H4,%'(1*&'4<*4<(*+,#'!*(%*+#"4('.*%(5 #'@/4'@(*"=*<,-/.*/.%*/@'/.*4,3(#2,$'.1*'.*H",4<*N,((.1$/.%>*8,5 2#(/1(1)*789*%'1(/1(*'1*/$1"*$'C($0*4"*'.2#(/1(>*8<(#(="#()* 3(#2$(>*PQUREWQfPQXhXTQ>*aeLf*PT>PTPUJTTWP5VRiQ6UR:QTTXX5U 4<(*'-+/24*"=*4<'1*%'1(/1(*".*<(/$4<*3,%;(41*'1*&"#4<0*"=* PX>** B-(#'2/.* 8<"#/2'2* H"2'(40>* a'/;."1'1* /.%* 4#(/4-(.4* "=* %'1(/1(* =,#4<(#*/44(.4'".>* 2/,1(%* 30* .".4,3(#2,$",1* -02"3/24(#'/>* B-* n* Y(1+'#* O#'4* O/#(* 9(%>*PQQiEPSUfHPhXS> PV>** B,14#/$'/.* M,#(/,* "=* H4/4'14'21>* B,14#/$'/.* a(-";#/+<'2* H4/4'15 Acknowledgments 4'21* n,.(* N,/#4(#* PQQQ>* O"--".&(/$4<* "=*B,14#/$'/E* PQQQ* b2'4(%* 8<(*/,4<"#*4* <44+fJJ&&&>/31>;"@>/,JBmHH8B8HJ/31q>.1=J 4<(*+#"@'1'".*"=*$/3"#/4"#0*'1"$/4(1*/.%*O<#'14"+<(#*9/2%(#-"44)* a(4/'$1?/;(JVPTP>Tn,.\XTPQQQre+(.a"2,-(.4>* F'#/.4<'*!/$+"$/)*O<#'14'.(*Z";/.)*9/#C*H4'2C$(0)*/.%*B./14/5 PW>** B,14#/$'/.* M,#(/,* "=* H4/4'14'21>* B,14#/$'/.* a(-";#/+<'2* H4/4'15 4'21* a(2(-3(#* G,/#4(#* XTTS>* O"--".&(/$4<* "=* B,14#/$'/E* XTTU* 1'"1* l".14/.4'."1* ="#* /11'14/.2(* '.* -/'.4/'.'.;* 4<(* 789* %/4/5 b2'4(%*XTPT*B,;*PXc>*<44+fJJ&&&>/31>;"@>/,JBmHH8B8HJ/31q>.1=J 3/1(>* 8Ta(2\XTXTTSre+(.a"2,-(.4 &<"* @"$,.4((#(%* '.="#-/4'".* #(;/#%'.;* '.%'@'%,/$* +/4'(.41* ="#* PS>* *I#'=!*4<*a`)*BC1/-'4*8)*M#"&.5`$$'"44*MB)*O/4/.A/#"*B)*a/$(0*O)* 789*%/4/*2"$$(24'".*/.%*4<(*+#'@/4(*$/3"#/4"#'(1*6H,$$'@/.*/.%* I"#%'.*d)*(4*/$>*B.*"=!*2'/$*B8HJLaHB*14/4(-(.4f*%'/;."1'1)*4#(/45 -(.4)* /.%* +#(@(.4'".* "=* .".4,3(#2,$",1* -02"3/24(#'/$* %'1(/1(1>* 7'2"$/'%(1* ?/4<"$";0* /.%* N,((.1$/.%* 9(%'2/$* Z/3"#/4"#0:* ="#* B-* n* Y(1+'#* O#'4* O/#(* 9(%>* XTTiEPiSfVUihWPU>* aeLf* PT>PPUWJ /%%'4'"./$*'.="#-/4'".*+#"@'%(%*'.*PQQQ>* #22->XTTUTW5SiPH8 PU>** eKM#'(.*a?)*O,##'(*Mn)*l#/,1(*oZ>*7".4,3(#2,$",1*-02"3/24(#'/$* L.*PQQQ*Y>8>*&/1*1,++"#4(%*30*/*#(1(/#2<*12<"$/#1<'+*=#"-* %'1(/1(*'.*."#4<(#.*B,14#/$'/f*/*2/1(*1(#'(1*/.%*#(@'(&*"=*4<(*$'4(#/5 4<(*?#'.2(11*B$(D/.%#/*F"1+'4/$*Y(1(/#2<*d",.%/4'".>* 4,#(>*O$'.*L.=(24*a'1>*XTTTE*6W:fQSRhUi>*aeLf*PT>PTRUJVPRPVU Pi>** `%&/#%1*dI>*a'1(/1(*2/,1(%*30*s/40+'2/$K*6"++"#4,.'14'2:*-02"3/25 a#* 8<"-1".* '1* /* 4<"#/2'2* +<01'2'/.* /4* 4<(* N8MOO)* 8<(* 4(#'/f*/*&<"$(*+"+,$/4'".*#(@'(&>*8,3(#2$(>*PQiTESPfXRShQS>*aeLf* ?#'.2(*O*H<(* PT>PTPUJTTWP5VRiQ6iT:QTTXP5R '1*/$1"*/*?** O/##,4<(#1*ln)*`%&/#%1*dI>*B40+'2/$*-02"3/24(#'/*'.*!(14(#.*B,15 4#/$'/>*B-*Y(@*Y(1+'#*a'1>*PQUSEQPfRRihQS> 3(#2,$",1*-02"3/24(#'/*#($/4(%*4"*+,$-"./#0*'.=(24'".1>* PQ>** ?/.;* HO>* Mycobacterium kansasii* '.=(24'".1* '.* !(14(#.* B,15 4#/$'/>* Y(1+'#* 9(%>* PQQPERSfXPVhR>* aeLf* PT>PTPUJHTQSW5UPPP5 6TU:RTTRX5t References XT>** ?/$-(#*O`>*8,3(#2,$'.*1(.1'4'@'40*/.%*2".4/24*&'4<*4,3(#2,$"1'1>*B-* Y(@*8,3(#2>*PQSVEURfUiRhQW> **P>** F(.#0* 98)* L./-%/#* Z)* eKY'"#%/'.* a)* H2<&(';(#* 9)* !/41".* n?>* XP>** `%&/#%1* ZM)*B2G,/@'@/* dB)*Z'@(1/0*o8)*O#"11* d!)*?/$-(#* O`>* 7".4,3(#2,$",1* -02"3/24(#'/* '.* .".5FLo* +/4'(.41f* (+'%(-'"$5 B.*/4$/1*"=*1(.1'4'@'40*4"*4,3(#2,$'.)*??a5M)*/.%*<'14"+$/1-'.*'.*4<(* ";0)* 4#(/4-(.4* /.%* #(1+".1(>* `,#* Y(1+'#* n>* XTTWEXVfiWPhU>* aeLf* m.'4(%*H4/4(1>*B-*Y(@*Y(1+'#*a'1>*PQUQEQQfPhPVX> PT>PPRVJTQTVPQVU>TW>TTPPWTTW XX>** l*7".4,3(#2,$",1*-02"3/24(#'/$*1(.5 **X>** 9/##/1*8l)*a/$(0*OZ>*`+'%(-'"$";0*"=*<,-/.*+,$-"./#0*'.=(24'".* 1'4'A/4'".*'.*4<(*m.'4(%*H4/4(1f*./4'"./$*4#(.%1*"@(#*4<#((*%(2/%(1>* &'4<*.".4,3(#2,$",1*-02"3/24(#'/>*O$'.*O<(14*9(%>*XTTXEXVfSSVh B-* n* Y(1+'#* O#'4* O/#(* 9(%>* XTTiEPiUfVTUhPV>* aeLf* PT>PPUWJ Ui>*aeLf*PT>PTPUJHTXiX5SXVP6TX:TTTPQ5Q #22->XTTiTX5XTPeO **V>** 9/##/1*8l)*O<(%"#(*?)*p'.;*B9)*n/-'(1".*d>*L1"$/4'".*+#(@/$(.2(* XV>** d/$C'.* `.@'#".-(.4/$* 1",#2(1* "=* Mycobacterium avium* "=*+,$-"./#0*.".54,3(#2,$",1*-02"3/24(#'/*'.*e.4/#'")*PQQihXTTV>* $'.C(%*4"*#",4(1*"=*(D+"1,#(>*L.f*?(%$(0*H*Mn)*Y((1*I)*a,=",#*B)* 8<"#/D>*XTTiEUXfUUPhU>*aeLf*PT>PPVUJ4XTTU>TiTiQi O"4#,@"*n)*(%'4"#1>*?/4<";(.'2*-02"3/24(#'/*'.*&/4(#>*Z".%".f*L!B* **W>** 8<"-1".*Y9)*p(&*!!>*!<(.*/.%*<"&*4"*4#(/4*+,$-"./#0*.".5 ?,3$'1<'.;E*XTTW>*+>*XUhVR> 4,3(#2,$",1* -02"3/24(#'/$* %'1(/1(1>* Y(1+'#"$";0>* XTTQEPWfPXhXU>* XW>** 9/##/1*8l)*!/$$/2(*Yn)*n#)*l"4<*ZZ)*H4,$3/#;*9H)*O"&$*O8)*a/$(0* aeLf*PT>PPPPJ_>PWWT5PRWV>XTTR>TPWTR>D OZ>*F0+(#1(.1'4'@'40*+.(,-".'4'1*#(/24'".*4"*Mycobacterium avium* **S>** Z(@'.*aZ>*Y/%'"$";0*"=*+,$-"./#0*Mycobacterium avium-intrac- '.*<",1(<"$%*&/4(#>*O<(14>*XTTSEPXifUUWhiP>* ellulare*2"-+$(D>*O$'.*O<(14*9(%>*XTTXEXVfUTVhPX>*aeLf*PT>PTPUJ XS>** d(/A($*Z9)*M/,-;/#4.(#*Zl)*?(4(#1".*lZ)*d#/.C*a7)*F/##'1*nl)* HTXiX5SXVP6TX:TTTTQ5U ?/2(* 7Y>* e++"#4,.'14'2* +/4<";(.1* (.#'2<(%* '.* 1<"&(#<(/%* 3'"5 **U>** l'-*Ya)*I#((.3(#;*a`)*`<#-/.4#/,4*9`)*I,'%(*Ho)*a'.;*Z)*H<(/* !*$-1>*?#"2*7/4$*B2/%*H2'*m*H*B>*XTTQEPTUfPUVQVhQ>*aeLf*PT>PTiVJ p)*(4*/$>*?,$-"./#0*.".4,3(#2,$",1*-02"3/24(#'/$*%'1(/1(f*+#"1+(25 +./1>TQTRWWUPTU 4'@(*14,%0*"=*/*%'14'.24*+#((D'14'.;*10.%#"-(>*B-*n*Y(1+'#*O#'4*O/#(* XU>** B<.*OF)*Z"&($$*nY)*e.14/%*Ia)*H<,="#%*`F)*F,#14*IB>*B*%(-"5 9(%>*XTTREPiRfPTUUhiW>*aeLf*PT>PPUWJ#22->XTTRTS5URUeO ;#/+<'2*14,%0*"=*%'1(/1(*%,(*4"*Mycobacterium kansasii*"#*M. in- **i>** d/$C'.;*d/24"#1*'."*,5 tracellulare-avium*'.*8(D/1>*O<(14>*PQiQEiSfPXThS>*aeLf*PT>PViRJ (.2'.;*.,-3(#1*"=*Mycobacterium avium, Mycobacterium intracel- 2<(14>iS>X>PXT lulare,* /.%* "4<(#* -02"3/24(#'/* '.* %#'.C'.;* &/4(#* %'14#'3,4'".* 1015 Xi>** d#((-/.*n)*9"##'1*B)*M$/2C-"#(*8)*F/--(#*a)*9,.#"(*H)*92l5 4(-1>* B++$* `.@'#".* 9'2#"3'"$>* XTTPEUifPXXShVP>* aeLf* PT>PPXRJ .';<4*Z>*L.2'%(.2(*"=*.".4,3(#2,$",1*-02"3/24(#'/$*%'1(/1(*'.*7(&* B`9>Ui>V>PXXS5PXVP>XTTP u(/$/.%)*XTTW>*7*u*9(%*n>*XTTiEPXTfSThU> **R>** eKM#'(.*Yn)*I('4(#*Zn)*H.'%(#*a`*n#>*8<(*(+'%(-'"$";0*"=*.".4,3(#5 XR>** ?#'.2(*aH)*?(4(#1".*aa)*H4('.(#*Y9)*I"44$'(3*n`)*H2"44*Y)*L1#/($* 2,$",1*-02"3/24(#'/$*%'1(/1(1*'.*4<(*m.'4(%*H4/4(1>*Y(1,$41*=#"-*/* FZ)*(4*/$>*L.=(24'".*&'4<*Mycobacterium avium*2"-+$(D*'.*+/4'(.41* ./4'"./$*1,#@(0>*B-*Y(@*Y(1+'#*a'1>*PQRiEPVSfPTTihPW> &'4<",4*+#(%'1+"1'.;*2".%'4'".1>*7*`.;$*n*9(%>*PQRQEVXPfRUV>

EIbG, L-'(C0#C,R#>')$0"%+,S0+'/+'+,T,???:)5):C"BU'05,T,V"*:,EQ8,!":,EH8,W)$"&'(,GHEH

333

N;/#C0#C,L=05'-0"*"C.,">,!23,R#>')$0"#+

XQ>** F,/.;* nF)* l/"* ?7)*B%'* o)*Y,"11* Hn>* Mycobacterium avium-in- B%%#(11* ="#* 2"##(1+".%(.2(f* Y/2<($* 9>* 8<"-1".)* N,((.1$/.%* 8,3(#5 tracellulare*+,$-"./#0*'.=(24'".*'.*FLo5.(;/4'@(*+/4'(.41*&'4<",4* 2,$"1'1* O".4#"$* O(.4#()* XWhXR* O"#.&/$$* H4)* B..(#$(0)* M#'13/.()* +#((D'14'.;* $,.;* %'1(/1(f* %'/;."14'2* /.%* -/./;(-(.4* $'-'4/4'".1>* N,((.1$/.%*WPTV)*B,14#/$'/E*(-/'$f*#>4<"-1".q,G>(%,>/, O<(14>*PQQQEPPSfPTVVhWT>*aeLf*PT>PViRJ2<(14>PPS>W>PTVV

, L-'(C0#C,R#>')$0"%+,S0+'/+'+,T,???:)5):C"BU'05,T,V"*:,EQ8,!":,EH8,W)$"&'(,GHEH, EIbJ,

334