University of Florida Thesis Or Dissertation Formatting

NONTUBERCULOUS MYCOBACTERIA IN FLORIDA

STEPHANIE CASSANDRA YARNELL

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2011

To my parents, Howard and Donna Yarnell

ACKNOWLEDGMENTS

I would like to thank the members of the Emerging Pathogens Institute for their help- Youliang Qiu, Jason Blackburn, Kevin Fennelly, Michael Lauzardo, Judith

Johnson, Glenn Morris, and Andrew Kane, as well as, the present and past members of the Marine Animal Disease Laboratory for their help - Linda Archer, Jim Wellehan, and

Hendrik Nollens. This work could not have been done without them. Stephen Hsu and

Skip Harris deserve additional thanks for the immense moral support. Funding for this project was provided, in part, by an Opportunity Fund Award (Project Number

00091337) from the University of Florida, Office of Research, the University of Florida

Emerging Pathogens Institute, the College of Public Health and Health Professions, and the MD/PhD Program. Further technical support was provided by Steve McNulty,

Barbara Elliott, Linda Bridge, Ravikiran, and Richard Wallace of the University of Texas,

Tyler, Texas. I thank my committee for their mentorship- my chair, Judith Johnson, and

Glenn Morris, Andrew Kane, Michael Lauzardo, as well as the IDP representative,

Richard Condit. My parents, Howard and Donna, deserve special thanks for being incredibly supportive as I continue down my seemingly endless educational path.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 8

LIST OF FIGURES ...... 9

LIST OF ABBREVIATIONS ...... 10

ABSTRACT ...... 13

CHAPTER

1 INTRODUCTION ...... 15

General Background ...... 15 Cell Structure ...... 16 Medical ...... 19 Respiratory Infections ...... 21 Hypersensitivity Pneumonitis ...... 25 Cystic Fibrosis (CF) ...... 26 Disseminated Disease ...... 28 Immunology ...... 31 Soft tissue/ Cutaneous Manifestations ...... 33 Buruli Ulcer ...... 34 Traumatic Wound/ Surgery ...... 36 Chronic Bowel Disease ...... 38 Cervical Lymphadenitis ...... 42 Wildlife and Veterinary Diseases ...... 43 Intracellular Growth ...... 45 Environmental Distribution ...... 46 Water ...... 48 Water Quality ...... 50 DOC ...... 51 Salt Tolerance ...... 51 Dissolved Oxygen ...... 52 Temperature ...... 53 pH ...... 54 Drinking Water ...... 55 Geographical Mapping ...... 65 Final Thoughts ...... 67

2 NONTUBERCULOUS MYCOBACTERIA DETECTED USING NOVEL QPCR PROBES ...... 69

Background ...... 69 Materials and Methods ...... 71 Primer Development ...... 71 Serial Dilutions ...... 73 Extraction by Physical Disruption ...... 74 Extraction by Chemical Disruption ...... 75 Comparison of qPCR and Traditional PCR ...... 75 Results ...... 75 Primer Sensitivity and Specificity ...... 75 Chemical versus Physical Disruption ...... 76 Culture Confirmation ...... 76 Comparison of qPCR and Traditional PCR ...... 77 Discussion ...... 77

3 NONTUBERCULOUS MYCOBACTERIA IN FLORIDA SURFACE AND MUNICIPAL WATERS ...... 86

Background ...... 86 Materials and Methods ...... 87 Sampling for Mycobacterial DNA ...... 88 DNA Extraction ...... 88 Quantitative PCR (q-PCR) ...... 88 Conformation of Near-Threshold qPCR Data ...... 89 Culture Confirmation ...... 90 Water Chemistry ...... 91 Statistics ...... 91 Results ...... 91 Quantification ...... 91 PCR and Sequencing ...... 92 Culture Confirmation ...... 93 Water Quality ...... 93 Discussion ...... 94

4 DEMOGRAPHICS OF NONTUBERCULOUS MYCOBACTERIA CASES IN FLORIDA, 2006-2008 ...... 104

Background ...... 104 Materials and Methods ...... 106 Data Source ...... 106 Data Analysis ...... 107 Multivariate Analysis ...... 107 Results ...... 108 Pulmonary NTM ...... 109 Disseminated NTM ...... 111

Discussion ...... 112 Age ...... 112 Gender ...... 114 Race ...... 115 Seasonality ...... 116 Payer Status ...... 117 Length of Stay ...... 117 Co-illnesses ...... 118 Incidence ...... 122 Multivariate Analysis ...... 123 Final Thoughts ...... 124

5 SPATIO-TEMPORAL PATTERNS OF REPORTED NONTUBERCULOUS MYCOBACTERIA CASES IN FLORIDA, 2006-2008 ...... 136

Background ...... 136 Overview ...... 136 Geospatial Analysis ...... 138 Materials and Methods ...... 140 Data Source ...... 140 Data Analysis ...... 141 Confidentiality ...... 141 Smoothing ...... 143 Spatial Analyses ...... 143 Results ...... 144 Spatial Distribution of Pulmonary NTM ...... 144 Spatial Distribution of Disseminated NTM ...... 145 Clustering of NTM Hospitalization ...... 146 Pulmonary NTM Hospitalizations ...... 146 Disseminated NTM Hospitalizations ...... 148 Discussion ...... 149 Incidence ...... 150 Cluster Analysis ...... 153 Final Thoughts ...... 156

6 FINAL CONCLUSIONS ...... 163

Overview ...... 163 Final Thoughts ...... 174

APPENDIX: SEQUENCE RESULTS OF CULTURED BACTERIA ...... 176

LIST OF REFERENCES ...... 180

BIOGRAPHICAL SKETCH ...... 236

LIST OF TABLES

Table page

2-1 CFU and calculated CFU for serial dilutions of mycobacterial cultures...... 80

3-1 Water quality data, pan-bacterial counts, mycobacteria counts and percentage mycobacteria from municipal and environmental water samples. .... 99

3-2 Comparison of estimated total bacterial and mycobacterial counts from aquifer source water (pre-treatment) and municipally-distributed water (post- treatment) from two municipal distribution systems...... 100

3-3 Range of environmental bacterial counts (total bacteria and mycobacteria) reported from previous studies...... 101

4-1 Co-illnesses associated with pulmonary NTM disease ...... 126

4-2 Co-illnesses associated with disseminated NTM disease...... 127

4-3 Significant results of the multivariate analysis...... 128

5-1 Caseload, zip codes (zips) reporting, incidence for both raw and smoothed data, and maximum incidence noted in any given zip code for pulmonary and disseminated NTM ...... 157

LIST OF FIGURES

Figure page

2-1 Genus Specificity Curves...... 81

2-2 MAC Specificity Curve...... 82

2-3 Standard Curves...... 83

2-4 Traditional PCR Results...... 84

2-5 qPCR Results...... 85

3-1 Estimated counts of total bacteria, mycobacteria, and MAC based on qPCR determination of number of DNA copies...... 102

4-1 Racial distribution of pulmonary versus disseminated NTM hospitalizations. ... 129

4-2 Racial distribution of pulmonary versus disseminated NTM normalized yearly incidence rates...... 130

4-3 Age and gender distributions of pulmonary NTM ...... 131

4-4 Normalized yearly incidence distributions broken down by age and gender for pulmonary NTM ...... 132

4-5 Age and gender distributions of disseminated NTM ...... 133

4-6 Normalized yearly incidence distributions broken down by age and gender for disseminated NTM ...... 134

4-7 Payer status for patients hospitalized with either pulmonary or disseminated NTM disease...... 135

5-1 Reference map of Florida showing major roadways, cities, and water systems...... 158

5-2 Incidence maps of pulmonary NTM disease, 2006-2008, with and without HIV co-illness...... 159

5-3 Incidence maps of disseminated NTM disease, 2006-2008, with and without HIV co-illness...... 160

5-4 LISA maps showing cluster analyses of pulmonary NTM disease, 2006-2008, with and without HIV co-illness...... 161

5-5 LISA maps showing cluster analyses of disseminated NTM disease, 2006- 2008, with and without HIV co-illness...... 162

LIST OF ABBREVIATIONS

A adenosine

AHCA Agency for Healthcare Administration

AIDS acquired immune deficiency syndrome

AOC assimilable organic carbon

APHA American Public Health Association

BALF bronchoalveloar lavage fluid

BCG Bacille Calmette-Guérin vaccine

Bp base pairs

C cytosine

CD Crohn‘s disease

CD4 cluster of differentiation 4, T-helper cells

CDC Center for Disease Control and Prevention

CET cryo-electron tomography

CF cystic fibrosis

CFU colony forming unit

CO2 carbon dioxide

COPD chronic obstructive pulmonary disease

CPC cetylpyridinium chloride

Ct threshold cycle

DMSO dimethyl sulfoxide

DNA deoxyribonucleic acid

DO dissolved oxygen

DOC dissolved organic carbon dNTP deoxynucleotide triphosphates

E. coli Escherichia coli

ELISA enzyme-linked immunosorbent assay

EPA Environmental Protection Agency

ESDA exploratory spatial data analysis

6-FAM 6-carboxyfluorescein label

FL Florida

G guanine

GIS geographic information systems

GPS global positioning system

HAART highly active antiretroviral therapy

HCl hydrochloric acid

HIPAA Health Insurance Portability and Accountability Act

HIV human immunodeficiency virus

HRCT high resolution computerized tomography

Hsp heat shock protein

ICD-9 international statistical classification of diseases and related health problems, 9th revision

IFN-γ interferon gamma

IRB institutional review board

JD Johne‘s disease

K+ potassium ion

LISA local indicator of spatial analysis

MAC Mycobacterium avium complex

MAP Mycobacterium avium subspecies paratuberculosis

Na+ Sodium ion

NaCl Sodium chloride

NaOH Sodium hydroxide

NTM nontuberculous mycobacteria

PCR polymerase chain reaction

PPD-B purified protein derivative B qPCR quantitative polymerase chain reaction (a.k.a. real-time PCR)

RNA ribonucleic acid rRNA ribosomal ribonucleic acid

SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electophoresis

SNTC Southeastern National Tuberculosis Center

Ssp. subspecies

T thymine

TB tuberculosis

TBB transbroncial biopsy

TNF-α tumor necrosis factor alpha

TE Buffer tris, EDTA Buffer

US United States

WHO World Health Organization

YO year old

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

NONTUBERCULOUS MYCOBACTERIA IN FLORIDA

Stephanie C. Yarnell

August 2011

Chair: Judith Johnson Major: Medical Sciences- Immunology and Microbiology

Nontuberculous mycobacteria (NTM) are ubiquitous in the environment, including within municipal water distribution systems. Incidence of NTM infections in humans has dramatically increased over the past 20 years, with the number of NTM cases exceeding that of tuberculosis cases within the United States. NTMs such as M. avium,

M. kansasii, M. xenopi, M. scrofulaceum, and M. marinum, have been associated with pulmonary disease, bone and soft tissue disease, as well as disseminated diseases.

Disseminated infection is primarily due to Mycobacterium avium complex (MAC), an opportunistic bacterial infection. Disseminated MAC is most commonly found in immunocompromised patients, such as those with HIV-1, organ transplant recipients, persons with kidney or liver failure, or the elderly and is associated with significant mortality in these persons. Incidence of pulmonary NTM infections is also increasing in otherwise healthy subjects with the majority of clinical isolates associated with environmental sources. The primary route of NTM infection is believed to be water through ingestion or inhalation of aerosolized water colonized with mycobacteria. This lead me to hypothesize that some environmental parameter(s) are likely affecting the species distribution and density of mycobacteria in the environment, and that changes

in these parameters may affect the number and diversity of mycobacteria in an area thereby altering the relative risk to the general population for NTM infections. In this study, a novel qPCR assay was designed for both the genus mycobacteria and MAC organisms and was used for detection of NTM prevalence and load. Field samples from both surface waters and municipal systems pre- and post-treatment demonstrated the ubiquitous presence of NTMs and MAC within these systems, although MAC loads were very low. Water quality analysis indicated various parameters may play a role in the diversity of these organisms within both surface and municipal water distribution systems.

Demographical trends of persons hospitalized with pulmonary or disseminated

NTM were also analyzed in this study. Our results indicate pulmonary NTM disease was more common in elderly, white females while disseminated NTM disease was more common in middle- aged, African American males. To analyze these trends further, home zip codes of persons hospitalized with pulmonary or disseminated NTM disease were mapped. Results of this exercise demonstrate the majority of persons with either pulmonary or disseminated NTM disease lived in urban areas. Combining this with the presence and growth of these bacteria in municipal sources, it stands to reason that municipal sources may provide a means for the distribution of these organisms to households throughout the study area, and may serve as a source of infection for high- risk patients.

CHAPTER 1 INTRODUCTION

General Background

Non-tuberculous bacteria, or NTMs are mycobaceria of public heath and clinical importance. The genus Mycobacterium contains gram-positive microaerophilic bacteria belonging to the family Mycobacteriaceae and is one of several mycolic-acid-containing genera within the order Actinomycetes.

Mycobacteria characteristically have genomic DNA with a high G-C content, and the tendency toward causing chronic infections in humans (Howard and Byrd, 2000;

Falkinham, 2003). More than 130 species have already been identified (Euzéby, 1997).

Each isolate can be categorized into one of three groups based upon clinical significance. The first grouping includes the obligate pathogens; Mycobacterium tuberculosis complex (M. africanum, M. bovis, M. canettii, M. caprae, M. microti, M. pinnipedii, and M. tuberculosis), M. leprae, and M. lepraemurium, none-of-which are generally found in the environment. The second grouping contains those, such as M. avium (MAC), which are potentially pathogenic to humans and animals and are generally derived from aquatic and terrestrial environments. The third grouping consists of low virulence species which are often saprophytic. Together, groups 2 and 3 encompass the environmentally derived mycobacteria and are often referred to as

‗atypical mycobacteria‘, ‗anonymous mycobacteria‘, ‗facultative pathogenic environmental mycobacteria‘, ‗potentially pathogenic environmental mycobacteria‘,

‗mycobacteria other than tuberculosis‘ or ‗non-tuberculous mycobacteria‘ (NTMs)

(Vaerewijck et al., 2005).

Mycobacteria can be divided into fast-growing (i.e., colony formation in less than 7 days) or slow-growing (i.e., colony formation requiring 7 days or more and including M. tuberculosis) (Primm et al., 2004). Fast-growing mycobacteria grow significantly slower than typical bacteria, as NTMs often have generation times of one day or more in rich media (Bercovier et al., 1986). The slow growth of mycobacteria is directly related to possession of only one or two 16S rRNA cistrons, impermeability of the lipid-rich cell wall, and the synthetic energy costs of the long-chain mycolic acids, waxes, and lipids

(C60-C80) (Rostogi et al., 1981; Brennan and Nikaido, 1995; Primm et al., 2004). These growth-limiting characteristics are not, however, without benefit. A single 16S rRNA cistron, for example, allows more time to adapt to a stressful environment and eases accumulation of resistance mutations to ribosomal targeting antibiotics (Primm et al.,

2004).

Cell Structure

Mycobacteria are the most hydrophobic bacteria (van Oss et al., 1975) and that serves as a major determinant of the environmental distribution (Primm et al., 2004).

Their innate hydrophobicity attracts nutrient sources and small organic compounds, allows for adherence and thereby biofilm production, allows for phagocytosis by macrophages and protozoa, and allows for easy aerosolization which can then grant pulmonary access to animals and humans (van Oss et al., 1975; Strahl et al., 2001;

Primm et al., 2004).

The impermeability of the cell wall endows mycobacteria with an innate resistance to heavy metals and many antimicrobial agents, including antibiotics and disinfectants

(Rastogi et al., 1981; Falkingham et al., 1984; Sanfranek et al., 1987; Pelletier et al.,

1988; Best et al., 1990; Taylor et al., 2000; Miyamoto et al., 2000; Cangelosi et al.,

2001). NTMs, such as M. avium and M. intracellulare, exhibit heavy metal resistance, and are able to populate habitats that metalosensitive microorganisms cannot, allowing them to attach to the metal surface and serve as biofilm pioneers (Falkingham, 2010;

Cook et al., 2010). The natural resistance to many antimicrobial compounds explains the presence of mycobacteria after standard disinfection procedures (Taylor et al.,

2000). The unique mycobacterial cell wall is also believed to play a crucial role in the infection, intracellular growth, and survival within animals and protozoa (Primm et al.,

2004).

Mycobacteria posses a unique cell envelope consisting of a cytoplasmic membrane and a cell wall composed largely of mycolic acids and long chain fatty acids

(Minnikin, 1982; Brennan and Nikaido, 1995; Barry, 2001; Rastogi et al., 2001). The mycolic acids are covalently linked to peptidoglycan by arabinogalactan forming a hydrophobic layer with other lipids (Nikaido et al., 1993; Minnikin et al., 2002). As it is typically arranged, linear galactan molecules are substituted into the peptidoglycan network with branched arabinose chains each ending in four arabinose dimers

(Minnikin, 1982; Brennan and Nikaido, 1995). These arabinose dimers, in turn, form the head for two mycolic acid molecules (Minnikin, 1982). The inner leaflet of these mycolic acids are thought to function much like a second lipid bilayer (Minnikin, 1982), and, in deed, recent studies involving cryo-electron tomography (CET) and electron microscopy of ultra-thin cryosections have confirmed the three-dimensional structure of the bilayer structure of the cell envelope in mycobacteria (Hoffman et al., 2008; Zuber et al., 2008).

This second, hydrophobic bilayer imparts impermeability to the cell envelope thereby excluding crystal violet dyes from staining the peptidoglycan layers. Thus, despite

being gram positive, mycobacteria are not gram stain positive (Barksdale and Kim,

1977). Upon heating, mycobacteria will retain acidic organic dyes, such as carbol fuchsin, and so mycobacteria exhibit acid-fast staining (Barksdale and Kim, 1977;

Minnikin, 1982).

Since the introduction of Gram‘s stain in 1884, the bacterial world has been divided into two broad groupings, gram-positive and gram-negative organisms (Gram,

1884). Gram-positive bacteria have typically been thought to have a thick cell wall composed of one layer consisting primarily of peptidoglycan (Bevridge, 2001). Gram- negative bacteria, on the other hand, have a cell wall composed of two layers, a thin peptidoglycan layer and an overlying lipoprotein bilayer known as the outer membrane with an intervening periplasmic space (Bevridge, 2001). Mycobacteria seem to defy this breakdown, as they are classified as gram-positive organisms based on their taxonomy, and yet their cell wall composed of mycolic acid-arabinogalactan-peptidoglycan polymers form a outer membrane similar to that of a gram-negative bacteria (Hoffman et al., 2008; Zuber et al., 2008). This observation means that mycobacteria have two subcellular compartments, a periplasmic space and an outer membrane, that other gram-positives do not (Hoppert, et al., 1999; Morita et al., 2005), and is of importance functionally, as it produces complications in both uptake and secretion processes

(Niederweis et al., 2010). Gram-negative bacteria address this by embedding porins and other transport proteins into their membranes to ease the transfer of nutrients from the external environment into the cell (Bevridge, 2001). Indeed, Escherichia coli, a well- studied gram-negative bacteria, employs more than 60 outer membrane proteins, most forming channels for the transport of nutrients (Molloy et al., 2000; Nikaido, 2003).

Similarly, mycobacteria have been found to contain porins (Trias et al., 1992;

Niederweis et al., 1999), and genome studies of M. tuberculosis have revealed more than 140 putative outer membrane proteins (Song et al., 2008). Thus, mycobacteria appear to have characteristics of both gram-positive and gram-negative bacteria, and like gram-negative bacteria, this outer membrane serves as a functional barrier to the uptake of exogenous compounds, including antimicrobials and disinfectants (Connell and Nikaido, 1994).

The mycobacterial cell wall is characteristic for its complex array of hydrocarbon chains perforated by porins, as well as the electron-dense peptidoglycan layer that is surrounded by a hydrophobic arabinogalactan-peptidoglycan-mycolic acid layer (Belisle et al, 1991; Inderlied et al., 1993; Wayne et al., 1993; Belisle and Brennan, 1994). The presence of fatty acids, lipids, and waxes in the cell wall are also determinants of mycobacterial hydrophobicity.

Medical

The best known pathogens in this group are M. tuberculosis and M. leprae.

Known for over a century, M. tuberculosis, the causative agent of tuberculosis (TB), is responsible for more human deaths than any other microbe (Koch, 1890; Kaufmann,

2011). Despite being curable by antibiotics in most cases, TB affects more than two billion people, with 9.3 million new infections and 1.3 million deaths each year (WHO,

2010; Kaufmann, 2011). Similarly, leprosy, caused by M. leprae, was first identified in the 19th century and is largely treatable with antibiotics, yet approximately 250,000 new cases are reported every year (WHO, 2008).

NTMs too have their share in human and animal morbidity and mortality (Primm et al., 2004). The incidence of infection with NTMs in humans is on the rise, with the

number of isolates of NTMs from infectious material exceeding those of M. tuberculosis within the United States (Sood et al., 2007). The increase in number of reported cases has been attributed to a number of factors, including increased awareness of these microbes as human pathogens, improved detection and culture techniques, increased proportion of the population that is elderly or immunocompromised, the ubiquitous presence of NTMs, increased exposure to heated water, increased selection of mycobacteria by certain human activities, and the widespread use of disinfectants

(Sood et al., 2007). Numerous studies have demonstrated the clinical significance of these NTMs in both immunocompromised and healthy patients (Vaerewijck et al.,

2005). Members of the Mycobacterium avium complex (MAC), (comprising M. avium and M. intracellulare), are responsible for the majority of NTM infections in developed countries with the main presentations being: lymphadenitis in children, respiratory infections in the elderly, and intestinal and disseminated disease in the HIV-infected population (Horsburgh, 1996; Vaerewijck et al., 2005). NTMs also cause nosocomial outbreaks and mini-epidemics, hypersensitivity pneumonitis, arthritis, keratitis, tenosynovitis, otitis media, corneal infections, endocarditis, immunologic dysfunction, and chronic disease states (Moore, 1993; Howard and Byrd, 2000; Vaerewijck et al.,

2005). The clinical scenario is complicated, because NTM infections often present with vague symptoms, such as fever, anorexia, weight loss, and night sweats. The vagueness of symptoms often delays diagnosis, which is problematic because NTM infections are a primary cause of death among immunocompromised persons (Howard and Byrd, 2000).

Diagnosis is further complicated by the ability of these organisms to colonize individuals. NTMs can be isolated from saliva, sputum, urine, and gastric washings even when no disease is present. This appears to be particularly true for respiratory isolates, as most persons from whom environmental mycobacteria were recovered from the airways of did not appear to be harmed (Howard and Byrd, 2000). Mycobacteria can also be found in the stool of healthy persons (Portaels et al., 1988). Further evidence for the transient, regular exposure of humans to NTMs is the finding that over

60% of men from the Southeastern coastal region of the United States tested positive on PPD-B, but showed no signs of disease. Thus, it is believed that they were exposed and produced an immune response to mycobacterial antigens, but never developed outright disease (Edwards et al, 1969; Reed et al., 2006). Further, it has been demonstrated that United States medical personal have a high rate of environmental mycobacterial skin test reactivity with no history of disease (von Reyn et al., 2001). The regular rates of exposure may be due largely to the ubiquitous nature of NTMs; humans are constantly and continuously exposed to environmental mycobacteria at low levels, but overt disease generally does not occur without one or more underlying medical conditions (Primm et al., 2004). Despite their growing importance, little is known about how these infections are spread.

Respiratory Infections

M. tuberculosis is well known as a respiratory pathogen and some NTMs are also capable of causing tuberculous-like lesions (Vaerewijck et al., 2005). Indeed, the second most common presentation of NTM infection is pulmonary manifestations and the incidence around the globe is increasing (Moore et al., 2010; Lai et al., 2010;

Winthrop et al., 2010). The majority of reported pulmonary cases of NTM infection within

the United States were from Southern coastal states, such as Florida (Falkingham,

1996; Bilinger et al., 2009), which correlates with the increased rates of skin test reactions seen in this region (von Reyn et al., 2001). In addition to granulomas and cavitary disease similar to that seen in TB, pulmonary disease from NTMs can take on a variety of clinico-pathologic presentations, including asymptomatic-indolent disease, interstitial disease, or hypersensitivity pneumonitis-like granulomatous lung disease.

Lesions occur predominantly in the upper lobes of the lung especially in sites of pre- existing mycobacterial or fungal disease or bronchiectasis (Howard and Byrd, 2000;

Sood et al., 2007). Although the clinical presentation of pulmonary NTM resembles that of TB, this disease is very different in that the host acquires the infection from environmental sources such as water, which may explain the predilection for warm coastal regions (Pedley et al., 2004). Indeed, NTMs are ubiquitous in the environment and exposure so common, that a finding of NTMs in respiratory secretions requires supportive clinical symptoms prior to beginning treatment (Griffith et al., 2007). These cases have caused many to wonder if colonization independent of disease occurs or if these findings simply demark subclinical disease likely to progress (Wang et al., 2009).

Regardless of location, pulmonary NTM infections can be quite severe, carrying a mortality rate of 14% (Howard and Byrd, 2000).

The classic presentation for NTM-induced pulmonary infection is cavitary lung disease in white males above the age of 60, who have some form of predisposing lung disease and a history of smoking and alcohol abuse (Wolinsky, 1979). These predisposing conditions include pneumoconiosis, chronic obstructive pulmonary disease, cavitary lesions from previous TB or bronchiectasis, smoking, black lung,

occupations where large amounts of dust and particulates are inhaled, cystic fibrosis, radiation therapy to the lung, or malignancies of the lung (Falkinham, 2003). Many of these conditions result in scarring and fibrosis of the lung, which is believed to allow subsequent colonization by NTMs (McGrath et al., 2010). The classic presenting signs are cough with sputum production, but those may be accompanied by shortness of breath, dyspnea, hemoptysis, fatigue, lassitude, weight loss, fever, or night sweats.

Radiographically, these cases typically manifest with pulmonary cavities, pleural thickening, nodular infiltrates, consolidation, and various forms of bronchiectasis (Griffith et al., 2007; Arend et al., 2009). Historically, these cases have been mistaken for TB, but the patient is not infectious to relatives or clinical staff. This does not diminish the seriousness of these infections, as they can often progress to respiratory failure and death (Field and Cowie, 2006). Delays in proper diagnosis often occur because of difficulty differentiating symptoms due to underlying lung disease, such as sputum production in bronchiectasis or dyspnea in emphysema, and symptoms derived from the superimposed mycobacterial infection (Griffith et al., 2007). Further complicating the care of these patients, respiratory secretions often harbor an array of organisms including mixed mycobacterial species, fungi, and other pathogenic bacteria making appropriate antibiotic selection more difficult (Pedley et al., 2004; Arend et al., 2009).

Once identified, NTM-induced pulmonary disease can be quite challenging to treat and may require surgical removal of the diseased tissue (Howard and Byrd, 2000).

The epidemiology of NTM pulmonary disease appears to be shifting toward non- smoking older females with no underlying lung disease. There is an increasing number of sub-acute pulmonary cases presenting with respiratory complaints, dyspnea and

cough being most common, occurring in this population with a lack of occupational exposure to dusts or other known predisposing conditions (Falkinham, 2003; Sood et al., 2007; Cassidy et al., 2009). This trend first described in 1989 when Prince et al. observed that 81% of the patients were female, 86% were white, and the mean age was

66 years (Prince et al., 1989), has been noted multiple times and represents a shift from male-dominated secondary infection to a primary infection in these patients (von Reyn et al., 2001; Chalermskulrat et al., 2002; Cassidy et al., 2009). Infected persons are often thin, and correlations have been drawn to particular body phenotypes including narrowing of the anterior-posterior diameter, pectus excavatum, and mitral valve prolapse (Iseman et al., 1991). These patients typically show no cavitary lung disease, but instead show fibronodular bronchiectasis (Field and Cowie, 2006). In reported cases of this nature, pulmonary function tests showed a restrictive lung disease, chest radiographs showed diffuse interstitial or nodular opacities, high resolution computerized tomography (HRCT) showed diffuse ground glass opacities and centrilobular ill-defined micronodules in both lungs, bronchoalvelolar lavage fluid (BALF) showed modest lymphocytic predominance with a high CD4/CD8 ratio, culture yielded primarily M. avium, M. kansasii, or M. abscessus, and histopathologic findings of transbrochial biopsy (TBB) showed well-formed, centrilobular and bronchiolocentric non-necrotizing granulomas (Aksamit, 2002; Chalermskulrat et al., 2002; Sood et al.,

2007). The increase in fibronodular bronchiectasis is thought to be, in part, due to fewer cases of TB, hence fewer pre-existing cavitations for the establishment of NTM disease, as well as the increase in elderly persons within the United States (Field and Cowie,

2006). This disease should not be taken lightly, however, as it can be fatal (Prince et

al., 1989). Left untreated the nodular bronchiectactic disease can cavitate causing severe, often permanent, lung damage (Griffith et al., 2007). Treatment can be difficult and relapse is common when therapy is stopped (Saleeb and Olivier, 2010). Given the general increase in incidence and susceptible patients, as the population continues to age, the prevalence of fibronodular bronchiectasis can be expected to continue to increase (Field and Cowie, 2006).

Hypersensitivity Pneumonitis

NTMs can infect recreational water supplies creating a risk for hypersensitivity pneumonitis. For decades it has been recognized that persons exposed to long-term aerosolization from swimming pools (i.e. lifeguards) were at risk for granulomatous pneumonitis (Havelaar et al., 1985; Emde et al., 1992; Rose et al., 1998). This form of

NTM disease often affects healthy, immunocompetent people with aerosol exposure to mycobacteria such as lifeguards and people using aquatic sources for recreation or work (Havelaar et al., 1985; Emde et al., 1992; Rose et al., 1998; Embil et al., 1997;

Mangione et al., 2001; Mery and Horan, 2002; Kang et al., 2010). While M. avium- complex is the most commonly reported NTM in hot tub lung, it should be noted that cases due to M. fortuitum have also been reported (Thompson and Yew, 2009). Hot tubs, indoor swimming pools, and spas, in particular, are linked to human infection

(Embil et al., 1997; Kahana et al., 1997; Khoor et al., 2001; Kang et al., 2010).

Mycobacteria are not only capable of surviving and reproducing in swimming pools and hot tubs, but may even comprise a significant fraction of the microbes in these locations

(Angenent et al., 2005). Aerosolization is believed to be the major route of infection from these sources, as NTMs have been found in higher quantities in the air above pools than in the pool water itself (Angenent et al., 2005).

It is unclear at the present time whether this is due to a hypersensitivity to M. avium complex, is associated with an actual infection, or is an immune-mediated illness

(Vaerewijck et al., 2005). The pathology is believed to be due to pulmonary macrophages processing and presenting the NTM antigens to T lymphocytes, inducing a clonal expansion and proliferation, which in turn induces granuloma formation within the lung (Sood et al., 2007). This extensive granulomatous reaction indicates that the host‘s immunologic response may be a crucial component contributing to the morbidity in this disease (Howard and Byrd, 2000). Subsequent symptoms include cough, dyspnea, fatigue, impaired exercise tolerance, and sputum production; and chest radiograph and computed tomography scans indicate hazy or ground glass opacities and small peripheral nodules (Rickman et al., 2002). Acid-fast smears of patient sputum are insensitive and should not be used for diagnosis (Khoor et al., 2001). A similar hypersensitivity reaction has been noted in persons exposed to metal working fluid (Hodgson et al., 2001; Wallace et al., 2002). In either case, avoidance of re- exposure is key to treatment (Field and Cowie, 2006).

Cystic Fibrosis (CF)

It has long been known that the lungs of patients suffering from cystic fibrosis are colonized with bacteria. It is believed that the viscous respiratory secretions, characteristic of CF, contribute to recurring pulmonary infections (Gilligan, 1991).

Mycobacteria were first isolated from patients suffering from CF in 1980, but the severity of disease has only been recognized in more recent years (Boxerbaum, 1980; LiPuma,

2010). The prevalence of NTMs recovered from the sputum of CF patients was previously reported to be 13% (Olivier et al., 2002). However, determining the exact burden of disease in this population has proven difficult as overgrowth of sputum

samples with other bacteria may lead to an underestimate of the infection rate of these patients with NTMs (Bange and Bottger, 2002).

Better treatment of CF has lead to increased longevity largely thanks to better medications, inhaled formulations, improved nutrition, and improved sputum clearance techniques (Simmonds et al., 2010). Ironically, however, studies have indicated that older CF patients (>10 years of age) were more likely to have NTM infections (Field and

Cowie, 2006). Thus, there appears to be a correlation between age and risk for NTM disease, and is therefore reasonable to believe that more cases of NTM lung disease in

CF patients will be reported over the coming decades as patients survive longer. The increased age of CF patients has also led to an increase in other pulmonary pathogens in these patients, including drug-resistant Staphylococcus aureus (Razvi et al., 2009).

Interestingly, studies have found CF patients were more likely to have NTM infections if they were co-infected with Staphylococcus aureus (Field and Cowie, 2006). This co- infected state makes it more difficult to determine the true extent of the damage derived from either disease state, so the contribution of NTMs to respiratory decline in CF patients is still an area of debate. Studies have suggested that NTMs produced no effect on clinical lung function tests in CF patients, but HRCT scans of the chest suggest NTM presence is predictive of disease progression (Griffith, 2003).

Additionally, CF patients whose sputum grows NTMs were more likely to show a progression of their pulmonary parenchymal disease (Field and Cowie, 2006).

Therefore, it is recommended that any CF patient with repeated, persistent, infectious symptoms despite traditional antibiotics should be evaluated for a mycobacterial disease (Vaerewijck et al., 2005).

Disseminated Disease

Having an immune deficiency carries the greatest likelihood for contracting disseminated disease due to NTMs or TB (Howard and Byrd, 2000). Disseminated NTM infection has been identified in persons with leukemia (especially hairy cell), lymphoma, collagen vascular disease, bone marrow and solid organ transplants, familial immunodeficiences, chronic corticosteroid use, and with advanced AIDS (Weinstein et al., 1981; Zakowski et al., 1982; Horsburgh et al., 1985; Bennett et al., 1986; Newport et al., 1996; Roy and Weisdorf, 1997; Barcat et al., 1998; Nagy and Rubin, 2001; Griffith et al., 2007; Eneh et al., 2010). Of the immunocompromised states, AIDS carries the worst prognosis in the event of a disseminated disease due to an NTM, and occurs most commonly when CD4 counts drop below 75-100/mm3 (Nightingale et al., 1992;

Falkinham, 2003). The most common presenting symptoms are quite vague and include fever, anorexia, malaise, weight loss, and night sweats. Abdominal pain and diarrhea are not uncommon (Horsburgh, 1991; Benson and Ellner, 1993; Griffith et al.,

2007). Commonly there is also widespread involvement of the reticuloendothelial system with accompanying hepatosplenomegaly, lymphadenopathy, and severe anemia, frequently requiring transfusions (Havlik et al., 1992). The organism burden is often quite high in these individuals and is consistent with persistent bacteremia (Wong et al., 1985; Torriani et al., 1996). Pathology studies indicate involvement of the liver, spleen, bone marrow, and gastrointestinal tract often to levels as high as 107 to 1010

CFU/g, as well as a possible association with nephrocalcinosis of the kidneys and peritoneal inflammation (Falkoff et al., 1987; van der Reijden et al., 1989; Perazella et al., 1993). Diagnosis is typically confirmed through culture of blood or biopsied tissues and staining of tissues for acid-fast organisms within foamy macrophages (Wong et al.,

1985; Torriani et al., 1996; Eneh et al., 2010). Ironically, pulmonary symptoms are not believed to occur in these patients (Field and Cowie, 2006).

The risk of disseminated NTM in AIDS patients is high in developed countries and low or absent in developing nations (von Reyn et al., 1996; Gopinath, 2010). NTMs are isolated from the environment in developing nations and healthy Africans have been shown to have significant reactivity to MAC antigens implying exposure does occur in

Africa and other developing nations (von Reyn et al., 1993). Despite this, disseminated

NTM disease is relatively uncommon in developing countries (von Reyn et al., 1996;

Gopinath, 2010). The low rate of disseminated NTM has been largely attributed to the death from TB of people with AIDS before the CD4 count drops low enough to allow for disseminated NTM infection or from a potential mycobacterial immunity conferred by cross-protection from prior TB or BCG vaccinations (von Reyn et al., 2002; Field and

Cowie, 2006). The high prevalence of TB may also complicate diagnosis of NTM infection, especially in resource limited countries. In the terminal stages of AIDS when the CD4 count falls below 100, nontuberculous mycobacteria induced disease begins to manifest, and NTMs can be isolated from blood, tissue, sputum, and fecal samples

(Falkinham, 2003; Dos Santos et al., 2010). M. avium is isolated most frequently, specifically serotypes 4 and 8 within the United States (Hoffner et al., 1990; Tsang et al., 1992; Falkingham, 2003). Complicating the issue, as many as 25% of persons infected with disseminated MAC within the AIDS population harbor two or more strains simultaneously (von Reyn et al., 1995). This raises the question of whether infection stems from exposure to mixed populations within a single source or from multiple exposures to single bacterial populations. An additional form of MAC has been

recognized within HIV infected individuals harboring previously unrecognized or subclinical MAC infection. In these persons, a local inflammatory syndrome involving culture-positive lymphadenitis or other granulomatous disease with draining sinuses and negative blood cultures can occur within two months of beginning antiretroviral therapy

(HAART), and so is termed immune reconstitution syndrome (DeSimone et al., 2000;

Aramaki et al., 2010).

Disseminated MAC cases have been decreasing in recent years. NTM disease is not reportable within the United States, so true infection rates must be passively deduced (Gopinath, 2010). Despite this, trends suggest that cases of disseminated

MAC were increasing and outnumbered the annual cases of TB prior to the introduction of effective prophylaxis against disseminated NTM in 1993 (Horsburgh et al., 2001).

The peak incidence is believed to have occurred in 1994 when an estimated 37,000 persons were infected with disseminated MAC (Horsburgh et al., 2001). Indeed, disseminated MAC infection used to be one of the most common bacterial opportunistic infection in adults infected with HIV-1 in the developed world (Karakousis, et al., 2004).

At that time, NTM disease in AIDS was second only to AIDS Wasting Syndrome as the most common cause of death in these patients with MAC being an independent predictor of mortality, even after adjustment for CD4 lymphocyte counts (Covert et al.,

1999; Karakousis, et al., 2004). During this period, the burden within the American HIV population for NTM infection was as high as 50%, but HAART has reduced this rate in recent years (Horsburgh, 1991; Aberg et al., 1998; Horsburgh et al., 2001; CDC, 2002).

Cases of disseminated NTM are now considered less common and thought to occur primarily in patients with advanced AIDS not under medical care or in persons non-

compliant with HAART or MAC prophylaxis (Kovacs and Masur, 2000). Prior to these measures, disseminated MAC infection was associated with a four to five month reduction in survival (Horsburgh et al., 1991; Chin et al., 1994). Prophylactic treatment and treatment of disseminated MAC includes administration of clarithromycin or azithromycin and ethambutol, as well as HAART therapy to reconstitute the immune system (Havlir et al., 1996; Aberg et al., 1998; Benson et al., 2000; CDC, 2002; Griffith et al., 2007). Treatments must be continued indefinitely in patients with advanced

AIDS, but may be safely discontinued after 12 months in those persons achieving immune reconstitution, as defined by an increase in CD4 counts to >100/mm3 for at least six months (Aberg et al., 1998; CDC, 2002; Griffith et al., 2007). Prophylaxis is important as, despite aggressive antibiotic therapy, disseminated NTM disease has reported to have only a 10% survival rate (Howard and Byrd, 2000).

Though the dominant source of infection is unknown in many cases, studies have implicated water from hospitals, patient homes, and recirculating hot water systems as sources of infection, with several identifying the identical isolates from water and patient samples (von Reyn et al., 1994; Aronson et al., 1999; von Reyn et al., 2002). Thus, it is likely that water may be a potential source of infection in these persons. It is recommended that any water given immunocompromised patients be filtered, and that showering be avoided due to risk of aerosolization of these organisms (Vaerewijck et al., 2005).

Immunology

While only a very small percentage of human-mycobacterial interactions progress to outright infection, it has been noted that this progression is far more common in the immunocompromised, especially AIDS patients (Arasteh et al., 2000). NTMs primarily

enter the body through the bronchial or intestinal mucosa after either inhalation or ingestion. Once inside, the NTMs are taken up into host macrophages via compliment or C3-mediated phagocytosis (Swarz et al., 1988; Schlesinger et al., 1990; Schlesinger and Horwitz, 1991; Bermudez et al., 1991). While most bacteria are killed within macrophages, NTMs inhibit phago-lysosomal fusion, thereby preventing the oxidative burst (Frehel et al., 1986; Frehel et al., 1991; Crowle et al., 1991; de Chastellier et al.,

1993; Sturgill-Koszycki et al., 1994). Avoiding normal lytic functions of the macrophage,

NTMs are able to reproduce within the phagosomal compartment (Frehel et al., 1991).

Eventually the macrophage lyses releasing the bacteria. Whether they are taken up by non-activated or activated macrophages determines whether the bacteria is killed or undergoes a continued cycle of reproduction (Crowle et al., 1991; Sturgill-Koszycki et al., 1994). Ultimately, a cell-mediated immune response is needed for infection control.

Severely immunocompromised persons, such as people who have AIDS, cannot mount an effective immune response, so the infection remains uncontrolled (Nightingale et al.,

1992; Falkingham, 2003). In the majority of persons, the immune system is able to clear these bacteria, but the subsequent release of potent immunomodulators derived from the NTM‘s cell wall, such as trehalose dimycolate, may have downstream effects, such as allergies, pulmonary viral infections, and irritations of the bowel (Black, 2001; Primm et al., 2004). Mycobacterial infections can also disturb the immune system.

Hypersensitivity pneumonitis induced by mycobacterial infections has been shown to cause a strong deregulation of pulmonary immunity, particularly by affecting either TNF-

α, IL-12, or IFN-γ, thereby predisposing persons to constant assault by aerial viruses

(Howard and Byrd, 2000; Kitaura et al., 2001; Field and Cowie, 2006). Increased

susceptibility to NTMs infection has been associated with mutations in five genes:

IFNGR1, IFNGR2, IL12RB1, IL12B, and STAT1 (Vaerewijck et al., 2005). Other reported risk factors of increased susceptibility include: rural residence, living in the

Southeastern United States, Black race, birth outside of the United States, history of bronchoscopy, swimming in an indoor pool, 6 or more years of cumulative occupational exposure to soil, consumption of raw or partially cooked fish or shellfish, treatment with granulocyte colony-stimulating factor, and owning or working with an aquarium (von

Reyn et al., 1996; Vaerewijck et al., 2005; Reed et al., 2006).

Soft tissue/ Cutaneous Manifestations

NTMs are a known cause of soft tissue diseases including tenosynovitis, panniculitis, fasciitis, synovitis, osteomyelitis, septic arthritis, meningitis, and sarcoidosis

(Jacob et al., 1993; Hellinger et al., 1995; Li et al., 1999; Frosch et al., 2000; Weitzul et al., 2000; Arend et al., 2001; Cheung et al., 2010). The most common manifestation within the United States of NTM infection is granulomatous skin lesions often called ‗fish fancier‘s finger,‘ ‗fish tank granuloma,‘ ‗mariner‘s TB‘ or ‗swimming pool granuloma‘

(Philpott et al., 1963; Wolinsky, 1979; Littlejohn and Dixon, 1984; Vincenzi et al., 1992;

Dobos et al., 1999; Cheung et al., 2010). These lesions occur when environmental mycobacteria gain entry into human skin via scrapes, punctures, or other forms of trauma. Once inside a human host, the NTMs cause a localized infection manifesting as papules, nodules, plaques, ulcers, or panniculitis-like lesions (Streit et al., 2008).

These skin lesions occur most commonly on the extremities including elbows, hands, fingers, knees, feet, and toes (Collins et al., 1985). Biopsy and drainage material is usually negative for acid-fast bacteria and usually heal spontaneously, though patients often remain positive on tuberculin skin tests long after the lesion has healed (Judson

and Feldman, 1974). In rare cases, satellite lesions occur that can ascend up the extremity via lymphatic spread (Wolinksky et al., 1972; Cheung et al., 2010). In the

United States and Europe, these cases are usually caused by M. marinum through contact with infected water or infected animals, especially fish (Zeligman, 1972; Aubry et al., 2002; Cheung et al., 2010). M. avium-complex, M. kansasii, M. terrae, M. malmoense, M. xenopi, M. abscessus, M. chelonae, M. smegmatis, and M. fortuitum- complex are examples of other NTM species known to cause cutaneous infections, and have similarly been associated with exposure to water including hot tubs, public baths, and footbaths (Aubuchon et al., 1986; Zenone et al., 1999; Lee et al., 2000; Winthrop et al., 2002; Streit et al., 2008). These cases occur in both fresh and salt water and certain activities, such as aquatic farming and owning an aquarium, carry a greater risk of infection (Huminar et al., 1986). Compared with historical rates, it appears that the incidence of such diseases is increasing (Dobos et al., 1999). Worldwide, buruli ulcer caused by M. ulcerans is the most prevalent, but occurs exclusively in tropical locations

(George et al., 1999).

Buruli Ulcer

One of the most common forms of NTM is the buruli ulcer caused by M. ulcerans.

Located in equatorial regions around the world with a particularly high prevalence in

West and Central Africa, buruli ulcers are a debilitating disease characterized by large necrotic skin ulcers involving massive necrosis of the subcutaneous fat that are often incurable (George et al., 1999). Death is an unusual result of buruli ulcers, but permanent deformities are common (Buntine et al., 2002). Treatment often involves a combination of surgery and multiple antimycobacterial drugs, placing a significant financial burden on infected persons (Drummond, 1998; Asiedu, 1998). Buruli ulcer

most often infects healthy individuals with no known immunological dysfunction, and is usually acquired from environmental sources, particularly water and soil (George et al.,

1999). The incidence of buruli ulcers is increasing in certain parts of the world with rates in some African villages reaching levels greater than 22% with reported annual incidences rates higher than that of TB (Amofah et al., 1993; Marston et al., 1995;

Ragunathan et al., 2001). Thus, buruli ulcers are an important NTM disease as the incidence is increasing, it is expensive to treat, and is most common in regions lacking advanced medical facilities (Asiedu, 1998).

M. ulcerans likely evolved from M. marinum. M. ulcerans is closely related to M. marinum, the primary cause of fish tank granuloma, showing >99.8% similarity in their full 16S rRNA sequences and similarly high levels of sequence conservation between other genes (Stinear et al., 2000). Despite their genetic similarities, they are very distinct phenotypically, including slower growth in M. ulcerans, differing host specificity, differing temperature requirements, and differing immunological responses to the two organisms. In M. marinum, as with most NTMs, the bacterium replicates within the host macrophages and provokes the formation of a granuloma, whereas in M. ulcerans histopathology shows a marked absence of a host inflammatory immune response and massive numbers of bacilli are found extracellularly (WHO, 2001). This unusual pathology has been attributed to a peptide, mycolactone, produced by M. ulcerans, but not M. marinum (George et al., 1999). This mycolactone causes the characteristic necrosis of the skin and also has immunosuppressive properties (George et al., 1999).

M. ulcerans was recently found to contain a large plasmid having all the genes necessary for mycolactone synthesis (Stinear et al., 2004). This has led to the belief

that M. ulcerans may be a clone of M. marinum that acquired a plasmid from another organism (Yip et al., 2007). Though the main role of mycolactone in M. ulcerans has yet to be determined, it is thought that the molecule may confer fitness for the survival within the salivary glands of aquatic insects (Masollier et al., 2002). If M. ulcerans did truly evolve from M. marinum, this would be an excellent example of a significant human pathogen developing from environmental forebears.

Traumatic Wound/ Surgery

Contaminated water may lead to mycobacterial skin infections and nosocomial infections. Despite being a recognized problem for decades, the number of reports of pathogenic mycobacterial disease from the use of contaminated devices or invasive procedures has been increasing in recent years (De Groote et al., 2002; Cortesial et al.,

2010). This is unsurprising given that some of the highest rates of contamination of drinking water have been reported in hospitals, hemodialysis units, anesthesia units, and dental offices (Carson et al., 1988; Schulze-Robbecke et al., 1995; Fujita et al.,

2002; Vaerewijck et al., 2005; Langevin et al., 2008). Indeed, the linkage between contaminated hospital waters and mycobacterial infections has been recognized for many years, and it is currently believed that the growth and persistence of these organisms in municipal and hospital water supplies combined with their inherent resistance to disinfectants provides a likely explanation for these cases (Graham et al.,

1988; du Moulin et al., 1988; Peters et al., 1995; Wallace et al., 1998; El Sahly et al.,

2002; Fujita et al., 2002; Lombardi et al., 2002; Falkingham, 2002). Important reservoirs for these bacteria within the hospital setting include: potable water, bedside drinking- water carafes, water fountains, sinks, faucets aerators, showers, immersion tubs, toilets, dialysis water, anesthesia breathing circuits, ice and ice machines, water baths, flower

vases, eyewash stations, dental unit stations, biofilms on equipment, and contaminated solutions and disinfectants (Carson et al., 1978; Safranek et al., 1987; Vaerewijck et al.,

2005; Langevin et al., 2008; Cortesial et al., 2010). Despite this known risk, epidemiological tracking has proven problematic. There are 3 main factors that complicate investigation of waterborne transmission of NTMs. First, though some outbreaks do occur, infections are typically sporadic. Secondly, there are many potential sources of exposure other than water. Thirdly, the typing schemes routinely used are not sufficiently discriminating to confidently identify whether an environmental isolate is the same as one derived from a patient (Pedley et al., 2004).

Most of these cases can be traced back to contaminated devices and solutions or water used for washing instruments. M. abscessus is one of the most commonly reported causes of post-surgical mycobacterial infection, especially after plastic surgery, but M. fortuitum, MAC, M. gordonae, and M. xenopi have been implicated in skin infections following iatrogenic procedures such as liposuction, facial blepharoplasty, augmentation mammmoplasty, and other cosmetic procedures (Wolinksky, 1992;

Wallace et al., 1998; Murillo et al., 2000; Phillips and von Reyn, 2001; Vaerewijck et al.,

2005). As with most mycobacterial infections, the presentation is often indolent with long incubation times, thus, the original exposure may have been long before the clinical presentation, which complicates diagnosis and epidemiological studies (Howard and Byrd, 2000; Griffith et al., 2007). Clinically, most cases present with localized abscesses often accompanied by purulent drainage, fistula formation, fever, and cellulitis (Wallace et al., 1998; Murillo et al., 2000). Once suspected, these organisms are typically readily culturable, but can be very difficult to eradicate (Wallace et al.,

1998; Phillips and von Reyn, 2001). In addition to considering the treatment of individual patients in these situations, care must be taken to determine if these cases represent real or pseudo-outbreaks from a shared contaminated device or solution

(Phillips and von Reyn, 2001). Examples of contaminated equipment include endoscopes, bronchoscopes, injection needles, artificial breasts, vascular devices, pacemakers, and orthopedic devices (Robicsek et al., 1988; CDC, 1991; Gubler et al.,

1992; Sniadack et al., 1993; Maloney et al., 1994; Griffiths et al., 1997; Wallace et al.,

1998; Verghese et al., 1998; Galil et al., 1999; Katz et al., 2000; Murillo et al., 2000;

Astagneau et al., 2001; Kressel and Kidd, 2001). The contamination of these devices is often attributed to inadequate cleaning and disinfecting or to the water used in the process, and it is not uncommon for equipment such as automated bronchoscope disinfecting machines, to become heavily contaminated (Sniadack et al., 1993; Wallace et al., 1998). Due to the nature of these outbreaks, clinicians, nurses, laboratory workers, and infection control and public health professionals often play important roles in the evaluation of individual cases and the determination of cause.

Chronic Bowel Disease

M. avium subsp. paratuberculosis (MAP) has been suggested as a cause of

Crohn's disease (CD) in humans, and has been a subject of much investigation over the last few decades. MAP was first identified at the Veterinary Pathology Institute in

Dresden in 1895 as the causative organism in a case of chronic inflammation of the intestinal tract of a cow (Johne and Frothingham, 1895). This disease became known as

Johne‘s disease (JD) and is characterized by steadily declining condition, weight loss, decreased milk yield, and diarrhea (Doyle, 1956; Buergelt et al., 1978; Riemann and

Abbas, 1983; Chiodini et al., 1984b; Cocito et al., 1994; Clarke, 1997; Beth Harris and

Barletta, 2001; Manning and Collins, 2001). Though originally identified in cows, JD can affect ruminants, deer, bison, elk, dogs, pigs, horses, chickens, rabbits, birds, foxes, insects, worms, and primates and is invariably fatal (Greig et al., 1999; Buergelt et al.,

2000; Ferroglio et al., 2000; Nebbia et al., 2000; Pavlik et al., 2000; Beard et al., 2001;

Fischer et al., 2001; Fischer et al., 2003). Internally, MAP causes chronic inflammation of the intestinal tract, specifically, the terminal ileum and colon with segmental lesions and rectal involvement. The gut wall becomes thickened, mucosal surfaces swell producing ulcers, and the regional mesenteric lymph nodes become enlarged (Buergelt et al., 1978; Cocito et al., 1994; Manning and Collins, 2001). In vivo, MAP falls into two categories: a pluribacillary form with abundant acid-fast bacilli within intestinal macrophages (similar to lepromatous leprosy) and a paucimicrobial form with no visible acid-fast bacteria but with chronic granulomatous inflammation (similar to tuberculoid leprosy) (Hermon-Taylor, 1998). This paucimicrobial form of JD closely resembles CD in humans leading many to speculate MAP as the cause of CD (Hermon-Taylor, 1998;

Timms et al., 2011).

Given the similarity in disease state between JD of animals and CD in humans,

MAP has been proposed as a cause of Crohn‘s disease in humans (Harris and

Lammerding, 2001). There has been some microbiological and clinical data suggesting

M. avium subsp. paratuberculosis (MAP) as the cause of CD (Chiodini et al., 1984a;

Chiodini et al., 1986; Chiodini et al., 1989; Hermon-Taylor et al., 1998).

The main characteristics of chronic inflammatory bowel syndrome seem to fit well with the notion of mycobacterial involvement. Oral infection of M. avium into immunocompromised mice results in a strong inflammatory response and necrosis of

the intestinal mucosa, similar to chronic bowel disease, and M. genavense infections in

HIV patients often result in abdominal wall thickening, lymphadenopathy, and ulceration, all of which are symptoms of chronic inflammatory bowel disease (Kim et al., 1998;

Monill et al., 2001).

Proving these linkages, however, has proven to be more difficult than expected.

For starters, CD appears to have a genetic component and so causation is likely to be multifactorial (Orholm et al., 1991; Yang et al., 1993; Peeters et al., 1996; Koets et al.,

2000; Orholm et al., 2000; Brest et al., 2011). Secondly, MAP grows slowly even for mycobacteria, requiring at least 16 weeks in culture before visible colonies can be noted

(Zhang and Zhang, 2011). Further, MAP requires the supplementation of exogenous mycobactin, an iron-transport protein, for in vitro growth (Merkal and McCullough, 1982;

Zhang and Zhang, 2011). Given the slow growth and unique media requirements, it is easy to see how MAP could fail to be grown in many cases. Confounding the issue even further, the culture conditions used can dramatically change phenotype and resistance of MAP isolates (Sung and Collins, 2003). MAP, for example, can switch between the traditional acid-fast positive cells to a tough, acid-fast negative form that is invisible by ordinary light microscopy (Van Kruningen et al., 1986; Hermon-Taylor, 1998;

Thomas et al., 1998; Pedley et al., 2004). MAP is also capable of switching between activated cells and latent cells, which differ in regards to resistance to lysis and immunological interactions (Hermon-Taylor et al., 2000; Naser et al., 2000; Odumeru et al., 2001). For all these reasons, MAP is a difficult organism to isolate, and many MAP strains cannot be grown at all (Scheibl and Gerlach, 1997; Pillai et al., 2001).

Therefore, conventional laboratory culture is not a consistently reliable method of

detection for MAP, which is problematic, as fecal culture remains the gold standard for detection despite the development of new ELISA and IFN-γ based diagnostic tests

(Eamens et al., 2000; Kalis et al., 2000; Whittington et al., 2000).

People may be exposed to MAP by food and contaminated water. MAP has been isolated in high numbers from pasteurized cow‘s milk and is excreted in large numbers in the feces of cattle and other animals, and these are pathways associated with high risk of transmission to humans (Doyle, 1954; Streeter et al., 1995; Millar et al., 1996;

Hermon-Taylor et al., 2000; Gao et al., 2002; Corti and Stephan, 2002; Moghadam et al., 2010). MAP organisms introduced through contamination from runoff of heavily grazed pastures are believed to exist in both planktonic form and within protozoa. This growth within protozoal species such as, Acanthamoeba castellanii, is likely the explanation for the ability of MAP organisms to survive in the environment for months to years (Barker and Brown, 1994; Falkingham, 1996; Ford, 1999; Hermon-Taylor et al.,

2000). Also, MAP organisms grown in protozoa have an increased capacity to infect human colonic cells, as well as an enhanced virulence in mice (Cirillo et al., 1997).

Thus, cycling through unicellular organisms may increase the virulence of these organisms once in the environment, and any MAP organisms surviving removal of suspended solids in water treatment are not likely to be killed by chlorination (Taylor et al., 2000; Whan et al., 2001). Indeed, conspicuous clusters of CD have been noted in areas with exposure to waters whose watersheds included heavily grazed pastures with significant fecal runoff (Mayberry and Hitchens, 1978; Allan et al., 1986; Hermon-Taylor,

1993; Van Kruiningen and Freda, 2001; Pierce, 2009). Given the wide range of domestic and wild-animal reservoirs for MAP and the ability of these organisms to

survive for long periods in the environment, it is easy to see how the excretion of such high numbers could translate into an infection risk for humans (Pierce, 2009).

Interestingly, geographical regions with acidic soils rich in humic and fulvic acids, such as those in Florida, have been suggested to favor accumulation of MAP in the environment (Kopecky, 1977; Kazda et al., 1990; Johnson-Ifearulundu and Kaneene,

1997; Iivanainen et al., 1999; Kirschner et al., 1999; Reviriego et al., 2000).

Cervical Lymphadenitis

NTMs are a major cause of cervical lymphadenitis in children. In developing nations, M. tuberculosis is the most common cause of sub-acute and chronic lymphadenitis. In North America, NTMs are the more common cause, with a cervical presentation being most common, followed by inguinal and axillary adenitis (Wallace et al., 1990; Inderlied, 1993; Chesney, 2002; Griffith et al., 2007). The linkage between

NTMs and cervical lymphadenitis has been known since 1957 (Prissick et al., 1957).

Since that time, many mycobacterial species have been implicated with M. avium- intracellulare, M. scrofulaceum, and M. kansasii most commonly reported; in recent years, however, other mycobacterial species have emerged, including M. malmoense

(McCrossin and Mailman, 2006). Cervical lymphadenitis is purported to be a result of exposure to environmental mycobacteria from drinking water (Primm et al., 2004). The age range for the majority of cases is between 6 months and 2 years with a mean age of 3.4 years, a male-to-female ratio of 1:1.5, and often coincides with the eruption of teeth and the proximity of the home to water (Wolinsky, 1979; Wolinsky, 1995;

McCrossin and Mailman, 2006; Thegerstrom et al., 2008). Infection is generally limited to cervical and mandibular lymph nodes and accompanied by gradual onset of painless, unilateral cervicofacial lymphadenopathy often lacking fever and is unresponsive to anti-

staphylococcal antibiotics. If left untreated, the nodes will gradually enlarge producing a skin discoloration followed closely with fistula formation and external drainage

(McCrossin and Mailman, 2006). Surgical removal of the infected lymph nodes has the best prognosis for cure, as antituberculous drugs and most antibiotics are ineffective

(Schaad et al., 1979; Starke et al., 1995; Primm et al., 2004; Griffith et al., 2007). The most important issue for diagnosis is the differentiation between NTM and TB-induced disease. Skin testing for TB (PPD) often shows positive results in both cases, so it does not differentiate between tuberculous disease and MAC cervical adenitis (Chesney,

2002). The gold standard of diagnosis remains isolation in culture which carries only a

50% sensitivity and requires weeks of culture (American Thoracic Society Statement,

1997; Pedley et al., 2004; Griffith et al., 2007). Thus, the mean time to correct diagnosis is often longer than 3 months (15 weeks), and the low level of suspicion for

NTM lymphadentitis often causes diagnostic delays, multiple bouts of antibiotics, and delayed referral for surgery (McCrossin and Mailman, 2006). Moreover, it has been noted that the incidence of cervical lymphadenitis in children caused by environmental mycobacteria has increased in western countries, especially since the cessation of the

BCG vaccination, leading many to wonder if vaccination was protective against this as well (Katila et al., 1987; Romanus et al., 1995; Thegerstrom et al., 2008).

Wildlife and Veterinary Diseases

NTMs have the ability to infect mammals, birds, fish, and humans, and have evolved mechanisms by which they can invade and grow within host cells. In addition to causing Johne's disease, NTMs have been isolated from a wide range of organisms including: aquatic insects, nematodes, earthworms, intestinal helminthes, water buffalo, pigs, deer, horses, cats, dogs, armadillos, cynomolgus and rhesus macaques,

marsupials, ferrets, chickens, white carneaux pigeons, emus, rheas, flamingos, and various fish species (Pond and Rush, 1981; Mann et al., 1982; Fleischman et al., 1982;

Morse and Hird, 1984; Bellinger and Bullock, 1988; Goodwin et al., 1988; Odiawo and

Mukurira, 1988; Robinson et al., 1989; Shackelford and Reed, 1989; Sills et al., 1990;

Dhople et al., 1992; Schoon et al., 1993; Shane et al., 1993; Sanford et al., 1994;

Fawcett et al., 1995; Miller et al., 1995; Kaufman et al., 1995; Hunter, 1996; Helie and

Higgins, 1996; Leifsson et al., 1997; Montali et al., 1998; Bollo et al., 1998; Portaels et al., 1999; O‘Grady et al., 2000; Buddle and Young, 2000; Cross et al., 2000; Horn et al.,

2000; Pavlik et al., 2000; Astrofsky et al., 2000; Heckert et al., 2001; Ramasoota et al.,

2001; Lloyd et al., 2001; Diniz et al., 2001; Freitas et al., 2001; Whittington et al., 2001;

Marsollier et al., 2002; Fischer et al., 2003; Kane et al., 2007; Stine et al., 2010).

Because the number of organisms shed back into the environment from infected animals can be relatively small, and heavy and widespread colonization of some environments occurs, it remains unclear what role animal/human infection plays in the life-cycle of many of NTMs (Pedley et al., 2004). However, NTMs such as M. avium subspecies paratuberculosis, the causative agent of Johne‘s disease in cattle, although difficult to recover from environmental samples, are excreted in large numbers in the feces of infected animals and are likely to be present in watersheds used for drinking water (Mayberry and Hitchens, 1978; Allan et al., 1986; Hermon-Taylor, 1993; Van

Kruiningen and Freda, 2001). Thus, NTMs may exert a significant economic impact on agriculture (Johne‘s disease), and the animals themselves may serve as potential reservoirs for human disease (Primm et al., 2004).

Intracellular Growth

Mycobacteria can live intracellularly. Macrophages are important in spread of

NTMS within the host. M. avium has also been shown to survive and grow in phagocytic protozoa (Tetrahymena pyriformis) and amoeba (Acanthamoeba polyphaga and

Acanthamoeba castellanii). Intracellular habitats provide a safe haven when environmental conditions deteriorate, but this intracellular growth cycle also imparts other advantages and is important for several reasons (Cirillo et al., 1997; Steinert et al.,

1998; Miltner and Bermudez, 2000; Skriwan et al., 2002). First, many protozoal species phagocytize and consume bacteria, so surviving phagocytosis is a major advantage to a water-borne bacterium. NTMs, such as M. avium, M. intracellulare, and M. scrofulaceum achieve this anti-phagocytic behavior by inhibiting lysosomal fusion within the amoeba (Primm et al., 2004; Vaerewijck et al., 2005).

Avoiding production of the phagolysosome also imparts a unique virulence to these bacteria. When compared with culture grown M. avium, amoeba grown M. avium have been demonstrated to be more invasive towards human epithelial cells and macrophages, as well as mouse intestinal cells, thereby demonstrating increased virulence after living intracellularly (Cirrillo et al., 1997). Moreover, growth rates for M. avium, M. intracellulare, and M. scrofulaceum were shown to be four to forty fold increased when grown within protozoa as compared to free-living mycobacteria, and mycobacteria that are grown intracellularly are more resistant to antibacterial agents than those derived from culture (Miltner and Bermudez, 2000; Stahl et al., 2001).

Further, NTMs can survive the encystment process, and may use protozoal encystment as a means to survive starvation and toxic stresses (Steinert et al., 1998; Strahl et al.,

2001). It follows then that this parasitic cycle may help explain the opportunistic nature

of NTM disease. Situations that allow these protozoal hosts to thrive may increase the

NTMs in the environment. Such was the case with an epizootic of MAC in flamingos that was found to be coincident with an algal bloom in water (Cirillo et al., 1997). Few studies have examined the risk of these intercellular mycobacteria to human health or the water supply, but efforts to control biofilm levels in distribution lines would also likely result in lower amoeba and protozoal levels, as these organisms feed on biofilms (Loret and Greub, 2010). In summary, it is believed that protozoans may have played a crucial role in the evolution of pathogenicity of mycobacteria and the selection for NTMs that can infect protozoans has most likely resulted in the ability of mycobacteria to become intracellular pathogens in animals as well (Primm et al., 2004).

Environmental Distribution

Mycobacteria have been isolated from an extraordinarily large, ubiquitous range of habitats. NTMs are widespread in the environment with little evidence of person-to- person transmission, so it is assumed that infections are derived from water, food, animals, or the environment (Pedley et al., 2004). The presence of NTMs has been reported in such varied environments as deionized sterile water, hot tap water, pools, spas, ice machines, municipal water sources, dental equipment, bronchoscopes, anesthesia breathing circuits, metalworking fluid, various animal species, Sphagnum vegetation, peat-rich soils, brown water swamps of the Southeastern United States, water and soils with low dissolved oxygen content, and dust (Stine et al., 1987;

Kirschner et al., 1992; Dawson et al., 1992; Katila et al., 1995; Langevin et al., 1999;

Shelton et al., 1999; Falkingham, 2003; Primm et al., 2004; Vaerewijck et al., 2005;

Field and Cowie, 2006; Sood et al., 2007).

The ubiquitous nature of NTMs may be due to their ability to utilize a wide array of carbon and energy sources such as hydrocarbons, and humic and fulvic acids, which allows them to survive in environments inhospitable to other organisms. Thus, the apparent ubiquitous nature of NTMs in water, soil, and other environments is likely due to the ability of mycobacteria to exploit niches unoccupied by other, faster-growing organisms (Pedley et al., 2004). NTMs metabolize a broad range of hydrophobic hydrocarbons including paraffin and chlorinated hydrocarbon pollutants, and the importance of these hydrocarbons in mycobacteria is emphasized by the large number of genes involved in lipid catabolism within the genomes of NTMs (Ollar et al., 1990;

Rafii et al., 1994; Primm et al., 2004). While mycobacteria are able to survive in water that is sterile or that contains only trace amounts of nutrients, drinking water derived from environments rich in humic and fulvic acids have much higher numbers of mycobacteria and more of these NTMs tend to be in the form of biofilms (Vaerewijck et al., 2005). The growth of NTMs from environmental sources also appears to be stimulated by humic and fulvic acids, and the number of MAC organisms recovered from environmental waterways was similarly found to correlate with concentrations of humic and fulvic acids (Kirschner et al., 1999). These compounds are further thought to be of significance as they are the principal organic compounds draining out of two locations whose high mycobacterial levels have been well documented — peat-rich boreal forests and acidic, brown water swamps, such as those present in Florida

(Iivanainen et al., 1997; Kirschner et al., 1999). In addition to humic and fulvic acids, water chemistry such as oxygen content, temperature, pH, or the presence of inorganic

molecules also appears to affect the survival of NTMs (Brooks et al., 1984; Kirschner et al., 1992).

Water

There is significant regional variability in the number of reported NTM diseases around the country. Significantly more NTM cases seem to be reported in persons residing in the Southeastern United States (Reed et al., 2006). In fact, the CDC found that over 41% of all reported isolates in the US were from the Southeast (Dobos et al.,

1999). Not surprisingly, M. avium complex organisms can be found in largest numbers in brackish swamps and estuaries of the Southeastern coastal United States (Kirschner et al., 1992; Kirschner et al., 1999). Further, members of the M. avium complex, M. scrofulaceum, M. fortuitum, and M. chelonae, all human pathogens, all grow in water, including environmental waters with and without salt and drinking-water distribution systems throughout the world (George et al., 1980; Covert et al., 1999; Falkingham et al., 2001). Moreover, exposure to water is the primary route of human infection by M. scrofulaceum and M. marinum (Falkingham, 1996). It is not unreasonable, then, to believe that water may be the main exposure route to all of these NTMs.

Numerous situations exist in which the geographic and environmental distributions of humans and mycobacteria overlap, leading to ecological impact on the mycobacteria themselves and exposure of humans to these potential pathogens. The most notable overlap is water. Humans are routinely exposed to environmental mycobacteria through activities such as drinking water, swimming and bathing. Indeed, persons who work in and around water, such as fishery managers, anglers, and commercial fishermen, are at increased risk of mycobacterial skin lesions (Panek and Bobo, 2006). In Sweden, persons living close to water have a higher proportion of positive reactions to NTM

sensitins (Thegerstrom et al., 2008). Within the United States, living or working in water damaged buildings or reconstruction of these buildings is associated with major outbreaks of NTM induced pulmonary disease (Falkingham, 2003). In fact, the risk associated with NTMs as an emerging pathogen in water systems was acknowledged by the U.S. Environmental Protection Agency (EPA) by the placement of M. avium on the contaminate candidate list for possible regulation in drinking water (Torvinen et al.,

2007).

It naturally follows then that the primary mode of infection of humans with NTMs is via ingestion or aerosolization of water (Primm et al., 2004). Unsurprisingly, an individual‘s pattern of water use and consumption, as well as diet, is believed to vary exposure levels (Vaerewijck et al., 2005). Consistent with the oral theory, M. avium has been shown to exhibit high bile salt and acid tolerance, thereby allowing M. avium to pass through the intestinal tract to invade through the apical surface of epithelial cells in the terminal ileum of the human intestine (Bodmer et al., 2000; Bermudez and Sangari,

2000; Sangari et al., 2000; Sangari et al., 2001). The invasion of human intestinal cells by M. avium is enhanced at 37°C, high osmolarity, and low oxygen tension, but appears unaffected by iron limitation or acidic pH (Bermudez et al., 1997). Person-to-person contact is not believed to be a mode of transmission, but inhalation of dusts and soils, inoculation from environmental sources, food, traumatic exposure, and tobacco have all been proposed as other sources of human exposure to environmental mycobacteria

(Witty et al., 1994; Yajko et al., 1995; Eaton et al., 1995; Falkingham, 1996; Yoder et al.,

1999; Argueta et al., 2000; Pedley et al., 2004; Primm et al., 2004; Vaerewijck et al.,

2005). Water, however, appears to be the most probable source of infections.

Aerosolization of contaminated water may be critical in allowing respiratory exposure to NTMs. Aerosolization occurs as a result of the high hydrophobicity leading to adsorption to rising air bubbles (van Oss et al., 1975; Blanchard and Hoffman, 1978;

Weber et al., 1983). Adsorption to bubbles leads to concentration of mycobacteria within droplets and when these bubbles reach the surface, they burst resulting in several droplets 8-10 cm above the water surface (Blanchard and Szydek, 1978).

These aerosolized particles associated with mycobacteria are within the size range capable of entering the human alveoli, and it is possible that significant numbers of

NTMs can be transferred to the air by this mechanism (Falkingham et al., 1990; Pedley et al., 2004). Thus, it is possible aerosolization could provide a source of infection, particularly of the respiratory tract. By the same process of preferential adsorption to rising air, the high hydrophobicity also leads to mycobacteria concentrating at the air/water interface where organic matter is typically concentrated (Blanchard and

Hoffman, 1978; Wendt et al., 1980). This positions NTMs in an environment rich in organic matter where there are few competitors, but many opportunities for human interaction (Pedley et al., 2004).

Water Quality

Environmental parameters appear to have an impact on mycobacterial growth rates, with some conditions more ideal than others for optimized NTM growth. A review of the literature suggests parameters such as temperature, pH, dissolved oxygen, dissolved organic carbon, zinc, iron, turbidity, and phosphorous may all play a role in optimizing the growth of mycobacteria (Kirschner et al., 1992; Howard and Byrd, 2000;

Primm et al., 2004; Vaerewijck et al., 2005; Bland et al., 2005; Hilborn et al., 2006; Reed et al., 2006; Torvinen et al., 2007; Sood et al., 2007; Langevin et al., 2008). While each

of these may contribute to the overall growth optimum, certain parameters such as dissolved organic carbon, salt tolerance, dissolved oxygen, temperature, and pH, are believed to be more important.

DOC

M. avium and other NTMs can grow in natural waters and municipal sources containing low dissolved organic carbon (DOC) (George et al., 1980; Falkingham et al.,

2001). The presence of biodegradable organic matter in water is associated with mycobacterial growth (LeChevallier, 2003; van der Kooij 2003). Biodegradable organic carbon is often measured by AOC (assimilable organic carbon), and AOC levels have been found to range between 20 to 214µg/L with a median of 100µg/L in North

American drinking water systems (LeChevallier et al., 1996; Volk and LeChevallier,

2000). Studies of AOC on mycobacterial biofilm formation have indicated that at increasing AOC levels the density of biofilms increase, and, more importantly, that M. avium biofilms are capable of forming at levels as low as 40µg/L (Falkingham et al.,

2001; Norton et al., 2004). Further, increases in M. avium levels in drinking water were found to correlate with AOC levels (Falkingham et al., 2001). Thus, it is possible that decreasing the amount of available biodegradable carbon from a water source may decrease the relative number of mycobacterial biofilms, but is not likely to prevent their occurrence. Conversely low DOC may provide a competitive advantage to NTMs over faster growing bacteria.

Salt Tolerance

Mycobacteria are capable of growth under an array of salinities. While individual species may exhibit their own varying tolerance levels, mycobacteria have been isolated from both fresh and brackish waters (George et al., 1980). Indeed, faster growth has

been reported in waters containing 1% NaCl (brackish) than in fresh water (Gruft et al.,

1981; George et al., 1994; Pedley et al., 2004). This ability to grow in brackish waters helps explain the high numbers of NTMs isolated from the tidal waters of large estuaries, like the Gulf of Mexico and the Chesapeake Bay (Gruft et al., 1975; Gentry,

2004; Panek and Bobo, 2006; Kane et al., 2007). Contrarily, despite reaching appreciable numbers in brackish waters, mycobacteria do not appear to grow in high salinity situations, such as seawater (3% NaCl) (Falkingham et al., 1980). Thus, it appears that mycobacteria grow best over low- to mid-range salinity, but every species has its own unique optimum (Gruft et al., 1981). The salinity correlation is not limited to

NaCl, however. Studies have suggested that NTMs are capable of shifting from Na+ rich environments, such as estuaries, to K+ rich environments, like intracellular conditions of macrophages and protozoa, without loss of viability (Amin et al., 2008).

Dissolved Oxygen

Higher mycobacterial loads are recovered from waters and soils with reduced oxygen levels. Mycobacteria are obligate aerobes requiring oxygen for growth, but paradoxically, they also have the ability to survive and metabolize under hypoxia

(Berney and Cook, 2010). While NTMs cannot grow anaerobically, they are capable of surviving rapid shifts in anaerobiosis (Pedley et al., 2004). Though surviving such shifts, mycobacteria experience energetic challenges when placed in hypoxic conditions, and changing the oxygen level can dramatically affect the doubling time for mycobacteria (Berney and Cook, 2010). The most rapid growth is said to occur between 12% and 21% oxygen, but growth has been reported at levels as low as 2.5% oxygen (Palamino et al., 1998; Pedley et al., 2004). To survive, mycobacteria employ three strategies under hypoxia: high affinity cytochrome oxidases scavenge oxygen,

they switch to NAD+/NADH independent enzymes, and they upregulate hydrogenases

(Berney and Cook, 2010). In fact, studies have previously found that in some mycobacterial species, specifically M. ulcerans, growth stimulation was directly proportional to decreases in oxygen concentrations (Palamino et al., 1998). Thus, it is not surprising that waters and soils that have low oxygen levels often yield the highest numbers of MAC (Brooks et al., 1984; Kirschner et al., 1992).

Temperature

Mycobacteria can grow over a wide range of temperatures, and individual species may survive either high heat or freezing conditions (George et al., 1980). Research has demonstrated that more NTMs tend to be isolated in warmer months of the year, and the ideal temperature range appears to be approximately 15.5-20°C though temperature sensitivities vary by species (Torvinen et al., 2007). Further, studies have indicated that temperature is an important factor influencing bacterial growth, and in climates where water temperatures are warm, bacterial growth may be very rapid (LeChevallier, 2003; van der Kooij, 2003). It follows then that the ability to grow at temperatures as high as

45°C helps explain NTM presence in hot water systems as well as in coastal brown- water swamps of the Southeastern United States where summer temperatures can surpass 45°C (du Moulin et al., 1988; Schulze-Robbecke and Buchholtz, 1992; Mijs et al., 2002). Alternatively, the higher NTM numbers isolated from brown-water swamps, as with peat-rich boreal forests, may be due to the principal organic compounds draining from these waters, humic and fulvic acids (Kirschner et al., 1999; Primm et al.,

2004; Vaerewijck et al., 2005). Further emphasizing the dynamic nature of these organisms, mycobacteria can survive freezing conditions and may actually have higher

counts after freezing, presumably as a result of disaggregation of bacterial clumps

(Iivanainen et al., 1995). pH

NTMs have an acidic growth optimum of pH 4.5 to 5.5 (Portaels and Pattyn, 1982;

George and Falkingham, 1986). It is still unclear, however, whether increased mycobacterial growth under acidic conditions is due to the utilization of acidic compounds in the water or whether these bacteria actually grow better in acidic environments independent of the organic acids. Of particular note, this acidic growth optimum noted for many mycobacteria falls below the pH of commonly used media

(Middlebrook 7H9 and 7H10), and may be limiting in the recovery of certain mycobacterial species (Falkingham, 2009). Regardless, mycobacterial growth and tolerance within these low pH environments provides another possible explanation for the higher numbers of NTMs reported from soils and waters of peat-rich boreal forests and brown-water swamps (Kirschner et al., 1992; Iivanainen et al., 1999; De Groote et al., 2006). Interestingly, increased survival through water treatment procedures was noted in M. fortuitum samples when the pH was increased from 5.7 to 10.1, suggesting a physiologic change with pH (Farooq et al., 1977). On the other hand, mycobacteria are also capable of surviving low pH extremes such as those found in the human stomach (Bodmer et al., 2000).

It is likely that mycobacterial growth is not driven by any one of these parameters alone, but by a unique interplay between them all. Brown water swamps of southeastern

US, such as those in Florida, have four of the major growth optima, demonstrating how difficult it can be to tease out which, if any, parameter drives the numbers to higher levels in these places. Regardless of cause, these higher numbers are concerning

because these waters are used as sources for drinking water, and present a risk of introducing NTMs into the municipal water system (Kirschner et al., 1992; Iivanainen et al., 1999). Given mycobacteria‘s high levels of resistance to disinfectants, once allowed to establish, NTMs may grow and persist in drinking water systems (Falkingham et al.,

2000; Taylor et al., 2000).

Drinking Water

Efforts to control mycobacteria have proven difficult. The primary focus of public health efforts to control mycobacteria has been directed toward disease treatment rather than prevention (Saleeb and Olivier, 2010). One area of possible intervention is treatment of drinking water. Multiple studies demonstrate the health hazard of NTM exposure via pools, spas, whirlpools, footbaths, showers, and aquariums (Aubuchon et al., 1986; Embil et al., 1997; Kahana et al., 1997; Zenone et al., 1999; Lee et al., 2000;

Khoor et al., 2001; Winthrop et al., 2002; Angenent et al., 2005; Streit et al, 2008;

Feazel et al., 2009; Cheung et al., 2010). These exposures most commonly result in skin and soft tissue disease, but pulmonary disease has been linked to inhalation of contaminated aerosols in many of these settings (Collins et al., 1984; Embril et al.,

1997; Shelton et al., 1999; Griffith et al., 2007; Feazel et al., 2009; Cheung et al., 2010).

Additionally, nosocomial infections have also been reported from contaminated tap water and de-ionized water used in procedures or to clean equipment for procedures

(Stine et al., 1987; Graham et al., 1988; Falkingham, 2002; De Groote et al., 2002;

Cortesial et al., 2010). Thus, there is an increasing body of work linking contaminated water sources and NTM disease. Given this trend, it is increasingly more important to understand and control for these infections.

Various methods have been proposed for the control of NTMs within drinking water systems. Current water treatment strategies are unlikely to provide adequate protection due to the high level of resistance of NTMs to disinfectants and their ability to grow as biofilms within distribution lines (Hall-Stoodley et al., 1999; Taylor et al., 2000;

Falkingham et al., 2001; Le Dantec et al., 2002; Steed and Falkingham, 2006;

Falkingham, 2009). Previously, it was shown that some NTMs, such as M. kansasii, could colonize cold-water distribution systems, whereas M. xenopi and M. avium are more commonly associated from hot water systems (du Moulin et al., 1988; Falkingham,

2002; Pedley et al., 2004). Further, MAC organisms have been recovered from drinking-water systems before and after treatment, from distribution lines, and from raw source waters (du Moulin et al., 1988; von Reyn et al., 1993; Glover et al., 1994; Peters et al., 1995; Covert et al., 1999; Ristola et al., 1999; Falkingham et al., 2001). One

MAC clone was isolated repeatedly from a water distribution system over an 18-month study, implying that MAC may not be a contaminant, but a normal inhabitant of drinking water systems (von Reyn et al., 1994). Indeed, NTM numbers were found to be higher within the distribution system than immediately pre- or post-treatment, suggesting that

NTMs are replicating within such systems (du Moulin et al., 1988; Falkingham et al.,

2001). Mycobacterial growth within these systems is believed to be primarily in the form of biofilms along the length of distribution lines, and the number of such mycobacteria isolated from drinking water can be very high, up to 700,000 CFU/L (Falkingham et al.,

2001). Given the length of water pipes in an average distribution system can easily be hundreds even thousands of miles, the potential risk of human exposure is immense

(Falkingham et al., 2001).

Mycobacterial growth in municipal systems is attributed to environmental factors and the inherent resistances of mycobacteria. Municipal water contamination takes place in three phases. First, NTMs derived from source waters must survive the treatment process. Secondly, NTMs must re-grow from relatively low numbers. Thirdly, these NTMs must grow within the water distribution pipelines typically in the form of biofilms, thereby re-contaminating the treated water (Pedley et al., 2004). The treatment of water does not entirely eliminate mycobacteria from municipal waters

(Falkingham et al., 2001). Most microbes are removed at treatment plants by two principal mechanisms, physical removal via coagulation, sedimentation or filtration and chemical inactivation via disinfectants (LeChevallier and Au, 2003; Dash et al., 2010;

Falkingham, 2010). The hydrophobic nature of the mycobacterial cell wall promotes attachment to surfaces such as suspended particulate material, thereby aiding in the removal of mycobacteria via physical means (Falkingham et al., 2001; Mazumder et al.,

2010). The use of coagulants, however, can destabilize NTM adherence to these particles by neutralizing or reducing the surface electrical charges, thereby releasing the mycobacteria into the water or into a flocculent. Fortunately, flocculated particles often have sufficient settling velocities to allow for sedimentation, but not all mycobacteria are collected through this process (Harrington et al., 2001; Bell et al., 2002; Bagga et al.,

2008; Li et al., 2010). Significant reduction of mycobacteria after water filtration has been demonstrated, but mycobacteria were not completely removed and the filter materials themselves were implicated as potential sources of NTM re-contamination of the water distribution systems post-filtration (Le Dantec et al., 2002; LeChevallier and

Au, 2003; Falkingham, 2010).

The second form of microbial control, inactivation through the use of disinfectants, can also be troublesome in mycobacteria, which are among the most resistant microbes

(Jacangelo et al., 2002). Increased resistance to disinfectants within strains has previously been reported in NTMs derived from low-nutrient sources, such as tap water or diluted media, as compared with rich laboratory media (LeChevallier et al., 1988;

Taylor et al., 2000; Le Dantec et al., 2002; Steed and Falkingham, 2006). Additionally, clumping, a natural mycobacterial property, imparts strain variability in resistance largely by decreasing the surface area and access to porins (Woelk et al., 2003; Danilchanka et al., 2008; Steinhauer et al., 2010). Given their inherent resistance to disinfectants, it is easy to see how many NTMs would be able to survive water treatment procedures

(Taylor et al., 2000). The number of mycobacteria immediately post-treatment, however, is believed to be low, suggesting the occurrence of re-growth within distribution systems (Feazel et al., 2009).

Tap water is known to be a low-nutrient system, and for most microbes this creates an inhospitable environment which prevents re-growth (Taylor et al., 2000;

Radomski et al., 2010; Lautenschlager et al., 2010). Mycobacteria, however, are capable of growing in low nutrient environments by utilizing trace levels of minerals and organic carbon (van Ingen et al., 2010; Radomski et al., 2010; Lautenschlager et al.,

2010). Indeed, stored distilled water, the epitome of a low nutrient source, has been demonstrated to harbor NTMs at levels as high as 105 to 106 mycobacterial cells/mL

(Carson et al., 1978; Falkingham et al., 2001; LeChevallier, 2003). Further, studies involving eight water distribution systems indicated that M. avium and M. intracellulare

(both MAC species) were frequently isolated from biofilms within municipal samples

(Falkingham et al., 2001; LeChevallier et al., 2001; Konstantinos et al., 2010). This finding implies that NTMs, even potentially pathogenic species, are regular inhabitants of water distribution systems, and are not likely the result of experimentally derived contamination (Falkingham et al., 2001; Thompson et al., 2008; Konstantinos et al.,

2010). The human activity of treating potable water with chlorine or other disinfectants, such as ozone, has led to selection for environmental mycobacteria by providing a unique ecological niche. For example, MAC, like many NTMs, are relatively resistant to chlorine, monochloramine, chlorine dioxide, and ozone (Taylor et al., 2000). In the absence of disinfection, M. avium, cannot compete effectively for limited nutrients, but disinfection kills competitors permitting growth of M. avium on the available nutrients

(Falkingham et al., 2001). This phenomenon is most likely responsible for the growth of

MAC within municipally distributed waters, as well as hot tubs, spas, and showerheads

(Embil et al., 1997; Falkingham et al., 2001; Feazel et al., 2009). Moreover, different mycobacteria have differing susceptibility to chlorine and other chemical disinfectants which may lead to human derived shifts in which mycobacteria species are in greatest abundance in water and thereby alter the risk of human exposure to these. For example, before 1970, the majority of cases of cervical lymphadenitis in children were due to M. scrofulaceum, which is fairly chlorine sensitive (Wolinsky, 1979). Since 1975, however, M. avium has predominated (Colville, 1993; Wolinsky, 1995). The change is believed to be due to the Clean Water Act of the United States that increased chlorination rates of waters beginning in 1978 (Primm et al., 2004). Similar trends were also noted in England and Australia following changes in their water treatment procedures (Colville, 1993; Pedley et al., 2004). In short, the implementation of

improved methods for disinfection of drinking-water combined with mycobacterial resistance to disinfectants and their presence in the source waters led to selection for

MAC and other mycobacteria within distribution systems. Of the drinking water sources analyzed thus far, higher mycobacterial numbers were encountered from systems using surface water and applying ozonation as an intermediate treatment or post-treatment than those using non-ozonated surface waters, ground water, or mixed water

(Vaerewijck et al., 2005).

Once through the treatment system, the re-growth of NTMs in municipal water systems is believed to be influenced by the concentration of biodegradable organic matter, sediment accumulation/ turbidity, microbially available phosphorous and nutrients, concentration of free residual disinfectants, residence time, microbial interactions, pH, temperature, diameter of the pipes, hydraulics, and the characteristics

(composition, porosity, and roughness) of the pipes themselves (Thofern et al., 1987;

LeChevallier et al., 1993; Vaerewijck et al., 2005). In general, it is accepted that old, corroded pipes containing dead ends and spaces favor mycobacterial growth, as does having brass, bronze, iron or galvanized pipes as compared with PVC or copper surfaces, but all these composites yield significantly lower growth than hot water silicon tubing (Schulze-Robbecke and Fischeder, 1989; Falkingham et al., 2001; LeChevallier et al., 2001; Vaerewijck et al., 2005). Pipelines may also favor the formation of biofilms due to their hydrophobic character enhancing bacterial adherence. NTMs have been known to form biofilms for decades (Schulze-Robbecke and Fischeder, 1989; Iivanainen et al., 1999; Falkingham et al., 2001). The intermittent use of many of these lines leads to stagnation of an entire water column for extended periods of the day, which allows for

undisturbed bacterial proliferation and potential aerosolization when used (Vaerewijck et al., 2005). This makes it possible for high numbers of NTMs to be swallowed, inhaled, or inoculated into wounds which in turn leaves potential for colonization, infection, and immunization (Vaerewijck et al., 2005). As such, the presence of MAC in biofilms helps to explain how groundwater-derived drinking water systems could test negative for MAC in their source waters, but the distribution lines themselves could test positive for MAC

(Falkingham et al., 2001). If one considers the size of pipes used for distributing drinking water and the length and number of pipes within the system, the contribution of these biofilms to the microbial flora of drinking water could be substantial (Falkingham et al., 2001; Pedley et al., 2004). More importantly, the presence of NTMs in biofilms represents not only a mechanism for the persistence and growth of mycobacteria within these tubes, but also a reservoir for these organisms (Schelonka et al., 1994). Thus, biofilms associated with piping, such as those in showers, catheters, etc, serve as a potential mechanism for direct human exposure to NTMs.

Recontamination of the distribution system can occur via cross-connections, backflows, and intrusion events (Ford, 1999; Craun and Calderon, 2001; EPA, 2002;

Sadiq et al., 2003). Most recognized microbial outbreaks resulting from such recontamination events involve acute gastrointestinal disorders (Swerdlow et al., 1992;

Semenza et al., 1998; Mermin et al., 1999; Yassin et al., 2006; Reynolds et al., 2008;

Colford et al., 2009). Thus, slowly developing infections, such as those caused by

NTMs, would not be easily identified, and so little is known about the risk these pose

(Hoffman et al., 2009). Increasingly people are becoming aware of the opportunities for re-introduction of microbes into the system, especially through transient negative

pressure events (Karim and LeChevallier, 2005; Teunis et al., 2010; Besner et al.,

2011). These occur as the result of sudden changes in water velocity, usually as the result of main breaks, fire flows, or pump shutdowns, which allow contaminated water and soil to enter the distribution system through cracks, seals, or pipe leaks

(LeChevallier et al., 2003; Karim et al., 2003; Karim and LeChevallier, 2005; Teunis et al., 2010; Besner et al., 2011). While no studies have examined NTMs specifically in these situations, fecal coliforms and virus introductions are well documented, as is the introduction of organic material, which would promote NTM growth (Kirschner et al.,

1992; Iivanainen et al., 1997; Karim et al., 2003; LeChevallier et al., 2003; Gullick et al.,

2005; Teunis et al., 2010). When soil or water enter the pipe-lines, such as during the repair of main breaks, the traditional decontamination step in these situations, chlorine swabs, likely do little to inactivate any NTMs that may have been introduced (Taylor et al., 2000; Teunis et al., 2010). High velocity flushing is likely to remove any introduced sediment from the system, but carries the added risk of biofilm shearing, thereby resulting in a net increase of NTMs within the water column itself (Geldreich and

LeChevallier, 1999; Lehtola et al., 2006). While this is not believed to be the source of free mycobacteria, the current study proposes this as a potential means of introducing mycobacteria into distribution systems. It is likely that NTMs could be introduced into water distribution systems in this manner, but given the dilutional effect of the water, the relevance of this is more likely as an opportunity to seed biofilms already existing within the system rather than posing a direct public health concern for single inoculums.

Regardless of source, controlling the growth of NTMs within distribution systems is

believed to be necessary to ensure the safety of municipally distributed waters (van

Ingen et al., 2010).

In summary, efforts to control mycobacteria in water distribution systems have proven difficult. Current water treatment strategies are unlikely to provide adequate protection due to the high level of resistance of NTMs to disinfectants and their ability to grow as biofilms within distribution lines (Hall-Stoodley et al., 1999; Taylor et al., 2000;

Falkingham et al., 2001; Le Dantec et al., 2002; Steed and Falkingham, 2006;

Falkingham, 2009). Source minimization of NTMs is difficult given the ubiquitous nature and variety of sources for contamination (EPA, 2009; Saleeb and Olivier, 2010). Thus, new strategies must be developed to control the risk to the water supply. One suggested strategy to reduce MAC and other NTMs from water-distribution systems is to reduce the particulates (turbidity) in raw and treated waters (Falkingham et al., 2001).

Mycobacteria are known to aggregate on particulate matter and turbidity rates of 2 NTU or greater have been demonstrated to enhance the detection rate of MAC from raw water, thereby indicating the concentrations of NTMs, notably MAC, are significantly associated with these higher turbidity levels (Falkingham et al., 2001). Therefore, filtration could be used to reduce turbidity, and has previously been shown to minimize some NTM related pseudo-outbreaks in hospitals (Stine et al., 1987; Graham et al.,

1988). As discussed, however, MAC and other mycobacteria can grow to high numbers on filters turning the filters into sources of NTM contamination (Ridgeway et al., 1984;

Rodgers et al., 1999; Falkingham, 2009; Falkingham, 2010). Thus, any filters used must be sampled and replaced often (< 3 weeks) (Falkingham, 2009). If filters are to be used, it has been proposed that the filters be coated in a hydrophobic material, such as

paraffin, to help selectively remove NTMs (Falkingham, 2009). Based on the hydrophobicity of NTMs, it is believed NTMs would aggregate on a hydrophobic surface, and indeed previous work demonstrated that almost all (>99.9%) NTMs were removed through the use of one hydrophobic substance, hexadecane (Stormer and Falkingham,

1989). Hydrophobic coatings, however, may reduce the number of non-mycobacterial species adhering to the filters (Falkingham, 2009). Other approaches could involve the identification of novel disinfectants active against mycobacteria or the identification of factors contributing to disinfectant resistance. Disinfection, however, would likely kill off most other organisms thereby selecting for any NTMs that survive, thereby allowing them to flourish in a niche lacking competitors (Falkingham, 2009). Alternatively, short- wave length ultraviolet irradiation (UVC) may be an option. Mycobacteria, like most bacteria, are susceptible to UVC, and UV irradiation of aquarium tanks has previously been demonstrated to effectively reduce the numbers of NTMs (Collins, 1971; David et al., 1971; David, 1973; McCarthy and Schaefer, 1974; Agbalika et al., 1984). Thus,

UVC may be a means of reducing the numbers of waterborne NTMs from drinking water systems and plumbing, but unfortunately, UV irradiation is also mutagenic, risking the selection for of resistant mutants (Konickova-Radochova and Malek, 1968; Falkingham et al., 2009). Another option is the use of copper-silver ions as an anti- mycobacterialcidal agent. Systems generating copper-silver ions are already in place in some hospitals as a means of reducing Legionella, and fortunately, NTMs, such as M. avium, have previously been demonstrated to be susceptible to these ions, although at higher levels than those required for Legionella (Liu et al., 1998; Lin et al., 1998;

Kusnetsov et al., 2001; Wolschendorfa, et al., 2011). The utility of these ions, however,

has not been explored for NTMs. For recreational waters, it has been suggested that careful maintenance and cleaning as per manufactures instructions be strictly adhered to, as well as periodic shock treatments or daily superheating of water to over 70°C to control for NTMs, and immunocompromised persons and those with scratches and scrapes should be advised to avoid swimming pools, hot tubs, and the like (Embil et al.,

1997; Freije, 2000; WHO, 2000).

Geographical Mapping

With the number of cases of NTM related disease increasing every year, the distribution and epidemiology of NTM disease is of critical importance. While NTM disease affects both humans and animals, its ecology is complex and its distribution remains poorly understood (Pedley et al., 2004; Vaerewijck et al., 2005; Winthrop

2010). A number of studies have evaluated the physiological niche of these organisms within the environment, but few studies have addressed the geographical distribution of infections or the distribution of environmental parameters that promote mycobacterial growth (Edwards et al, 1969; von Reyn et al., 2001; Reed et al., 2006; Sood et al.,

2007; Langevin et al., 2008; Winthrop, 2010). In order to investigate the geographical distribution of NTMs and any areas where persons may be at an increased risk, studies focusing on the relationships between the biological requirements of NTMs themselves, the ecological conditions that support their survival and growth, and the areas where these overlap are needed. Understanding these fields is crucial to developing or implementing a disease control program (Blackburn, 2010). Determining the spatial distribution of NTM disease requires spatially explicit predictive models, as well as available data on the known occurrence of disease (Blackburn, 2010). For the current study, geographical data on disease occurrence was provided by the Florida‘s Agency

for Health Care Administration (AHCA) database and spatial models and analyses were produced using Geographic Information Systems (GIS).

GIS is a fast growing field in international research, particularly research focusing on epidemiology, ecology, or disease distribution (Clarke et al., 1996; Rogers, 2006;

Blackburn et al., 2007; Blackburn, 2010). GIS entails a combination of computer software, computer hardware, and databases for storing, editing, visualizing, and analyzing spatial data (Curtis et al., 2007). The strength of a GIS is the ability to establish relationships between data sets and analyze them spatially (Blackburn, 2010).

Disease distributions can easily be reviewed in a GIS, as associations intrinsically exist between the locations of disease occurrences and other factors, such as ecologies promoting long-term survival, environmental reservoirs, vectors that transmit the disease, or at risk populations (Smith et al., 2000; Rogers, 2006; Eisen et al., 2006;

Blackburn, 2010). Within a GIS, dissimilar data sets are linked through relational databases to produce dynamic maps portraying spatial associations (Blackburn, 2010).

Stated simply, GIS evaluates data in space through integration of databases and map visualizations. Visualizing data in the form of maps can be considered the first step in determining spatial patterns (Anselin, 2005). While these maps contain no statistical examination, they are informative for localizing disease outbreaks. The second step in the assessment process is the use of spatial autocorrelation techniques to identify statistically significant patterns within the data sets (Anselin, 2005). These statistics are often done through local indicators of spatial autocorrelation (LISA) to determine whether statistically significant spatial patterns exist within the distribution of the data

(Anselin, 1995). Places where more occurrences happen than would be expected by

random chance alone are defined as hot spots, and similarly, areas with significantly fewer occurrences than would be expected by random chance are called cold spots

(Getis and Ord, 1992; Anselin, 1995). These tests involve iterative algorithms examining the association between the occurrence of disease and neighboring rates

(Anselin, 1995; Anselin, 2005). For the current study, neighbors were defined as those polygons sharing a border (rook matrix) (Anselin, 2005).

Final Thoughts

NTMs, such as M. avium, have historically been isolated in the highest numbers in the southeastern US (Falkingham, 1996; Primm et al., 2004), and significantly more

NTM infections, approximately 41% of the US total, were reported from persons residing in that region (Dobos et al., 1999; Reed et al., 2006). In fact, approximately 15% of pulmonary mycobacterial infections are diagnosed annually in the state of Florida alone

(Dobos et al., 1999; Lauzardo, pers. comm.) and the incidence is rising (Bilinger et al.,

2009). This led to the hypothesis that some environmental parameters are likely affecting the species distribution and density of mycobacteria in the environment, and that changes in these parameters may affect the number and diversity of mycobacteria in an area, thereby altering the relative risk to the general population for NTM infections.

Given the ever-increasing population of at-risk individuals (Day, 1996; Rappaport,

2007), it is more important than ever to understand factors that govern the spread of

NTM disease. This study will provide important insights into the nature of mycobacterial infections within the state of Florida, and will provide a base for further research into epidemiological trends surrounding NTM infections. The following chapters include the development of novel molecular probes for the rapid and accurate detection of NTMs directly from clinical or environmental samples, the investigation of selected surface

waterways and municipal distribution systems in Florida for environmental parameters governing NTM occurrence and quantities, the evaluation of hospital discharge data for demographical trends of persons hospitalized with NTM disease, as well as spatial analysis of home zip-codes of persons hospitalized with pulmonary or disseminated

NTM disease. This project was the first to provide field-based data examining the functional association between environmental parameters, NTM presence in the environment, and clinical incidence in Florida, and has made much needed inroads to provide linkages to ultimately discern NTM transmission and risk factors in future studies.

CHAPTER 2 NONTUBERCULOUS MYCOBACTERIA DETECTED USING NOVEL QPCR PROBES

Background

The genus Mycobacterium contains gram-positive microaerophilic bacteria, and is one of several mycolic acid-containing genera within the order Actinomycetales

(Howard and Byrd, 2000; Falkingham, 2003). There are currently thought to be over 130 species of in the genus, with many others still unidentified. Although this genus of bacteria is best known for the causative agents of tuberculosis and leprosy (M. tuberculosis and M. leprae, respectively), the vast majority of mycobacteria are free- living environmental species that are opportunistic pathogens. Mycobacteria can be divided into three non-phyletic groups based upon clinical significance. The first grouping is an assemblage of obligate pathogens including the Mycobacterium tuberculosis complex (M. africanum, M. bovis, M. canettii, M. caprae, M. microti, M. pinnipedii, and M. tuberculosis), and M. leprae and M. lepraemurium, that are usually not found in the environment. Non-tuberculous mycobacteria, NTMs, comprise the remaining two groups; the facultative pathogens (M. avium, M. marinum, M. ulcerans, etc.), commonly found in aquatic and terrestrial environments (Covert et al., 1999;

Vaerewijck et al., 2005), and the non-pathogenic or rarely pathogenic species that are generally saprophytic. (Covert et al., 1999; Vaerewijck et al., 2005)

NTMs are associated with a variety of chronic, debilitating human and animal diseases. Those within the Mycobacterium avium complex (MAC, i.e., M. avium and M. intracellulare), are important human pathogens in non-immunocompromised persons as well as HIV co-infected individuals (Karakousis et al., 2004; CDC, 2005; Vaerewijck et al., 2005). M. avium ssp. paratuberculosis, the agent of Johne's disease in cattle and

other hoofstock, also has a significant economic impact on agriculture (Primm et al.,

2004). Other human pathogens include M. abscessus, M. kansasii, M. marinum, and

M. xenopi, which have been associated with pulmonary disease, hypersensitivity pneumonitis, cervical lymphadentitis, osteomyelitis, arthritis, keratitis, tenosynovitis, disseminated disease, otitis media, corneal infections, endocarditis, immunologic dysfunction, and chronic disease states (Howard and Byrd, 2000; Vaerewijck et al.,

2005).

NTM infection in humans is on the rise, with the number of isolates exceeding those of M. tuberculosis within the United States (Sood et al., 2007). This includes otherwise healthy subjects, particularly middle-aged to older, below average weight, caucasian women (Moore, 1993). NTMs are also associated with nosocomial outbreaks and mini-epidemics (Vaerewijck et al., 2005).

Clinical diagnosis of NTM disease is complicated because patients typically present with nonspecific symptoms including fever, anorexia, weight loss, and night sweats. The vagueness of the symptoms often leads to misdiagnosis. Delay in accurate diagnosis is problematic because NTM infections are one of the primary reasons for death in immune compromised populations (Howard and Byrd, 2000).

Despite this, the epidemiology and ecology of NTM disease remains poorly discerned.

The current gold standard for clinical detection is based on identification of mycobacteria using culture-based methods (Wagner and Young, 2003; Griffith et al.,

2007). Many mycobacteria, however, are fastidious and difficult to culture due to slow growth characteristics and specific nutrient requirements (Griffith et al., 2007). These techniques obviously provide bias for organisms that are readily cultured (Crosby and

Criddle, 2003). As an example, based on culture results, the primary agent of mycobacteriosis in pet birds was previously considered to be M. avium, which is relatively easy to culture. Non-culture based methods, however, have shown that M. genavense, which is difficult to culture, is responsible for up to 96% of companion bird mycobacterioses (Manarolla et al., 2009). Traditional PCR-based techniques require a growth enrichment step prior to extraction, and therefore do not allow direct detection of mycobacteria from environmental or clinical samples (Wagner and Young, 2003).

Extraction of DNA from mycobacteria is also difficult due to their thick, waxy cell wall

(Kotlowski et al., 2004; Griffith et al., 2007), further complicating sensitive detection of these microorganisms in both clinical and environmental samples.

Tools for assessing NTMs are needed. The goal of this study was to develop a non-culture based, quantitative PCR (qPCR) method of detection for mycobacteria, and to compare physical and chemical DNA extraction techniques. The proposed qPCR techniques do not require culture of the bacteria, and provide selectivity, better sensitivity, and the ability to quantitatively measure relative number of copies from environmental samples (Fang et al., 2002; Crosby and Criddle, 2003). Probes developed through this study will be valuable for determining the relative abundance of

NTMs in environmental reservoirs, which is expected to correlate with risks to humans, domestic animals, and wildlife.

Materials and Methods

Primer Development

Mycobacterial genus-specific primers were developed based on the 16S rRNA gene sequence (Genbank, National Center for Biotechnology Information, Bethesda,

MD) from 23 mycobacterial species and 6 closely related species. Sequences were

aligned and visualized using MUSCLE software (Edgar, 2004). The species of mycobacteria represented all major clades from within the genus-level phylogenetic tree

(Migard and Flandrois, 2008). DNA sequences from closely related genera in the order

Actinomycetales, including Corynebacterium, Rhodococcus, Nocardia, and

Actinobacter, were also used to represent non-Mycobacterium sequences. Aligned sequences were reviewed for regions of similarity within all mycobacteria that bore inherent differences from the other Actinomycetales. Regions matching such parameters were analyzed with Primer3 software (Howard Hughes Medical Institute,

Chevy Chase, MD) to ensure the feasibility of these primers based on tendency to self- bind, melting temperature, size, and G/C content. Sequences were then tested for specificity using Primer-BLAST (National Center for Biotechnology Information,

Bethesda, MD) by scanning all sequences in Genbank and identifying all DNA sequences that might be amplified with the selected primers and probe.

The designed primers and probe were synthesized (Applied Biosystems, Foster

City, CA). The forward primer (GAAGAACCTTACCTGGGTTTGACATG) and reverse primer (ACAGCCATGCACCACCTGC) were used to amplify an 88-bp portion of the

16S rRNA gene from roughly positions 920 to 1007 of the published M. fortuitum sequence. The probe (CAGGCCACAAGGGAAYVSMYATCTCT) contains a 6‘-FAM fluorescent label.

This development process was repeated to develop primers and probe for MAC using the hsp65 gene. The forward primer (ACCTGCTCAAGGCCGGYGT), reverse primer (GCCTCGGTSGTCAGGAACA), and probe (CCGGTGARGGTGWCCCGT) were

used to amplify a 96-bp portion of the hsp65 gene from roughly positions 830 to 926 of the published MAC sequence.

Specificity of both primer sets was confirmed in the laboratory using M. fortuitum,

M. sengalense, M. chelonae, M. goodii, and M. avium complex (MAC), as well as the closely related species, Rhodococcus equi, Rhodococcus corynebacteroides, Nocardia nova, Nocardia brasiliensis, and unrelated bacteria (Escherichia coli and Vibrio parahaemolyticus), as negative controls. Mycobacterium fortuitum 16S rRNA DNA was used to discern sensitivity of the genus primer/probe set, and Mycobacterium avium subspecies avium Hsp65 DNA was used to examine both the genus and the MAC primer/probe sets. The standard curve used a 10-fold serial dilution of a Mycobacterium fortuitum 16S rRNA PCR amplicon or a Mycobacterium avium subspecies avium Hsp65

PCR amplicon, ranging from 106 to 10 copies. The copies per µL was calculated by multiplying the sample concentration by Avogadro‘s number and dividing this total by the total amplicon length multiplied by the average base pair weight (656.6x109). Each

20µL reaction was run in triplicate, and included 0.9 μM of each primer, 0.25 μM of probe, 1 µL DNA extract and 10µL of a commercial universal qPCR mix (TaqMan Fast

Universal PCR Master Mix, Applied Biosystems, Carlsbad, CA). A 7500 Fast Real-Time

PCR System (Applied Biosystems, Carlsbad, CA) was used to amplify the reaction with the standard fast protocol: initial denaturation at 95°C for 20 sec; 40 cycles of 95°C for 3 sec followed by 60°C for 30 sec. Data was analyzed using the 7500 Software v2.0.3

(Applied Biosystems, Foster City, CA).

Serial Dilutions

Mycobacterium fortuitum (ATCC 6841) and M. avium subspecies avium (ATCC

700898), cultured on Middlebrook 7H10 plates at 35˚C, were used to provide master

standards for subsequent dilutions and extraction protocols. Inocula from each of the two mycobacteria cultures were used to prepare a 1.0 McFarland standard in TE buffer

(Promega, Madison, WI). Suspensions were mixed thoroughly to minimize clumping, and a series of ten-fold dilutions were derived. Aliquots from each dilution were frozen at -20C for subsequent DNA extraction via physical methods with traditional PCR amplification, or extraction via chemical methods with qPCR amplification. A separate aliquot of 75µL from each dilution series was plated onto replicate Middlebrook 7H10 plates and incubated at 35C for 7 days to determine colony counts. Two plates from each dilution series were used to generate approximate colony forming units (CFU) numbers. After 7 days of growth at 35C, the plates from the dilution series were examined. CFU numbers were generated for each plate (Table 1). Standard calculations were then performed to determine the approximate CFU/mL for each dilution in the dilution series.

Extraction by Physical Disruption

Aliquots of bacterial dilutions from the two mycobacteria species were placed into a pre-heated 95C water bath for 15 minutes. The heat-killed solutions were vortexed at low speed, and 400µL of each dilution was added to a new tube containing 25µL of sterile 0.1 mm glass beads. Suspensions were agitated at high speed via vortexer

(Fischer Scientific, Pittsburgh, PA) for 5 minutes, and then centrifuged at 13,500 x g for

15 minutes. Supernatant (~350µL) from each dilution was transferred to a new tube and re-centrifuged at 13,500 x g for 15 minutes. Aliquots of supernatant were frozen at -

20C. This physical DNA extraction is standard operating protocol for handling clinical samples at a NTM diagnostic reference laboratory.

Extraction by Chemical Disruption

DNA was also extracted through chemical disruption via a commercially available kit, Qiagen ® DNeasy Blood and Tissue (Valencia, CA). Cells were lysed using proteinase K and DNA was purified through a silica-membrane. Frozen, serially-diluted bacterial samples were brought to room temperature and centrifuged for 15 minutes at

21,000 x g. The supernatant was carefully removed from each tube to avoid disturbing the pellet. DNA was extracted from the pellets using the Qiagen ® DNeasy Blood and

Tissue Extraction Protocol as per manufacturer‘s instructions (Valencia, CA). DNA extracts were held at 4C until ready for use. Fast qPCR reactions were performed as described above to compare the efficiency of physical versus chemical extraction methods.

Comparison of qPCR and Traditional PCR

To demonstrate the utility of these novel primers, we examined detection limits of

DNA isolated via physical disruption (as described above) via both qPCR and traditional

PCR. 3 µL and 5 µL starting quantities of each of these reactions were subjected to 1% gel electrophoresis PCR was performed using the mycobacterial Tb11 and Tb12 primers as previously described (Shinnick, 1987; Ringuet et al., 1999) to verify successful DNA extraction. Each dilution was subsequently subjected to qPCR testing as described above.

Results

Primer Sensitivity and Specificity

Specificity of the genus specific primers was confirmed by positive qPCR results on DNA from M. fortuitum, M. sengalense, M. chelonae, M. goodii, and M. avium

Complex (MAC), as well as negative results on the closely related species,

Rhodococcus equi, Rhodococcus corynebacteroides, Nocardia nova, and Nocardia brasiliensis, and the unrelated species Escherichia coli and Vibrio parahaemolyticus

(Figure 2-1). DNA from all bacteria within the Mycobacterium genus were amplified above the threshold of detection, while DNA from other species failed to cross the threshold of detection. Similarly, the MAC qPCR assay specificity was confirmed by positive results with MAC and negative results with M. fortuitum, M. goodii, M. sengalense, N. nova, N. brasiliensis, R. equi, R. corynebacteroides, E. coli, and V. parahaemolyticus (Figure 2-2). Sensitivity testing found that both the genus and MAC standard curves demonstrated the ability to detect down to 10 DNA copies (Figure 2-3).

We did not test for detection below this level. The standard curve for the genus-specific qPCR assay had a slope of -3.42 and a high correlation coefficient (R2) of 0.998, and the standard curve for the MAC qPCR assay had a slope of -3.33 and a high correlation coefficient (R2) of 0.996.

Chemical versus Physical Disruption

Since DNA extraction from mycobacteria is known to be challenging, we compared the physical versus the chemical extraction methods of the same dilution series of 2 different mycobacterial species, M. fortuitum and MAC. The limit of detection for the physical disruption extraction of both M. fortuitum and MAC was 103 CFU/mL (Figure 2-

5). Chemical disruption of M. fortuitum and MAC, however, were both detected at the level of 1 CFU/mL (Figure 2-5), a 3-fold increase in the level of detection.

Culture Confirmation

After 7 days of growth, the dilution series of M. fortuitum and MAC colonies were assessed based on colony counts from replicate culture plates. Colonies were

enumerated and calculations were made from the dilution with counts between 50-300 to estimate the number of bacteria throughout the series (Table 2-1).

Comparison of qPCR and Traditional PCR

Results of traditional PCR indicated that only dilution 1 (~107) was visualized for

M. fortuitum, and dilutions 1-3 (~105) of MAC are clearly visualized with partial visualization of dilution 4 (~104) for MAC (Figure 2-4). Thus, the limit of detection for

DNA extracted using physical disruption, visualized using traditional PCR, is 104-107 copies.

For direct comparison, the same DNA was tested using the qPCR technology as described above. This method demonstrated reliable detection down to 103 CFU/mL for both M. fortuitum and MAC (Figure 2-5). Partial detection was also observed at 102 in both M. fortuitum and MAC; at this level, however, with 1/3 of M. fortuitum and 2/3 of

MAC triplicate reactions crossing the threshold of detection. In summary, the level of detection for the physical disruption extraction was above 103 using qPCR and above

105 using traditional PCR.

Discussion

Data from this study demonstrates that the qPCR probes have a notably better sensitivity for NTMs than traditional PCR. DNA extracted by physical disruption and then subjected to standard PCR amplification required at least 106 bacteria or more for detection. Further, BLAST analysis (National Center for Biotechnology Information,

Bethesda, MD) of the Tb11 and Tb12 primers revealed that they can amplify mycobacteria as well as closely related actinomycetales (Steingrube et al., 1997), thus suggesting the potential for false positives using this method.

The qPCR method used in this study provided excellent specificity for mycobacteria (Figure 2-1). The graph of these results show that only mycobacteria were amplified, and that neither the closely related actinomycetales, R. equi, R. corynebacteroides, N. nova, and N. brasiliensis, nor the unrelated bacteria, E. coli and

V. parahaemolyticus, were amplified using these primers. This increased specificity is critical for detecting low levels of mycobacteria. Environmental samples, for example, often contain a host of bacterial species, and being able to say with confidence that the bacteria in a sample are actually mycobacteria could yield new insights into the true composition of the flora. Additionally, the use of qPCR in the analysis of environmental samples allows for quantization of the relative abundance of mycobacteria, which a standard PCR and gel electrophoresis cannot provide.

An additional benefit of qPCR is the added sensitivity. The detection limit of the standard PCR and gel electrophoresis was approximately 106 bacteria per mL, whereas the qPCR data showed detection down to approximately 102 bacteria per mL from the same extraction. This added sensitivity could be of major importance in a clinical diagnostic lab. Mycobacteria are often notoriously difficult to culture and detect, allowing for many inherent delays in the time before proper diagnosis and treatment can begin.

This study also finds chemical disruption to be a better extraction method than physical disruption, detecting two orders of magnitude lower CFU counts. Further, the physical disruption method showed variability of detection at the 102 CFU/mL level; at this level, only 1/3 of M. fortuitum and 2/3 of MAC triplicate samples crossed threshold.

Thus, we can only show confidence in detection down to the 103 CFU/mL level using

the physical disruption method. Chemical disruption provided a 3 fold increase in our detection level, ~1 CFU/mL (Figure 2-5), as compared with physical disruption. It is important, however, to note that mycobacteria are known to clump. Thus, one CFU may not correlate with one mycobacterial cell. Thus, we are not implying that we can detect 1 bacterial cell, despite our results indicating the ability to detect 1 CFU/mL.

Non-culture based, molecular probes provide an improved approach to diagnosing clinical mycobacterial cases. Current standards require culture, which can prove exceedingly difficult given the variety of temperature requirements and media requirements for the different mycobacteria (Falkingham 2003; Griffith et al., 2007).

These specimens are then subjected to PCR amplification using primers shown to amplify non-mycobacterial Actinomycetales as well as mycobacteria (Steingrube et al.,

1997). Here we demonstrate a method for detection that allows for higher sensitivity and specificity, without the biases and need for weeks of growth involved in standard culture-based methods for detection of mycobacteria.

Table 2-1. CFU and calculated CFU for serial dilutions of mycobacterial cultures. Dilution Series Plate Count* Estimated CFU/mL M. fortuitum 1 TNC 1.7 x 107 M. fortuitum 2 TNC 1.7 x 106 M. fortuitum 3 TNC 1.7 x 105 M. fortuitum 4 616 1.7 x 104 M. fortuitum 5 53 1.7 x 103 M. fortuitum 6 3 1.7 x 102 M. fortuitum 7 0 17 M. fortuitum 8 1 1.7 MAC 1 TNC 1.3 x 107 MAC 2 TNC 1.3 x 106 MAC 3 TNC 1.3 x 105 MAC 4 464 1.3 x 104 MAC 5 48 1.3 x 103 MAC 6 1 1.3 x 102 MAC 7 0 13 MAC 8 0 1.3 Note: * TNC indicates a plate with colonies too numerous to count.

Figure 2-1. Genus Specificity Curves. Amplification of bacterial DNA showing specificity of genus-specific primers. Only mycobacteria (M. fortuitum, M. goodii, M. sengalense, and MAC) crossed the threshold for detection, showing specificity of the primers and probe for mycobacteria. Nocardia nova, Nocardia brasiliensis, Rhodococcus equi, Rhodococcus corynebacteroides, and Esherichia coli were also tested, but failed to cross threshold of detection. Additionally, Vibrio parahaemolyticus was also tested, but fluorescence was not detected at any point measured. Error bars = SD from triplicate PCR runs.

Figure 2-2. MAC Specificity Curve. Amplification of bacterial DNA showing specificity of MAC-specific primers. Only MAC DNA crossed the threshold of detection, while M. fortuitum, M. goodii, M. sengalense, Nocardia nova, Nocardia brasiliensis, Rhodococcus equi, Rhodococcus corynebacteroides, Esherichia coli, Vibrio parahaemolyticus did not. Error bars = SD from triplicate PCR runs.

Figure 2-3. Standard Curves. A) Mycobacterium genus-specific and B) MAC-specific standard curves. Curves generated from dilution series ranging from 10,000,000 copies down to 10 copies of DNA. Data indicate detection limit of 10 copies of DNA per mL sample. Error bars = SD from triplicate PCR runs.

Figure 2-4. Traditional PCR Results. Gel electrophoresis to discern limit of detection of A) M. fortuitum and B) MAC. In each gel, lanes 1-2 represent dilution 1, 3-4 dilution 2, etc with the ladder inserted centrally for optimal visualization. For M. fortuitum, only dilution 1 is visualized. For MAC, dilutions 1-4 are visualized. The positive control for both was run in lane 18 on the MAC gel.

Figure 2-5. qPCR Results. qPCR results from dilution series of M. fortuitum and MAC extracted using A) physical and B) chemical disruption methods. The physical disruption method showed reliable detection down to approximately 103 CFUs/mL with partial success at 102 CFUs/mL. Partial success was defined only one or two of the triplicate amplifications crossing threshold. Threshold of detection (0.016921) was the same for each of the 4 datasets presented. The chemical disruption method showed reliable detection for both M. fortuitum and MAC down to approximately 1 CFU/mL in each of the triplicate samples amplified. Error bars = SD.

CHAPTER 3 NONTUBERCULOUS MYCOBACTERIA IN FLORIDA SURFACE AND MUNICIPAL WATERS

Background

Mycobacteria are a highly diverse, species-rich assembledge of bacteria within the genus Mycobacterium. Although mycobacteria are commonly associated with tuberculosis and leprosy (cased by M. tuberculosis and M. leprae, respectively), the vast majority of mycobacteria are free-living environmental species that are opportunistic pathogens of humans and wildlife. Environmental mycobacteria, collectively known as nontuberculous mycobacteria (NTMs), are ubiquitous throughout a broad range of habitats ranging from soils and surface waters, to hot tubs, swimming pools, fish tanks, and municipal water distribution systems (Kirschner et al., 1992; Katila et al., 1995; Langevin et al., 1999; Falkingham, 2003; Primm et al., 2004; Vaerewijck et al., 2005; Field and Cowie, 2006; Sood et al., 2007).

Water chemistry has been reported to influence the presence and distribution of

NTMs in surface waters (Falkinham et al., 2001; Primm et al., 2004), although there is a lack of published data to substantiate this. Notably, Florida and the southeastern US are associated with study sites that have recovered many NTM isolates (Falkingham

1996), and this region appears to have a relatively high incidence of human NTM disease (Dobos et al., 1999; Reed et al., 2006).

NTMs relevant to human disease include species and subspecies within the

Mycobacterium avium complex (MAC), as well as M. kansasii, M. fortuitum, M. chelonae, M. gastrii, M. scrofulaceum, M. ulcerans, and others (Howard and Byrd, 2000;

Vaerewijck et al., 2005; Griffith et al., 2007). Infections with these bacteria are associated with pulmonary and disseminated diseases, hypersensitivity pneumonitis,

cervical lymphadentitis, osteomyelitis, arthritis, keratitis, tenosynovitis, otitis media, corneal infections, endocarditis and immunologic dysfunction (Horsburgh, 1996; Howard and Byrd, 2000; Vaerewijck et al., 2005; Griffith et al., 2007). The frequency of NTM infections appears to be increasing in both healthy and immunocompromised subjects, particularly in the southeastern US (Primm et al., 2004).

Human exposure to NTMs is associated with water exposure, involving some combination of drinking water or ingestion of field-irrigated leafy vegetables, inhalation of aerosols during showering, use of hot tubs, and contact with fish or natural surface waters (Witty et al., 1994; Yajko et al., 1995; Eaton et al., 1995; Falkingham, 1996;

Yoder et al., 1999; Argueta et al., 2000; Pedley et al., 2004; Primm et al., 2004). A variety of mycobacteria, including M. avium, M. chelonae, M. fortuitum, M. gordonae, M. kansasii, and M. xenopi, have been reported from municipal water distribution systems

(Fischeder et al., 1991; Glover et al., 1994; Covert et al., 1999; Falkingham et al., 2001;

Vaerewijck et al., 2005; Falkingham, 2011), and treatment with chlorination or chloramination is believed to have little effect on their presence (Taylor et al., 2000;

Vaerewijck et al., 2005). Water-use patterns and diet may alter individual exposure risk

(Vaerewijck et al., 2005).

This study applied novel Mycobacterium genus- and MAC-specific qPCR probes to discern the presence and density of mycobacteria in surface and municipally- distributed water in Florida, and examine possible water quality relationships that might relate to mycobacterial density in these waters.

Materials and Methods

De-identified areas in Florida were chosen for water sampling based on regional proximity to municipal water distribution systems, housing developments that

incorporate landscape-scale surface water features, and irrigation canals associated with agriculture. The two municipal water distribution systems were sampled to represent source water from the Floridan and the Biscayne aquifer.

Sampling for Mycobacterial DNA

Water samples for mycobacterial quantification were all collected on February 16th,

2010 using sterile 50 mL conical tubes (VWR Scientific, West Chester, PA). Surface waters were collected from the air/water interface. Samples from the two municipal water distribution facilities were taken from sampling valves, immediately up- and downstream from the treatment/distribution facilities, as initial boluses without flushing.

Filled tubes were capped, immediately placed on ice, transported to the laboratory and frozen at -20C until processing, within 48 hours. Duplicate sample tubes were taken from each sampling site for water chemistry that, depending on the analysis, were processed at the time of colletion in the field or laboratory-analyzed the same day.

DNA Extraction

DNA was extracted from thawed water samples using a combination of both chemical and physical disruption. Upon thawing to room temperature and mixing, 1mL of sample was centrifuged at 21,000xg for 15 minutes. DNA was extracted from the resulting pellet using the PowerBiofilm DNA Isolation Kit as per manufacturer‘s directions (MO-BIO Laboratories Inc., Carlsbad, CA). DNA extraction products were held at 4C until ready to use.

Quantitative PCR (q-PCR)

Three primer/probe combinations were used to quantify pan-bacterial DNA, genus- specific mycobacterial DNA, and MAC-specific DNA. For pan-bacterial DNA, the universal primer/probe set targeted the 16s rDNA gene as described by Nadkarni et al.,

(2002). Genus-specific mycobacteria primers and MAC-specific primers targeted the mycobacterial 16s rDNA and hsp65 genes, respectively (Yarnell et al., 2011). Briefly, each 20µL reaction was run in triplicate, and consisted of 0.9 μM for each primer and

0.25 μM for the probe, and contained 1µL of extracted DNA, and 10µL of a commercial universal qPCR mix (TaqMan ® Fast Universal PCR Master Mix, Applied Biosystems,

Carlsbad, CA) using a standard fast protocol. Mycobacterium fortuitum 16s rDNA was used as a control for the genus primer/probe set, and Mycobacterium avium subspecies avium hsp65 DNA was used as the control for the MAC primer/probe set. A standard curve, ranging from 10 to 106 copies was run on each plate, and contained a 10-fold dilution series of a Mycobacterium fortuitum 16s rRNA PCR amplicon or a

Mycobacterium avium subspecies avium hsp65 PCR amplicon. A 7500 Fast Real-Time

PCR System (Applied Biosystems, Carlsbad, CA) was used to amplify the reaction with cycling conditions as follows: initial denaturation at 95°C for 20 sec; 40 cycles of 95°C for 3 sec followed by 60°C for 30 sec (see Figure 3-1).

Conformation of Near-Threshold qPCR Data

Samples with a cycle threshold (Ct) less than 10 copies were confirmed using hemi-nested PCR (this applied to MAC samples only, based on data collected). The confirmatory nested PCR used forward primer (5‘-CGCCACCGGTGAGTACGAG-3‘) and reverse primer (5‘-CGCCTTCTCCGGCTTGTC-3‘) for the first round, and forward primer MAC-FP and reverse primer MAC-RP as a second round (Yarnell et al., 2011).

Amplifications were performed using an Eppendorf AG Mastercycler (Eppendorf,

Hamburg, Germany), using Platinum Taq DNA Polymerase (Invitrogen, Carlsbad, CA) with the following conditions: 15 min denaturation at 95°C, followed by 40 cycles of denaturation at 95°C (30 sec), annealing at 59°C (30 sec), extension at 72°C (30 sec),

with a final elongation step at 72°C for 10 min. Amplification products, along with, positive control DNA, positive environmental DNA, a no template control, and an open air negative control were run in 1% agarose gels, and bands of interest were excised from the gel and extracted using the Qiaquick gel extraction kit (Qiagen ®, Valencia,

CA). Direct sequencing was performed using the Big-Dye Terminator Kit (Applied

Biosystems ®, Foster City, CA) and ABI automated sequencers (DNA Sequencing

Facilities, Center for Mammalian Genetics, University of Florida). Sequences received two-fold coverage in each direction. Sequence results were reviewed using 4Peaks

Software (Mekentosj, Amsterdam, NL), and aligned and visualized using MUSCLE

Software (Edgar, 2004).

Culture Confirmation

Culture techniques were used to confirm the presence of mycobacteria in water samples analyzed using molecular probes. A subset of water samples, representing surface water collection sites as well as municipal water distribution lines, were plated onto selective 7H11 medium plates containing 10µg mL-1 malachite green, 36µg mL-1 cycloheximide, 50µg mL-1 ampicillin, and 25 U mL-1 nystatin, and incubated at 35°C for

> 2 weeks (Bland et al., 2005). Subsequent colonies were subcultured onto standard

Middlebrook 7H11 medium plates, incubated for 10 days, and then acid-fast stained.

Acid fast positive colonies were extracted for DNA, amplified using PCR, and sequenced. Confirmatory nested PCR targeted the ITS1 region of the 16s rDNA gene, using forward primer ITS16F (Whipps et al., 2003) and reverse primer SP2R (Roth et al., 2000) for first round, and forward primer SP1F (Roth et al., 2000) and reverse primer SP2R (Roth et al., 2000) for the second round. Amplifications were performed on an Eppendorf AG Mastercycler (Eppendorf, Hamburg, Germany) using Platinum Taq

DNA Polymerase (Invitrogen, Carlsbad, CA) as follows: 15 min denaturation at 95°C, followed by 40 cycles of denaturation at 94°C (1 min), annealing at 58°C (1 min), extension at 72°C (1 min), with a final elongation step at 72°C for 10 min. Products were prepared for sequencing as described in the previous section.

Water Chemistry

Temperature, pH, conductivity, and dissolved oxygen were measured in the field using a YSI-pH10 meter, a YSI-33 SCT conductivity meter, and a YSI-55 dissolved oxygen meter (YSI Incorporated, Yellow Springs, OH), respectively. Other analyses, discerned using standard EPA laboratory methods (EPA, 1993; Clescerl et al., 1999), included dissolved organic carbon by (SM 5310-C method), nitrates+nitrites (EPA

353.2 method), total and dissolved phosphorous (EPA 365.1 method), total and dissolved Kjeldahl nitrogen (EPA 351.2 method), and total and dissolved nitrogen by calculation.

Statistics

Statistical differences between sampling sites were assessed for water chemistry parameters and relative bacterial loads using a Mann-Whitney test, and correlations were evaluated using Pearson correlation. Statistical tests were run in SPSS (IBM

Corporation, Somers, NY). P values of < 0.05 were considered statistical significant.

Results

Quantification

Bacterial counts were estimated based on qPCR reactions against a standard curve. Estimated bacterial counts, based on number of DNA copies, revealed that the pan-bacterial count >>> mycobacteria genus count > MAC count (Figure 3-1). The majority of the bacteria detected were not mycobacteria. Mycobacteria, including MAC,

however, were detected in all samples tested from surface waters associate with housing developments, agricultural irrigation canals, and municipal water distribution systems. Several patterns emerged from the bacterial count estimates from municipal water distribution systems: (1) the number of total bacteria in Floridan and Biscayne aquifer sources was reduced by municipal treatment prior to distribution (P=0.002)

(Table 3-2); (2) distribution water from Municipal System 1 had higher mycobacterial genus counts than Municipal System 2 distribution water (P=0.027), and (3) estimated counts for mycobacteria were not reduced by municipal treatment (P=0.395; Table 3-2), unlike total bacteria counts.

Additionally, while treated municipal samples had lower estimated total bacterial and mycobacteria compared with environmental surface water samples, municipal samples had a notably higher percentage of mycobacteria in the total bacterial load compared to environmental samples (0.03 –1.07%, compared with 0.51-2.63%;

P=0.001). This trend appeared to be correlated with pH (P<0.001) across all sample sites (Figure 3-2).

MAC was detected in all municipal and surface water samples tested, albeit near the limit of detection. Therefore, to differentiate whether these data represented true low-level detection (between 1 and 10 DNA copies) or false reads, MAC samples were screened using heminested PCR.

PCR and Sequencing

To rule out false negatives due to limitations of threshold of detection, samples with a cycle threshold (Ct) less than 10 copies were confirmed with heminested PCR.

Only MAC qPCR products showed Ct) < 10 copies, hence hsp65 gene nested primer sets were used for targeting. Heminested PCR products (electrophoresis bands) were

extracted and sequenced, and aligned and compared against published hsp65 sequences in Genbank (National Center for Biotechnology Information, Bethesda, MD) for MAC and other closely related Actinomycetales. Sequences from extracted bands matched published MAC sequences, and were confirmed as MAC using both BLAST analysis (National Center for Biotechnology Information, Bethesda, MD) and manual comparison. These results suggest that bands visualized from the heminested PCR confirmed the presence of MAC in these samples, many at low copy rates.

Culture Confirmation

Mycobacteria were isolated from 4/8 municipal water samples, representing both treatment facilities yielded 10 colonies, and 13/15 surface water samples yielded 21 colonies. Sequence results from these cultured isolates were confirmed to be M. intracellulare, M. abscessus, M. chelonae, M. avium subspecies avium, M. fortuitum, M. immunogenum, M. gordonae, and M. flavescens (Appendix).

Water Quality

Water chemistry data from all sample sites are presented in Table 3-1, alongside estimated bacterial counts. All water quality parameters were graphed (data not shown) and were observed for trends in estimated mycobacteria (genus-level) counts. No trend, positive or negative, from any single water quality parameter was observed in relation to the presence or density of genus-level mycobacteria.

There were, however, differences in water chemistry between discharge (i.e., post- treatment) water from ―Municipal 1‖ and ―Municipal 2‖ distribution systems. Municipal 1 discharge had higher total and dissolved phosphorous (P<0.001) and lower dissolved organic carbon (P<0.001) compared to Municipal 2 discharge samples (Table 3-1). It is not clear, however, if these differences are related to the positive 2-fold difference in

estimated mycobacterial counts (4,239 ± 1,634 versus 2,532 ± 1,119, respectively;

Table 3-1) since other factors could be associated with these differences.

Discussion

This study demonstrated the presence and relative density of NTMs in environmental and municipal water samples in Florida using quantitative PCR methodology. Analyses from surface water samples taken proximal to housing communities, agricultural irrigation canals, as well as municipally-distributed potable water, all revealed the presence of mycobacteria including species within the

Mycobacterium avium complex. Estimated total bacterial counts and genus-level mycobacteria counts from environmental samples were within the range of observations previously reported (Table 3-3).

In general, qPCR-based estimates of bacterial counts are typically greater than culture-based CFU inventories due to the comprehensive detection of all DNA by qPCR

(Boehm et al., 2009; Walters et al., 2009; Bae and Wuertz 2009; Lavender and

Kinzelman 2009; Byappanahalli et al., 2010). Thus, it is reasonable that our estimates are at the higher end of the range of reported culture-based data (Table 3-3).

Growth of mycobacteria within potable water distribution systems has been well- documented (Fischeder et al., 1991; Glover et al., 1994; Covert et al., 1999; Falkingham et al., 2001; Falkingham, 2011). This study demonstrated the presence of mycobacteria and MAC species within two drinking water distribution systems in South Florida. These data also suggest that the effectiveness of chloramine treatment of municipal distribution water for removing or reducing mycobacteria is relatively poor compared with the overall bacterial population (Table 3-2).

Resistance to chemical disinfectants such as chlorine, monochloramine (the disinfectant used in these facilities), chlorine dioxide, and ozone, is associated with the relatively high concentration of mycolic acids in mycobacterial cell walls that imparts a reduced efficacy of these treatments on mycobacteria (Taylor et al., 2000; Vaerewijck et al., 2005). Thus, differences noted between the municipal post-treatment waters may be associated with (1) differences in the mycobacterial species/strain composition in the respective (Floridan versus Biscayne aquifer pre-treatment) source waters, and (2) the species composition of mycobacteria post-treatment based on a combination of both species- and strain susceptibility to disinfectants (Taylor et al., 2000). These differences are reasonable since the two municipal systems are supplied by two different aquifers, Floridan and Biscayne, that vary in water quality (Fernald and

Purdum, 1998). Municipal samples, regardless of the aquifer water source, had a higher relative proportion of mycobacteria than environmental samples (Table 3-1), associated primarily with the effective reduction of the pan-bacterial population associated with chloramine disinfection.

Small sample sizes in this initial study precluded comparison of mycobacterial loading between pre-treatment sources. In an effort to better understand post-treatment differences, however, water quality parameters were examined for differences between the two municipal distribution facilities. Municipal 1 post-treatment samples revealed higher levels of total and dissolved phosphorous compared with Municipal 2 post- treatment samples (Table 3-1).

The source of phosphorous in Municipal 1 distribution water was the addition of

AquaCros (Harcrocs Chemicals Inc., Kansas City, Kansas) at the water treatment

facility. This additive, used to reduce pipe corrosion within older neighborhoods, was not added to Municipal 2 post-treatment waters, and elevated phosphorous was not noted in these waters. Phosphorous has previously been linked to enhanced microbial growth (Appenzeller et al., 2001; Vaerewijck et al., 2005; Torvinen et al., 2007), and so mycobacterial growth may be aided by added phosphorous, especially in municipal sources where general bacterial competition is reduced by chlorinamination.

Mycobacterial numbers relative to total bacteria counts may be influenced by pH.

When we evaluated water quality parameters and the percentage of mycobacteria relative to the total bacterial counts in both municipal and surface waters, a higher mycobacterial contribution to the pan-bacterial counts was positively correlated with higher pH (Figure 3-2). This is consistent with M. fortuitum studies conducted by

Farooq et al. (1977) where NTM survival was higher at pH 10.1 compared with pH 5.7 to 10.1.

Phosphorus levels and the relationship between pH and the relative contribution of mycobacteria to the total bacterial load may be important constructs to focus on in follow-up studies. However, water quality and biological factors driving the growth of environmental mycobacteria remain unclear. Regardless, this study contributes to a growing body of literature demonstrating the growth of mycobacteria within municipal distribution lines (Falkingham et al., 2001; Briancesco et al., 2010; Falkingham, 2011); this time in South Florida.

Previous studies have demonstrated high exposure rates to MAC in South Florida, based on antigenic skin reactivity (Reed et al., 2006), implying there must be an environmental source for these in this area. Our study found NTMs and MAC to be

ubiquitous in the environment, albeit with MAC at very low levels. The apparent low levels found in our study does not eliminate these as possible sources of infection.

The minimum infectious dose, defined as the minimum amount of bacteria within a sample required to cause disease, is frequently used to determine the infectious level of a certain quantity of disease organisms. Exposure to doses below this level will not result in disease, whereas doses in excess will, with increasing probability, result in disease (French et al., 2002), such that a large inoculum of organisms increases the chance of infection (McCue, 1990). This has been demonstrated in organisms such as

Salmonella, E. coli, Foot-and-mouth disease virus, and mycobacteria (Shepard et al.,

1965; Blaser and Newman, 1982; Tuttle et al., 1999; Alexandersen et al., 2002; French et al., 2002). The infectious dose levels also vary between modes of transmission

(French et al., 2002; Alexandersen et al., 2002), which is particularly significant for pathogen transmission from the environment, where numbers are often low (Smith and

Rose, 1990).

Mycobacteria are thought to grow along the length of distribution lines in the form of biofilms (Falkingham et al., 2001; Briancesco et al., 2010; Falkingham, 2011).

Indeed, mycobacterial numbers were previously shown to be on average 25,000-fold higher in distribution systems than immediately post-treatment (Briancesco et al., 2010) consistent with the idea of growth along the length of the pipeline. Therefore, NTM numbers at the household level could potentially be higher than those detected close to the treatment facility (Falkingham et al., 2001). Within the home, NTMs may form biofilms in stagnant water lines, washing machines, whirlpools, showerheads, etc., and

may provide exposure levels leading to infection (Primm et al., 2004; Griffeth et al.,

2007; Feazel et al., 2009).

Biofilms undergo a continuous process of growth and detachment of aggregates, which can lead to the ingestion or inhalation of condensed infective doses (Hall-

Stoodley and Stoodley, 2005). This detachment process is well-studied under fluid flow and is affected by changes in flow rate (Stoodley et al., 1998; Stoodley et al., 2001).

Detached particles can contain high numbers of organisms within a small area such that pathogen cell densities may reach 107 cells/cm2 (Hall-Stoodley and Lappin-Scott, 1998).

Such detachments have been previously noted in mycobacteria (Asuncion et al., 1999;

Stoodley et al., 2001), and so it has been suggested that NTM biofilms may disperse large clumps of mycobacteria into waters and aerosols (Hall-Stoodley and Stoodley,

2005) thereby potentially contributing to NTM transmission by facilitating a sufficient infective dose to be ingested, inhaled, or abraded into the skin (Hall-Stoodley and

Stoodley, 2005).

South Florida is densely populated, has an ever-increasing high-risk population of children, elderly, and immunocompromised persons, and is a major tourist destination.

Therefore, understanding the ecology of NTMs in south Florida is important and requires further exploration. This study attempted to identify variables regulating the growth of NTMs in both environmental and municipal sources using a quantifiable method of detection. This study suggests mycobacteria are ubiquitous in surface and municipal distribution waters, and that further studies are essential in order to discern environmental reservoirs, exposure pathways and risk factors associated with NTMs and associated infections.

Table 3-1. Water quality data, pan-bacterial counts, mycobacteria counts and percentage mycobacteria from municipal and environmental water samples. A B C D E F G H I J K L M Mun1 2.49 243,526 6,068 439 0.13 6.06 8.89 0.44 0.44 0.95 1.00 20.1 Mun1 1.91 213,372 4,079 446 0.13 6.92 8.98 0.43 0.44 1.06 0.83 22.3 Mun1 1.47 213,247 3,141 428 0.18 7.06 8.64 0.43 0.42 1.04 1.05 21.0 Mun1 0.99 265,742 2,639 429 0.10 6.95 8.94 0.41 0.42 1.10 0.91 21.5 Mun1 2.63 231,582 6,080 411 0.14 6.51 9.01 0.48 0.46 0.98 0.98 18.1 Mun1 2.35 265,819 6,234 423 0.16 7.13 8.91 0.42 0.41 1.02 0.70 20.8 Mun1 1.28 195,934 2,500 407 0.18 4.67 9.00 0.41 0.40 0.97 0.90 20.9 Mun1 2.02 157,000 3,168 405 0.15 5.91 8.99 0.41 0.40 0.93 0.94 21.4 Avg 1.89 223,278 4,239 424 0.15 6.40 8.92 0.43 0.42 1.01 0.91 20.8 Stdev 0.60 36,705 1,634 15 0.03 0.84 0.12 0.02 0.02 0.06 0.11 1.2

Mun2 1.31 295,539 3,885 195 1.58 9.34 9.06 0.03 0.03 1.29 1.19 19.0 Mun2 0.99 157,839 1,557 193 1.42 7.65 9.10 0.02 0.03 1.20 1.18 20.5 Mun2 0.99 150,679 1,490 193 1.78 7.81 9.09 0.03 0.03 1.13 1.12 19.4 Mun2 1.29 194,524 2,515 204 1.39 5.45 9.10 0.02 0.02 1.04 1.07 20.5 Mun2 2.48 181,850 4,517 228 1.80 6.46 9.13 0.04 0.02 1.14 0.94 17.9 Mun2 0.70 250,870 1,764 198 1.57 5.07 8.87 0.03 0.03 1.24 1.03 15.0 Mun2 1.11 233,261 2,595 183 1.67 5.06 8.87 0.04 0.03 1.03 1.06 15.0 Mun2 0.51 379,868 1,929 180 1.67 5.42 9.05 0.02 0.02 1.04 0.79 23.2 Avg 1.17 230,554 2,532 197 1.61 6.53 9.03 0.03 0.03 1.14 1.05 18.8 Stdev 0.60 77,713 1,119 15 0.15 1.58 0.10 0.01 0.01 0.10 0.13 2.8

SurfA 0.09 15,437,705 13,355 2,110 23.80 6.50 7.80 0.06 0.03 2.83 2.78 18.0 SurfA 0.34 1,641,078 5,586 1,150 29.40 7.60 8.10 0.05 0.05 3.50 3.49 16.0 SurfA 0.47 3,030,670 14,147 1,490 29.70 12.50 8.50 0.03 0.03 2.85 2.64 16.5 SurfA 1.07 14,919,250 160,337 1,250 33.10 6.20 8.80 0.79 0.15 6.83 3.75 20.0 Avg 0.49 8,757176 48,356 1,500 29.00 8.20 8.30 0.23 0.07 4.00 3.17 17.6 Stdev 0.42 7,439,363 74,754 431 3.85 2.93 0.44 0.37 0.06 1.91 0.54 1.8

SurfB 0.32 4,278,803 13,604 425 12.50 9.40 8.20 0.09 0.08 0.94 0.76 18.5 SurfB 0.50 1,291,208 6,425 415 12.50 8.80 8.10 0.09 0.08 0.97 0.84 18.0 SurfB 0.39 2,829,216 10,983 410 13.20 8.90 8.30 0.09 0.08 0.59 0.53 18.0 Avg 0.40 2,799,742 10,338 417 12.73 9.03 8.20 0.09 0.08 0.83 0.71 18.2 Stdev 0.09 1,494,016 3,633 8 0.40 0.32 0.10 0.00 0.00 0.21 0.16 0.3

SurfC 0.37 2,753,082 10,099 355 12.20 7.50 7.50 0.12 0.11 0.95 0.66 17.0 SurfC 0.03 13,072,934 3,901 360 12.20 9.70 7.70 0.09 0.08 0.73 0.55 16.5 SurfC 0.28 6,578,069 18,141 405 11.10 7.70 7.50 0.17 0.12 0.93 0.83 18.0 SurfC 0.17 7,711,563 13,395 230 7.52 9.70 7.60 2.41 2.35 0.65 0.52 15.6 SurfC 0.51 1,899,708 9,626 240 12.90 9.40 8.00 0.05 0.02 1.05 0.69 17.0 SurfC 0.51 2,409,751 12,394 320 12.50 8.80 8.10 0.06 0.03 1.42 1.13 19.0 SurfC 0.50 1,079,535 5,408 350 9.33 9.70 8.20 0.02 0.02 1.29 1.30 18.0 SurfC 0.57 2,684,834 15,254 285 8.18 8.70 8.10 0.04 0.03 0.67 0.64 19.0 Avg 0.34 5,072,092 10,423 323 11.11 8.93 7.80 0.42 0.39 1.00 0.81 17.3 Stdev 0.19 4,083,158 4,798 62 2.11 0.90 0.29 0.83 0.81 0.28 0.28 1.2 Note: A denotes sampling location, B denotes percentage mycobacteria, C denotes pan-bacterial count, D denotes genus count, E denotes conductivity (umhos/cm), F denotes DOC (mg/L), G denotes DO (mg/L), H denotes pH, I denotes total phosphorous (mg/L), J denotes dissolved phosphorous (mg/L), K denotes total nitrogen (mg/L), L denotes dissolved nitrogen, and M denotes temperature (C°).

Table 3-2. Comparison of estimated total bacterial and mycobacterial counts from aquifer source water (pre-treatment) and municipally-distributed water (post- treatment) from two municipal distribution systems. System treatment (chloramination) appears to uniformly reduce estimated counts for total bacteria, but not estimated counts for mycobacteria. Sample source Est. total bacterial count/mL Mean Est. Mycobacterial Mean count/mL Municipal 1 Pre-treatment 562,534 711,726 2,604 2,960 860,918 3,316 Municipal 1 Post-treatment 157,000 223,278 2,500 4,239 195,934 2,639 213,247 3,141 213,372 3,168 231,582 4,079 243,526 6,068 265,742 6,080 265,819 6,234 Municipal 2 Pre-treatment 1,111,995 6,502,420 2,976 8,532 11,892, 845 14,088 Municipal 2 Post-treatment 150,679 230,554 1,490 2,532 157,839 1,557 181,850 1,764 194,524 1,929 233,261 2,515 250,870 2,595 295,539 3,885 379,868 4,517

100

Table 3-3. Range of environmental bacterial counts (total bacteria and mycobacteria) reported from previous studies. Reported Bacterial Grouping Estimated CFUs Sample source Methodology Reference Heterotrophic bacterial plate 100 -106 CFU/mL Municipal Water Culture Payment et counts al., 1997 Heterotrophic bacterial plate 105 -106 CFU/mL Municipal Water Culture Norton et al., counts 2004 NTM 10-1 - 102 CFU/mL Municipal Water Culture Falkingham et al., 2001 NTM 104 CFU/cm2 Municipal Water, Culture Falkingham, biofilm swabs 2011 NTM 102-103 CFU/mL Environmental qPCR Jacobs et al., Water 2009 MAC 101- 103 CFU/mL Municipal Water Culture Fattorini et al., 2002 MAC 104 -105 CFU/mL Municipal Water Culture Norton et al., 2004 MAC 10-1- 103 CFU/mL Municipal and Culture von Reyn et Environmental al., 1993 Waters M. cheloneae, M. fortuitum 104–105 CFU/mL Commercial Culture Carson et al., distilled water 1978 M. xenopi 101-105 CFU/mL Municipal Water qPCR Hussein et al., 2009 Total Bacterial Count 105 cells/mL Municipal Water qPCR Yarnell et al., 2011 NTM 103 cells/mL Municipal Water qPCR Yarnell et al., 2011 MAC 100-102 cells/mL Municipal Water qPCR Yarnell et al., 2011

101

Figure 3-1. Estimated counts of total bacteria, mycobacteria, and MAC based on qPCR determination of number of DNA copies.

102

Figure 3-2. Effect of pH on estimated bacteria counts for surface and municipal waters.

103

CHAPTER 4 DEMOGRAPHICS OF NONTUBERCULOUS MYCOBACTERIA CASES IN FLORIDA, 2006-2008

Background

Mycobacteria represent a highly diverse, species-rich group of microorganisms within the genus Mycobacterium. Although M. tuberculosis and M. leprae, the causative agents of tuberculosis and leprosy, are obligate pathogens, the vast majority of mycobacteria are free-living environmental species that can be opportunistic pathogens and are known as nontuberculous mycobacteria (NTM). Over the past 20 years, the number of clinical NTM isolates has exceeded that of tuberculosis isolates within the

United States (US) (American Thoracic Society, 1997; Falkingham, 2002; Sood et al.,

2007). NTMs, such as M. avium, M. kansasii, M. xenopi, M. scrofulaceum, and M. marinum, have been associated with pulmonary disease, bone and soft tissue disease, as well as disseminated diseases (Howard and Byrd, 2000; Vaerewijck et al., 2005).

Disseminated infection with Mycobacterium avium complex (MAC) had been one of the most common bacterial opportunistic infections in adults infected with HIV-1 in the developed world, and at one point was second only to AIDS Wasting Syndrome as the most common cause of death in those patients (Covert et al., 1999; Karakousis et al.,

2004). With the availability of highly active antiretroviral therapy, there has been a dramatic decline in mortality in the US HIV population due to NTM disease (Horsburgh,

1991; Aberg et al., 1998; Horsburgh et al., 2001; CDC, 2002).

However, the incidence of NTM infections appears to be increasing in otherwise healthy subjects (Moore, 1993; Griffith et al., 2007; Marras et al., 2007; Thomson,

2010), with the majority of clinical isolates associated with environmental sources

(Fordham et al., 1993). The majority of these immunocompetent persons also have

104

underlying structural lung disease, such as bronchiectasis or COPD (Griffith et al.,

2007), and so it remains to be seen if disease prevalence is truly increasing in ‗healthy‘ persons, that is persons with no known risk factors. The primary modes of transmission of NTMs to humans appears to be ingestion of water or inhalation of aerosols containing mycobacteria, though more studies are needed to conclusively link the occurrence of

NTM disease and these exposure routes (Falkingham, 1996; Primm et al., 2004; Feasel et al., 2009; Winthrop, 2010). Given the increasing prevalence of NTM infections, a growing population of susceptible, elderly and immunocompromised persons, there is a critical need to better understand the ecology of this group of pathogens, as this may help explain the environmental distribution, and thereby the transmission routes.

There appears to be substantial regional variability in the number of reported NTM diseases across the US, with notably more NTM cases and higher isolation prevalence being reported in persons residing in the Southeastern US, particularly along the

Atlantic and Gulf coasts (Reed et al., 2006; Bilinger et al., 2009). In the 1960‘s, the

Navy found that 60% of men from the Southeastern coastal regions of the US tested positive on Purified Protein Derivative-Battery (PPD-B) antigen tests, but showed no signs of disease (Edwards et al., 1969). To put that in context, only 2-7% of persons from the Northeast were found to react positively to MAC PPD-B tests (von Reyn et al.,

1993). The PPD-B test is used to discern skin reactivity to MAC antigens, and is used to detect antibodies present within a person associated with previous immunologic reaction. Thus, it was believed that these young men were exposed and produced an immune response to mycobacterial antigens, namely MAC, but never developed outright disease (Edwards et al., 1969; von Reyn et al., 1993; Reed et al., 2006). The

105

CDC reported that 41% of NTM isolates nationwide were from the Southeast US

(Dobos et al., 1999). Further, previous studies have documented finding the highest numbers of M. avium complex organisms from brackish swamps and estuaries of the

Southeastern coastal US (Kirschner et al., 1992; Kirschner et al., 1999). The state of

Florida, in particular, was reported to have the highest rate of hospital admissions for

NTM disease out of 11 states surveyed (Bilinger et al., 2009). Bilinger et al. (2009) also reported that prevalence of pulmonary NTM-associated hospitalizations was increasing in Florida for both men and women, a trend they noted only in Florida (Bilinger et al.,

2009), though other studies have indicated similar trends elsewhere (Marras et al.,

2007; Cassidy et al., 2009; Thomson, 2010).

Based on this background, it is relevant to gain an understanding of the epidemiology of NTM disease in Florida. This study evaluated hospital discharge data for trends in gender, race, age, seasonality, payer status, or associated co-illness relative to NTM disease, to determine any potential risk factors. This study begins to unravel the epidemiology of NTM disease within the state of Florida, and will begin to provide linkages to ultimately discern NTM transmission and risk factors in future studies.

Materials and Methods

Data Source

De-identified data were provided by Florida‘s Agency for Health Care

Administration (AHCA) database of hospital discharge diagnoses for all hospital admissions, 2006-2008. The AHCA database contains records of all hospital discharges within the state of Florida (Bean et al., 1992). Records containing an

International Classification of Diseases, 9th Revision, Clinical Modification code (ICD-9

106

code) for NTMs (031.0, 031.1, 031.2, 031.8, and 031.9) were included in the initial screening. Due to low numbers of reported cases for the other NTM codes, only pulmonary NTM, 031.0, and disseminated NTM, 031.2, were included in the final analyses. The study population included all hospitalized persons within the state of

Florida with either pulmonary or disseminated NTM disease listed as a primary or secondary discharge diagnosis.

Data Analysis

I describe NTM cases for year of hospitalization, zip code of home residence, quarter of the year when hospitalized, length of stay, age when hospitalized, sex, race, state of residence, payer status, and co-illness ICD-9 codes associated with the hospitalized individual. Incidences were determined by averaging the caseload across all three years (2006-2008), dividing by the state population estimates for various age and racial categories (US Census Bureau, 2009; Florida Demographic Estimating

Conference, 2010), and multiplying by 100,000. Statistical differences between pulmonary NTM disease observations and disseminated NTM disease observations were assessed for age, race, gender, and payer status, using a chi-square test and linear regressions in SPSS ® (IBM Corporation, Somers, NY). P values of < 0.05 were considered statistically significant.

Multivariate Analysis

Variables of interest for the multivariate analysis were selected based on a review of the literature and prior clinical experience. The variables included in this study were demographic characteristics including age, gender, race, length of stay, payer status, year of hospitalization, quarter of the year hospitalized, as well as co-illnesses including:

HIV, anemia unspecified, candidiasis, cachexia, anemia of other chronic illness, atrial

107

fibrillation, bronchiectasis without acute exacerbation, coronary atherosclerosis of the native coronary artery, obstructive chronic bronchitis with acute exacerbation, heart failure, acute respiratory failure, chronic airway obstruction not elsewhere classified, unspecified acquired hypothyroidism, nondependent tobacco use disorder, fever and other physiologic disturbances of temperature, diarrhea, pancytopenia, acute kidney failure, esophageal reflux, and unspecified protein-calorie malnutrition. The primary outcome of interest was hospitalization with pulmonary NTM disease as compared with all other forms of NTM disease and likewise hospitalization with disseminated NTM disease as compared with all other forms of NTM disease. To determine significant, clinically relevant risk factors for hospitalization with either pulmonary or disseminated

NTM disease, these outcomes were dichotomized and variables compared between the two outcomes. Factors with a P value <0.05 were considered significant. Adjusted odds ratios and corresponding 95% confidence intervals were calculated. Analyses were performed with SAS ® version 9.1.3 (SAS Institute Inc, Cary, NC).

Results

Demographic profiles for pulmonary and disseminated NTM were examined. This study attempted to examine key demographic components of NTM disease, including age, gender, race, quarter of the year hospitalized, payer status, length of stay, and associated co-illnesses within hospitalized patients in the state of Florida. These were examined separately for both pulmonary and disseminated NTM disease. It is my belief that these profiles may help to shed light on why these disease states appear to be increasing in Florida (Bilinger et al. 2009). Data from this study were categorically associated with either disseminated or pulmonary NTM disease based on observations through the SNTC (Michael Lauzardo, personal communication) as well as previous

108

studies suggesting that one type of NTM does not convert to the other. Our dataset does, however, include a limited number of persons (n=49, 0.96%) with both pulmonary and disseminated diseases. Our data set lacked significant year to year variability

(χ2=3.25, df=2, P=0.20), and so our results were aggregated to show totals across all three years of the study.

Pulmonary NTM

A total of 2,535 cases of pulmonary NTM were reported in Florida through the

ACHA database from 2006-2008. I examined potential contribution of race, gender, age, quarter of year, payer status, length of hospitalization, and most common co-illness from these cases.

The racial distribution of NTM hospitalizations was examined for pulmonary NTM cases. White and Black/African Americans had significantly more cases, 70.85 and

19.21%, respectively than persons of other races (Figure 4-1). Normalizing hospitalization incidences by race suggests differences between racial categories

(χ2=24.47, df=5, P<0.001), and indeed, incidence rates were 2-7 fold higher for whites and black/African Americans (Figure 4-2).

Gender was found to differ statistically for normalized hospitalization incidence in older age groups (≥70yo) (χ2=45.94, df=1, P<0.001), although the biological relevance of this difference across all age groups (47.81% male; 52.19% female) is not clear.

Females make up the majority of cases (64.88%) in older (≥70yo) pulmonary NTM hospitalizations (Figure 4-3) and carry the highest incidence rates, up to 22.11/100,000 population (Figure 4-4).

Age distribution for pulmonary NTM cases was evaluated based on decadal age categories. The distribution of normalized hospitalization incidence among age

109

categories was not homogeneous (χ2=17180.54, df=8, P<0.001), and was positively correlated with age (Pearson R=0.91, P=0.001). While both linear and exponential models demonstrated a correlation with increased age and pulmonary NTM caseload, the exponential model was a better fit (R2=0.96), suggesting an exponential increase of risk with age. The highest caseloads appeared in ≥70 year olds (44.14%; Figure 4-3), as did normalized hospitalization incidence (χ2=71.87, df=1, P<0.001) with incidences as high as 19.59/100,000 for both sexes combined (Males ranged from 11.50 to

15.72/100,000; females ranged from 14.98 to 22.11/100,000 population) (Figure 4-4).

Quarter of the year did not appear to influence the number of hospitalized cases reported, roughly 25% for each quarter.

Payer status: the majority of patients were insured by Medicare, 57.84%, surpassing other methods of insurance or no insurance (χ2=3481, df=5, P<0.001)

(Figure 4-7).

Average length of stay of these patients was 11.25 days. My calculated statewide incidence rate of hospitalization of persons with pulmonary NTM disease was 4.41 per

100,000, but dramatic differences were noted in age adjusted categories, up to

22.11/100,000 population (Figure 4-4).

Commonly associated co-illnesses were primarily associated with underlying lung conditions, although other conditions were not uncommon (Table 4-1). To understand the contribution chronic obstructive pulmonary disease (COPD) as a pre-existing condition, data for obstructive chronic bronchitis with and without exacerbation (ICD-9-

CM 491.21 and 491.22), emphysema not elsewhere classified (492.8), chronic obstructive asthma (493.20), and chronic airway obstruction not elsewhere classified

110

(496) were analyzed together. Overall, at least one of these conditions was present in

40.28% of pulmonary NTM cases. An additional 20.4% of patients had a personal history of tobacco use.

Disseminated NTM

A total of 2,326 cases of disseminated NTM were reported through the ACHA database from 2006-2008. I examined potential contribution of race, gender, age, quarter of year, payer status, length of hospitalization, and most common co-illness from these cases.

The racial distribution of NTM hospitalizations was examined for disseminated

NTM cases. White and black/African Americans had significantly more cases, 41.23 and 47.29%, respectively than persons of other race (Figure 4-1). Normalizing hospitalization incidences by race suggests differences between racial categories

(χ2=86.049, df=5, P<0.001) with black/African Americans carrying the highest incidence of all races, 12.47/100,000 population (Figure 4-2). Indeed, incidence rates were 5-11 fold higher for black/African Americans (Figure 4-2).

Gender was found to differ statistically for normalized hospitalization incidence in middle-aged persons (30-49 year old) (χ2=39.24, df=1, P<0.001), although the biological relevance of this difference across all age groups (52.41% male; 47.59% female) is not clear. Males make up the majority of cases (57.39%) in middle-aged, disseminated

NTM hospitalizations (Figure 4-5) and carry the highest incidence rates, up to

9.14/100,000 population (Figure 4-6).

Age distribution for disseminated NTM cases was evaluated based on decadal age categories. The distribution of normalized hospitalization incidence among age categories was not homogeneous (χ2=4246, df=8, P<0.001). Hospitalization caseload

111

was relatively low for <20 year old patients ranging from (0.08-1.38% of cases; 0.03-

0.69/100,000 population), while the majority of cases from other age groups ranged from 8.21-10.96% (Figure 4-5). The exception was in the 30-49 year old groups which had a higher percentage of the total caseload, 23.52-25.32% (Figure 4-5), and represented the highest normalized incidence rates, 7.29-7.77/100,000 population

(χ2=45.10, df=1, P<0.001) (Figure 4-6). Within gender adjusted categories, males aged

40-49 had the highest incidence noted, 9.14/100,000 population (Figure 4-6).

Quarter of the year did not appear to influence the number of hospitalized cases reported, roughly 25% for each quarter.

Payer status: the majority of patients were insured by either Medicare or

Medicaid, 39.46% and 33.62%, respectively, surpassing other methods of insurance or no insurance (χ2=1178.12, df=4, P<0.001) (Figure 4-7).

Average length of stay of these patients was 11.34 days. Our calculated incidence rate of hospitalization of persons with disseminated NTM was 4.03 per 100,000, though incidences as high as 9.14/100,000 population were noted in age/sex adjusted populations (Figure 4-6).

Commonly associated co-illnesses were primarily associated with HIV/AIDS, however other co-illnesses were not uncommon (Table 4-2).

Discussion

Age

Pulmonary NTM disease appears more common in the later decades of life, consistent with data from other studies (Bilinger et al., 2009; Cassidy et al., 2009;

Thomson, 2010). My study found the highest percentage of cases occurred in persons aged 70-79 (22.09%), but with similarly high rates in the 80+ grouping (22.05%). Age

112

adjusted incidences were also found to be the highest in persons greater than 70

(19.59/100,000), consistent with data from other studies (Bilinger et al., 2009). This increased caseload in persons greater than 50 years of age suggests a disease process with an onset in the 5th or 6th decade of life. This may be due to some underlying genetic susceptibility or to the onset of underlying illness, such as COPD (Griffith et al.,

2007). As NTM pulmonary disease is considered a chronic disease (Griffith et al.,

2007), the highest incidence in the elderly likely reflects the accumulation of old cases as well as new cases. For this reason, no further conclusions could be drawn regarding age of onset. Additionally, as persons are more likely to be hospitalized later in the course of the disease, our data may be skewed toward an older population (Bilinger et al., 2009), and as our study was derived from hospitalized patients only, this potential bias can be difficult to rule out. Previous studies from outpatient data, however, have shown a similar trend toward the elderly, making this less likely (Prince et al., 1989;

Huang et al., 1999; Kim et al., 2008). Further, Florida has long been associated with the number of retirees that choose the warmer climate in their later years, so the possibility exists that pulmonary NTM is more common in the elderly due to immigration of retirees from other areas. This is unlikely the sole reason for the noted increases in

Florida as young naval recruits from the southeast were found to have a higher prevalence of NTM exposures as measured by skin hypersensitivity reactions to purified protein derivative (PPD-B), as well as multiple studies linking the southeast to higher exposure rates (Smith, 1967; Edwards and Smith, 1969; Dobos et al., 1999; Reed et al.,

2006; Bilinger et al., 2009).

113

Conversely, disseminated NTM disease appears to be more common in the middle years of life, ages 30-49, and age adjusted incidence rates over this range were similarly found to be highest. Disseminated NTM is closely linked to immunocompromised states including those persons on long term steroid therapy or, most notably, those persons living with HIV (Horsburgh et al., 1985). In HIV negative persons, disseminated NTM is primarily linked to defects in cellular immunity, while in the HIV population it is linked to CD4 counts of less than 75-100/mm3 (Nightingale et al.,

1992). Given that the highest numbers of both new cases of HIV/AIDS and the highest number of cumulative cases of HIV/AIDS occur in persons aged 35-44 and 30-44 respectively (CDC, 2006), it follows that disseminated NTM would also follow this trend.

I believe the younger distribution of disseminated cases of NTM to be correlated with the younger pool of susceptible persons, particularly the HIV infected population.

Gender

Gender was statistically different in pulmonary NTM disease associated hospitalized patients. Older studies from the 1970s-1980s historically documented a male predominance (O‘Brien et al., 1987). Recent studies including both hospitalized discharge data and single site studies, however, have reported a female predominance for pulmonary NTM disease (Prince et al., 1989; Kennedy et al., 1994; Huang et al.,

1999; Kim et al., 2008; Bilinger et al., 2009). Our study found a modest female predominance, 52.19%, within pulmonary NTM cases. Thus, our finding does follow previously reported trends, but the degree of female predominance was not strong enough to draw any definitive conclusions. Looking at age adjusted figures, however, females over the age of 70 were found to have the highest incidence of all age/sex categories examined, up to 22.11/100,000. It is still unclear at this time if pulmonary

114

NTM disease predominates in more recent clinical studies because women outnumber men in the older age groups or if this modest increase in females may be linked to a bias inherent to the data. AHCA data, for example, does not include VA hospitals, and so may be missing large numbers of male subjects, particularly elderly males. Further, a large proportion of patients had COPD as an associated diagnosis (40.3%) and an additional percentage (20.4%) used tobacco products. Since males predominated in the earliest studies of NTM and the majority of these had COPD, it is possible that the risk of hospitalization with NTM is increased for men because of an increased incidence of

COPD and the associated increased risk of hospitalization (Timpe and Runyon, 1981).

Gender was significantly different in disseminated NTM disease in the middle years of life (30-49 year olds). My study found that 52.41% of patients hospitalized with disseminated NTM were male, but 57% were male in the middle-aged groups. As the highest normalized incidence was found in persons in the middle years of life and the majority of HIV infected persons are in the middle years of life (CDC, 2006), I expected gender would follow HIV status, and, indeed, the majority of persons infected with HIV are male (CDC, 2006). Thus, I believe HIV status may be governing this trend.

Race

Pulmonary NTM disease is more common in persons identified as white and disseminated NTM is more common in black/African Americans. Studies have long suggested that NTM disease is increasing in otherwise healthy persons, especially elderly, white females (Moore, 1993; Griffith et al., 2007). I found persons hospitalized with pulmonary NTM within the state of Florida to be white 70.85% of the time, thus supporting these observations, although normalized yearly incidence implied a higher risk for black/African Americans than whites. Disseminated NTM had a more even

115

mixture between black/African Americans (47.29%) and white persons (41.23%), but normalized incidence rates show an increased risk for black/African Americans

(12.47/100,000 population). As HIV incidence rates for African Americans are 6-7 times higher than for whites, the African American predominance was not unexpected (CDC,

2006; Hall et al., 2008). It has previously been hypothesized that African Americans are at an increased risk of tuberculosis (Stead et al., 1985) and here I demonstrate increased risk for both NTM states in black/African Americans, suggesting there may be some underlying genetic susceptibility within African Americans to mycobacterial disease. Further, the concentration within these two races, white and black, implies a possible genetic contribution to susceptibility, but as regional variability is also reported

(Kirschner et al., 1992; Kirschner et al., 1999; Dobos et al., 1999; Bilinger et al., 2009), it is likely that NTM disease has both a genetic and ecological component.

Seasonality

Quarter of the year was monitored as a marker for seasonality. NTM related diseases were reported to be higher in the southeastern United States (Edwards et al.,

1969; Reed et al., 2006), and it has been hypothesized that the higher average temperature and humidity in these areas may favor mycobacterial growth and/or survival within aerosol droplets (Bilinger et al., 2009). These conditions imply that exposure should be roughly year round and so no major difference was expected between quarters (von Reyn et al., 2001). My results demonstrate that NTM caseloads are similar year-round, roughly 25% of the caseload per quarter. These data support the theory that year round exposure may be correct (von Reyn et al., 2001). It does not, however, dismiss the notion that hospitalizations are similar throughout the year, but the timing of exposure(s) may be independent.

116

Payer Status

Payer status was analyzed as a marker for socioeconomic status. In both pulmonary (16.13%) and disseminated (15.00%) NTM diseases, only a small subset of the population carried commercial health insurance. The vast majority of persons relied on government issued insurance such as Medicare and Medicaid. Persons with pulmonary NTM primarily paid with Medicare (57.84%), whereas most persons with disseminated NTM paid with either Medicare (39.46%) or Medicaid (33.62%) (Figure 7).

Medicare is a program designed to provide health coverage for the individuals aged 65+ years of age, and so the increased number of persons with pulmonary NTM paying with

Medicare was not unexpected as the majority of persons with pulmonary NTM are of advanced age. The younger demographics associated with disseminated NTM may explain why fewer persons paid with Medicare, and why there were more Medicaid claims for disseminated NTM patients compared with pulmonary NTM patients.

Interestingly 39.46% of patients with disseminated NTM disease paid with Medicare while only 27.82% were over the age of 60. This implies that a portion of persons with disseminated NTM disease are qualifying for Medicare for reasons other than age.

Ultimately, the majority of persons hospitalized with NTM disease rely on government insurance programs.

Length of Stay

A patient‘s length of stay was analyzed as a marker for how ill these patients truly were. There was no difference between length of stay for pulmonary versus disseminated NTM disease patients (11.25 vs 11.34 days, Z=-0.006, P=0.50). Based on previous studies (DeFrances et al., 2003), however, patients admitted for either pulmonary NTM or disseminated NTM stayed on average twice as long as patients

117

admitted for any reason (4.8 days) or for an infectious disease (6.4 days). This suggests that persons admitted with an NTM disease are likely to be much sicker than the average person admitted for any cause or admitted for infectious diseases. As the admitted time was extended for both pulmonary and disseminated NTM, which have different age distributions, it is not believed that this finding is based solely on age, and may instead be related to the immunocompromised state of individuals infected with

NTM disease. The immune system often fails in the elderly leading to immunocompromised states (Kendall et al., 2003), and persons with disseminated NTM are likely to have HIV or some other underlying immunocompromised state (Howard and Byrd, 2000; Falkingham, 2003). Thus, it is likely the extended hospital stays reflect that NTM patients are often far sicker than average, as many are immunocompromised.

Co-illnesses

Heart disease and COPD were the most commonly associated co-illnesses with pulmonary NTM. Analysis of the most common ICD-9 codes associated with pulmonary

NTM revealed that most patients had underlying heart or lung disease. A previous study similarly found underlying heart and lung disease to be the most commonly associated co-illnesses with pulmonary NTM (Bilinger et al., 2009). Combining percentages from the underlying heart conditions such as hypertension, atrial fibrillation, atherosclerosis of the coronary artery, and heart failure we found that 72.97% of patients admitted with pulmonary NTM had underlying heart disease. It is not clear at this time if the correlation of pulmonary NTM and heart disease is additive, as it seems more likely that the heart disease component is explained by the increased age of persons admitted with pulmonary NTM, as heart diseases are more common in the elderly (Kannel et al., 1998). Underlying lung disease was also common in patients

118

admitted for pulmonary NTM. Bronchiectasis and other underlying structural damage

(COPD) have long been linked to pulmonary NTM disease (Griffith et al., 2007), and were both found as common co-illnesses in previous studies of pulmonary NTM disease

(Bilinger et al., 2009). Similarly, we found bronchiectasis in over 14.79% and COPD in

40.28% of persons admitted for pulmonary NTM. Pneumonia was previously reported as the most common reason for admission (Bilinger et al., 2009), and we found pneumonia in 24.97% of our patients, thereby supporting this as a common co-illness in these patients. Taken together, underlying heart and lung disease appear to be common co-illnesses to pulmonary NTM. It is possible these co-morbidities may drive physicians to hospitalize these patients, and so may be over-inflated in our hospitalized population, as well as skewing towards an elderly population base. The lack of one predominant co-illness is not insignificant, as it supports the possibility that NTM disease is caused by diverse etiologies, and is consistent with single site studies suggesting an increasing proportion of cases with no known risk factors (Prince et al.,

1989; Kennedy et al., 1994; Huang et al., 1999; Kim et al., 2008).

Disseminated NTM was closely linked to HIV. Persons with immune compromising conditions have the greatest risk of contracting disseminated NTM with AIDS carrying the worst prognosis in these patients (Howard and Byrd, 2000; Falkingham, 2003).

NTM disease often begins to manifest in AIDS patients when CD4 counts drop below

100/mm3 with NTM organisms often isolated from the blood, tissue, sputum, and fecal material of these patients (Nightingale et al., 1992; Falkingham 2003). My study found

65.56% of persons with disseminated NTM were co-infected with HIV. Given the decline in both overall health as well as immune failures associated with HIV/AIDS,

119

especially late stage AIDS (Horsburgh, 1991; Karakousis et al., 2004), it was unsurprising to find opportunistic diseases and other diseases associated with general poor health such as candidiasis, diarrhea, cachexia, and dehydration were among the most commonly reported co-illnesses in this study. Other associated co-illnesses, such as anemia (41.79%), electrolyte imbalances (32.94%), and malnutrition (10.45%), are likely to be related to the overall decline. Each of these disease states are too general to make any concrete conclusions, but taken together, it seems that the co-illnesses associated with disseminated NTM are largely driven by the HIV positive status of most disseminated NTM patients.

Several disease states were common to both pulmonary and disseminated NTM.

Hypertension, anemia, HIV, pneumonia, chronic airway obstruction, tobacco use, esophageal reflux, hypoosmolality and/or hyponatremia, hypopotassemia, cachexia, dehydration, protein-calorie malnutrition, and candidiasis were common to both pulmonary and disseminated NTM. The presence of hypertension may be explained by the high prevalence rate within the US population, but this does not hold true for the other co-illnesses common to both NTM diseases. Wasting syndromes have long been associated with AIDS, but in the last decade a growing number of studies have reported pulmonary disease in elderly, thin, white, women (Griffith et al., 2007). It is possible then that being underweight lends some increased risk for the acquisition of NTM disease. It is unlikely that weight-loss itself would have this effect, but cachexia may help explain some of the other states, such as malnourishment, dehydration, and electrolyte imbalances, which may contribute to increased susceptibility.

120

Anemia has previously been reported in both disseminated and pulmonary NTM patients (Havlik et al., 1992; Field and Cowie, 2006; Bilinger et al., 2009), but to my knowledge this is the first study reporting this in both disease states together. Given this long-standing trend, confirmed in the present study, anemia may be a risk factor for all NTM diseases.

Esophageal reflux was found to be common to both NTM diseases in this study.

Anectodal reports suggest that persons taking proton pump inhibitors are at increased risk of pulmonary NTM disease (David Ashkin, AG Holly Tuberculosis Hospital, personal communication). Possible rationale associating reflux disease with NTM disease may include aspiration of regurgitated stomach contents or through irritations in the esophageal lining.

Pneumonia, chronic airway obstruction, and tobacco use have long been documented risk factors for pulmonary NTM disease (Griffith et al., 2007), but it has long been suggested that persons with disseminated NTM did not show lung manifestations (Field and Cowie, 2006). Given these were common in both NTM disease states, I question the validity of observations suggesting persons with disseminated NTM do not show signs of pulmonary disease.

Finally, HIV is common to both pulmonary and disseminated NTM. HIV is commonly associated with disseminated NTM, as described above, but it is far less commonly studied in the context of pulmonary disease. My finding that over 20% of persons with pulmonary NTM also have HIV supports the idea that this group of persons, those co-infected with pulmonary NTM and HIV, are an understudied group of clinical importance.

121

Incidence

Incidence rates for persons hospitalized with pulmonary NTM appears higher than previously reported. Previously, non-AIDS related mycobacterial diseases was estimated to infect around 1.8 out of 100,000 individuals every year in the US (O‘Brien et al., 1987). Statewide incidence of NTM infection, however, cannot be reasonably estimated based on nationwide data since there appear to be regions where differences in health status (e.g., HIV status, nutritional status) may increase susceptibility to disease (Pradhan et al., 2003; Deaton, 2008; Tan et al., 2010). Also, substantial regional variation in reported infections with different NTM organisms (i.e., MAC, M. scrofulaceum, etc.) makes geographic generalization problematic (Marion et al., 2010;

Tan et al., 2010).

To help address these discrepancies, a follow up study by Bilinger et al. (2009) focused on regional incidence rates for 11 individual states. The average incidence in

Florida was 2.1 hospitalizations per 100,000 persons for pulmonary NTM in 1998, increasing to 2.4 hospitalizations in 2005, the highest for all states observed. Incidence of hospitalizations per 100,000 persons for pulmonary NTM from the present study

(2006-2008) was 4.41, 1.8-fold higher compared with 2005 Florida data from the

Bilinger et al. (2009) study. Incidence rates of hospitalization for disseminated NTM were also found to be high in my study, with a statewide incidence of 4.03 per 100,000 hospitalizations. As my rates are not uniform across the state, there are likely some external factors controlling distributions of NTMs within the state and will be explored further in a follow up study.

Multiple hospitalizations for a patient in a given year in the present study cannot be ruled out, based on the de-identified data that was analyzed. Therefore, data reported

122

in this study may slightly overestimate incidence if patients were admitted multiple times for the same disease in a given year.

Nevertheless, the present study does represent data from a hospitalized population. It is important to note that most pulmonary NTM diseases are diagnosed and managed in an outpatient setting, not in hospitals, with approximately 10:1 ratio of outpatient to inpatient NTM cases (Michael Lauzardo and Kevin Fennelly, SNTC, personal communication). Therefore, this study likely grossly underestimates the actual caseload of NTM disease in the state of Florida, whereby ‗total incidence‘ may be some ten-fold higher than ‗hospitalized incidence.‘ This may be particularly true of children with Cystic Fibrosis (CF), a known risk factor of NTM disease (Griffith et al., 2007), as they are likely managed in an outpatient setting. Additionally, the use of outpatient data might allow for inclusion of other mycobacterial diseases such as cutaneous manifestations, which are believed to be the most common NTM manifestation (Griffith et al., 2007). Without a database of outpatient diagnoses, such a study would be extremely difficult at this time, as these are not reportable diseases and rarely report for inpatient treatment. Additionally, the validity of IDC-9 codes for reporting NTM diseases is unclear, and therefore may underrepresent hospitalization incidence since these diseases are rarely observed in hospital evaluations (Bilinger et al., 2009).

Multivariate Analysis

With the multivariate analysis 11 factors were found to be risk factors for hospitalization with pulmonary NTM disease (age, payer status, HIV, candidiasis, bronchiectasis, bronchitis, respiratory failure, tobacco use, fever, pancytopenia, and chronic airway obstruction) and 5 factors were risk factors for hospitalization with

123

disseminated NTM disease (payer status, HIV, cachexia, fever, and pancytopenia)

(Table 4-3).

This study had significant limitations as we only had information on persons hospitalized with NTM disease and not information on all hospitalized persons.

However, The data revealed some interesting results. The results of the multivariate analysis demonstrated that age increases the odds of hospitalization with pulmonary

NTM disease, but only to a small effect. Payer status is significant. The lack of a diagnosis of HIV, fever, and pancytopenia increases your chances of being hospitalized with pulmonary NTM disease as compared with other NTM diseases while a lack of candidiasis, bronchiectasis, bronchitis, respiratory failure, chronic airway obstruction, and a history of tobacco use may be protective against hospitalization for pulmonary

NTM as opposed to other NTM diseases. This suggests these predisposing lung conditions may increase your chances of developing pulmonary NTM disease and is consistent with previous clinical observations (Griffith et al., 2007). Similarly, the payer status was significant for hospitalization with disseminated NTM over other NTM disease while a lack of conditions such as HIV, cachexia, fever, and pancytopenia may be protective against hospitalization with disseminated NTM disease. Again, this suggests that having these conditions would likely increase your chances of developing disseminated NTM disease. Though limited, this study begins to tease out the important risk factors differentiating between hospitalization with pulmonary and disseminated NTM disease within the state of Florida.

Final Thoughts

In summary, NTM cases are likely increasing within the United States, and in particular, Florida (Bilinger et al., 2009). The present study suggests that statewide

124

incidence for pulmonary (4.41/100,000) and for disseminated (4.03/100,000) NTM diseases in Florida was over 1.8-fold higher than previously reported (2.4/100,000;

Bilinger et al., 2009). This study documented the demographic profile of persons hospitalized with NTM disease in Florida.

This study demonstrated that hospitalizations with pulmonary NTM disease were more common in elderly (≥70 year old) white females, whereas hospitalizations with disseminated NTM disease were more common in middle-age (30-49 year old) white and black/African American males. Normalized incidence data for both diseases, however, implies that race, specifically black/African American, is a risk factor. The most common co-illnesses associated with pulmonary NTM disease related to either underlying heart or lung conditions. The most common co-illnesses associated with disseminated NTM disease appear to be driven primarily by health status associated with HIV infection. Coverage of hospitalizations for both diseases was primarily through government-based programs, Medicare and Medicaid, and length of stay was longer than hospitalizations associated with other infectious diseases or any other reason.

125

Table 4-1. Co-illnesses associated with pulmonary NTM disease, Florida‘s Agency for HealthCare Administration (AHCA) database of hospital discharge diagnoses, 2006-2008. ICD-9 Code Description Percentage of cases with co-reported illness 401.9 Unspecified essential hypertension 32.58 486 Pneumonia organism unspecified 24.97 42 Human immunodeficiency virus (HIV) 20.63 285.9 Anemia unspecified 19.57 Chronic airway obstruction not 496 elsewhere classified 17.48 305.1 Nondependent tobacco use disorder 17.04 530.81 Esophageal reflux 15.66 427.31 Atrial fibrillation 14.83 Bronchiectasis without acute 494 exacerbation 14.79 Coronary atherosclerosis of native 414.01 coronary artery 14.16 272.4 Other and unspecified hyperlipidemia 14.12 Obstructive chronic bronchitis with 491.21 (acute) exacerbation 14.00 276.1 Hyposmolality and/or hyponatremia 13.69 276.8 Hypopotassemia 12.35 276.51 Dehydration 11.76 428 Heart failure 11.40 518.81 Acute respiratory failure 11.28 799.4 Cachexia 11.16 244.9 Unspecified acquired hypothyroidism 10.89 263.9 Unspecified protein-calorie malnutrition 10.77 285.29 Anemia of other chronic illness 10.57 112 Candidiasis 10.22

126

Table 4-2. Co-illnesses associated with disseminated NTM disease, Florida‘s Agency for HealthCare Administration (AHCA) database of hospital discharge diagnoses, 2006-2008. ICD-9 Code Description Percentage of cases with co-reported illness 42 Human immunodeficiency virus (HIV) 65.56 285.9 Anemia unspecified 23.13 112 Candidiasis of mouth 19.78 799.4 Cachexia 19.65 285.29 Anemia of other chronic illness 18.66 486 Pneumonia organism unspecified 18.40 401.9 Unspecified essential hypertension 18.27 276.8 Hypopotassemia 17.89 305.1 Nondependent tobacco use disorder 17.46 276.51 Dehydration 15.56 276.1 Hyposmolality and/or hyponatremia 15.05 Fever and other physiologic disturbances 780.6 of temperature regulation 13.03 787.91 Diarrhea 12.94 284.1 Pancytopenia 12.21 584.9 Acute kidney failure, unspecified 11.13 530.81 Esophageal reflux 10.58 263.9 Unspecified protein-calorie malnutrition 10.45 Chronic airway obstruction not 496 elsewhere classified 10.06

127

Table 4-3. Significant results of the multivariate analysis. Variable Chi Square P value Odds Ratio 95% Confidence Limit Pulmonary NTM Disease Age 112.8132 <0.0001 1.025 1.02-1.03 Payer Status 37.0119 0.0007 0.14 0.039-0.560 HIV 80.0749 <0.0001 2.481 2.033-3.027 Candidiasis 4.1407 0.0419 0.823 0.682-0.993 Bronchiectasis 21.1161 <0.0001 0.627 0.513-0.765 Bronchitis 5.0370 0.0248 0.799 0.657-0.972 Respiratory Failure 4.7863 0.0287 0.792 0.643-0.976 Tobacco use 15.1422 <0.0001 0.727 0.619-0.853 Fever 8.2495 0.0041 1.396 1.112-1.752 Pancytopenia 7.3785 0.0066 1.447 1.108-1.890 Chronic Airway Obstruction 4.5173 0.0336 0.830 0.699-0.986

Disseminated NTM Disease Payer Status 28.2359 0.0132 2.98 0.599-14.622 HIV 408.2841 <0.0001 0.115 0.093-0.141 Cachexia 6.2671 0.0123 0.793 0.661-0.951 Fever 4.0926 0.0431 0.809 0.658-0.993 Pancytopenia 13.5337 0.0002 0.617 0.477-0.798

128

Figure 4-1. Racial distribution of pulmonary versus disseminated NTM hospitalizations. White persons predominate in pulmonary NTM while both white and African American persons contribute significantly to disseminated NTM caseloads. Data from Florida‘s Agency for Health Care Administration (AHCA) database of hospital discharge diagnoses, 2006-2008.

129

Figure 4-2. Racial distribution of pulmonary versus disseminated NTM normalized yearly incidence rates. African Americans and Whites have elevated risk in pulmonary NTM disease and African Americans are at increased risk of disseminated NTM disease. Data from Florida‘s Agency for Health Care Administration (AHCA) database of hospital discharge diagnoses, 2006-2008.

130

Figure 4-3. Age and gender distributions of pulmonary NTM cases based on Florida‘s Agency for Health Care Administration (AHCA) database of hospital discharge diagnoses, 2006-2008. Pulmonary NTM cases appear to increase with age (P<0.001).

131

Figure 4-4. Normalized yearly incidence distributions broken down by age and gender for pulmonary NTM disease based on Florida‘s Agency for Health Care Administration (AHCA) database of hospital discharge diagnoses, 2006-2008. Pulmonary NTM disease risk increases with age (P=0.001) with the highest risk in elderly females (P<0.001).

132

Figure 4-5. Age and gender distributions of disseminated NTM cases based on Florida‘s Agency for Health Care Administration (AHCA) database of hospital discharge diagnoses, 2006-2008. Disseminated NTM disease caseload appears highest in middle-age persons, 30-49 years old (P<0.001).

133

Figure 4-6. Normalized yearly incidence distributions broken down by age and gender for disseminated NTM disease based on Florida‘s Agency for Health Care Administration (AHCA) database of hospital discharge diagnoses, 2006-2008. Disseminated NTM disease risk appears highest in middle-age persons, 30- 49 years old (P<0.001), particularly males (P<0.001).

134

Figure 4-7. Payer status for patients hospitalized with either pulmonary or disseminated NTM disease, based on Florida‘s Agency for Health Care Administration (AHCA) database of hospital discharge diagnoses, 2006-2008. The majority of persons admitted with pulmonary NTM paid with Medicare, whereas the majority of persons admitted with disseminated NTM paid with either Medicare or Medicaid.

135

CHAPTER 5 SPATIO-TEMPORAL PATTERNS OF REPORTED NONTUBERCULOUS MYCOBACTERIA CASES IN FLORIDA, 2006-2008

Background

Overview

The genus Mycobacterium represents a highly diverse, species-rich group of microorganisms. Although best known as obligate pathogens and the causative agents of tuberculosis and leprosy, the vast majority of mycobacteria are free-living environmental species known as nontuberculous mycobacteria (NTM), which can be opportunistic pathogens (Vaerewijck et al., 2005). Over the past 20 years, the number of clinical NTM isolates has exceeded that of tuberculosis isolates within the United

States (US) (American Thoracic Society, 1997; Falkingham, 2002; Sood et al., 2007).

NTMs such as M. avium, M. kansasii, M. xenopi, M. scrofulaceum, and M. marinum have been associated with pulmonary disease, bone and soft tissue disease, and disseminated diseases (Howard and Byrd, 2000; Vaerewijck et al., 2005). Indeed, disseminated infection with Mycobacterium avium complex (MAC) was one of the most common bacterial opportunistic infections in HIV-1-infected adults in the developed world, and at one point was second only to AIDS Wasting Syndrome as the most common cause of death in such patients (Covert et al., 1999; Karakousis et al., 2004).

The advent of highly active antiretroviral therapy (HAART) has led to a dramatic decline in mortality attributable to NTM disease among HIV-infected individuals in the US

(Horsburgh, 1991; Aberg et al., 1998; Horsburgh et al., 2001; CDC, 2002).

The incidence of NTM infections appears to be increasing among otherwise healthy subjects (Moore, 1993; Griffith et al., 2007; Marras et al., 2007; Thomson,

2010). But since most of these immunocompetent persons also have underlying

136

structural lung disease such as bronchiectasis or COPD (Griffith et al., 2007), it remains to be seen if disease prevalence is truly increasing in ‗healthy‘ persons, who have no known risk factors.

Most clinical NTM isolates are associated with environmental sources (Fordham et al., 1993). The primary modes of transmission NTMs to humans appears to be ingestion of water or inhalation of aerosols containing mycobacteria, though more studies are needed to conclusively link these exposure routes to the occurrence of NTM disease

(Falkingham, 1996; Primm et al., 2004; Feasel et al., 2009; Winthrop, 2010). Given the increasing prevalence of NTM infections, a growing population of susceptible, elderly and immunocompromised persons, there is a critical need to better understand the ecology of this group of pathogens, as this may help explain the environmental distribution, and, in turn, the transmission routes.

Immunologic, epidemiologic and environmental microbiology studies together suggest that the southeastern US may have higher NTM exposure risk compared with other parts of the country (Falkingham, 1996; Dobos et al., 1999; Reed et al., 2006), but do not provide the basis for definitive conclusions. Nevertheless, this literature sets the stage for questions about risk factors in the Southeast.

Clinical NTM cases have a particularly high prevalence along the Atlantic and Gulf coasts compared with other parts of the country (Reed et al., 2006; Bilinger et al.,

2009). Purified Protein Derivative-Battery (PPD-B) antigen tests used by the Navy in the

1960s revealed that 60% of men from the southeastern coastal regions of the US tested positive for MAC exposure (Edwards et al., 1969), compared with 2-7% in men in the

137

Northeast (von Reyn et al., 1993). Further, the majority (41%) of diagnostic laboratory

NTM isolates nationwide were from the Southeast US (Dobos et al., 1999).

The state of Florida, in particular, has higher NTM-related hospital admissions compared with other parts of the country, and hospital admissions continue to increase

(Bilinger et al., 2009). That might be attributable to prevailing environmental conditions such as fresh waters rich in humic and fulvic acids, and brackish coastal swamps and estuaries, which appear to foster the growth of NTMs and MAC (Kopecky, 1977; Kazda et al., 1990; Kirschner et al., 1992; Johnson-Ifearulundu and Kaneene, 1997; Kirschner et al., 1999; Reviriego et al., 2000).

Nonetheless, other geographic regions with differing environmental conditions, such as Ontario, Canada (Marras et al., 2007), Oregon (Cassidy et al., 2009) and

Queensland, Australia (Thomson, 2010). also have seen increasing numbers of cases.

Geospatial Analysis

The ability to discern the distribution and epidemiology of NTM disease is critical, given the increasing numbers of cases in various regions (Marras et al., 2007; Bilinger et al., 2009; Cassidy et al., 2009). Spatial statistics and geographic information systems

(GIS) were used in this study to assess regional patterns of NTM-related hospitalization.

Blackburn (2010) provided a framework for employing Exploratory Spatial Data

Analysis (ESDA) to evaluate spatial patterns of disease case data. ESDA is a set of tools to describe and visualize spatial distributions, identify atypical localizations, detect spatial associations such as hot or cold spots, and suggest spatial heterogeneity

(Haining, 1990; Bailey and Gatrell, 1995; Anselin, 1998; Anselin, 1998).

Spatial autocorrelation, an important ESDA, is used to discern coincidence of value similarity with locational similarities (Anselin, 2000). Tests of spatial

138

autocorrelation are inferential and used to identify spatial patterns. For example, a positive spatial correlation occurs when high or low values for a given variable cluster in space. Likewise, a negative spatial correlation occurs when a geographical area is surrounded by neighbors with dissimilar values. Patterns that exhibit neither positive nor negative autocorrelation can be defined as spatially random. Spatial autocorrelation analyses can be global, aimed at determining if spatial dependence is present across a study region; or local, identifying where on the landscape spatial dependence occurs

(Anselin, 1995). Local spatial autocorrelations can be discerned using local indicators of spatial association, LISA (Anselin, 1995), which evaluates the spatial relationship between attribute values of a local geographic unit (e.g., county, zip code, etc.) with those of its neighboring units in a hypothesis testing framework that compares local observations against a spatially random distribution. Anselin (1995) introduced the local

Moran‘s I statistic, which decomposes the global Moran‘s I to determine the contribution of each aerial unit to the global measure, identifying where on the landscape patterns of autocorrelation are occurring.

While NTM disease affects both humans and animals, its ecology is complex and its distribution remains poorly understood (Pedley et al., 2004; Vaerewijck et al., 2005;

Winthrop, 2010). A number of studies have evaluated the physiological niche of these organisms within the environment (Edwards et al., 1969; von Reyn et al., 2001; Reed et al., 2006; Sood et al., 2007; Langevin et al., 2008; Winthrop, 2010), but few have addressed the geographical distribution of infections or the distribution of environmental parameters that may promote mycobacterial growth (Bloch et al., 1998; Claudia et al.,

2004; Duker et al., 2004; Judge et al., 2005; Chan-Yeung et al., 2005; Rushton et al.,

139

2007; Jacobs et al., 2009; Charti and Kazemnegad, 2010). Other studies focusing on the geographical distribution of M. tuberculosis or M. leprae caseloads (e.g., Claudia et al., 2004; Chan-Yeung et al., 2005; Charti and Kazemnegad, 2010), do not aid discerning of patterns for NTM, because those organisms are not environmental, and they have very different modes of transmission (Vaerewijck et al., 2005).

Geospatial studies of NTMs, while limited, give interesting results. Environmental studies suggest that NTMs may be influenced by arsenic levels in soils (Duker et al.,

2004), water chemistry conditions (Jacobs et al., 2009), and host species habitat (Judge et al., 2007). Human NTM diseases have been associated with areas of poverty and high HIV prevalence (Bloch et al., 1998), and general health deprivation (Rushton et al.,

2007). Extensive efforts are clearly warranted to provide a better understanding of environmental risk factors associated with NTM disease. Thus, the purpose of the current study was to: 1) evaluate the spatial distribution of NTM hospitalization incidence in Florida across a three year period for pulmonary and disseminated cases in both HIV-positive and HIV-negative persons, and 2) employ LISA to test for spatial autocorrelation of NTM cases of each group in the state.

This is the first study to examine the geographical distribution of NTM disease in

Florida using geospatial modeling techniques. These efforts are aimed at identifying local hotspots of cases that may inform hypotheses on NTM transmission and risk factors for future studies.

Materials and Methods

Data Source

De-identified data were provided by the Florida Agency for Health Care

Administration (AHCA) database of hospital discharge diagnoses from 2006-2008

140

(Bean et al., 1992). Records containing an International Classification of Diseases, 9th

Revision, Clinical Modification code (ICD-9 code) for NTMs (031.0, 031.1, 031.2, 031.8, and 031.9) were included in the initial screening. Only pulmonary NTM, 031.0, and disseminated NTM, 031.2, were included in the final analyses because of low numbers of reported cases for other NTM codes. The study population included all hospitalized persons within the state of Florida with either pulmonary or disseminated NTM disease listed as a primary or secondary discharge diagnosis.

Data Analysis

A queryable database was constructed and linked to a zip code data layer in a

GIS. To estimate annual zip code-level incidence, case counts were identified to the zip code by year, divided by US Census Bureau population estimates for 2008 (US Census

Bureau, 2008), then multiplied by 100,000. Population estimates for 2008 were available for most zip codes, and when they were not, federal estimates for 2005 (US

Census Bureau, 2005) were used. Incidence of NTM-related hospitalizations was determined using the field calculator within ArcGIS software (ESRI, Redlands, CA), and subsequently mapped using ArcGIS software to the zip code level. A guide map was created to visualize major roadways, cities and waterways (Figure 5-1).

Confidentiality

Confidentiality was maintained through HIPPA standards and approved protocols reviewed by the University of Florida Institutional Review Board (IRB). Data from any zip codes with fewer than 20,000 persons were masked (Leitner and Curtis, 2006). It was determined that 273/946 zip codes (29%) required masking. One method is to dissolve (i.e., adjoin) zip codes with neighboring zip codes to form larger units with >

20,000 persons (Curtis et al., 2011). Since the primary focus of this project was to use

141

local measures of autocorrelation to evaluate local (zip code level) spatial patterns and identify hot spots of NTM hospitalizations (Anselin, 1995), such modifications in spatial aggregation would greatly affect the outcome of the analyses (Grubesic, 2008).

Instead, a spatial masking strategy was employed that preserved approximate spatial relationships while removing specific identifiers of a zip code, such as exact shape and size, through the use of a Thiessen polygon surface (Curtis and Leitner, 2006; Curtis et al., 2007). This method works by assigning a centroid (center point of the spatial unit) to the geographic center of the zip code in a GIS layer, creating a series of vertices at the midpoint of any two points in the distribution, and connecting those vertices to create unrecognizable polygons in the approximate location of the zip codes (Brassel and Reif,

1979; Anselin et al., 2006).

To prevent re-engineering of specific zip codes (Curtis et al., 2006; Brownstein et al.,

2006; Curtis et al., 2011), an additional layer of masking was needed. A stratified random sampling approach was used to generate a randomly located point within each zip code to replace the true centroid (Epstein et al., 2002). These points do not directly correspond to the center of a zip code boundary, but may lie anywhere within the zip code. Thiessen polygon boundaries were then drawn around the new, masked points using the Proximity tool in ArcGIS Toolbox 9.3.1. Thus, this polygon surface maintained the spatial patterns of the zip code data with relative polygon locations, but the locations were not directly linked to the actual zip code of reporting. Figure 5-1 illustrates the

Thiessen polygon surface and references several landmarks used to describe cluster patterns.

142

Smoothing

Data sets for many diseases are based on relatively low case numbers and variable population data across a study area (Beroll et al., 2007), and accommodations must be made for this. Bayesian methods have been employed to reduce the heterogeneity in variance associated with disease reporting data prior to performing hypothesis test- based analyses (Happold et al., 2008; Chen et al., 2010). Spatial empirical Bayes smoothing, also referred to as a shrinkage metric, applies the global incidence estimate as the ‗prior‘ probability. Areas with low variance, associated with larger populations, are adjusted little, while high variance estimates associated with small samples sizes

(rural zip codes) are adjusted toward the regional mean. The resulting smoothed incidence rates are arguably better representations of true spatial patterns as compared with raw incidence rates (Haddow and Odoi, 2009). In the absence of smoothing, it is plausible that the Local Moran‘s I calculation could overestimate spatial outliers, such as single zip codes with a high NTM incidence surrounded by zip codes with low or no incidence, as those NTM rates may exceed estimates of the actual occurrence of disease (Owusu-Edusei and Owens, 2009). The spatial Bayes smoothing routine in

GeoDa 0.9.5-i (GeoDa Center, Tempe AZ) was used to adjust incidence rates from low case numbers in some zip codes (Bernardinelli and Montomoli, 1992; Clayton and

Bernardinelli, 1997; Bithell, 2000; Odoi et al., 2003). Our results indicated similar calculated average and maximum incidences from raw and smoothed data, suggesting that this method did not significantly change our results (Table 5-1).

Spatial Analyses

To evaluate spatial autocorrelation and dependency for pulmonary and disseminated NTM hospitalizations both with and without HIV co-illness for each year,

143

the univariate Local Moran‘s I statistic (LISA) (Anselin, 1995) was employed in GeoDa

0.9.1-i. A 1st order rook weights matrix was used to define neighbors as Thiessen polygons that shared a common boundary. Significance was tested with 999 perumations and α = 0.05. Outputs were defined as spatial clusters when polygons with high or low incidence are associated with neighbors of similar high or low incidence

(high-high and low-low clusters, respectively). Conversely, spatial outliers are defined as a polygon of high or low incidence surrounded by opposite values (Anselin, 1995).

Twelve incidence maps and 12 LISA analyses were generated.

Results

This study determined the incidence of pulmonary and disseminated NTM disease with and without HIV co-illness in Florida based on hospital discharge diagnoses. HIV co-illness occurs in 20.63% and 65.56% of pulmonary and disseminated NTM disease cases, respectively. The incidence of pulmonary and disseminated NTM hospitalizations were individually estimated for all Florida zip codes and were geospatially analyzed. Geographic location for NTM patients was mapped based on home zip code.

Spatial Distribution of Pulmonary NTM

Incidence of pulmonary NTM hospitalization ranged from 0 to 130.28 per 100,000 persons (Table 5-1). Distribution was wider than for disseminated NTM hospitalization, with approximately 65% of all zip codes reported at least one case of pulmonary NTM for all three years combined (Table 5-1). Cases with HIV co-illness were reported in

21% of zip codes, and without HIV co-illness in 60% of zip codes. Yearly ranges were

9-12% with HIV co-illness and 34-37% without HIV co-illness.

144

Pulmonary NTM hospitalizations with HIV co-illness tended to occur around urban areas such as Jacksonville, Tallahassee, Gainesville, Orlando, Tampa, St. Petersburg,

Sarasota, Naples, Fort Pierce, Boca Raton, Fort Lauderdale, and Miami, as well as around Lake Okeechobee.

Cases without HIV co-illness had a more widespread distribution that included urban areas (Jacksonville, Daytona Beach, Orlando, Tampa, St. Petersburg, Bradenton,

Fort Meyers, Naples, Melbourne, Port St. Lucie, Fort Lauderdale, Miami, and

Homestead), as well as major roadways (I-4, I-10, I-75, I-95, and the Florida Turnpike) and waterways. Waterways included both the Atlantic and Gulf of Mexico coastal areas from central Florida southward, Lake Okeechobee, the Santa Fe River, the Aucilla

River, and in 2008 the addition of the St. John‘s River System (Figure 5-2).

Spatial Distribution of Disseminated NTM

Disseminated NTM hospitalizations ranged from 0 to 121.46 per 100,000 persons

(Table 5-1). Approximately 55% of all zip codes reported at least one case of disseminated NTM disease for all three years combined. Cases occurred in 32% and

41% of zip codes with and without HIV co-illness, respectively. Yearly ranges for disseminated NTM hospitalizations ranged from 17-20% regardless of HIV co-illness status.

Disseminated NTM hospitalizations with HIV co-illness occurred almost exclusively in metropolitan areas such as Jacksonville, Fort White, Tallahassee, Orlando, Tampa,

St. Petersburg, Sarasota, Naples, Fort Lauderdale, Miami, and Homestead, rarely occurring outside of cities. One Fort Lauderdale zip code consistently reported the highest case counts and incidence rates in all three years.

145

Cases without HIV co-illness had a more widespread distribution that included primarily urban areas (Pensacola, Jacksonville, Tallahassee, Orlando, Clearwater, St.

Petersburg, Tampa, Sarasota, Fort Meyers, West Palm Beach, Fort Lauderdale, Miami, and Homestead), as well as major roadways (I-4, I-75, I-95, and the Florida Turnpike) and waterways (St. John‘s River system, the Santa Fe River, the Aucilla River, and

Lake Okeechobee)(Figure 5-3).

Clustering of NTM Hospitalization

Incidence data for hospitalizations for both pulmonary and disseminated NTM disease were analyzed for spatial clustering trends. To rule out the possibility that incidence distributions noted above were due to random chance, a LISA statistic was used to determine non-stochasticity (Anselin, 1995; Fosgate et al., 2002). Geographical trends for both pulmonary (Figure 5-4) and disseminated (Figure 5-5) NTM hospitalizations were analyzed with and without HIV co-illness for 2006-2008, looking for any spatial clusterings, commonly referred to as ‗hot-spots‘ or ‗cold-spots‘ (Pfeiffer et al., 2008; O‘Sullivan and Unwin, 2010). Clusters were noted in all four disease states, suggesting that the distribution of hospitalizations for pulmonary and disseminated

NTM, with and without HIV, are not due to chance alone.

Pulmonary NTM Hospitalizations

Cluster analysis for pulmonary NTM hospitalizations with HIV co-illness demonstrated a linkage to cities (Figure 5-4). Cases with HIV co-illness revealed hot spots, i.e., areas with high incidence surrounded by areas of low incidence, in major metropolitan areas, including Jacksonville, Tallahassee, Orlando, Tampa, West Palm

Beach, Fort Lauderdale, Miami, and Homestead. No hot outliers, i.e., areas with high incidence surrounded by areas of low incidence, were found. Cold spots, i.e., areas

146

with low incidence surrounded by other areas of low incidence, were found in areas containing state or nationally-protected lands, including Four Creeks State Forest,

Nassau Wildlife Management Area, Pumpkin Hill Creek Preserve State Park, Big Talbot

Island State Park, Thomas Creek Conservation Area, Jennings State Forest, as well as areas of the state that are scarcely populated. Cold outliers, i.e., areas with lower incidence rates than the surrounding areas, were found on the edges of cities such as

Jacksonville, Tallahassee, Fort Lauderdale, and Miami.

Pulmonary NTM hospitalizations without HIV co-illness also appeared to cluster primarily around cities (Figure 5-4). Cases without HIV co-illness revealed hot spots in metropolitan areas, primarily coastal cities on bays or estuaries, including Melbourne,

Palm Bay, Port St. Lucie, Stuart, Clearwater, Palmetto, Bradenton, Punta Gorda, Port

Charlotte, Naples, Ocala, and Orlando Suburbs, as well as at the intersection of I-75 and the Florida Turnpike and along the Aucilla River. Two hot outliers were found that demonstrated no clear correlation. Cold spots were noted in areas containing state and nationally protected lands including: Ecofina Creek Water Management Area, Lake

Talquin State Forest, Apalachicola National Forest, Lake Kissimmee State Park,

Kissimmee Prairie Preserve State Park, Everglades National Park, Osceola National

Forest, Ocala National Forest, Greenswamp, Greenswamp Wildlife Management Area,

Withlacoochee State Forest, Waccasassa Bay Preserve State Park, Crystal River

Preserve State Park, Fakahatchee Strand Preserve State Park, Osceola National

Forest, Payne‘s Praire Preserve, and the Lochloosa Wildlife Conservation Area, as well as scarcely populated areas particularly those in the pan-handle. Cold outliers were

147

found on the edges of cities such as Ocala, Orlando, Sarasota, Port Charlotte, Naples, and West Palm Beach.

Disseminated NTM Hospitalizations

Spatial clustering analysis (Figure 5-5) revealed hot spots of disseminated NTM hospitalizations with HIV co-illness in metropolitan areas including Jacksonville,

Tallahassee, Orlando, Tampa, West Palm Beach, Boca Raton, Fort Lauderdale, Miami, and Homestead. No hot outliers were identified. Cold spots were noted in areas containing state and nationally protected lands including the Biscayne National Park,

Everglades National Park, Lake Talquin State Forest, and the Apalachicola National

Forest, as well as portions of the state that are relatively unpopulated or scarcely populated. Cold outliers were found on the edges of cities such as Jacksonville, Lake

Butler, Tallahassee, Orlando, Fort Lauderdale, and Homestead.

For cases without HIV co-illness, cluster data revealed hot spots in cities and towns, including Vernon, Fernandina Beach, Jacksonville, Starke, Lake Butler, Crystal

River, Ocala, Orlando, Sarasota, Port Charlotte, Venice, Vero Beach, Fort Pierce, Port

St. Lucie, and Homestead, as well as at the intersection of points of I-75, I-4, and the

Florida Turnpike. Only one hot outlier was noted, corresponding to the town of High

Springs. Cold spots were noted in areas containing state and nationally protected lands, including Ecofina Creek Water Management Area, Lake Butler Wildlife

Management Area, Santa Fe Swamp Conservation Area, Seminole State Forest,

Crystal River State Preserve, Escambia River Wildlife Management Area, Blackwater

River State Forest, Yellow River Water Management, Three Forks Marsh Conservation

Area, Blue Cypress Conservation Area, Big Talbot Island State Park, Pumpkin Hill

Creek Preserve State Park, and Four Creeks State Forest, as well as scarcely

148

populated areas of the state, many in the northern half of the state. Cold outliers were found on the edges of cities such as Lake Butler, Starke, Orlando, Melbourne, Sarasota,

Port St. Lucie, and West Palm Beach.

Discussion

The main objective of this study was to evaluate the spatial and temporal patterns of pulmonary and disseminated NTM hospitalization rates, respectively, across Florida between 2006 and 2008. Our results clearly show that both pulmonary and disseminated NTM tend to cluster around urban areas. There may be a correlation with roadways and coastal living that needs to be explored further.

The use of hospital discharge data may grossly underestimate the number of cases of NTM disease across the state of Florida. First, our data represents a hospitalized population, whereas most pulmonary NTM diseases are diagnosed and managed in an outpatient setting and that information is not reported to the AHCA database. Based on clinical observations, there may be approximately a 10:1 ratio of outpatient to inpatient NTM patients (M. Lauzardo and K. Fennelly, Southeastern

National Tuberculosis Center, personal communication). Thus, incidence trends are likely to be different for outpatient settings, and may be up to tenfold higher than rates based on hospitalization. Additionally, the use of outpatient data might allow for inclusion of other mycobacterial diseases such as cutaneous manifestations, which are believed to be the most common NTM manifestation (Griffeth et al., 2007).

Additionally, since pulmonary NTM is not considered a common condition, hospitalizations identified by use of ICD-9 codes likely underestimate the impact of pulmonary NTM (Bilinger et al., 2009). The dataset used in the study was de-identified, preventing determination of whether any patients had multiple hospitalizations within a

149

given year. Given the relative rarity of the disease, it is unlikely that this would result in a substantial overestimation of the true impact of pulmonary NTM, however. Taking all these factors into account, it is likely that using hospital discharge billing codes to estimate incidence rates may significantly underestimate the true incidence of disease.

Incidence

Zip code level incidence of hospitalization for pulmonary NTM ranged from 0-

130.28/100,000 persons, and disseminated NTM hospitalizations ranged from 0-

121.46/100,000 persons. Rates are not uniform across the state, so there are likely some external factors controlling the distribution of NTM disease within the state.

Hospitalization incidence rates and maps were generated from reported home zip codes of persons admitted for NTM disease as a means to visualize where these persons reside, hence where they were most likely exposed to their NTM.

Both pulmonary NTM and disseminated NTM hospitalizations demonstrated a fairly broad distribution, representing 65% and 55% of zip codes, respectively, across the three years of this study (Table 5-1). Pulmonary NTM hospitalizations without HIV co-illness had the broadest distribution noted, with 60% of zip codes reporting at least one case in the three-year study period and up to 37% in a one-year period.

Hospitalizations for other NTM states had lower statewide distributions. This was expected, given the close ties of these diseases to immunocompromised states such as

HIV (Covert et al., 1999; Karakousis et al., 2004), and the tendency of immunocompromised persons to settle around cities (Butler, 1969; Wasser et al., 1993;

Glaeser et al., 2000).

HIV is a major co-illness in both pulmonary (20%) and disseminated (65%) NTM hospitalizations (Chapter 4), so spatial distributions were visualized both with and

150

without HIV co-illness to test for inherent trends independent of HIV status.

Differentiation by HIV status demonstrated common as well as unique trends in both pulmonary NTM hospitalizations (Figure 5-2) and disseminated NTM hospitalizations

(Figure 5-3).

Pulmonary NTM hospitalizations with HIV co-illness occurred primarily around metropolitan areas. Relatively few hospitalizations occurred outside of urban areas, and most of those followed interstate patterns or occurred around Lake Okeechobee.

The proximity to urban centers is likely explained by the geographic distribution of HIV- positive individuals who often reside in urban areas, and is not controlled for by normalizing the population.

Pulmonary NTM hospitalizations without HIV co-illness also occurred primarily in metropolitan areas. Unlike pulmonary NTM hospitalizations with HIV co-illness, however, they also spread outside of these urban areas, primarily following interstate patterns, coastal regions from central Florida southward, and the Aucilla River. Coastal living may be a risk factor for pulmonary NTM disease, as almost every coastal zip code from central Florida southward has reported hospitalizations. As this pattern begins in central Florida, there is likely some environmental change occurring at this level.

Temperature differences between northern and southern Florida may explain this trend. Mycobacteria can grow over a wide range of temperatures (George et al., 1980), but many grow best at warmer temperatures (Griffith et al., 2007), often preferring tropical regions of the world (Fine, 1995). The higher average temperatures and humidity in southern Florida are thought to favor mycobacterial growth and survival in aerosol droplets (Bilinger et al., 2009), thereby contributing to increased infection rates

151

and subsequent hospitalizations. The panhandle and northern Florida, however, experience much cooler temperatures than central and south Florida (Greller, 1980;

Cathey et al., 1998), and so may not provide year round growth conditions as favorable as south Florida.

Disseminated NTM hospitalizations also occurred primarily in metropolitan areas.

Cases with HIV co-illness were localized to major cities. This trend is, again, most likely due to the population distribution of HIV infected individuals, as over 65% of persons with disseminated NTM are co-infected with HIV. Removing HIV co-illness appears to widen the distribution of disseminated NTM cases, supporting the above hypothesis.

The majority of persons hospitalized with disseminated NTM without HIV co-illness, however, still resided in cities, with non-urban hospitalizations primarily following interstate patterns with a possible association with select waterways.

Together, several common trends were noted. First, hospitalizations for both pulmonary and disseminated NTM, with and without HIV co-illness, occurred primarily around population centers.

Secondly, non-urban hospitalizations both with and without HIV co-illness appeared to follow interstate patterns. Roads themselves are not thought to pose a risk of infection, but the tendency of people to settle near transportation routes (Lichter and

Fuguitt, 1980) likely explains this effect, suggesting a correlation with areas of increased population. This may be especially true for elderly and immunocompromised persons who would likely live near routes capable of transporting them to and from care centers.

Thirdly, environmental factors may play a role in the distribution of hospitalizations for NTM disease. Hospitalizations may occur more regularly around bodies of water,

152

particularly lake Okeechobee, the Aucilla River, and both coastal regions. These trends need to be further analyzed, but they are not entirely unexpected, as NTMs, including pathogenic strains, have long been known to grow in water (George et al., 1980; von

Reyn et al., 1993; Falkingham et al., 2001; Falkingham, 2011). Temperature may also contribute to the distribution of hospitalizations, as previously discussed.

Cluster Analysis

The trends noted, while interesting, required statistical analysis to determine if they were true patterns or the result of random distribution. LISA cluster analysis tools were used to determine if any areas held significant spatial clustering patterns, and so were non-random.

Pulmonary NTM hospitalizations cluster primarily around urban areas. Pulmonary

NTM hospitalizations with HIV co-illness hotspots tended to occur around major cities, such as Miami. This was expected, given the trends noted on the incidence maps and the increased HIV population in cities (Wasser et al., 1993). Cold spots localized to areas around state or nationally protected lands or to areas of the state that were scarcely populated. This trend does not seem to exemplify a lack of mycobacterial growth in these areas, but more likely reflects the fact that fewer people live and travel to these areas. An inherent assumption of this method is that people are more likely infected close to home. Cold outliers occurred at the edge of cities, emphasizing the settling trends of infected persons toward cities. Taken together, the distribution of hospitalized persons with pulmonary NTM and HIV co-illness appears to be largely driven by human populations, most likely based on settling patterns of those persons infected with HIV.

153

Pulmonary NTM hospitalizations without HIV co-illness clustered in coastal cities on bays and estuaries beginning in central Florida and extending south, as well as at the intersection point of two major interstates and along select rivers. The waterways noted were the same ones noted in the incidence maps, and should be further evaluated. Highway intersection points again likely reflect settling and migratory patterns. Many of these urban hotspots occurred along bays and estuaries, which tend to have more brackish waters. NTMs grow in both fresh and brackish waters (George et al., 1980) with higher growth rates reported in waters containing 1% NaCl (brackish) than in fresh water (George et al., 1980; Pedley et al., 2004), likely explaining the high numbers of NTMs isolated from the tidal waters of large estuaries, like the Chesapeake

Bay (Kane et al., 2007). Contrarily, NTMs do not grow in high salinity situations such as seawater (3% NaCl) (Falkingham et al., 1980), likely explaining the occurrence of hotspots in cities along bays. Cold spots and cold outliers showed the same trends noted above, likely for the same reasons. In summary, hospitalizations for pulmonary

NTM without HIV co-illness appear to positively correlate with major cities, many along bays and estuaries, and to negatively correlate with protected lands or unpopulated areas.

Hospitalizations for disseminated NTM with HIV co-illness appeared again to correlate with the settlement tendencies of infected persons. Hotspots for disseminated

NTM hospitalizations with HIV co-illness, tended to cluster once again around cities, particularly those in south Florida. Cold spots were found to localize to areas containing state and nationally protected lands and relatively unpopulated areas of the state, and cold outliers were again found to correlate with the edges of major cities. Taken

154

together, it appears that disseminated NTM in HIV infected persons also seems to be largely driven by the human population.

Types of hot and cold spots and outliers for disseminated NTM hospitalizations without HIV co-illness were similar to those described for disseminated NTM with HIV co-illness, except that cold spots occurred more often in the northern portion of the state and panhandle.

Persons with an immune deficiency have the greatest likelihood of contracting disseminated disease due to NTMs (Howard and Byrd, 2000), with AIDS carrying the worst prognosis for disseminated NTM disease (Falkingham, 2003). As most immunocompromised persons tend to live in cities (Wasser et al., 1993), the linkage of disseminated NTM to cities is not surprising. It does, however, call to question some of assumptions pertaining to highly active antiretroviral therapy (HARRT). Disseminated

NTM disease used to be a major cause of death in immunocompromised patients

(Covert et al., 1999; Howard and Byrd, 2000; Karakousis, et al., 2004), but it rarely manifests at CD4 counts greater than 100/mm3 (Falkingham, 2003). HARRT therapy is believed to have reduced the number of persons reaching low CD4 counts, and therefore decreased the number of persons experiencing disseminated NTM disease

(Horsburgh, 1991; Aberg et al., 1998; Horsburgh et al., 2001; CDC, 2002). Prior to advanced HAART therapy, M. kansasii was found to spatially track with the HIV-positive population (Bloch et al., 1998). It appears, however, disseminated NTM disease may still track closely with the HIV-infected population despite advances in HARRT therapy.

This leads us to question, despite lowering mortality, has HARRT therapy truly decreased the prevalence? It is possible that immunocompromised persons, especially

155

HIV infected persons, may be colonized with these organisms, rendering them more susceptible to NTM disease.

Final Thoughts

In summary, incidence of NTM hospitalizations at the zip code level demonstrated variability in the occurrence of cases across zip codes with some areas reporting substantially higher incidence rates, up to 124.00/100,000 persons, than previously published rates (O‘Brien et al., 1987; Bilinger et al., 2009) (Table 5-1). Spatial analysis of both pulmonary and disseminated NTM hospitalizations, both with and without HIV co-illness, suggest that pulmonary NTM patients cluster around major cities, most near the coast, while disseminated NTM patients clustered around large metropolitan areas.

Both NTM states likely cluster in urban areas as more immunocompromised persons live within cities (Butler, 1969; Glaeser et al., 2000). For both states, regardless of HIV status, cold spots occurred in state and national protected areas, as well as scarcely populated areas. Taken together, living in urban areas may be a risk factor for increased hospitalizations for NTM disease. Further studies are needed to determine if coastal living or living closer to water is a risk factor, as coastal regions and particular waterways were observed near higher incidence areas in this study.

156

Table 5-1. Caseload, zip codes (zips) reporting, incidence for both raw and smoothed data, and maximum incidence noted in any given zip code for pulmonary and disseminated NTM yearly and aggregated data. Data was derived from Florida‘s Agency for HealthCare Administration (AHCA) database of hospital discharge diagnoses, 2006-2008. NTM category Total Number Percent Florida Spatial Raw max SB cases of zips of zips statewide Bayes incidence smoothed reporting reporting incidence smoothed max incidence incidence Pulmonary NTM 172 86 09.1% 0.93 0.84 108.01 097.40 with HIV 06 Pulmonary NTM 645 329 34.8% 3.36 3.65 105.37 055.30 without HIV 06 Pulmonary NTM 159 92 09.7% 0.83 0.77 046.57 033.70 with HIV 07 Pulmonary NTM 638 330 34.9% 3.30 3.46 051.18 042.90 without HIV 07 Pulmonary NTM 192 115 12.2% 1.01 0.86 075.59 023.70 with HIV 08 Pulmonary NTM 729 358 37.8% 3.78 4.18 121.46 090.70 without HIV 08 Disseminated 543 169 17.9% 2.81 2.48 130.28 124.00 NTM with HIV 06 Disseminated 246 165 17.4% 1.27 1.31 048.36 032.80 NTM without HIV 06 Disseminated 499 170 18.0% 2.63 2.10 053.04 070.80 NTM with HIV 07 Disseminated 249 167 17.7% 1.30 1.43 048.36 034.40 NTM without HIV 07 Disseminated 483 164 17.3% 2.47 2.09 061.46 079.70 NTM with HIV 08 Disseminated 306 193 20.4% 1.59 1.70 080.97 074.30 NTM without HIV 08

157

Figure 5-1. Reference map of Florida showing major roadways, cities, and water systems.

158

Figure 5-2. Incidence maps of pulmonary NTM disease, 2006-2008, with and without HIV co-illness.

159

Figure 5-3. Incidence maps of disseminated NTM disease, 2006-2008, with and without HIV co-illness.

160

Figure 5-4. LISA maps showing cluster analyses of pulmonary NTM disease, 2006- 2008, with and without HIV co-illness.

161

Figure 5-5. LISA maps showing cluster analyses of disseminated NTM disease, 2006- 2008, with and without HIV co-illness.

162

CHAPTER 6 FINAL CONCLUSIONS

Overview

The number and diversity of mycobacteria may be altered regionally by the chemistry of natural waterways, and these effects may influence the relative risk for

NTM in the general population. Existing methods for measuring Mycobacterium spp. have proven inadequate, so a novel qPCR assay was designed for detection and quantitation of NTM. Field samples from surface waters and municipal systems pre- and post-treatment demonstrated the ubiquitous presence of NTMs and MAC. Water quality analyses yielded information on potential correlations between various parameters and the diversity of the organisms.

On the basis of hospitalization records for 2006-2008 from Florida‘s Agency for

Health Care Administration (AHCA), pulmonary NTM disease appears to be more common in elderly, white females, whereas disseminated NTM disease is more common in middle-aged, African American males. For both NTM states, most infected persons were found to live in urban areas.

Municipal sources may provide a means for the distribution of these organisms to households and serve as a source of infection for high-risk patients. Findings from the current study suggest that variations in the quality of municipal waters may affect the growth of NTMs and, in turn, the infection risk.

This study contributes to the field of mycobacterial research through the development of a new screening mechanism for NTM in environmental sources that is more sensitive, specific, and rapid than traditional methods. Culture-based techniques have been used historically (von Reyn et al., 1993; Iivanainen et al., 1993; Iivanainen et

163

al., 1999; Bland et al., 2005), but are not without shortcomings. Only a small fraction of the total NTMs are recovered or enumerated using these methods, and the efficiency varies among species (Steingrube et al., 1997; Griffith et al., 2007; Heidari et al., 2010).

Since there is no one medium or culture condition that works best for all Mycobacterium species, results obtained by different researchers have lacked reproducibility (Redmond and Ward, 1966; Hunter et al., 2001; Yeboah-Manu et al., 2004; Griffith et al., 2007; El

Bohy et al., 2009). In addition, the relatively long mycobacterial incubation time of weeks to months opens the door for other bacteria to overgrow cultures, and the process of decontamination can adversely affect the quantity and richness of the species recovered (Iivanainen, 1995; Iivanainen et al., 1997; Jacobs et al., 2009).

Further, NTM colony counts are often tenfold lower than cell counts (Angenent et al.,

2005).

Many of these issues can potentially be overcome through the use of DNA detection methods without the need for culture (Nadkarini et al., 2002; Ghosh et al.,

2006). The general approach involves isolating all DNA present in a sample and measuring specific mycobacterial sequences from the subsequent pool of DNA

(Schochetman et al., 1988; Nadkarini et al., 2002). Since the concentration of mycobacterial DNA from environmental sources is often low, PCR can be used to amplify target sequences to detectable levels (Wang et al., 1996; Stinear et al., 2000;

Marsollier et al., 2002; Pedley et al., 2004).

In recent years qPCR technologies have gained favor as a way to rapidly enumerate organisms or genes from environmental samples (Takai and Horikoshi,

2000; Smythe et al., 2002; Brinkman et al., 2003; He and Jiang, 2005). Through the use

164

of fluorescent reporter molecules, the technique allows for continuous reaction monitoring and evaluation at the peak of the exponential phase. This reduces variations based on reagent depletion or assay efficiency related to end point reads. While PCR- based methods show potential, they remain limited by the efficiency and reproducibility of cell lysis and the selectivity, sensitivity, and availability of primers (Steingrube et al.

1997; Heidari et al., 2010; Kay et al., 2011).

Effective PCR requires species-specific sequence-based primers (Schochetman et al., 1988). Despite a strong need in the clinical laboratory, definitive markers to differentiate each of the subspecies within the highly homogeneous Mycobacterium tuberculosis complex were not developed until 2002 (Brosch et al., 2002). It is not surprising that progress has been even slower for the less well-characterized NTMs. In many ways this was due to a lack of sequence information from which to design sensitive and specific primers (Kirschner et al., 1993). In recent years, more whole genome sequences of NTMs have been completed (Garnier et al., 2003; Stinear et al.,

2008; Castellanos et al., 2009; Kaser et al., 2009; Kaser et al., 2009; Wynne et al.,

2010).

Genomic comparison of different mycobacterial species enabled identification of a region of the 16S rRNA gene that appears to be genus-specific. This discovery is important, as previously published primer and probe sets for the genus mycobacteria often also amplify closely related Actinomycetales (Taylor et al., 1997; Gaafar et al.,

2001; Sheng et al., 2009). The MAC complex subgroup was chosen as a target for primer development due to the significant levels of associated morbidity and mortality,

165

as well as the existence of several published complete genomes for each of the species within this complex (Paustian et al., 2008; Castellanos et al., 2009; Wynne et al., 2010).

These methods show promise for environmental measurements, although samples can be difficult to work with because of high levels of background organic material that often co-purifies with DNA (Tebbe et al., 1993; Wilson, 1997).

Environmental compounds such as humic and fulvic acids are capable of inhibiting Taq polymerase and thereby inhibiting the PCR reaction (Tebbe et al., 1993; Wilson, 1997).

Several companies have designed kits to remove such contaminants at the time of DNA extraction or prior to PCR (Rochelle, MO-BIO, etc). Despite these limitations, direct

DNA methods provide an absolute indication of the presence or absence of mycobacteria and can be quantitative (qPCR) (MacGregor et al., 1999; Stinear et al.,

2000).

The new primers were used to test water samples collected in south Florida.

Eubacterial universal primers detected more bacteria than genus-specific primers, which, in turn, detected more than MAC primers, as expected. These results were confirmed by culture and by sequencing of the PCR amplicons. The study demonstrated the ubiquitous nature of these organisms in the environment and their presence in municipally distributed water. Mycobacteria and MAC were present in all samples, and mycobacterial numbers appear to increase post-treatment. These findings suggest that mycobacteria live within the source waters, at least a portion survive water treatment procedures, and the surviving NTMs go on to repopulate the distribution system along its length, most likely in the form of biofilms.

166

This study contributes to a growing body of evidence that people may be acquiring

NTM infections from organisms present in water, possibly municipal waters. Exposure to NTMs without the development of overt symptoms, but with the occurrence of an immunological response, is relatively common (Edwards and Smith, 1965; Smith, 1967).

Combining this with the apparent ubiquitous nature of NTMs, it appears the majority of people must be relatively resistant to most environmental NTMs. For this reason, further reductions in environmental exposure are unlikely to significantly decrease risk.

Still, for vulnerable populations such as children, the elderly, pregnant women, and immunocompromised persons, there is value in reducing potential sources of exposure from municipally distributed waters through devices such as whirlpools, indoor pools, and showers (Feazel et al., 2009; Ingen et al., 2010; Falkingham, 2010).

This study was the first attempt to identify variables in both environmental and municipal sources that contribute to mycobacterial growth using a quantifiable, molecular method. Phosphorous likely has an effect on mycobacterial growth in municipal samples, but this effect is likely the product of the stimulatory effect of phosphorous on microbial growth in an environment with decreased levels of competitors. Further work is needed to clarify parameters affecting mycobacterial distribution.

Why do some persons become infected with NTMs while others evade disease?

Unfortunately, there have been few published studies that describe the populations infected by these organisms (Winthrop, 2010). To determine whether there is a link between water sources and infections, the demographic characteristics of persons infected with pulmonary or disseminated NTM disease were examined. It had been

167

estimated that 1.8 out of 100,000 individuals are infected by non-AIDS-related mycobacterial diseases every year in the U.S. (O‘Brien et al., 1987), but precise incidence and prevalence data is not available. Laboratory surveys across the United

States and Europe have been limited by the fact that they are based on cultures rather than on patients (Abe et al., 1992; Woods et al., 1996). Also, unlike tuberculosis, NTM disease is not reportable in the U.S. (Arend et al., 2009; Winthrop et al., 2009), so the incidence likely remains largely underestimated (Gopinath, 2010; Benwill et al., 2010).

A statewide reporting system of hospital discharge diagnoses was established in 1993, but NTM disease as a whole remains non-reported.

A 1987 survey of state and city public health departments revealed that over 90% of all NTM isolates from these locations were respiratory in origin, but likely gave a substantial underestimate of incidence and prevalence. TB controllers have estimated prevalence at 1.8/100,000 persons (O‘Brien et al., 1987), but that study represented an incomplete sampling from across the U.S., excluded private laboratories, and took place at a time when there was under-recognition of the nodular bronchiectatic form of disease. More recent laboratory-based surveys across Europe, North America, Asia,

Africa, and Australia suggest a rising proportion of potentially pathogenic mycobacteria

(Marras and Daley, 2002). Across Europe and in regions such as Massachusetts and

British Columbia, incidence rates were found to be increasing exponentially. Pooled data approximated the incidence rate to be between 1.7-4.5/100,000 persons, but regional values or variability were not reported (Marras and Daley, 2002). Incidence rates for NTM isolations in Ontario, Canada rose significantly from 9.1/100,000 in 1997 to 14.1/100,000 in 2007 (Marras et al., 2007). Increases of this nature are concerning,

168

and the possibility that more people are becoming infected cannot be ruled out (Khan et al., 2007). Still, the increases could be a reflection of heightened awareness leading to more requests for sputum cultures and computed tomography lung scans, thereby allowing for increased recognition of the classic patterns, and subsequently more positive cultures (Winthrop, 2010). Alternatively, improved laboratory culture techniques, such as the use of liquid media, may have resulted in higher recovery rates than with traditional media such as Lowenstein-Jenson agar (Reddacliff et al., 2003;

Reddacliff et al., 2010). The true reason is likely multifactorial in nature (Winthrop,

2010).

Multiple factors contribute to making the southeastern U.S. an ideal site for NTM growth (Kirschner et al., 1992; Kirschner et al., 1999; Primm et al., 2004; Vaerewijck et al., 2005). One national study of hospitalization discharge codes found Florida to have the highest incidence rates of pulmonary NTM infection (2.1-2.4/100,000), and this rate increased over the study period 1998-2005 (Bilinger et al., 2009).

In the current study, hospital discharge data for more recent years, 2006-2008, put the incidence rate for pulmonary disease in Florida at 4.41/100,000, representing continued increase and supporting suggestions that in the U.S., rates of MAC, the primary pulmonary NTM isolate, tended to be highest in states bordering the Atlantic

Ocean and the Gulf of Mexico (>4.8/100,000) (Good and Snider, 1982; Griffith et al.,

2007; Huang et al., 2009). Florida abuts both of these waterways, so it is easy to see why the state would be expected to have similarly high rates. The difference compared with rates reported by O‘Brien et al. and Marras et al. are likely attributable to survey method. Those studies used laboratory isolates, which may represent colonization

169

rather than infection, as a measure of disease burden, whereas Bilinger et al. and the present study used hospital discharge codes.

Until NTMs become reportable, it remains unclear which of these approximation methods, if any, is best. The use of hospital discharge codes is likely to seriously underestimate the true incidence of NTM disease because outpatient cases are not included. This problem is compounded by the difficulties in diagnosis.

Pulmonary NTMs appear to be more common in elderly females. Using the microbiologic component of the American Thoracic Society/Infectious Diseases Society of America's pulmonary NTM disease criteria to define cases of pulmonary NTM, one study estimated the statewide prevalence rate within Oregon to be 5.6/100,000 persons, and found a correlation with pulmonary NTM and age (Cassidy et al., 2009). The prevalence was 15.5/100,000 for those over 50 years of age, and greater than

20/100,000 in those over 70 years of age. Almost 60% of these patients were female

(6.4/100,000 for women and 4.7/100,000 for men). Similarly, Bilinger et al. reported the highest annual prevalence rates in women aged 70 years and older (9.4/100,000) compared with similarly aged men (7.6/100,000) (Bilinger et al., 2009). Both these studies portray this disease as disproportionally affecting elderly females, and demonstrate the clinical significance of this disease with incidence rates surpassing those of TB (5.2/100,000) (Wolinsky, 1979; CDC, 2009; Winthrop, 2010). Likewise, the current study found a predominance of elderly (61.18% aged 60+) females (52.19%), with the highest reported incidence in women over 70 (22.11/100,000). Together, these studies demonstrate that pulmonary NTM is not an uncommon disease amongst the elderly and document what many experts and clinicians have suggested for over a

170

decade: this is no longer a male-dominated disease. These findings suggest that incidence will continue to rise given projections that as many as 1 in 5 individuals will be classified as elderly by 2020 (Day, 1996), and that women have longer life expectancies and tend to make up the majority of the elderly population (CIA, 2010). Further, the current trend in the United States and other countries for the elderly to retire to warm, moist environments such as Florida, may place these persons at further risk for NTM disease (Rappaport, 2007). Additional risks are associated with particular diseases, such as immunocompromised states, CF, or COPD, most of which are more common in the elderly (Barnes and Celli, 2009).

The profile for disseminated NTM disease largely matches trends within the HIV- infected population. Currently, two-and-a-half million people in industrial countries

(United States, Canada, Australia, New Zealand, Japan, and Western and Eastern

Europe) are estimated to be living with HIV, most of whom have or will have access to antiretroviral therapy (Wood et al., 2003; CDC, 2008; Stall et al., 2009). Within the

United States, the majority of these infected persons tend to be African American males in their 30s and 40s (CDC, 2008; CDC, 2010). Given the strong historical ties between disseminated NTM and HIV (Nunn and McAdam, 1988), it is unsurprising that our population-based data indicated that the majority of persons hospitalized for disseminated NTM disease tended to match this profile. Given the widespread use of

HAART in developed nations, it is increasingly less likely that a person‘s CD4 count will drop sufficiently to allow for disseminated NTM disease (Shepherd et al., 2010).

Despite this, hospitalization rates for disseminated NTM (4.03/100,000) still appear to be relatively high, approaching state TB levels (5.2/100,000) (CDC, 2009). It is therefore

171

possible that the view of disseminated NTM disease in developed nations, especially in the HIV-infected has been oversimplified, and carrier or unrecognized infections could be occurring in this population.

The current study contributes to a growing body of evidence linking pulmonary disease to elderly, white females, and disseminated disease to HIV profiles, and goes on to further describe these persons with regard to insurance status, length of stay, and commonly associated co-illnesses. Patients paid primarily with government insurance, providing possible linkages to lower socio-economic status. Additionally, patients stayed in the hospital on average over 11 days over two times the national average for any reason or for an infectious disease (DeFrances et al., 2003), thereby demonstrating the true severity of illness in these patients. Associated co-illnesses varied with NTM disease, but were largely associated with immunocompromised states such as pancytopenia and HIV.

There are virtually no population-based studies documenting basic epidemiological trends, including the ‗where‘ of NTM disease (Winthrop, 2010). Mapping of zip codes for persons discharged from the hospital with a diagnosis of NTM disease revealed that while pulmonary NTM may have a wider overall distribution than disseminated NTM, both disease states seem to occur primarily in or around urban areas. The tendency for elderly and immunocompromised people to settle in these areas may explain why cases focus in these areas, but does not account for why people become infected there

(Butler, 1969; Wasser et al., 1993; Glaeser et al., 2000).

The distribution of NTM disease in urban areas may be driven by exposure to municipal waters. Cassidy et al. previously determined that in Oregon, significantly

172

higher NTM rates were detected in the western, more urban portion of the state

(Cassidy et al., 2009). Urban areas are more likely to use large, municipal water systems than are rural areas whose primary water source may be wells. Water in municipal systems can be stored for long periods in both pipes and reservoirs, thereby providing opportunities for NTM growth (Cassidy et al., 2009). Further, given the hundreds to thousands of miles of pipe that may lie between a susceptible person‘s home and the treatment facility, as well as the general lack of competition within these sites, it is conceivable that NTMs may grow to high levels in biofilms within distribution lines (Falkingham et al., 2001). Indeed, the growth of NTMs within distribution systems was previously hypothesized in the 1970s and 1980s as an explanation for the higher incidence of MAC found in urban communities (DuMoulin et al., 1985).

Diagnostic bias and climate may also explain, in part, this trend toward NTM disease in urban areas, as persons residing in urban areas may be more likely to seek medical care than persons residing in rural areas (Sudha et al., 2003), and facilities in urban areas may have more experience in diagnosing these infections. Further, patients with predisposing conditions such as COPD or HIV more commonly reside in urban areas with better access to medical treatment facilities (Butler, 1969; Wasser et al., 1993; Glaeser et al., 2000). In terms of climate, wetter and more temperate, conditions in the more urban western Oregon promote NTM growth over that in the more arid, dry eastern part of the state (Kirschner et al., 1999; Falkingham, 2009;

Cassidy et al., 2009). On the other hand, the entire state of Florida is ‗hot and humid‘ and yet increased rates are noted in urban areas (Moon and Han, 2011). Despite this, climatic factors may help explain continued presence of NTMs in certain surface waters

173

that are subsequently used as municipal source waters. Environmental sampling has indicated that NTM organisms are present in the environments of most regions (von

Reyn et al., 1993), so geographic variation in the prevalence of MAC infection and disease is more likely the result of differences in the opportunities for exposure to environmental sources rather than the result of presence or absence of MAC in the environment. Pulmonary NTM infections appear to have increased incidence along the coastline in southern Florida but not in the northern part of the state. The milder climate in the south provides more opportunity for growth in source waters, as the ground is not frozen during the winter months and the agricultural season is longer (von Reyn et al.,

2001). These surface waters may then be used as municipal sources, thereby allowing a fairly constant flow of organisms into municipal supplies. This may help explain the increased rates of infection of NTMs long noted in the southeastern United States

(Edwards and Smith, 1965; Smith, 1967). There is also a cluster of cases with potential linkage to the Aucilla River in the panhandle. Confirmation of the significance of these environmental linkages will require further analysis and are hampered by the low numbers of cases of NTM infection in this study and a lack of information on water usage habits. This is further complicated by the likelihood that each species of NTM needs to be analyzed separately and that only diagnoses from hospitalized patients were available.

Final Thoughts

This study contributes to a growing body of evidence linking NTMs to water.

Having developed a fast, quantitative way to detect all mycobacteria and the MAC subgroup, the study went on to demonstrate the growth of these mycobacteria in municipal and surface waters, the demographic profile of those hospitalized with NTM

174

disease, and the distribution of hospitalizations around the state. The majority of mycobacterial hospitalizations tend to occur in urban areas and along the coast of south

Florida. This linkage to urban areas is likely driven in large part by the growth of NTMs in municipal waters, although further studies will be needed to confirm this. Additional epidemiologic work is needed to determine disease risk factors and identify potentially modifiable exposures for high-risk populations. If it is true that most patients are infected from public water supplies, efforts should be made to better understand the factors contributing to NTM concentrations in municipal water supplies, the risk of these potential pathogens, and ways to reduce those risks (Winthrop, 2010).

175

APPENDIX SEQUENCE RESULTS OF CULTURED BACTERIA

Sequence Identitity Label Sequence ANNNNNNNNNNNGNNNNCGANCCGNNNNGGGTTCCCGT CTGTAGTGGACGGGGGCCGGGTGCACAACAGCAAATGA 109 TTGCCAGACACACTATTGGGCCCTGAGACAACACTCGGT M. intracellulare CGATCCGTGTGGAGTCCCTCCATCTTGGTGGTGGGGTGT GGTGTTTGAGTATTGGATAGTGGTTGCGAGCATCA NNNNNNNNNNNCCAANNNGGAGGGNNTCCNCACGGATC GACCGAGNGTTGTCTCAGGGCCCAATAGTGTGTCTGGCA 110 ATCATTTGCTGTTGTGCACCCGGCCCCCGTCCACTACAG M. intracellulare ACGGGAACCCCTCACGGCTCGCACCCCACCAATTGGAGT GCTTTTCGTGGTGCTCCTTAGAAAGGAGGTN NNNNNNNNNNNNNANATNNNNNNNNCNTCNNNANNGNAT ATCTNCNTGGNNNANNTNTNNNGNNAATCCTGANNNNNN CCNNGNNATGGNNANNCNGNNAAACNNCNNANTGTCNTA 117 M. abscessus AAAATTGAAACGCTNGNACACTGNTGGNTCCTGANGNAA CACNNTNNGTTGTCACCCTGCTTGGNGGNGGGGTGTGN ACTTTNACTTCTGGATAGTGNTTGCNAGCATCA NNNNNNNNNNNNGCNNNNNNCNACNCNNNNCGNNTGCC TCANGANNNNNCNNNGNNNCANANNTTCNNTTNTNATNA NNNNCCNACNNTTTGCCNACNACCCNNCCNCNGNGNGG 118 M. abscessus ACNNGNNTTACAAAACATNTTCNNNNNGTNGATNTCCNCT ACNGATGCCTACTTTANNTNNCCTANTTNNTTCNNCTGGG AAANGNNGNTNNNTANNANNNNNNTANGNN NNNNNNNNNNNNGNNANNNNNNNNGNGTTCCCGTCTGT AGNGGACGGGGGCCGGGTGCACAACAGCAAATGATTGC 121 CAGACACACTATTGGGCCCTGAGACAACACTCGGTNNAT M. intracellulare NNNNGTGGAGTCCCTCCATCTTGGTGGTGNNNNNNNNNN N CNCCNCNNNNNTGGNNGGNNTCCNCACGGNTCGACCGA GNGTTGTCTCAGGGCCCAATAGTGTGTCTGGCAATCATTT 122 GCTGTTGTGCACCCGGCCCCCGTCCACTACAGACGGGA M. intracellulare ACCCCTCACGGCTCGCACCCCACCAATTGGAGTGCTTTT CGTGGTGCTCCTTAGAAAGGAGGTA NNNNNNNNNNNNTNNNGNGAGTTTCTGTAGTGGTTACTC GCTTGGTGAATATGTTTTATAAATCCTGTCCACCCCGTGG ATAGGTAGTCGGCAAAACGTCGGACTGTCAATAGAATTG 125 M. chelonae AAACGCTGGCACACTGTTGGGTTCTGAGGCAACACATTG TGTTGTCACCCTGCTTGGTGGTGGGGTGTGGTCTTTGAC TTATGGATAGTGGTTGCGAGCATCN NNNCNCCNCNNNNAGGGNGANAACACAATGTGTTGCCTC AGAACCCAACAGTGTGCCAGCGTTTCAATTCTATTGACAG TCCGACGTTTTGCCGACTACCTATCCACGGGGTGGACAG 126 M. chelonae GATTTATAAAACATATTCACCAAGCGAGTAACCACTACAG AAACTCACTTTATGTTCCCAAGCTCATTCGGCTAGGAAAT GGTGCTCCTTAGAAAGGAGGTA NNNNNNNNNNNCGANCCGNNNNGGGTTCCCGTCTGTAG TGGACGGGGGCCGGGNGCGCAACAGCAAATGATTGCCA M. avium subsp. 131 GACACACTATTGGGCCCTGAGACAACACTCGGNCCGTCC avium GTGTGGAGTCCCTCCATCTTGGTGGTGGGGTGTGGTGTT TGAGTATTGGATAGTGGTTGCGAGCATCA

176

NNNNNCNCCAANNNGGNNGGANTCCACACGGNCGGACC GAGNGTTGTCTCAGGGCCCAATAGTGTGTCTGGCAATCA M. avium subsp. 132 TTTGCTGTTGCGCACCCGGCCCCCGTCCACTACAGACGG avium GAACCCCTCACGGCTCGCACCCCACCAGTTGGGGTGCTT TTCGTGGTGCTCCTTAGAAAGGAGGN NNNNNNNNNNNNCGANCCGNNNNGGGTTCCCGNCTGTA GTGGACGGGGGCCGGGTGCGCAACAGCAAATGATTGCC M. avium subsp. 135 AGACACACTATTGGGCCCTGAGACAACACTCGGTCCGTC avium CGTGTGGAGTCCCTCCATCTTGGTGGTGGGGTGTGGTGT TTGAGTATTGNA NNNCCACCNCNNNNNGGNNGGNNTCCNCACGGACGGAC CGAGNGTTGTCTCAGGGCCCAATAGTGTGTCTGGCAATC M. avium subsp. 136 ATTTGCTGTTGCGCACCCGGCCCCCGTCCACTACAGACG avium GGAACCCCTCACGGCTCGCACCCCACCAGTTGGGGTGC TTTTCGTGGTGCTCCTTAGAAAGGAGGTA CNNNNNNNNNNNNNNGGANCAGNGCGGTTGGGANATCA GTGCCGGGCCTGTAGTGGGTTTCCGGTGGGTGCACAAC AAACGTGAGAAGTGGTGTGGGAACACTGCTTTGAGGAAT 139 M. flavescens CATCAGACACACTATTGNGCTTTGAGGCAACAGGCCCGT TGTTTCCCTGGCCACTGTGTGTGGTGGGGGGTCTGGTGT CGCCCTGTCTTTGGTGGTGGGGTGTGGTGTTTG NNNNNCNNNNNNNNNGGNNCGANACCAGNCCCCCCACC ACACACAGNGGCCAGGGAAACAACGGGCCTGTTGCCTC AAAGCCCAATAGTGTGTCTGATGATTCCTCAAAGCAGNGT 140 TCCCACACCACTTCTCACGTTTGTTGTGCACCCACCGGAA M. flavescens ACCCACTACAGGCCCGGCACTGATCTCCCAACCGCACTG ATCCCACACACGTGGGGGCGGGGGAACAATAAATGGTG CTCCTTAGAAAGGAGGTA NNNNNNNNNGNNNGANCCGNGANGGGTTCCCGNCTGTA GTGGACGGGGGCCGGGTGCACAACAGCAAATGATTGCC 141 AGACACACTATTGGGCCCTGAGACAACACTCGGNCNNTC M. intracellulare CGTGTGGAGTCCCTCCATCTTGGTGGNGGGGTGTGGTGT TT NNNNNNCNCCNANNNGGAGGGANTCCNCACGGNTCGAC CGAGTGTTGTCTCAGGGCCCAATAGTGTGTCTGGCAATC 142 ATTTGCTGTTGTGCACCCGGCCCCCGTCCACTACAGACG M. intracellulare GGAACCCCTCACGGCTCGCACCCCACCAATTGGAGTGCT TTTCGTGGTGCTCCTTAGAAAGGAGNN NNNNNNNNNNNNNNNNNNNNNANNNTGCGGTTGGGANN NNNNTGCCNGGNCTGNTAGTGGGTTTCCGGNGGGTGCN CNACAAACGTGANAAGTGGTGTGGGAACACTGCTTTGAG 143 GANNNATCANACACACTNTNGNGNTTTGAGGCAACANGN M. flavescens CCGTTGNTNCCCTGGNNNCTGNNTGTGGTGGGGGGTCT GGTGTNNNCCTGTCTTTGGNGGTGGGGTGTGGTGTTTGA TTCGNGNANANTGGTTNCNNNNNNNNNN NNNNNNNNNNNNNNNNNNNNNNNNNNNNGGNNNANACC AGNACCCCCCACCACACACANNGGCCAGGGAAACAACG GGCCTGTTGCCTCAAAGCCCAATANTGTGTCTGATGATTC 144 CTCAAAGNNNNGTTCCCACACCACTTCTCACGTTTGTTGT M. flavescens GCACCCACCGNAAACCCACTACANGNCCGGNNCTGATCT CCCAACCGCACTGATCCCACACNCNTGGGGACGGGGGA ACAATAAATGGTGCTCCTTANAAAGGANGTA NNNNNNNNNNNGCGNNCCNNNNNGGGTTCCCNNCTGNA 145 GTGGACGGGGGCCGGNNGNACNACAGCAAATGATTGCC M. intracellulare AGACACACTATTGGGNCCTGAGACAACACTCNGNCNATC

177

CGTGTGGAGTCCCTCCATCTTGGNGGNGGGGNGTGGNG TTTG

CACNNCNNNNNGGNNGGNNTCCNCACGGNATCGACCGA GNGTTGTCTCAGGGCCCAATAGTGTGTCTGGCAATCATTT 146 GCTGTTGTGCACCCGGNCCCCGTCCACTACAGACGGGA M. intracellulare ACCCCTCACGGCTCGCACCCCACCAATTGGAGTGCTTTT CNTGNNGCTCCTTANNAAGGAGGTA NNNNNNNNNNNNNNANCCGNNNNGNNCCGGTTGCCTGT AGTGGGCACGGTTTGGTGCACAACAAACTTTNNNNANTG NNNNACACNCTATTGGNNTTTCAGACNNNNNNNCCNTGC 147 M. fortuitum CCCTTTTGGNNGGNGGNNTCCNGNTGCNNNNGTCGGNN NGNTGNTNCCTCANTTTGNNGGNGGGGTGTGGNGTNNG ATTTGTGGATAGNGGTNGNGANCNTCANNTG CNNNCNCNNNNNNAGGCNNCAACACGCCGACACCCGCA ACCGGATGCCACCCCCCAAAAGGGGCACGGGCCTGTTG TCTCAAAGCCCAATAGTGTGTCTGGCAGTCNNATNANCN 148 NGNAGTGCNTTGNACNNNNNCNNNTGNTGNGNACCAGA M. fortuitum NCNTGNCNNCTANNGGAAAACNNNNNGTCTCNNCNNTTT TNNNANCGNNNGTTNNAANNNNNNTANNGGTGCTCCTTA GAAAGGAGGTAN NNNNNNNNNNNANNANNNNNNNNNNNNNNNTCNGTCTG NNNNGNANNAAGANNNNNNNNNNNACANCNAGCNNNNN CNNANNNACTATTGNNNCCTGANGNAACNCCCTCNNNNT 151 M. chelonae GNTNTCCCCCCNTCTNGNNGNTGGGGTGNGNNGTTTNAN AACTGNNTNNNGNTTGCNAGCNTCANGGTGGGGTGTGGT CTTTGACTTATGGATAGTGGTTGCGAGCATCAA NNNNCNNNNNNCNNNNNNGGNNNNNNNNCNNNNNNNNN NNNNNNNNNNNNGACCNANNANNNNNNNNNGGNTTTNN TTGNNNNCNTGNACCCNGNNNNCANNNCCTACNNACGAT 152 M. chelonae GANNNCNCNNNNNTTGNACCCNANNNNTTGNANNGNTTC NCNNGNNGNTNNTTANANAGNNGNNNNNAAGCTCATTCG GCTGGGAAATGGTGCTCCTTAGAAAGGAGGTAN NNNNNNNNNNNNNTNNNGCNNNNNTCTGNNNNTGNTTTC CTGCTNGNNNNNNNTGTTTTATAAATCCTGTCCGTTCTCG TTATCNAGGTGGATGGGTAGTCGGCAAGACGTCNGACTG 153 TCAAAAGAATTGNAATGCTGGCACACTGTTGGGTCCTGA M. immunogenum GGCAACACATTGTGTTGTCGCCCTGCTTGGCGGTGGGGT GTGGACTTTGACTTCTGGATAGTGGTTGCGAGCATCANN N CNNNNNNNNNGNNNGANACNCANTGCTGNTTGCCTCAG GACCCAACAGTGTGCCAGCATTCCAATTCTTTTGACAGTC CGACGTCTTGCCGACTACCCATCCACCTCGATAACGAGA 154 M. immunogenum ACGGACAGGATTTATAAAACATATTCACCAAGCAGGAAAC CACTACAGAAACCCGCTTTATGTTCCCACGCTCATTCGGC GGGGAAATGGTGCTCCTTANAAAGGAGGTAN NNNNNANNNNNNNNNANNNNNNNCGNNNNTNNNGTNNN GNNTNNNCTGNNNGGNGAATATGNTNTATAAATCCTGTC CNNTCNCNNTATCNANGTGNATGGGNANNCNGCAANACG 155 TCGGACTGCCNAAAGAATTGNAATGCTGGNNCNCTGNTT M. immunogenum GNGNCCTGAAGCAACNCATNNNNNNNNNNNCCTGCTTNN NGGTGGGGTGTGGACTTTGACTTCTGGATAGTGGTTGCG AGCATCA

178

NNNCNNNNNNNNNNNGGNNNNCAACNCANNGNNGTNGC CTCNNGACCNNACAGTGTGNCNNCNTTNCNNTTCTTNNG NCAGNNNNACGTCTTGCCGACTACCCATCCACCTCGATA 156 M. immunogenum ACGAGANCGGNGNGGATTTATAAAAGATNTNCNCCAAGN NNGNNNCCACTACNGAANCNNNNTTTATGTTCCCACGCT CATTCGGCGGGGAAATGGTGCTCCTTAGAAAGGAGGTA NNNNNNNNNNNNNNNNNNNNNNNNTNNNNNCNGCTNTN NTGNNNNNNNANNATNNNNNNNANANCCNGNNNNNTNN NNNNATNNNNGNGNANGGNNANNCNNCCNTACGTCGNN 157 NTGNCNNNANANTNGNNATGNTGGNNNNNNNNNTGNNN M. immunogenum NCTGANNNAANNNANNNNNNNNCANNCCTGNTTGNNGGT GGGGTGTGGACTTTGACTTCTGGATAGTGGTTGCGAGCA TCA NNNNNCNNNNNNNNNGNNNNNNANNCANNNNNNNTGNN NNCNNNNCNNNACNNNGNAGCNNNNNNTNNNNTTCTTTG NACNNNNNNACNNCTTGNCCACTACCCATCCACNNNGAN 158 NACNANNACNGNGNNNNNNNNTAAANGNTNTGCNCCNN M. immunogenum NNGNNNNNNNNNNNNNNNNNNNNNANTTTATGTTCCCAC GCTCATTCGGCGGGGAAATGGTGCTCCTTANAAAGGAGG TA NNNNNNNNGGGNNCAGCCGNNNNGGGTCATCGTCTGTA GTGGANGAAGACCGGGTGCACGACAACAAGCAAAGCCA 159 GACACACTATTGGGTCCTGAGGCAACACCCTCGGGTGCT M. gordonae GTCCCCCCATCTTGGTGGTGGGGTGTGGTGTTTGAGAAC TGGATAGTGGTTGCGAGCATCA NNNNNNNNNNNNNCCNNCCNCCNANNNGGGGGGANAGC ACCCGANGGNGTTGCCTCAGGACCCAATAGTGTGTCTGG 160 CTTTGCTTGTTGTCGTGCACCCGGTCTTCGTCCACTACAG M. gordonae ACGATGACCCCTCACGGCTTGCACCCCACCAATTGGAGT GCTTCTCGTGGTGCTCCTTAGAAAGGAGNN NNNNNNGNNNGNANCCNNGNNGNGTCATCGNTCTGTAN NGNANGAAGANCGGGTGCACNACAACNAGCAAAGCCAG ACACACTATTGGGTCCTGAGGCAACACCCTCNGGTGCTG 163 M. gordonae TCCCCCCATCTTGGTGGTGGGGTGTGGTGTTTGAGAACT GGATAGTGGTTGCGAGCATCANNGGTGGGGTGTGGTCTT TGACTTATGGATAGTGGTTGCGAGCNNN CNNNNNCNNANNNGGGNNNNNNNNCACCNNNANNGTGN NNGNCTCNNGACCCAATAGNGNGTCTGGNTTTGCTTGNT GNCNNGCACCCGGTCTTCANCCACTACANACGATGACCC 164 M. gordonae CTCNCNGNTTGCACCCCACCAATTGGAGTGCTTCTCGNG GNGCTCCTTANAAAGGAGGTACCAAGCTCATTCGGCTGG GAAATGGTGCTCCTTAGAAAGGAGGTAA

179

LIST OF REFERENCES

Abe, C., Hosojima, S., Fukasawa, Y., Kazumi, Y., Takahashi, M., Hirano, K., Mori, T. 1992. Comparison of MB-Check, BACTEC, and egg-based media for recovery of mycobacteria. J. Clin. Microbiol. 30, 878-881.

Aberg, J., Williams, P., Liu, T., Lederman, H., Hafner, R., Torriani, F., Lennox, J., Dube, M., MacGregor, R., Currier, J., AIDS Clinical Trial Group 393 Study Team. 2003. A study of discontinuing maintenance therapy in human immunodeficiency virus- infected subjects with disseminated Mycobacterium avium complex: AIDS Clinical Trial Group 393 Study Team. J. Infect. Dis. 187(7), 1046-1052.

Aberg, J., Yaijko, D., Jacobson, M. 1998. Eradication of AIDS-related Disseminated Mycobacterium avium complex after 12 Months of Antimycobacterial Therapy Combined with Highly Active Antiretroviral Therapy. J. Infect. Dis. 178, 1446-1449.

Agbalika, F., Dailloux, M., Escallier, G., Joret, J. 1984. Analyses bactériologiques et recherché de mycobactéries à l‘aquarium tropical de Nancy. Revue fr. Aquariol. 10, 113–124.

AIDS. 2002. AIDS: The Status and Trends of the HIV/AIDS Epidemics in the World. Provisional Report of the Barcelona MAP Symposium and the XIV International AIDS Conference. Barcelona, Spain.

Aksamit, T. 2002. Mycobacterium avium complex pulmonary disease in patients with pre-existing lung disease. Clin. Chest Med. 23, 643-653.

Allan, R., Pease, P., Ibbotson, J. 1986. Clustering of Crohn‘s disease in a Cotswold village. Q. J. Medicine. 59, 473-478.

Alexandersen, S., Brotherhood, I., Donaldson, A. 2002. Natural aerosol transmission of foot-and-mouth disease virus to pigs: minimal infectious dose for strain O1 Lausanne. Epidemiol. Infect. 128, 301-312.

American Thoracic Society Statement. 1997. Diagnosis and treatment of disease caused by nontuberculous mycobacteria. Am. J. Resp. Crit. Care Med. 156, S1- S25.

Amin, A., Angala, S., Chatterjee, D., Crick, D. 2008. Rapid Screening of Inhibitors of Mycobacterium tuberculosis Growth Using Tetrazolium Salts, pp. 465, 1-16. In: Parish, T., Brown, A. (ed.) Mycobacteria Protocols: Second Edition. Springer Science, New York.

Amofah, G., Sagoe-Moses, C., Adjei-Acquah, C., Frimpong, E. 1993. Epidemiology of Buruli ulcer in Amansie West district, Ghana. Trans. Royal Society Trop. Med. Hyg. 87, 644-645.

180

Andersen, S. 1993. Effects of waste water treatment on the species composition and Antibiotic resistance of coliform bacteria. Current Microbiol. 26(2), 97-103.

Angenent, L., Kelley, S., Amand, A., Pace, N., Hernandez, M. 2005. Molecular identification of potential pathogens in water and air of a hospital therapy pool. PNAS USA. 102(13), 4860-4865.

Anselin, L. 1986. MicroQAP: a microcomputer implementation of generalized measures of spatial association. Unpublished Manuscript. University of California Press, Santa Barbara.

Anselin, L. 1995. Local indicators of spatial association-LISA. Geograph. Anal. 27, 93- 115.

Anselin, L. 1998. Interactive techniques and exploratory spatial data analysis. In: Longley, P., Goodchild, M., Maguire, D., Wind, D. (ed.) Geographical information systems: principles, techniques, management and applications. Wiley, New York.

Anselin, L. 1998. Exploratory spatial data analysis in a geocomputational environment. In: Longley, P., Brooks, S., McDonnell, R., Macmillan, B. (ed.) Geocomputation, a primer. Wiley, New York.

Anselin, L. 2000. Spatial econometrics. In: Baltagi B (ed.) Companion to econometrics. Basil Blackwell, Oxford.

Anselin, L. 2005. Spatial statistical modeling in a GIS environment. In: Maguire, D., Batty, M., Goodchild, M. (ed.) GIS, spatial analysis, and modeling. ESRI Press, Redlands.

Anselin, L., Bera, A. 1998. Spatial Dependence in Linear Regression Models with an Introduction to Spatial Econmetrics, pp. 237-281. In: Ullah, A., Giles, D. (ed.). Handbook of Applied Economic Statistics. Marcel Dekker, New York.

Anselin, L., Syabri, I., Kho, Y. 2006. GeoDa: An Introduction to Spatial Data Analysis. Geogr. Anal. 38, 5–22.

Appenzeller, B., Batte, M., Mathieu, L., Block, J., Lahoussine, V., Cavard, J., Gatel, D. 2001. Effect of Adding Phosphate to Drinking Water on Bacterial Growth in Slightly and Highly Corroded Pipes. Water Res. 35(4), 1100-1105.

Aramaki, M., Silachamroon, U., Desakorn, V., Maek-a-nantawat, W., Waiwaruwut, J., Jutiwarakun, K., Hahn Kim, J., Pitisuttithum, P. 2010. Immune Reconstitution Inflammatory Syndrome in Adult Human Immunodeficiency Virus-Infected Patients in Thailand. Southeast Asian J. Trop. Med. Public Health. 41(1), 138-145.

181

Arasteh, K., Cordes, C., Ewers, M., Simon, V., Dietz, E., Futh, U., Brockmeyer, N., L‘Age, P. 2000. Human immunodeficiency virus-related nontuberculous mycobacterial infection: incidence, survival analysis and associated risk factors. Eur. J. Med. Res. 5, 424–430.

Arbeit, R., Slutsky, A., Barber, T., Maslow, J., Niemczyk, S., Falkingham, J., O‘Connor, G., von Reyn, C. 1993. Genetic diversity among strains of Mycobacterium avium causing monoclonal and polyclonal bacteremia in patients with AIDS. J. Infect. Dis. 167, 1384-1390.

Arend, S., Janssen, R., Gosen, J., Waanders, H., de Boer, T., Ottenhoff, T., van Dissel, J. 2001. Multifocal osteomyelitis caused by nontuberculous mycobacteria in patients with a genetic defect of the interferon-gamma receptor. Netherlands J. Med. 59, 140-151.

Arend, S., van Soolingen, D., Ottenhoff, T., Tom, H. 2009. Diagnosis and treatment of lung infection with nontuberculous mycobacteria. Current Opinion Pulm. Med. 15(3), 201-208.

Aronson, T., Holtzman, A., Glover, N., Boian, M., Froman, S., Berlin, O., Hill, H., Stelma, G. 1999. Comparison of large restriction fragments of Mycobacterium avium isolates recovered from AIDS and non-AIDS patients and those of isolates from potable water. J. Clin. Microbiol. 37, 1008-1012.

Asiedu, K., Etuaful, S. 1998. Socioeconomic implications of Buruli ulcer in Ghana: a three year review. Am. J. Tropical Med. Hyg. 59, 1015-1022.

Astagneau, P., Desplaces, N., Vincent, V., Chicheportiche, V., Botherel, A., Maugat, S., Lebascle, K., Leonard, P., Desenclos, J., Grosset, J., Ziza, J., Brucker, G. 2001. Mycobacterium xenopi spinal infections after discovertebral surgery: investigation and screening of a large outbreak. Lancet. 358, 747-751.

Astrofsky, K., Schrenzel, M., Bullis, R., Smolowitz, R., Fox, J. 2000. Diagnosis and management of atypical Mycobacterium spp. infections in established laboratory zebrafish (Brachydanio rerio) facilities. Compar. Med. 50, 666-672.

Asuncion, M., Torello, S., Kolter, R. 1999. Sliding Motility in Mycobacteria. J. Bacteriol. 181, 7331–7338.

Aubry, A., Chosidow, O., Caumes, E., Robert, J., Cambau, E. 2002. Sixty-three cases of Mycobacterium marinum infection. Arch. Intern. Med. 162, 1746-1752.

Aubuchon, C., Hill, J., Graham, D. 1986. Atypical mycobacterial infection of soft tissue associated with use of a hot tub. A case report. J. Bone Joint Surg. Am. 68, 766- 768.

182

Bae, S., Wuertz, S. 2009. Discrimination of viable and dead fecal Bacteroidales bacteria by quanitative PCR with propidium monoazide. Appl. Environ. Microbiol. 75(9), 2940-2944.

Bagga, A., Chellam, S., Clifford. D. 2008. Evaluation of iron chemical coagulation and electrocoagulation pretreatment for surface water microfiltration. J. Membrane Sci. 309, 82-93.

Bailey, T., Gattrell, A. 1995. Interactive spatial data analysis. Harlow, Longman.

Bange, F., Bottger, E. 2002. Improved decontamination method for recovering mycobacteria from patients with cystic fibrosis. Eur. J. Clin. Microbiol. Infect. Dis. 21, 546-548.

Barcat, D., Mercie, P., Constans, J., Triassas, T., LeClouerec, G., Texier-Maugein, J., Conri, C. 1998. Disseminated Mycobacterium avium complex infection associated with bifocal synovitis in a patient with dermatomyositis. Clin. Infect. Dis. 26, 1004- 1005.

Barker, J., Brown, M. 1994. Trojan Horses of the microbial world: protozoa and the survival of bacterial pathogens in the environment. Microbiol. 140, 1253-1259.

Barksdale, L., Kim, K. 1977. Mycobacterium. Bacteriolog. 41(1), 217-372.

Barnes, P., Celli, B. 2009. Systemic manifestations and comorbidities of COPD. Eur. Respir. J. 33, 1165-1185.

Barry, C. 2001. Interpreting cell wall ‗virulence factors‘ of Mycobacterium tuberculosis. Trends Microbiol. 9, 237-241.

Bean, N., Martin, S., Bradford, H. 1992. PHLIS: an electronic system for reporting public health data from remote sites. Am. J. Public Health. 82, 1273-1276.

Beard, P., Daniels, M., Henderson, D., Pirie, A., Rudge, K., Buxton, D., Rhind, S., Greig, A., Hutchings, M., McKendrick, I., Stevenson, K., Sharp, J. 2001. Paratuberculosis infection of nonruminant wildlife in Scotland. J. Clin. Microbiol. 39, 1517-1521.

Beard, P., Stevenson, K., Pirie, A., Rudge, K., Buxton, D., Rhind, S., Sinclair, M., Wildblood, L., Jones, D., Sharp, J. 2001. Experiemental paratuberculosis in calves following inoculation with a rabit isolate of Mycobacterium avium subsp. paratuberculosis. J. Clin. Microbiol. 39, 3080-3084.

Belisle, J., Brennan, P. 1994. Molecular basis of colony morphology in Mycobacterium avium. Res. Microbiol. 145, 237-242.

Belisle, J., Pascopella, L., Inamine, J., Brennan, P., Jacobs, W. 1991. Isolation and expression of a gene cluster responsible for biosynthesis of the glycopeptidolipid antigens of Mycobacterium avium. J. Bacteriol. 173, 6991-6997.

183

Bell, K., LeChevallier, M., Abbaszadegan, M., Amy, G., Sinha, S., Benjamin, M., Ibrahim, E. 2002. Enhanced and Optimized Coagulation for Particulate and Microbial Removal. AWWA Research Foundation and the American Water Works Association, Denver.

Bellinger, D., Bullock, B. 1988. Cutaneous Mycobacterium avium infection in a cynomolgus monkey. Lab. Animal Sci. 38, 85-86.

Bennett, C., Vardiman, J., Golomb, H. 1986. Disseminated atypical mycobacterial infection in pateints with hairy cell leukemia. Am. J. Med. 80, 891-896.

Benson, C., Ellner, J. 1993. Mycobacterium avium complex and AIDS: advances in theory and practice. Clin. Infect. Dis. 17, 7-20.

Benson, C., Williams, P., Cohn, D. 2000. Clarithromycin or rifabutin alone or in combination for primary prophylaxis of Mycobacterium avium complex disease in patients with AIDS: a randomized, double-blind, placebo-controlled trial. J. Infect. Dis. 181, 1289-1297.

Benwill, J., Babineaux, M., Sarria, J. 2010. Pulmonary Mycobacterium abscessus in an AIDS Patient. Am. J. Med. Sci. 339(5), 495-496.

Bercovier, H., Kafri, O., Sela, S. 1986. Mycobacteria possess a surprisingly small number of ribosomal RNA genes in relation to the size of their genome. Biochem. Biophys. Res. Com. 136, 1136-1141.

Bermudez, L., Petrofsky, M., Goodman, J. 1997. Exposure to low oxygen tension and increased osmolarity enhance the ability of Mycobacterium avium to enter intestinal epithelial (HT-29) cells. Infect. Immunol. 65, 3768–3773.

Bermudez, L., Sangari, F. 2000. Mycobacterial invasion of epithelial cells. Subcell. Biochem. 33, 231–249.

Bermudez, L., Young, L., Enkel, H. 1991. Interaction of Mycobacterium avium complex with human macrophages: roles of membrane receptors and serum proteins. Infect. Immun. 59, 1697-1702.

Bernardinelli, L., Montomoli, C. 1992. Empirical Bayes versus fully Bayesian analysis of geographical variation in disease risk. Stat. Med. 11, 983–1007.

Berney, M., Cook, G. 2010. Unique Flexibility in Energy Metabolism Allows Mycobacteria to Combat Starvation and Hypoxia. PLoS One. 5(1), 1-11.

Beroll, H., Berke, O., Wilson, J., Barker, I. 2007. Investigating the spatial risk distribution of West Nile virus disease in birds and humans in southern Ontario from 2002 to 2005. Population Health Metrics. 5, 3.

184

Besner, M., Prevost, M., Regli, S. 2011. Assessing the public health risk of microbial intrusion events in distribution systems: Conceptual model, available data, and challenges. Water Res. 45(3), 961-979.

Best, M., Sattar, S., Springthorpe, V., Kennedy, V. 1990. Efficacies of selected disinfectants against Mycobacterium tuberculosis. J. Clin. Microbiol. 28, 2234- 2239.

Beveridge, T. 2001. Use of the Gram stain in microbiology. Biotech. Histochem. 76(3), 111-118.

Bilinger, M., Olivier, K., Viboud, C., Montes de Oca, R., Steiner, C., Holland, S., Prevots, D. 2009. Nontuberculous mycobacteria-associated lung disease in hospitalized persons, United States, 1998-2005. Emerg Infect Dis. 15(10), 1562-1568.

Bithell, J. 2000. A classification of disease mapping methods. Stat. Med. 19, 2203– 2215.

Black, P. 2001. Why is the prevalence of allergy and autoimmunity increasing? Trends Immunol. 22, 354–355.

Blackburn, J., McNyset, K., Hugh-Jones, M., Curtis, A. 2007. Modeling the geographic distribution of Bacillus anthracis, the causative agent of anthrax disease, for the contiguous United States using predictive ecological niche modeling. Am. J. Trop. Med. Hyg. 77, 1103-1110.

Blackburn, J. 2010. Integrating Geographic Information Systems and Ecological Niche Modeling into Disease Ecology: A Case Study of Bacillus anthracis in the United States and Mexico. In: O‘Connell, K. (ed.) Emerging and Endemic Pathogens. Springer Science, New York.

Bland, C., Ireland, J., Lozano, E., Alvarez, M., Primm, T. 2005. Mycobacterial Ecology of the Rio Grande. Appl. Environ. Microbiol. 71(10), 5719-5727.

Blandchard, D., Hoffman, E. 1978. Control of jet drop dynamics by organic matter in seawater. J. Geophys. Res. 83, 6187-6191.

Blanchard, D., Syzdek, L. 1978. Seven problems in bubble and jet drop researches. Limnol. Oceanograp. 23, 389-400.

Blaser, M., Newman, L. 1982. A Review of Human Salmonellosis: I. Infective Dose. Rev. Infect. Dis. 4(6), 1096-1106.

Bloch, K., Zwerling, L., Pletcher, M., Hahn, J., Gerberding, J., Ostroff, S., Vugia, D., Reingold, A. 1998. Incidence and clinical implications of isolation of Mycobacterium kansasii: Results of a 5-year, population-based study. Ann. Intern. Med. 129(9), 698-704.

185

Boddinghaus, B., Rogall, T., Flohr, T., Blocker, H., Bottger, E. 1990. Detection and identification of mycobacteria by amplification of rRNA. J. Clin. Microbiol. 28, 1751- 1759.

Bodmer, T., Miltner, E., Bermudez, L. 2000. Mycobacterium avium resisits exposure to the acidic conditions of the stomach. FEMS Microbiol. Letters. 182, 45-49.

Boehm, A., Yamahara, K., Love, D., Peterson, B., McNeill, K., Nelson, K. 2009. Covariation and photoinactivation of traditional and novel indicator organisms and human viruses at a sewage-impacted marine beach. Environ. Sci. Technol. 43(21), 8046-8052.

Bollo, E., Guarda, F., Capucchio, M., Galietti, F. 1998. Direct detection of Mycobacterium tuberculosis complex and M. avium complex in tissue specimens from cattle through identification of specific rRNA sequences. Zentralbl. Veterinarmed. 45, 395-400.

Boxerbaum, B. 1980. Isolation of rapidly growing mycobacteria in pateints with cystic fibrosis. J. Pediatr. 96(4), 689-691.

Brassel, K., Reif, D. 1979. A Procedure to Generate Thiessen Polygons. Geogr. Anal. 11, 289–303.

Brennan, P., Nikaido, H. 1995. The envelope of mycobacteria. Ann. Rev. Biochem. 64, 29-63.

Brest, P., Lapaquette, P., Souidi, M., Lebrigand, K., Cesaro, A., Vouret-Craviari, V., Mari, B., Barbry, P., Mosnier, J., Hébuterne, X., Harel-Bellan, A., Mograbi, B., Darfeuille-Michaud, A., Hofman, P. 2011. A synonymous variant in IRGM alters a binding site for miR-196 and causes deregulation of IRGM-dependent xenophagy in Crohn‘s disease. Nature Genetics. In Press. Available online 2011 Mar 1.

Briancesco, R., Semproni, M., Libera, S., Sdanganelli, M., Bonadonna, L. 2010. Non- tuberculous mycobacteria and microbial populations in drinking water distribution systems. Ann. Ist. Super Sanita. 46(3), 254-258.

Brinkman, N., Haugland, R., Wymer, L., Byappanahalli, M., Whitman, R., Vesper, S. 2003. Evaluation of a rapid, quantitative real-time PCR method for enumeration of pathogenic Candida cells in water. Appl. Environ. Microbiol. 69, 1775-1782.

Brooks, R., George, K., Parker, B., Falkingham, J. 1984. Recovery and survival of nontuberculous mycobacteria under various growth and decontamination conditions. Canad. J. Microbiol. 30, 1112-1117.

Brooks, R., Parker, B., Falkingham, J. 1984. Epidemiology of nontuberculous mycobacteria. Numbers in eastern United States soils and correlation with soil characteristics. Am. Rev. Respir. Dis. 130, 630-633.

186

Brosch, R., Gordon, S., Mariesse, M., Brodin, P., Buchrieser, C., Eiglmeier, K., Garnier, T., Gutierrez, C., Hewinson, G., Kremer, K., Parsons, L., Pym, A., Samper, S., van Soolingen, D., Cole, S. 2002. A new evolutionary scenario for the Mycobacterium tuberculosis complex. PNAS. 99, 3684-3689.

Brown, J., McNeil, M. 2003. Actinomycetes, pp. 2322. In: Murray, P. (ed.) Manual of Clinical Microbiology. ASM Press, Washington.

Brownstein, J., Cassa, C., Kohane, I., Mandl, K. 2006. An unsupervised classification method for inferring original case locations from low-resolution disease maps. Int. J. Health Geogr. 5, 56.

Buergelt, C., Hall, C., McEntee, K., Duncan, J. 1978. Pathological Evaluation of Paratuberculosis in Naturally Infected Cattle. Vet. Path. 15, 196-207.

Buergelt, C., Layton, A., Ginn, P., Taylor, M., King, J., Habecker, P., Mauldin, E., Whitlock, R., Rossitier, C., Collins, M. 2000. The pathology of spontaneous paratuberculosis in the North American Bison (Bison bison). Vet. Path. 37, 428- 438.

Buddle, B., Young, L. 2000. Immunobiology of mycobacterial infections in marsupials. Develop. Compar. Immunol. 24, 517-529.

Bull, T., McKinn, E., Sidi-Boumedine, K., Skull, A., Durkin, D., Neild, P., Rhodes, G., Pickup, R., Hermon-Taylor, J. 2003. Detection and verification of Mycobacterium avium subspecies paratuberculosis in fresh ileocolonic mucousal biopsies from people with and without Crohn‘s disease. J. Clin. Microbiol. 41, 2915-2923.

Buntine, J., Crofts, K. 2002. Management of Mycobacterium ulcerans disease, pp. 17- 22. In: WHO (ed.), Fifty-seventh World Health Assembly Resolutions and Decisions. World Health Organization, Geneva. Available at http://www.who.int/gtb-buruli/publications/index.html.

Byappanahalli, M., Whitman, R., Shively, D., Nevers, M. 2010. Linking non-culturable (qPCR) and culturable enterococci densities with hydrometeorological conditions. Sci. Total Environ. 408(16), 3096-3101.

Cangelosi, G., Palermo, C., Bermudez, L. 2001. Phenotypic consequences of red-white colony type variation in Mycobacterium avium. Microbiol. 147, 527-533.

Carson, L., Bland, L., Cusick, L., Favero, M., Bolan, G., Reingold, A., Good, R. 1988. Prevalence of nontuberculous mycobacteria in water supplies of hemodialysis centers. Appl. Environ. Microbiol. 54, 3122-3125.

Carson, L., Petersen, N., Favero, M., Aguero, S. 1978. Growth characeristics of atypical mycobacteria in water and their comparative resistance to disinfectants. Appl. Environ. Microbiol. 36, 839-846.

187

Cassidy, P., Hedberg, K., Saulson, A., McNelly, E., Winthrop, K. 2009. Nontuberculous mycobacterial disease prevalence and risk factors: a changing epidemiology. Clin. Infect. Dis. 49(12), 124-129.

Castellanos, E., Aranaz, A., de Juan, L., Alvarez, J., Rodriguez, S., Romero, B., Bezos, J., Stevenson, K., Mateos, A., Dominguez, L. 2009. Single nucleotide polymorphisms in the IS900 sequence of Mycobacterium avium subspecies paratuberculosis are strain type-specific. J. Clin. Microbiol. 47(7), 2260-2264.

CDC. 1991. Nosocomial infection and pseudoinfection from contaminated endoscopes and bronchoscopes-Wisconsin and Missouri. JAMA. 266, 2197-2198.

CDC. 2002. Guidelines for Preventing Opportunistic Infections Among HIV-infected Persons 2002 Recommendations of the U. S. Public Health Service and the Infectious Diseases Society of America. MMWR. 51(8), 10-11.

CDC. 2008. HIV Prevalence Estimates-United States, 2006. MMWR. 57(39), 1073-76.

CDC. 2009. Reported tuberculosis in the United States, 2008. US Department of Health and Human Services, CDC, Atlanta. Available at: http://www.cdc.gov/tb/statistics/reports/2008/default.htm.

CDC. 2010. Diagnoses of HIV infection and AIDS in the United States and Dependent Areas, 2008. In: CDC (ed.), HIV Surveillance Report. 20. CDC, Atlanta. Available at: http://www.cdc.gov/hiv/surveillance/resources/reports/2008report/

Chalermskulrat, W., Gilbey, J., Donohue, J. 2002. Nontuberculous mycobacteria in women, young and old. Clin. Chest. Med. 23, 675-686.

Chan-Yeung, M., Yeh, A., Tam, C., Kam, K., Leung, C., Yew, W., Lam, C. 2005. Socio- demographic and geographic indicators and distribution of tuberculosis in Hong Kong: a spatial analysis. Int. J. Tuberculosis Lung Dis. 9(12), 1320-1327.

Charti, J., Kazemnegad, A. 2010. Spatial distribution of Tuberculosis in Mazandaran Province-Iran: Spatiotemporal modeling. Tanaffos. 9(3), 15-21.

Chen, X., Tang, N. 2010. Bayesian analysis of semiparametric reproductive dispersion mixed-effects models. Comp. Stat. Data Anal. 54(9), 2145-2158.

Chesney, P. 2002. Nontuberculous mycobacteria. Ped. Review. 23, 300-309.

Cheung, J., Fung, B., Wong, S., Ip, W. 2010. Review article: Mycobacterium marinum infection of the hand and wrist. J. Orthop. Surg. 18(1), 98-103.

Chilima B., Clark, I., Floyd, S., Fine, P., Hirsch, P. 2006. Distribution of environmental mycobacteria in Karonga District, Northern Malawi. Appl. Environ. Microbiol. 72, 2343–2350.

188

Chimara, E., Lucilaine, F., Ueky, S., Martins, M., Durham, A., Arbeit, R., Leao, S. 2008. Reliable identification of mycobacterial species by PCR-restriction enzyme analysis (PRA)-hsp65 in a reference laboratory and elaboration of a sequence- based extended algorithm of PRA-hsp65 patterns. BMC Microbiol. 8, 48

Chin, D., Reingold, A., Stone, E., Vittinghoff, E., Horsburgh, C., Simon, E., Yajko, D., Hadley, W., Ostroff, S., Hopewell, P. 1994. The impact of Mycobacterium avium complex bacteriemia and its treatment on survival of AIDS patients- a prospective study. J. Infect. Dis. 170, 578-584.

Chiodini, R., van Kruningen, H., Merkal, R., Thayer, W., Coutu, J. 1984a. Possible role of mycobacteria in inflammatory bowel disease. Dig. Dis. Sci. 29, 1073-1079.

Chiodini, R., van Kruningen, H., Merkal, R., Thayer, W., Coutu, J. 1984b. Characteristics of an unclassified Mycobacterium species isolated from patients with Crohn‘s disease. J. Clin. Microbiol. 20, 966-971.

Chiodini, R., van Kruningen, H., Merkal, R., Thayer, W., Coutu., J. 1986. Spheroplastic phase of mycobacteria isolated from patients with Crohn‘s disease. J. Clin. Microbiol. 24, 357-363.

Chiodini, R. 1989. Crohn‘s disease and Mycobacterioses: a review and comparison of two disease entities. Clin. Microbiol. Rev. 2, 90-117.

CIA. 2010. Sex Ratios. In: US Central Intelligence Agency, CIA (ed.) The World Factbook. CIA Publications, Langley. Available at https://www.cia.gov/library/publications/the-world-factbook/fields/2018.html.

Cirillo, J., Falkow, S., Tompkins, L., Bermudez, L. 1997. Interaction of Mycobacterium avium with Environmental Amoebae Enhances Virulence. Infect. Immunol. 65, 3759–3767.

Clarke, C. 1997. The Pathology and Pathogenesis of Paratuberculosis in Ruminants and Other Species. J. Comp. Pathol. 116, 217-261.

Clarke, K., McLafferty, S., Tempalski, B. 1996. On epidemiology and geographic information systems: a review and discussion of future directions. Emerg. Infect. Dis. 2, 85-92.

Clarke, S. 2002. Nucleotide sequence-based typing of bacteria and the impact of automation. Bioessays. 24, 858-862.

Claudia, A., Montenegro, D., Werneck, G., Kerr-Pontes, L., Barreto, M., Feldmeier, H. 2004. Spatial analysis of the distribution of leprosy in the state of Ceará, Northeast Brazil. Mem. Inst. Oswaldo Cruz. 99(7), 683-686.

189

Clayton, D., Bernardinelli, L. 1997. Bayesian methods for mapping disease risk, pp. 205-220. In: Elliott, P., Cuzick, J., English, D., Stern, R. (ed.). Geographical and environmental epidemiology: methods for small area studies. Oxford University Press, Oxford.

Clescerl, L., Greenberg, A., Eaton, A. 1999. Standard Methods for Examination of Water and Wastewater. APHA, Washington, DC.

Cocito, C., Gilot, P., Coene, M., De Kesel, M., Poupart, P., Vannuffel, P. 1994. Paratuberculosis. Clin. Microbiol. Rev. 7, 328-345.

Colford, J., Hilton, J., Wright, C., Arnold, B., Saha, S., Wade, T., Scott, J., Eisenberg, J. 2009. The Sonoma Water Evaluation Trial: A Randomized Drinking Water Intervention Trial to Reduce Gastrointestinal Illness in Older Adults. Am. J. Public Health. 99, 1988-1995.

Collins, F. 1971. Relative susceptibility of acid-fast and non-acid-fast bacteria to ultraviolet light. J. Appl. Microbiol. 21, 411–413.

Collins, C., Grange, J., Noble, W., Yates, M. 1985. Mycobacterium marinum infections in man. J. Hygiene London. 94, 135-149.

Collins, C., Grange, J., Yates, M. 1984. Mycobacteria in water. J. Appl. Bacteriol. 57, 193-211.

Colville, A. 1993. Retrospective review of culture-positive mycobacterial lymphadenitis cases in children in Nottingham, 1979-1990. Eur. J. Clin. Microbiol. Infect. Dis. 12, 192-195.

Connell, N., Nikaido, H. 1994. Membrane Permeability and transport in Mycobacterium tuberculosis, pp. 333-352. In: Bloom, B. (ed.). Tuberculosis: Pathogenesis, Protection, and Control. American Society for Microbiology, Washington, DC.

Cook, K., Britt, J., Bolster, C. 2010. Survival of Mycobacterium avium subsp. paratuberculosis in biofilms on livestock watering trough materials. Vet. Microbiol. 141(2), 103-109.

Cortesial, C., Lopez, G., de Waard, J., Takiff, H. 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. J. Antimicro. Chemo. 65(12), 2574-2581.

Corti, S., Stephan, R. 2002. Detection of Mycobacterium avium subspecies paratuberculosis specific IS900 insertion sequences in bulk-tank milk samples obtained from different regions throughout Switzerland. BMC Microbiol. 2,15.

190

Covert, T., Rodgers, M., Reyes, A., Stelma, G. 1999. Occurance of Nontuberculous Mycobacteria in Environmental Samples. Appl. Environ. Microbiol. 65(6), 2492- 2496.

Crosby L., Criddle, C. 2003. Understanding Biases in Microbial Community Analysis Techniques Due to rrn Operon Copy Number Heterogeneity. BioTechniques. 34(4), 2-9.

Cross, M., Labes, R., Mackintosh, C. 2000. Oral infection of ferrets with virulent Mycobacterium bovis or Mycobacterium avium: susceptibility, pathogenesis and immune response. J. Comp. Pathol. 123, 15-21.

Crowle, A., Dahl, R., Ross, E., May, M. 1991. Evidence that vesicles containing living, virulent Mycobacterium tuberculosis and Mycobacterium avium in cultured human macrophages are not acidic. Infect. Immun. 59, 1823-1831.

Curtis, A., Mills, J., Leitner, M. 2006. Spatial confidentiality and GIS: Reengineering mortality locations from published maps about Hurricane Katrina. Int. J. Health Geogr. 5, 44.

Curtis, A., Mills, J., Leitner, M. 2007. Katrina and vulnerability: the geography of stress. J. Health Care Poor Underserved. 18, 315-330.

Curtis, A., Mills, J., Agustin, L., Cockburn, M. 2011. Confidentiality risks in fine scale aggregations of health data. Comp. Environ. Urban. Sys. 35, 57-64. da Silva Rocha, A., Werneck Barreto, A., Dias Campos, C., Villas-Boas da Silva, M., Fonesca, L., Saad, M., Degrave, W., Suffys, P. 2002. Novel allelic variants of Mycobacteria isolated in Brazil as determined by PCR-restriction enzyme analysis of hsp65. J. Clin. Microbiol. 40, 4191-4196.

Daillux, M., Hartemann, P., Beurey, J. 1980. Study on the relationship between isolation of mycobacteria and classical microbiological and chemical indicators of water quality in swimming pools. Zentral. Bakteriol. Mikrobiol. Hyg. 171(6), 473-486.

Danilchanka, O., Pavlenok, M., Niederweis, M. 2008. Role of porins for uptake of antibiotics by Mycobacterium smegmatis. Antimicrob. Agents Chemother. 52, 3127-3134.

Dash, R., Prakash, E., Kumar, P., Mehrotra, I., Sandhu, C., Grischek, T. 2010. River bank filtration in Haridwar, India: removal of turbidity, organics, and bacteria. Hydrogeol. J. 18(4), 973-983.

David, H. 1973. Response of mycobacteria to ultraviolet light radiation. Am. Rev. Respir. Dis. 108, 1175–1185.

David, H., Jones, W., Newman, C. 1971. Ultraviolet light inactivation and photoreactivation in the mycobacteria. Infect Immun. 4, 318–319.

191

Dawson, D., Armstrong, J., Blacklock, Z. 1982. Mycobacterial Cross-Contamination of Bronchoscope Specimens. Am. Rev. Respir. Dis. 126, 1095-1097.

Day, J. 1996. Population Projections of the United States by Age, Sex, Race, and Hispanic Origin: 1995-2050, pp. 25-1130. In: US Bureau of the Census, Current Population Reports (ed.) US Government Printing Office, Washington, DC.

Deaton, A. 2008. Income, Health, and Wellbeing Around the World: Evidence from the Gallup World Poll. J. Econ. Perspect. 22(2), 53-72. de Chastellier, C., Frehel, C., Offredo, C., Skamene, E. 1993. Implication of phagosome-lysosome fusion in restriction of Mycobacterium avium growth in bone marrow macrophages from genetically resistant mice. Infect. Immun. 61, 3775- 3784.

DeFrances, C., Hall, M., Podgornik, M. 2003. National Hospital Discharge Survey. Advanced data from Vital and Health Statistics; no. 359. US Department of Health and Human Services, CDC, National Center for Health Statistics, Hyattsville. Available at http://www.cdc.gov/nchs/data/ad/ad359.pdf

De Groote, M., Pace, N., Fulton, K., Falkingham, J. 2006. Relationships between Mycobacterium isolates from patients with pulmonary mycobacterial infection and potting soils. Appl. Environ. Microbiol. 72, 7602-7606.

De Groote, M., Strausbaug, L., Jernigan, D., Liedke, L. 2002. Infections caused by rapidly growing mycobacteria: experience of the infectious disease consultants. In: CDC (ed.) International Conference on Emerging Infectious Diseases. CDC, Atlanta.

DeSimone, J., Pomerantz, R., Babinchak, T. 2000. Inflammatory reactions in HIV-1 infected persons after initiation of highly active antiretroviral therapy. Ann. Intern. Med. 133, 447-454.

Devallois, A., Goh, K., Rastogi, N. 1997. Rapid Identification of Mycobacteria to Species Level by PCR-Restriction Fragment Polymorphsm Analysis of the hsp65 Gene and Proposition of an Algorithm to Differentiate 34 Mycobacterial Species. J. Clin. Microbiol. 35(11), 2969-2973.

DeZuane, J. 1990. Handbook of Drinking Water Quality Standards and Controls. Van Nostrand Rein-hold, New York.

Dhople, A., Storrs, E., Lamoureux, L. 1992. Isolation of cultivable mycobacteria from feces and lungs of armadillos infected with Mycobacterium leprae. Inter. J. Leprosy Other Mycobacterial Dis. 60, 244-249.

Diniz, L., Zandonade, E., Dietze, R., Pereira, F., Ribeiro-Rodrigues, R. 2001. Short report: do intestinal nematodes increase the risk of multibacillary leprosy? Am. J. Trop. Med. Hyg. 65, 852-854.

192

Dixon, S., McDonald, S. 2002. The impact of HIV and AIDS on Africa‘s economic development. BMJ. 324(7331), 232-234.

Dobos, K., Quinn, F., Ashford, D., Horsburgh, R., King, C. 1999. Emergenceof a Unique Group of Necrotizing Mycobacterial Diseases. Emerg. Infect. Dis. 5, 367-378.

Dos Santos, R., Scheid, K., Goldani, L. 2010. Disseminated nontuberculous mycobacterial disease in patients with acquired immune deficiency syndrome in the south of Brazil. Trop Doct. 40(4), 211-213.

Doyle, T. 1954. Isolation of Johne‘s bacilli from the udders of clinically affected cows. British Vet. J. 110, 215-218.

Doyle, T. 1956. Johne‘s Disease. Vet. Rec. 68, 869-886.

Dra, M., Gaafar, A., Baranano, M., Notario, J., Cancer, R., Cebrian, F. 2005. Clinical and Epidemiological Study of Disease Caused by Mycobacterium kansasii in the Metropolitan Area of Bilbao, Spain. Arch. Bronconeumol. 41(4), 189-196.

Drancourt, M., Bollet, C., Carlioz, A., Martelin, R., Gayral, J., Raoult, D. 2000. 16s ribosomal sequence analysis of a large collection of environmental and clinical unidentifiable bacterial isolates. J. Clin. Microbiol. 38, 3623-3630.

Duker, A., Carranza, E., Hale, M. 2004. Spatial dependency of Buruli ulcer prevalence on arsenic-enriched domains in Amansie West District, Ghana: implications for arsenic mediation in Mycobacterium ulcerans infection. Int. J. Health Geogr. 3, 19.

Drummond, C. 1999. Victorian Infectious Diseases. Laboratory Report. 2, 1. Available at: http://www.dhs.vic.gov.au/phb/vidb/back.htm.

Du Moulin, G., Stottmeier, K., Pelletier, P., Tsang, A., Hedley-Whyte, J. 1988. Concentration of Mycobacterium avium by Hospital Hot Water Systems. JAMA. 260, 1599-1601.

Du Moulin, G., Sherman, I., Hoaglin, D., Stottmeier, K. 1985. Mycobacterium avium complex, an emerging pathogens in Massachusetts. J. Clin. Microbiol. 22, 9-12.

Du Moulin, G., Stottmeier, K. 1978. Use of cetylpyridinium chloride in the decontamination of water for the culture of mycobacteria. Appl. Environ. Microbiol. 36, 771-773.

Dundee, L., Grant, I., Ball, H., Rowe, M. 2001. Comparative evaluation of four decontamination protocols for the isolation of Mycobacterium avium subsp. paratuberculosis from milk. Letters Appl. Microbiol. 33, 173-177.

193

Eamens, G., Whittington, R., Marsh, I., Turner, M., Saunders, V., Kemsley, P., Rayward, D. 2000. Comparative sensitivity of various faecal culture methods and ELISA in dairy cattle herds with endemic Johne‘s disease. Vet. Microbiol. 77, 357- 367.

Eaton, T., Falkingham, J., von Reyn, C. 1995. Recovery of Mycobacterium avium from cigarettes. J. Clin. Med. 33, 2757-2758.

Eckstein, T., Inamine, J., Lambert, M., Belisle, J. 2000. A genetic mechanism for deletion of the ser2 gene cluster and formation of rough morphological variants of Mycobacterium avium. J. Bacteriol. 182, 6177-6182.

Edgar, R. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32, 1792-1797.

Edwards, L., Acquaviva, F., Livesay, V., Cross, F., Palmer, C. 1969. An atlas of sensitivity to tuberculin, purified protein derivative-B, and histoplasmin in the United States. Am. Rev. Respir. Dis. 99, 1–132.

Edwards L, Smith, D. 1965. Community-Wide Tuberculin Testing Study in Pamlico County, North Carolina. Am. Rev. Respir. Dis. 92, 43–54.

Eisen, R., Bearden, S., Wilder, A., Montenieri, J., Antolin, M., Gage, K. 2006. Early- phase transmission of Yersinia pestis by unblocked fleas as a mechanism explaining rapidly spreading plaque epizootics. PNAS USA. 103, 15380-15385.

Eishi, Y., Suga, M., Ishige, I., Kobayashi, D., Yamada, T., Takemura, T., Takizawa, T., Koike, M., Kudoh, S., Costabel, U., Guzman, J., Rizzato, G., Gambacorta, M., du Bois, R., Nicholson, A., Sharma, O., Ando, M. 2002. Quantitative Analysis of Mycobacterial and Propionibacterial DNA in Lymph Nodes of Japanese and European Patients with Sarcoidosis. J. Clin. Microbiol. 40, 198-204.

El Bohy, M., Raafat, H., Aly, F., El-Aziz, M. 2009. Comparison Between Mycobacteria Growth Indicator Tube (MGIT), BACTEC 460 TB System, and Lowenstein-Jensen Medium for Detection of Mycobacterium tuberculosis. Egyptian J. Bronch. 3(2), 102-109.

El Sahly, H., Septimus, E., Soini, H., Septimus, J., Wallace, R., Pan, X., Williams- Bouyer, N., Musser, J., Graviss, E. 2002. Mycobacterium simiae Pseudo-outbreak Resulting from a Contaminated Hospital Water Supply in Houston, Texas. Clin. Infect. Dis. 35, 802-807.

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.

Emde, K., Chomyc, S., Finch, G. 1992. Initial investigation on the occurrence of Mycobacterium species in swimming pools. J. Environ. Health. 54, 34-37.

194

Eneh, K., Zahir, M., Mora, M., Schmidt, F., Enriquez, D., Hammoudeh, F., Neupane, N., Quist, J. 2010. Disseminated Mycobacterium avium intracellulare Infection in an ―Immunocompetent‖ Host. Southern Med J. 103(7), 693-696.

Environmental Protection Agency. 2009. Mycobacteria: Health Advisory. United States Environmental Protection Agency. Washington, DC. EPA-822-B-01-007.

Environmental Protection Agency. 2002. Standard Operating Procedure for Dissolved Organic Carbon. United States Environmental Protection Agency. Washington, DC. EPA-8211-LG-03-002.

Environmental Protection Agency. 2002. Mycobacteria: Drinking Water Fact Sheet. United States Environmental Protection Agency. Washington, DC. EPA-822-F-02- 002.

Environmental Protection Agency. 1993. Methods for the determination of inorganic substances in environmental samples. United States Environmental Protection Agency. Washington, DC. EPA-600-R-93-100.

Epstein, J., Payne, K., Kramer, E. 2002. Techniques for Mapping Suburban Sprawl. Photogram. Engin. Remote Sensing. 63(9), 913-918.

Euzéby, J. 1997. List of bacterial names with standing in nomenclature: A folder available on the internet. www.bacterio.net. Intern. J. Systematic Bacteriol. 47, 590-592.

Falkingham, J., 1996. Epidemiology of Infection by Nontuberculous Mycobacteria. Clin. Microbiol. 9(2), 177-215.

Falkingham, J. 2002. Nontuberculous mycobacteria in the environment. Clin. Chest Med. 23, 529-551.

Falkingham, J., 2003. The Changing Pattern of Nontuberculous Mycobacterial Disease. Canad. J. Infect. Dis. 14(5), 281-286.

Falkingham, J., 2003. Mycobacterial Aerosols and Respiratory Disease. Emerging Infect. Dis. 9(7), 763-767.

Falkingham, J. 2009. Physiological Ecology of Environmnetal Saprophytic and Potentially Pathogenic Mycobacteria. In: J. Kazda (ed.)The Ecology of Mycobacteria: Impact on Animal‘s and Human‘s Health. Springer Science, New York.

Falkingham, J. 2009. Surrounded by mycobacteria: nontuberculous mycobacteria in the human environment. J. Appl. Environ. Microbiol. 107, 356–367.

Falkingham, J. 2010. Impact of human activities on the ecology of nontuberculous mycobacteria. Future Microbiology. 5(6), 951-960.

195

Falkingham, J. 2010. Hospital water filters as a source of Mycobacterium avium complex. J. Med. Microbiol. 59, 1198-1202.

Falkingham, J. 2011. Nontuberculous mycobacteria from household plumbing of patients with nontuberculous mycobacteria disease. EID. 17(3), 419-424.

Falkingham, J., Bruce, C., Gruft, H. 1980. Epidemiology of infection by nontuberculous mycobacteria. Am. Rev. Respir. Dis. 121, 931-937.

Falkingham, J., George, K., Ford, M., Parker, B. 1990. Collection and characteristics of mycobacteria in aerosols, pp. 71-81. In: Morey, P., Feeley, J., and Otten, J., (ed.) Biological Contaminants in Indoor Environments. American Society for Testing and Materials, Philadelphia.

Falkingham, J., George, K., Parker, B., Gruft, H. 1984. In vitro susceptibility of human and environmental isolates of Mycobacterium avium, M. intracellulare, and M. scrofulaceum to heavy metal salts and oxyanions. Antimicrob. Agents Chemotherap. 25, 137-139.

Falkingham, J., Norton, C., LeChevallier, M. 2001. Factors Influencing Numbers of Mycobacterium avium, Mycobacterium intracellulare, and other mycobacteria in Drinking Water Distribution Systems. Appl. Environ. Microbiol. 67, 1225-1231.

Falkoff, G., Rigsby, C., Rosenfield, A. 1987. Partial, combined cortical and medullary nephrocalcinosis: US and CT paterns in AIDS-associated MAI infection. Radiol. 162, 343-344.

Fang Y., Wu, W., Pepper, J., Larsen, J., Marras, S., Nelson, E., Epperson, W., Hennings, J. 2002. Comparison of Real-Time, Quantitative PCR with Molecular Beacons to Nested PCR and Culture Methods for Detection of Mycobacterium avium subsp. paratuberculosis in Bovine Fecal Samples. J. Clin. Microbiol. 40(1), 287-291.

Farooq, S., Chian, E., Engelbrecht, R. 1977. Basic concepts in disinfection with ozone. J. Water Pollution Control Fed. 49(8), 1818-1831.

Fattorini, L., Nisini, R., Fan, Y., Li, Y., Tan, D., Mariotti, S., Teloni, R., Iona, E., Orefici, G. 2002. Exposure to BALB/c mice to low doses of Mycobacterium avium increases resistnace to a subsequent high-dose infection. Microbiol. 148, 3173- 3181.

Fawcett, A., Goddard, P., McKelvey, W., Buxton, D., Reid, H., Greig, A., Macdonald, A. 1995. Johne‘s disease in a herd of farmed red deer. Vet. Rec. 136, 165-169.

Feazel, L., Baumgartner, L., Perterson, K., Frank, D., Harris, J., Pace, N. 2009. Opportunistic Pathogens Enriched in Showerhead Biofilms. PNAS. 106(38), 16393-16399.

196

Fernald, E., Purdum, E. 1998. Water Resources Atlas of Florida. Institute of Science and Public Affairs, Florida State University, Tallahassee.

Ferroglio, E., Nebbia, P., Robino, P., Rossi, L., Rosati, S. 2000. Mycobacterium paratuberculosis infection in two free-ranging Alpine ibex. Rev. Sci. Tech. 19, 859- 862.

Field, S., Cowie, R. 2006. Lung Disease Due to the More Common Nontuberculous Mycobacteria. Am. College Chest Physicians. 129, 1653-1672.

Findlay, S., Sinsabaugh, R. 2002. Aquatic Ecosystems: Interactivity of Dissolved Organic Matter. Academic Press, New York.

Fine, P. 1995. Variation in Protection by BCG: Implications of and for Heterologous Immunity. Lancet. 346, 1339–45.

Fischeder, R., Schulze-Robbecke, R., Weber, A. 1991. Occurrence of mycobacteria in drinking water samples. Zentralbl. Hyg. Umweltmed. 192, 154-158.

Fischer, O., Motlova, L., Dvorska, L., Svastova, P., Bartl, J., Melicharek, I., Weston, R., Pavlik, I. 2001. Diptera as vectors of mycobacterial infections in cattle and pigs. Med. Vet. Entomol. 15, 208-211.

Fischer, O., Matlova, L., Bartl, J., Dvorska, L., Svastova, P., du Maine, R., Melicharek, I., Bartos, M., Pavlik, I. 2003. Earthworms (Oligochaeta, Lumbricidae) and Mycobacteria. Vet. Microbiol. 25, 325- 338.

Fleischman, R., du Moulin, G., Esber, H., Ilievski, V., Bogden, A. 1982. Nontuberculous mycobacterial infection attributable to Mycobacterium intracellulare serotype 10 in two rhesus monkeys. J. Am. Vet. Med. Ass. 181, 1358-1362.

Florax, R., Rey, S. 1995. The impact of misspecified spatial structure in linear regression models, pp.111-135. In: Anselin, L., Florax, R. (ed.). New Directions in Spatial Econometrics. Springer, Berlin.

Florida Demographic Estimating Conference. 2010. Florida Office of Economic and Demographic Research, Florida Department of Health, Tallahassee. Available at: http://edr.state.fl.us/Content/population-demographics/data/index.cfm.

Ford, T. 1999. Microbiological safety of drinking water: United States and global perspectives. Environ. Health Persp. 107, 191-206.

Fordham von Reyn, C., Waddell, R., Eaton, T., Arbeit, R., Maslow, J., Barber, T., Brindle, R., Gilks, C., Lumio, J., Lahdevirta, J., Ranki, A., Dawson, D., Falkinham, J. 1993. Isolation of Mycobacterium avium Complex from Water in the United States, Finland, Zaire, and Kenya. J. Clin. Microbiol. 31(12), 3227-3230.

197

Fosgate, G., Carpenter, T., Chomel, B., Case, J., De Bess, E., Reilly, K. 2002. Time- Space Clustering of Human Brucellosis, California, 1973-1992. Emerg. Infect. Dis. 8(7), 672-678.

Frehel, C., de Chastellier, C., Lang, T., Rastogi, N. 1986. Evidence for inhibition of fusion of lysosomal and pre-lysosomal compartments with phagosomes in macrophages infected with pathogenic Mycobacterium avium. Infect. Immun. 52, 252-262.

Frehel, C., de Chastellier, C., Offredo, C., Berche, P. 1991. Intramacrophage growth of Mycobacterium avium during infection of mice. Infect. Immun. 59, 2207-2214.

Freije, M. 2000. Spas, Hot tubs, and Whirlpool Bathtubs: A Guide for Disease Prevention. HC Information Resources, Inc., Fallbrook.

Freitas, J., Panetta, J., Curcio, M., Ueki, S. 2001. Mycobacterium avium complex in water buffaloes slaughtered for consumption. Review Saude Publica. 35, 315-317.

French, N., Kelly, L., Jones, R., Clancy, D. 2002. Dose-response relationships for foot and mouth disease in cattle and sheep. Epidemiol. Infect. 128, 325-332.

Frosch, M., Roth, J., Ullrich, K., Harms, E. 2000. Successful treatment of Mycobacterium avium osteomyelitis and arthritis in a non-immunocompromised child. Scandivanian J. Infect. Dis. 32, 328-329.

Fujita, J., Nanki, N., Negayama, K., Tsutsui, S., Taminato, T., Ishida, T. 2002. Nosocomial contamination by Mycobacterium gordonae in hospital water supply and super-oxidized water. J. Hospital Infections. 51, 65-68.

Gaafar, A., Unzaga, M., Cisterna, R., Leal, M., Bustamante, V., Tirapu, J., Crespo, J. 2001. Separate Nocardia Infections in a Patient with Chronic Granulomatous Disease. J. Clin. Microbiol. 39, 3015-3016.

Galil, K., Miller, L., Yakrus, M., Wallace, R., Mosley, D., England, B., Huitt, G., McNeil, M., Perkins, B. 1999. Abscesses due to Mycobacterium abscessus linked to injection of unapproved alternative medication. Emerg. Infect. Dis. 5, 681-687.

Gao, A., Mutharia, L., Chen, S., Rahn, K., Odumeru, J. 2002. Effect of pasteurization on survival of Mycobacterium paratuberculosis in milk. J. Dairy Sci. 85, 3198-3205.

Garnier, T., Eiglmeier, K., Camus, J., Medina, N., Mansoor, H., Pryor, M., Duthoy, S., Grondin, S., Lacroix, C., Monsempe, C., Simon, S., Harris, B., Atkin, R., Doggett, J., Mayes, R., Keating, L., Wheeler, P., Parkhill, J., Barrell, B., Cole, S., Gordon, S., Hewinson, R. 2003. The complete genome sequence of Mycobacterium bovis. PNAS. 100(13), 7877-7882.

198

Geldreich, E., LeChevallier, M. 1999. Microbial Water Quality in Distribution Systems, pp. 18.1-18.49. In: Letterman, R., (ed.) Water Quality and Treatment, 5th edition. McGraw-Hill, New York.

Gentry, C. 2004. Atypical Mycobacteria. Pharmacotherapy Self-Assessment Program. 5, 99-126.

George, K., Chatterjee, D., Gunawardana, G., Welty, D., Hayman, J., Lee, R., Small, P. 1999. Mycolactone: a polyketide toxin from Mycobacterium ulcerans required for virulence. Science. 283, 854-857.

George, K., Falkinham, J. 1986. Selective medium for the isolation and enumeration of Mycobacterium avium-intracellulare and M. scrofulaceum. Canad. J. Microbiol. 32, 10–14.

George, K., Parker, B., Gruft, H., Falkingham, J. 1980. Epidemiology of infection by nontuberculous mycobacteria. II. Growth and survival in natural waters. Am. Rev. Respir. Dis. 122, 89-94.

Gerszten, E., Allison, M., Dalton, H., Brummer, D. 1968. An Epidemiologic Study of Unclassified Mycobacteria in Virginia. Southern Med. J. 61(3), 275-277.

Getis, A., Ord, J. 1992. The analysis of spatial association by use of distance statistics. Geo. Anal. 24, 189-260.

Ghosh, I., Stains, C., Ooi, A., Segal, D. 2006. Direct detection of double-stranded DNA:molecular methods and applications for DNA diagnostics. Molecular Biosystems. 2, 551-560.

Gilligan, P. 1991. Microbiology of airway disease in patients with cystic fibrosis. Clin. Microbiol. Rev. 4(1), 35-51.

Glover, N., Holtzman, A., Aronson, T., Froman, S., Berlin, O., Dominguez, P., Kunkel, K., Overturf, G., Stelma, G., Smith, C., 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. Intern. J. Environ. Health Research. 4, 63-72.

Gomila, M., Ramirez, A., Lalucat, J. 2007. Diversity of Environmental Mycobacterium Isolates from Hemodialysis Water as Shown by a Multigene Sequencing Approach. Appl. Environ. Microbiol. 73(12), 3787-3797.

Good, R., Snider, D. 1982. Isolation of nontuberculous mycobacteria in the United States, 1980. J. Infect. Dis. 146, 829-833.

Goodwin, B., Jerome, C., Bullock, B. 1988. Unusual lesion morphology and skin test reaction for Mycobacterium avium complex in macaques. Lab. Animal Sci. 38, 20- 24.

199

Gopinath, S. 2010. Non-Tuberculous Mycobacteria in TB-Endemic Countries: Are We Neglecting the Danger? PLoS Negl. Trop. Dis. 4(4), e615.

Graham, L. Jr., Warren, N., Tsang, A., Dalton, H. 1988. Mycobacterium avium Complex pseudobacteriuria from Hospital Water Supply. J. Clin. Microbiol. 26, 1034-1036.

Gram, C. 1884. Uber die isolirte Farbung der Schizomyceten in Schnitt- und Trockenpreparaten. Fortschr. Med. 2, 185-189.

Greig, A., Stevenson, K., Henderson, D., Perez, V., Hughes, V., Pavlik, I., Hines, M., McKendrick, I., Sharp, M. 1999. Epidemiological study of paratuberculosis in wild rabbits in Scotland. J. Clin. Microbiol. 37, 1746-1751.

Griffith, D. 1997. Nontuberculous mycobacteria. Current Opinions Pulm. Med. 3, 139- 145.

Griffith, D. 2003. Emergence of Nontuberculous Mycobacteria as Pathogens in Cystic Fibrosis. Am. J. Resp. Critical Care Med. 167, 810-812.

Griffith, D., Aksamit, T., Brown-Elliott, B., Catanzaro, A., Daley, C., Gordin, F., Holland, S., Horsburgh, R., Huitt, G., Iadmarco, M., Iseman, M., Olivier, K., Ruoss, S., Fordham von Reyn, C., Wallace, R., Winthrop, K. 2007. An Official ATS/IDSA Statement: Diagnosis, Treatment, and Prevention of Nontuberculous Mycobacterial Diseases. Am. J. Respir. Critical Care Med. 175, 367-416.

Griffiths, P., Babb, J., Bradley, C., Fraise, A. 1997. Glutaraldehyde-resistant Mycobacterium chelonae from endoscope washer disinfectors. J. Appl. Microbiol. 82, 519-526.

Grubesic, T. 2008. Zip codes and spatial analysis: Problems and prospects. Socio- Econ. Planning Sci. 42(2), 129-149.

Gruft, H., Falkingham, J., Parker, B. 1981. Recent Experience in the Epidemiology of Disease Caused by Atypical Mycobacteria. Rev. Infect. Dis. 3(5), 990-996.

Gruft, H., Katz, J., Blanchard, D. 1975. Postulated source of Mycobacterium intracellulare (Battey) infection. Am. J. Epidemiol. 102, 311.

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, 1245-1249.

Gullick, R., Lechevallier, M., Case, J., Wood, D., Funk, J., Friedman, M. 2005. Application of pressure monitoring and modeling to detect and minimize low pressure events in distribution systems. J. Water Supply: Res. Technol. 54(2), 65– 81.

200

Haining, R. 1990. Spatial data analysis in the social and environmental sciences. Cambridge University Press, Cambridge.

Hale, Y., Pfyffer, G., Salfinger, M. 2001. Laboratory Diagnosis of Mycobacterial Infections: New tools and Lessons Learned. Med. Microbiol. 33, 834-846.

Hall, H., Ruiguang, S., Rhodes, P. 2008. Estimation of HIV incidence in the United States. JAMA. 300, 520-529.

Hall-Stoodley, L., Lappin-Scott, H. 1998. Biofilm Formation by the rapidly growing mycobacterial species Mycobacteria fortuitum. FEMS Microbiol. Lett. 168, 77–84.

Hall-Stoodley, L., Keevil, C., Lappin-Scott, H. 1999. Mycobacterium fortuitum and Mycobacterium chelonae biofilm formation under high and low nutrient conditions. J. Appl. Microbiol. 85, 60S-69S.

Hall-Stoodley, L., Stoodley, P. 2005. Biofilm formation and dispersal and the transmission of human pathogens. Trends Microbiol. 13(1), 7-10.

Happold, J., Brunhart, I., Schwermer, H., Stark, K. 2008. Surveillance of H5 avian influenza virus in wild birds found dead. Avian Dis. 52(1), 100-105.

Harrington, G., Chen, H., Harris, A., Xagoraraki, I., Battigelli, D., Standridge, J. 2001. Removal of emerging waterborne pathogens. AWWA Research Foundation and the American Water Works Association, Denver.

Harris, B., Barletta, R. 2001. Mycobacterium avium subsp. paratuberculosis in Veterinary Medicine. Clin. Microbiol. Rev. 14, 489-512.

Harris, J., Lammerding, A. 2001. Crohn‘s disease and Mycobacterium avium subsp. paratuberculosis: current issues. J. Food Prot. 64, 2103– 2110.

Havelaar, A., Berwald, L., Groothuis, D., Baas, J. 1985. Mycobacteria in semi-public swimming pools and whirlpools. Zentralbl Bakteriol Mikrobiol Hyg B. 180, 505-514.

Havlik, J., Horsburgh, C., Metchcock, B., Williams, P., Fann, S., Thompson, S. 1992. Disseminated Mycobacterium avium complex infection: clinical identification and epidemiological trends. J. Infect. Dis. 165, 577-580.

Havlir, D., Dupe, M., Sattler, F., Furthal, D., Kemper, C., Dunne, M., Parenti, D., Lavelle, J., White, A., Witt, M., Bozzette, S., McCuthan, J. 1996. Prophylaxis against disseminated Mycobacterium avium complex with weekly azithromycin, daily rifabutin or both. New England J. Med. 335, 392-398.

Hay, S., Snow, R. 2006. The Malaria Atlas Project: Developing Global Maps of Malarial Risk. PLoS Med. 3(12), 2204-2208.

201

He, J., Jiang, S. 2005. Quantification of enterococci and human adenoviruses in environmental samples by real-time PCR. Appl. Environ. Microbiol. 71, 2250-2255.

Heckert, R., Elankumaran, S., Milani, A., Baya, A. 2001. Detection of a new Mycobacterium species in wild striped bass in the Chesapeake Bay. J. Clin. Microbiol. 39, 710-715.

Heidari, F., Farnia, P., Nowroozi, J., Majd, A., Masjedi, M., Velayati, A. 2010. Evaluating the sensitivity of three primers using PCR-restriction fragment length polymorphism analysis for rapid identification of Mycobacterium simiae isolated from pulmonary tuberculosis patients. Iranian J. Clin. Infect. Dis. 5(1), 30-35.

Helie, P., Higgins, R. 1996. Mycobacterium avium complex abortion in a mare. J. Vet. Diag. Invest. 8, 257-258.

Hellinger, W., Smilack, J., Greider, J., Alvarez, S., Trigg, S., Brewer, N., Edson, R. 1995. Localized soft-tissue infections with Mycobacterium avium/ Mycobacterium intracellulare complex in immunocompetent patients: granulomatous tenosynovitis of the hand or wrist. Clin. Infect. Dis. 21, 65-69.

Hermon-Taylor, J. 1993. Causation of Crohn‘s disease: The impact of clusters. Gastroenterology. 104, 643-646.

Hermon-Taylor, J., Barnes, N., Clarke, C., Finlayson, C. 1998. Mycobacterium paratuberculosis cervical lymphadenitis, followed five years later by terminal ileitis similar to Crohn‘s disease. BMJ. 316, 449- 453.

Hermon-Taylor, J., Bull, T., Sheridan, J., Cheng, J., Stellakis, M., Sumar, N. 2000. The causation of Crohn‘s disease by Mycobacterium avium subsp. paratuberculosis. Canad. J. Gastroenterol. 14, 521-539.

Hilborn, E., Covert, T., Yakrus, M., Harris, S., Donnelly, S., Rice, E., Toney, S., Bailey, S., Stelma, G. 2006. Persistence of Nontuberculous Mycobacteria in a Drinking Water System After Addition of Filtration Treatment. Appl. Environ. Microbiol. 72(9), 5864-5869.

Hodgson, M., Bracker, A., Yang, C., Storey, E., Jarvis, B., Milton, D., Lummus, Z., Bernstein, D., Cole, S. 2001. Hypersensitivity pneumonitis in a metal-working environment. Am. J. Industiral Med. 39, 616-628.

Hoffman, C., Leis, A., Niederweis, M., Plitzko, J., Engelhardt, H. 2008. Disclosure of the mycobacterial outer membrane: cryo-electron tomography and vitreous sections reveal the lipid bilayer structure. PNAS. 105(10), 3963-3967.

Hoffman, R., Marshall, M., Gibson, M., Rochelle, P. 2009. Prioritizing Pathogens for Potential Future Regulation in Drinking Water. Envrion. Sci. Technol. 43(14), 5165-5170.

202

Hoffner, S., Kallenius, G., Petrini, B., Brennan, P., Tsang, A. 1990. Serovars of Mycobacterium avium complex isolated from patients in Sweden. J. Clin. Microbiol. 28, 1105-1107.

Hoppert, M., Mayer, F. 1999. Principles of macromolecular organization and cell function in bacteria and archea. Cell. Biochem. Biophys. 31, 247-284.

Horn, B., Forshaw, D., Cousins, D., Irwin, P. 2000. Disseminated Mycobacterium avium infection in a dog with chronic diarrhoea. Australian Vet. J. 78, 320-325.

Horsburgh, C., Jr. 1991. Mycobacterium avium Complex Infection in Acquired Immunodeficiency Syndrome. New England J. Med. 324, 1332-1338.

Horsburgh, C., Jr. 1996. Epidemiology of Disease Caused by Nontuberculous Mycobacteria. Semin. Respiratory Infection. 11, 244-251.

Horsburgh, C., Jr., Gettings, J., Alexander, L., Lennox, J. 2001. Disseminated Mycobacterium avium Complex Disease Among Patients Infected with Human Immunodeficiency Virus, 1985-2000. Clin. J. Infect. Dis. 33, 1938-1943.

Horsburgh, C., Jr., Chin, D., Yajko, D., Hopewell, P., Nassos, P., Elkin, E., Hadley, W., Stone, E., Simon, E., Gonzalez, P. 1994. Environmental risk factors for acquistion of Mycobacterium avium complex in persons with human immunodeficiency virus infection. J. Infect. Dis. 170, 362-367.

Horsburgh, C., Jr., Havlik, J., Ellis, D., Kennedy, E., Fann, S., Dubois, R., Thompson, S. 1991. Survival of patients with acquired immune deficiency syndrome and disseminated Mycobacterium avium complex infection with and without antimycobacterial chemotherapy. Am. Rev. Respir. Dis. 144, 557-559.

Horsburgh, C., Mason, R., Farhi, D., Iseman, M. 1985. Disseminated infection with Mycobacterium avium-intracellulare: a report of 13 cases and a review of the literature. Med. 64, 36-48.

Hospodsky, D., Yamamoto, N., Peccia, J. 2010. Accuracy, precision, and method detection limits of quantitative PCR for airborne bacteria and fungi. Appl Environ Microbiol. 76(21), 7004-7012.

Howard, S., Byrd, T. 2000. The rapidly growing mycobacteria: saprophytes and parasites. Microbes Infect. 2, 1845-1853.

Huang, C., Tsai, Y., Shu, C., Lei, Y., Wang, J., Yu, C., Lee, L., Yang, P., the TAMI Group. 2009. Clinical Significnce of isolation of nontuberculous mycobacteria in pulmonary tuberculosis patients. Respir. Med. 103(10), 1484-1491.

Huang, J., Kao, P., Adi, V., Ruoss, S. 1999. Mycobacterium avium-intracellularae pulmonary infection in HIV-negative patients without pre-existing lung disease: disgnostic and management limitations. Chest. 115, 1033-1040.

203

Hughes, V., Stevenson, K., Sharp, J. 2001. Improved preparation of high molecular weight DNA for pulsed-field gel electrophoresis from mycobacteria. J. Microbiol Methods. 44, 209-215.

Huminer, D., Pitlik, S., Block, C., Kaufman, L., Amit, S., Rosenfeld, J. 1986. Aquarium borne Mycobacterium marinum skin infection. Report of a case and review of the literature. Arch. Dermatol. 122, 698-703.

Hunter, D. 1996. Tuberculosis in free-ranging, semi free-ranging, and captive cervids. Rev. Sci. Technol. 15, 171-181.

Hunter, P. 1997. Waterborne Disease: Epidemiology and Ecology. Wiley, New York.

Hussein, Z., Landt, O., Wirths, B., Wellinghausen, N. 2009. Detetction of non-tuberclous mycobacteria in hospital water by culture and molecular methods. Int. J. Med. Microbiol. 229(4), 281-290.

Iivanainen, E., Katila, M., Martikainen, P. 1999. Mycobacteria in drinking water networks; occurance in water and loose deposits, formation of biofilms. Proceedings of the European Society for Microbiology, Lucerne.

Iivanainen, E., Martikainen, P., Katila, M. 1995. Effect of freezing water samples on viable counts of environmental mycobacteria. Letters Appl. Microbiol. 21, 257-260.

Iivanainen, E., Martikainen, P., Katila, M. 1997. Comparison of some decontamination methods and growth media for isolation of mycobacteria from northern brook waters. J. Appl. Bacteriol. 82, 121-127.

Iivanainen, E., Martikainen, P., Räisänen, M., Katila, M. 1997. Mycobacteria in boreal coniferous forest soils. FEMS Microbiol. Ecol. 23, 325-332.

Iivanainen, E., Martikainen, P., Vaananen, P., Katila, M. 1999. Environmental factors affecting the occurrence of mycobacteria in brook sediments. J. Appl. Microbiol. 86, 673-681.

Iivanainen, E., Northrup, J., Arbeit, R., Ristola, M., Katila, M., von Reyn, C. 1999. Isolation of mycobacteria from indoor swimming pools in Finland. APMIS. 107, 193-200.

Iivanainen, E., Sallantus, T., Katila, M., Martikainen, P. 1999. Mycobacteria in runoff- waters from natural and drained peatlands. J. Environ. Quality. 28, 1226-1234.

Inderlied, C., Kemper, C., Bermudez, L. 1993. The Mycobacterium avium complex. Clin. Microbiol. Rev. 6, 266-310.

Ingen, J., Blaak, H., de Beer, J., de Roda Husman, A., van Soolingen, D. 2010. Rapidly growing nontuberculous mycobacteria cultured from home tap and shower water. Appl. Environ. Microbiol. In Press. Available online 2011 March 1.

204

Inyaku, K., Hiyama, K., Ishioka, S., Inamizu, T., Yamakido, M. 1993. Rapid detection and identification of mycobacteria in sputum samples by nested polymerase chain reaction and restriction fragment length polymorphisms of dnaJ heat shock protein gene. Hiroshima J. Med. Sci. 42, 21-31.

Iseman, M., Buschman, D., Ackerson, L. 1991. Pectus excavatum and scoliosis. Thoracic anomalies associated with pulmonary disease caused by Mycobacterium avium complex. Am. Rev. Respir. Dis. 144, 914-916.

Jacangelo, J., Patania, N., Trussell, R., Hass, C., Gerba, C. 2002. Inactivation of waterborne emerging pathogens by selected disinfectants. AWWA Research Foundation and the American Water Works Association, Denver.

Jacob, C., Henein, S., Heurich, A., Kamholz, S. 1993. Nontuberculous mycobacterial infection of the central nervous system in patients with AIDS. South. Med. J. 86, 638-640.

Jacobs, J., Rhodes, M., Sturgis, B., Wood, B. 2009. Influence of Environmental Gradients on the Abundance and Distribution of Mycobacterium spp. in a Coastal Lagoon Estuary. Appl. Env. Microbiol. 75(23), 7378-7384.

Johne, H., Frothingham, L. 1895. Ein eigenthumlicher Fall von Tuberculose beim Rind. Deutsche Zeitschr. Tierm. Path. 21, 438-454.

Johnson-Ifearulundu Y., Kaneene, J. 1997. Relationship between soil type and Mycobacterium paratuberculosis. JAVMA. 210, 1735-1740.

Jones, S. 1969. Mycobacterium kansasii in East Kent: A report of seven pulmonary infections with an environmental study. British J. Dis. Chest. 63(2), 83-95.

Jones, J., Falkingham, J. 2003. Decolorization of Malachite Green and Crystal Violet by Waterborne Pathogenic Mycobacteria. Antimicrob. Agents Chemo. 47(7), 2323- 2326.

Judge, J., Kyriazakis, I., Greig, A., Allcroft, D., Hutchings, M. 2005. Clustering of Mycobacterium avium subsp. paratuberculosis in rabbits and the environment: How hot is a hot spot? Appl. Environ. Microbiol. 71(10), 6033-6038.

Judson, F., Feldman, R. 1974. Mycobacterial skin tests in humans 12 years after infection with mycobacterium marinum. Am. Rev. Respir. Dis. 109, 544-547.

Kahana, L., Kay, J., Yakrus, M., Wasserman, S. 1997. Mycobacterium avium complex infection in an immunocompetent young adult related to hot tub exposure. Chest. 111, 242-245.

Kalis, C., Hesselink, J., Barkema, H., Collins, M. 2000. Culture of strategically pooled bovine fecal samples as a method to screen herds for paratuberculosis. J. Vet. Diagn. Invest. 12, 547-551.

205

Kamala, T., Paramasivan, C., Herbert, D., Venkatesan, P., Prabhakar, R. 1994. Evaluation of procedures for isolation of nontuberculous mycobacteria from soil and water. Appl. Environ. Microbiol. 60, 1021-1024.

Kane, S., Stine, C., Hungerford, L., Matsche, M., Driscoll, C., Baya, A. 2007. Mycobacteria as Environmental Portent in Chesapeake Bay Fish Species. Emerg. Infect. Dis. 13(2), 329-331.

Kang, C., Chung, Y., Lee, H., Kim, E., Hong, Y., Cho, S. 2010. Hot Tub Lung Due to Mycobacterium avium Complex in a Public Bath. Korean J Occup Environ Med. 22(2), 166-172.

Kannel, W., Wolf, P., Benjamin, E., Levy, D. 1998. Prevalence, incidence, prognosis, and predisposing conditions for atrial fibrillation: population-based estimates. Am. J. Cardiol. 16(82), 2N-9N.

Karakousis, P., Moore, R., Chaisson, R. 2004. Mycobacterium avium complex in Patients with HIV Infection In the Era of Highly Active Antiretroviral Therapy. Lancet Infect. Dis. 4, 557–65.

Karim, M., LeChevallier, M. 2005. Microbiological Concerns of Drinking Water Distribution Systems. Water Encyclopedia. 15, 341–346.

Karim, M., Pontius, F., LeChevallier, M. 2004. Detection of noroviruses in water- summary of an international workshop. J. Infect. Dis. 189(1), 21-28.

Kasai, H., Ezaki, T., Harayama, S. 2000. Differentiation of phylogenetically related slowly growing mycobacteria by their gyrB sequences. J. Clin. Microbiol. 38, 301- 308.

Kaser, M., Hauser, J., Pluschke, G. 2009. Single Nucleotide Polymorphisms on the Road to Strain Differentiation in Mycobacterium ulcerans. J. Clin. Microbiol. 47, 3647-3652.

Kaser, M., Hauser, J., Small, P., Pluschke, G. 2009. Large Sequence Polymorphisms Unveil the Phylogenetic Relationship of Environmental and Pathogenic Mycobacteria Related to Mycobacterium ulcerans. Appl Environ Microbiol. 75(17), 5667-5675.

Katila, M., Brander, E., Backman, A. 1987. Neonatal BCG vaccination and mycobacterial cervical adenitis in childhood. Tubercle. 68, 291–296.

Katila, M., Iivanainen, E., Torkko, P., Kauppinen, J., Martikainen, P., Vaananen, P. 1995. Isolation of potentially pathogenic mycobacteria in the Finnish environment. Scandanavian J. Infect. Dis. Supplmental. 98, 9–11.

206

Katz, V., Farmer, R., York, J., Wilson, J. 2000. Mycobacterium chelonae sepsis associated with long-term use of an intravenous catheter for treatment of hyperemesis gravidarum. A case report. J. Reproductive Med. 45, 581-584.

Kaufman, A., Greene, C., Rakich, P., Weigner, D. 1995. Treatment of localized Mycobacterium avium complex infection with clofazimine and doxycycline in a cat. J. Am. Vet. Med. Assoc. 207, 457-459.

Kaufmann, S. 2011. Tuberculosis vaccines- a new kid on the block. Nature Med. 17, 159-160.

Kay, M., Linke, L., Triantis, J., Salman, M., Larsen, R. 2011. Evaluation of DNA Extraction Techniques for Detecting Mycobacterium tuberculosis Complex Organisms in Asian Elephant Trunk Wash Samples. J. Clin. Microbiol. 49, 618- 623.

Kazda, J., Irgens, L., Kolk, A. 1990. Acid-fast bacilli found in sphagnum vegetation of coastal Norway containing Mycobacterium-leprae specific phenolic glycolipid-11. Intern. J. Leprosy. 58, 353-357.

Kelejian, H., Robinson, D. 1998. A suggested test for spatial autocorrelation and/or heteroskedasticity and corresponding Monte Carlo results. Reg. Sci. Urban Econ. 28, 389- 417.

Kendall, P., Medeiros, L., Hillers, V., Chen, G., Dimascola, S. 2003. Food handling behaviors of special importance for pregnant women, infants and young children, the elderly, and immuno-compromised people. J. Am. Dietetic Assoc. 103(12), 1646-1649.

Kennedy, T., Weber, D. 1994. Nontuberculous mycobacteria, an underappreciated cause of geriatric lung disease. Am. J. Respir. Critical Care Med. 149, 1654-1658.

Khan, K., Wang, J., Marras, T. 2007. Nontuberculous mycobacterial sensitization in the United States: national trends over three decades. Am. J. Respir. Crit. Care Med. 176, 306–313.

Khoor, A., Leslie, K., Tazelaar, H., Helmers, R., Colby, T. 2001. Diffuse pulmonary disease caused by nontuberculous mycobacteria in immunocompetent people (hot tub lung). Am. J. Clin. Pathol. 115, 755-762.

Kim, R., Greenberg, D., Ehrmantraut, M., Guide, S., Ding, L., Shea, Y. 2008. Pulmonary nontuberculous mycobacterial disease: a prospective study of a distinct pre- exisiting syndrome. Am. J. Respir. Critical Care Med. 178, 1066-1074.

Kim, S., Goodman, J., Petrofsky, M., Bermudez, L. 1998. Mycobacterium avium infection of gut mucosa in mice associated with late inflammatory response and intestinal cell necrosis. J. Med. Microbiol. 47, 725–731.

207

Kirschner, P., Springer, B., Vogel, U., Meier, A., Wrede, A., Kiekenbeck, M., Bange, F., Bottger, E. 1993. Genotypic identification of mycobacteria by nucleic acid sequence determination: report of a 2-year experience in a clinical laboratory. J. Clin. Microbiol. 31(11), 2882-2889.

Kirschner, R., Parker, B., Falkingham, J. 1999. Humic and fulvic acids stimulate the growth of Mycobacterium avium. FEMS Microbiol. Ecol. 30, 327-332.

Kirschner, R., Parker, B., Falkinham, J. 1992. Epidemiology of infection by nontuberculous mycobacteria. Mycobacterium avium, Mycobacterium intracellulare, and Mycobacterium scrofulaceum in acid, brown-water swamps of the southeastern United States and their association with environmental variables. Am. Rev. Respir. Dis. 145, 271–275.

Kitaura, H., Ohara, N., Kobayashi, K., Yamada, T. 2001. TNF-alpha mediated activation of human immunodeficiency virus-1 LTR in monocytoid cells by mycobacteria. FEMS Immunol. Med. Microbiol. 31, 97–103.

Koch, R. 1890. Remedy for Tuberculosis. Brit. J. Med. 22, 1193-1199.

Koets, A., Adugna, G., Janss, L., van Weering, H., Kalis, C., Wentink, G., Rutten, V., Schukken, Y. 2000. Genetic variation of susceptibility to Mycobacterium avium subsp. paratuberculosis infection in Dairy Cattle. J. Dairy Sci. 83, 2702-2708.

Konickova-Radochova, M., Malek, I. 1968. The induction of auxotrophic mutants of Mycobacterium phlei PA by ultraviolet radiation. Folia Microbiologica. 14(3), 251- 253.

Konstantinos, A., Neofitou, C., Pachiadaki, M., Koufostathi, E. 2010. Changes of the bacterial assemblages throughout an urban drinking water distribution system. Environ. Monitoring Assessment. 165, 27-38.

Kopecky, K. 1977. Distribution of paratuberculosis in Wisconsin, by soil regions. J. Am. Vet. Med. Assoc. 130, 320-324.

Kovacs, J., Masur, H. 2000. Prophlyaxis against opportunistic infections in patients with human immunodeficiency virus infection. New England J. Med. 342, 1416-1429.

Kotlowski R., Martin, A., Ablordey, A., Chemlal, K., Fonteyne, P., Portaels, F. 2004. One-tube Cell Lysis and DNA Extraction Procedure for PCR-Based Detection of Mycobacterium ulcerans in Aquatic Insects, Molluscs and Fish. J. Med. Microbiol. 53, 927-933.

Kressel A., Kidd, F. 2001. Pseudo-outbreak of Mycobacterium chelonae and Methylbacterium mesophilicum Caused by Contamination of an Automated Endoscopy Washer. Infect. Control Hospital Epidemiol. 22, 414-418.

208

Kusnetsov, J., Iivanainen, E., Elomaa, N., Zacheus, O., Martikainen, P. 2001. Copper and silver ions more effective against legionellae than against mycobacteria in a hospital warm water system. Water Res. 35, 4217–4225.

Labombardi, V., O‘Brien, A., Kislak, J. 2002. Pseudo-outbreak of Mycobacterium fortuitum due to contaminated ice machines. Am. J. Infect. Control. 30, 184-186.

Lai, C., Tan, C., Chou, C., Hsu, H., Liao, C., Huang, Y., Yang, P., Luh, K., Hsueh, P. 2010. Increasing incidence of nontuberculous mycobacteria, Taiwan, 2000–2008. Emerg Infect Dis. 16, 2.

Langevin, P., Rand, K., Layon, A. 2008. The Potential for Dissemination of Mycobacterium tuberculosis Through Anesthesia Breathing Circuit. Chest. 115(4), 1107-1114.

Laurent, J., Faske, S., Cangelosi, G. 2002. Characterization of IS999, an unstable genetic element in Mycobacterium avium. Gene. 294, 249-257.

Lautenschlager, K., Boon, N., Wang, Y., Egli, T., Hammes, F. 2010. Overnight stagnation of drinking water in household taps induces microbial growth and changes in community composition. Water Research. 44(17), 4868-4877.

Lavender, J., Kinzelman, J. 2009. A cross comparison of qPCR to agar-based or defined substrate test methods for the determination of Escherichia coli and enterococci in municipal water quality monitoring programs. Water Res. 43(19), 4967-4979.

LeChevallier, M., Au, K. 2003. Water Quality and Drinking-water Treatment: the impact of treatment processes on microbial water quality. World Health Organization, Geneva.

LeChevallier, M., Cawthon, C., Lee, R. 1988. Factors promoting survival of bacteria in chlorinated water supplies. Appl. Environ. Microbiol. 54, 649-654.

LeChevallier, M., Gullick, R., Karim, M., Friedman, M., Funk, J. 2003. The potential for health risks from intrusion of contaminants into the distribution system from pressure transients. J. Water Health. 1(1), 3–14.

LeChevallier, M., Lowry, C., Lee, R., Gibbon, D. 1993. Examining the relationship between iron corrosion and the disinfection of biofilm bacteria. J. Am. Water Works Assoc. 85, 111-123.

LeChevallier, M., Norton, C., Falkingham, J., Williams, M., Taylor, R., Cowan, H. 2001. Occurrence and Control of Mycobacterium avium Complex. AWWA Research Foundation and American Water Works Association, Denver.

LeChevallier, M., Welch, N., Smith, D. 1996. Full-scale studies of factors related to coliform regrowth in drinking water. Appl. Environ. Microbiol. 62, 2201-2211.

209

Le Dantec, C., Duguet, J., Monteil, A., Dumontier, N., Dubrou, S., Vincent, V. 2002. Chlorine disinfection of atypical mycobacteria isolated from a water distribution system. Appl. Environ. Microbiol. 68, 5318-5325.

Le Dantec, C., Duguet, J., Monteil, A., Dumontier, N., Dubrou, S., Vincent, V. 2002. Occurrence of mycobacteria in water treatment lines in water distribution systems. Appl. Environ. Microbiol. 68, 5318-5325.

Lee, H., Park, H., Cho, S., Bai, G., Kim, S. 2000. Species identification of mycobacteria by PCR-restriction fragment length polymorphism of the rpoB gene. J. Clin. Microbiol. 38, 2966-2971.

Lee, W., Kim, T., Shur, K., Kim, B., Kook, Y., Lee, J., Park, J. 2000. Sporotrichoid dermatosis caused by Mycobacterium abscessus from a public bath. J. Dermatol. 27, 264-268.

Lehtola M., Laxander, M., Miettinen, I., Hirvonen, A., Vartiainen, T., Martikainen, P. 2006. The effects of changing water flow velocity on the formation of biofilms and water quality in pilot distribution system consisting of copper or polyethylene pipes. Water Res. 40, 2151–2060.

Leifsson, P., Olsen, S., Larsen, S. 1997. Ocular tuberculosis in a horse. Vet. Rec. 141, 651-654.

Leitner, M., Curtis, A. 2006. A first step towards a framework for presenting the location of confidential point data on maps- results of an empirical perceptual study. Int. J. Geogr. Infor. Sci. 20(7), 797-811.

Li, G., Guo, S., Zhang, C., Liu, X. 2010. Treatment of oily sludge from floatation process: Floc breakage and dewatering. Adv. Materials Res. Vols. 113-114, 579- 583.

Li, N., Bajoghli, A., Kubba, A., Bhawan, J. 1999. Identification of mycobacterial DNA in cutaneous lesions of sarcoidosis. J. Cutaneous Pathol. 26, 271-278.

Lin, Y., Vidic, R., Stout, J., McCartney, C., Yu, V. 1998. Inactivation of Mycobacterium avium by copper and silver ions. Water Res. 32, 1997–2000.

LiPuma, J. 2010. The Changing Microbial Epidemiology in Cystic Fibrosis. Clin Microbiol Rev. 23, 299-323.

Littlejohn G., Dixon, P. 1984. Fish fancier‘s finger. J. Rheumatol. 11, 390-391.

Liu, Z., Stout, J., Boldin, M., Rugh, J., Diven, W., Yu, V. 1998. Intermittent use of copper-silver ionization for Legionella control in water distribution systems: a potential option in buildings housing individuals at risk for infection. Clin Infect Dis. 26, 138–140.

210

Lloyd, J., Whittington, R., Fitzgibbon, C., Dobson, R. 2001. Presence of Mycobacterium avium subspecies paratuberculosis in Suspensions of Ovine Trichostrongylid Larvae Produced in Faecal Cultures Artificially Contaminated with the Bacterium. Vet. Rec. 148, 261-263.

Loret, J., Greub, G. 2010. Free-living amoebae: Biological by-passes in water treatment. Intern. J. Hyg. Environ. Health. 213(3), 167-175.

Lyons, S., Griffen, A., Leys, E. 2000. Quantitative realtime PCR for Porphyromonas gingivalis and total bacteria. J. Clin. Microbiol. 38, 2362-2365.

Maartens, G. 2002. Opportunistic Infections Associated with HIV Infection in Africa. Oral Dis. 8(2), 76-79.

MacGregor, R., Dreyer, K., Herman, S., Hocknell, P., Nghiem, L., Tevere, V., Williams, A. 1999. Use of PCR in detection of Mycobacterium avium complex (MAC) bacteremia: sensitivity of the assay and effect of treatment for MAC infection on centrations of human immunodeficiency virus in plasma. J. Clin. Microbiol. 37, 90- 94.

Maloney, S., Welbel, S., Daves, B., Adams, K., Becker, S., Bland, L., Arduino, M., Wallace, R., Zhang, R., Buck, G. 1994. Mycobacterium abscessus Pseudoinfection Traced to an Automated Endoscope Washer: Utility of Epidemiologic and Laboratory Investigation. J. Infect. Dis. 169, 1166-1169.

Manarolla G., Liandris, E., Pisoni, G., Sassera, D., Grilli, G., Gallazzi, D., Sironi, G., Moroni, P., Piccinini, R., Rampin, T. 2009. Avian mycobacteriosis in companion birds: 20-year survey. Vet. Microbiol. 133(4), 323-327.

Mangione, E., Huitt, G., Lenaway, D., Beebe, J., Bailey, A., Figoski, M., Rau, M., Albrecht, K., Yakrus, M. 2001. Nontuberculous mycobacterial disease following hot tub exposure. Emerg. Infect. Dis. 7, 1039-1042.

Mann, P., Montali, R., Bush, M. 1982. Mycobacterial osteomyelitis in captive marsupials. J. Am. Vet. Med. Assoc. 181, 1331-1333.

Manning, E., Collins, M. 2001. Mycobacterium avium subspecies paratuberculosis: pathogen, pathogenesis and diagnosis. Rev. Sci. Tech. 20, 133-150.

Marras, T., Chedore, P., Ying, A., Jamieson, F. 2007. Isolation prevalence of pulmonary non-tuberculous mycobacteria in Ontario, 1997–2003. Thorax. 62, 661–666.

Marras, T., Daley, C. 2002. Epidemiology of human pulmonary infection with nontuberculous mycobacteria. Clin. Chest Med. 23, 553-567.

211

Marion E., Eyangoh, S., Yeramian, E., Doannio, J., Landier, J., Aubry, J., Fontanet, A., Rogier, C., Cassisa, V., Cottin, J., Marot, A., Eveillard, M., Kamdem, Y., Legras, P., Deshayes, C., Saint-Andre, J., Marsollier, L. 2010. Seasonal and Regional Dynamics of M. ulcerans Transmission in Environmental Context: Deciphering the Role of Water Bugs as Hosts and Vectors. PLoS Negl Trop Dis. 4(7), e731.

Marsollier, L., Robert, R., Aubry, J., Saint Andre, J., Kouakou, H., Legras, P., Manceau, A., Mahaza, C., Carbonnelle, B. 2002. Aquatic Insects as a Vector for Mycobacterium ulcerans. Appl. Environ. Microbiol. 68, 4623-4628.

Marston, B., Diallo, M., Horsburgh, C., Diomande, I., Saki, M., Kanga, J., Patrice, G., Lipman, H., Ostroff, S., Good, R. 1995. Emergence of Buruli ulcer disease in the Daloa region of Cote d‘Ivoire. Am. J. Trop. Med. Hyg. 52(3), 219-224.

Matsiota-Bernard, P., Zinzendorf, N., Onody, C., Guenounou, M. 2000. Comparison of Clarithromycin-resistant Mycobaceterium avium strains isolated from AIDS patients during therapy regimens including clarithromycin. J. Infect. 40, 49-54.

Mayberry, J., Hitchens, R. 1978. Distribution of Crohn‘s disease in Cardiff. Social Sci. Med. 12, 137-138.

Mazumder, S., Falkingham, J., Dietrich, A., Puri, I. 2010. Role of hydrophobicity in bacterial adherence to carbon nanostructures and biofilm formation. Biofouling. 26(3), 333-339.

McCarthy, C., Schaefer, J. 1974. Response of Mycobacterium avium to ultraviolet irradiation. Appl Microbiol 28, 151–153.

McCrossin, C., Mailman, T. 2006. First Canadian reports of Cervical Adenitis due to Mycobacterium malmoense and a 10-year Review of Nontuberculous Mycobacterial Adenitis. Canad. J. Infect. Dis. Med. Microbiol. 17(2), 123-127.

McCue, J. 1990. The Infectious Patient, pp. 1034-1037.In: Walker, H., Hall. W., Hurst, J. (ed.). Clinical Methods: The History, Physical, and Laboratory Examinations. Buttersworths, Boston.

McGrath, E., Blades, Z., McCabe, J., Jarry, H., Anderson, P. 2010. Nontuberculous Mycobacteria and the Lung: From Suspicion to Treatment. Lung. 188(4), 269-282.

Merkal, R., McCullough, W. 1982. A new mycobactin, mycobactin J, from Mycobacterium paratuberculosis. Current Microbiol. 7(6), 333-335.

Mermin, J., Villar, R., Carpenter, J., Roberts, L., Samaridden, A., Gasanova, L., Lomakina, S., Bopp, C., Hutwagner, L., Mead, P., Ross, B., Mintz, E. 1999. A massive epidemic of multidrug-resistant typhoid fever in Tajikistan associated with consumption of municipal water. J. Infect. Dis. 179(6), 1416–1422.

212

Mery, A., Horan, R. 2002. Hot tub-related Mycobacyerium avium intracellulare pneumonitis. Allergy Asthma Proced. 23, 271-273.

Migard, S., Flandrois, J. 2008. A Seven-Gene, Multilocus, Genus-wide Approach to the Phylogeny of Mycobacteria Using Supertrees. Intern. J. System. Evolut. Microbiol. 58, 1432-1441.

Mijs, W., de Haas, P., Rossau, R., Van Der Laan, T., Rigouts, L., Portaels, F., van Soolingen, D. 2002. Molecular evidence to support a proposal to reserve the designation Mycobacterium avium subspecies avium for bird-type isolates and ‗M. avium subsp. hominissuis‘ for human/ porcine type of M. avium. Intern. J. System Evolut. Microbiol. 52, 1505-1518.

Millar, D., Ford, J., Sanderson, J., Withey, S., Tizard, M., Doran, T., Hermon-Taylor, J. 1996. IS900 PCR to detect Mycobacterium paratuberculosis in retail supplies of whole pasteurized cow‘s milk in England and Wales. Appl. Environ. Microbiol. 62, 3446-3452.

Miller, M., Greene, C., Brix, A. 1995. Disseminated Mycobacterium avium-intracellulare complex infection in a miniature schnauzer. J. Am. Animal Hosp. Assoc. 31, 213- 216.

Miltner, E., Bermudez, L. 2000. Mycobacterium avium grown in Acanthamoeba castellanii is Protected from the Effects of Antimicrobials. Antimicro. Agents Chemo. 44, 1990–1994.

Minnikin, D. 1982. Lipids: complex lipids, their chemistry, biosynthesis, and roles, pp. 95-184. In: Ratledge, C., Standford, J. (ed.). The Biology of the Mycobacteria: Physiology, Identification, and Classification. Academic Press, Maryland Heights.

Minnikin, D., Kremer, L., Dover, L., Besra, G. 2002. The methyl-branched fortifications of Mycobacterium tuberculosis. Chem. Biol. 9(5), 545-553.

Mintz, E., Bartram, J., Lochery, P., Wegelin, M. 2001. Not Just a Drop in the Bucket: Expanding Access to Point-of-Use Water Treatment Systems. Am. J. Public Health. 91, 1565-1570.

Miyamoto, M., Yamaguchi, Y., Sasatsu, M. 2000. Disinfectants effects of hot water, ultraviolet light, silver ions, and chlorine on strains of Legionella and nontuberculous mycobacteria. Microbiol. 101, 7-13.

Moghadam, M., Sarv, S., Moosakhani, F., Badiie, A. 2010. Detection of Mycobacterium avium subspecies paratuberculosis in Milk and fecal samples in dairy cattle by PCR and Nested-PCR. J. Animal Vet. Adv. 9(24), 3055-3061.

Molloy, M., Herbert, B., Slade, M., Rabilloud, T., Nouwens, A., Williams, K., Gooley, A. 2000. Proteomic analysis of the Escherichia coli outer membrane. Eur. J. Biochem. 267(10), 2871-2881.

213

Monill, J., Franquet, T., Sambeat, M., Martinez-Noguera, A., Villalba, J. 2001. Mycobacterium genavense infection in AIDS: imaging findings in eight patients. Eur. Radiol. 11, 193–196.

Montali, R., Bush, M., Cromie, R., Holland, S., Maslow, J., Worley, M., Witebsky, F., Phillips, T. 1998. Primary Mycobacterium avium complex infections correlate with lowered cellular immune reactivity in Matschie‘s tree kangaroos (Dendrolagus matschiei). J. Infect. Dis. 178, 1719-1725.

Moon, J., Han, S. 2011. Thermostat strategies impact on energy consumption in residential buildings. Energy Buildings. 43, 338-346.

Moore, E. 1993. Atypical Mycobacterial Infection in the Lung: CT Appearance. Radiol. 187, 777-782.

Moore, J., Kruijshaar, M., Ormerod, L., Drobniewski, F., Abubakar, I. 2010. Increasing reports of non-tuberculous mycobacteria in England, Wales and Northern Ireland, 1995-2006. BMC Public Health. 10, 612.

Moran, P. 1950. Notes on continuous stochastic phenomena. Biometrika. 37, 17-23.

Morita, Y., Velasquez, R., Taig, E., Waller, R., Patterson, J., Tull, D., Williams, S., Billman-Jacobe, H., McConville, M. 2005. Compartmentalization of lipid biosynthesis in mycobacteria. J. Biol. Chem. 280(22), 21645-21652.

Morse, J., Hind, D. 1984. Bacteria isolated from lymph nodes of California slaughter swine. Am. J. Vet. Res. 45, 1648-1649.

Murga, R., Stewart, P., Daly, D. 1995. Biofilm thickness variability. Biotech. Bioengin. 45, 503-510.

Murillo, J., Torres, J., Bofill, L., Rios-Fabra, A., Irausquin, E., Isturiz, R., Guzman, M., Castro, J., Rubino, L., Cordido, M. 2000. Skin and wound infection by rapidly growing mycobacteria: an unexpected complication of liposuction and liposculpture. The Venezuelan Collaborative Infectious and Tropical Diseases Study Group. Arch Dermatol. 136, 1347-1352.

Muyzer, G., Smalla, K. 1998. Application of denaturing gradient gel electrophoresis (DGGE) and temperature gradient gel electrophoresis (TGGE) in microbial ecology. Antonie van Leeuwenhoek. 73,127-141.

Nagy, G., Rubin, R. 2001. Disseminated Mycobacterium avium-intracellulare in a kidney transplant recipient. Transplant Infect. Dis. 3, 220-230.

Nasser, S., Hulten, K., Shafran, I., Graham, D., El-Zaatari, F. 2000. Specific seroreactivity of Crohn‘s disease patients against p35 and p36 antigens of M. avium subspecies paratuberculosis. Vet. Microbiol. 77, 497-504.

214

Nadkarni, M., Martin, E., Jacques, N., Hunter, N. 2002. Determination of bacterial load by real-time PCR using a broad-range (universal) probe and primers set. Microbiol. 148, 257–266.

Nebbia, P., Robino, P., Ferroglio, E., Rossi, L., Meneguz, G., Rosati, S. 2000. Paratuberculosis in red deer (Cervus elaphus hippelaphus) in the Western Alps. Vet. Res. Comm. 24, 435-443.

Neumann, M., Schulze-Robbecke, R., Hagenau, C., Behringer, K. 1997. Comparison of methods for isolation of mycobacteria from water. Appl. Environ. Microbiol. 63, 547-552.

Newport, M., Huxley, C., Huston, S., Hawrylowicz, C., Oostra, B., Williamson, R., Levin, M. 1996. A mutation in the interferon receptor gene and susceptibility to mycobacterial infection. New England J. Med. 335, 1941-1949.

Newport, M. 2003. The Genetics of Nontuberculous Mycobacterial Infection. Expert Rev. Molecular Med. 5(28), 1-13.

Niederweis, M., Danilchanka, O., Huff, J., Hoffman, C., Engelhardt, H. 2010. Mycobacterial outer membranes: in search of proteins. Trends Microbiol. 18(3), 109-116.

Niederweis, M., Ehrt, S., Heinz, C., Klocker, U., Karosi, S., Swiderek, K., Riley, L., Benz, R. 1999. Cloning of the mspA gene encoding a porin from Mycobacterium smegmatis. Mol. Microbiol. 33(5), 933-945.

Nieminen, T., Pakarinen, J., Tsitko, I., Salkinoja-Salonen, M., Breitenstein, A., Ali- Vehmas, T., Neubauer, P. 2006. 16S rRNA Targeted Sandwich Hybridization Method for Direct Quantification of Mycobacteria in Soils. J. Microbiol Methods. 67(1), 44-55.

Nightingale, S., Byrd, L., Southern, P., Jockush, J., Cal, S., Wynne, B. 1992. Incidence of Mycobacterium avium intracellulare complex bacteriemia in human immunodeficiency virus-positive patients. J. Infect. Dis. 165, 1082-1085.

Nikaido, H. 2003. Molecular basis of bacterial outer membrane permeability revisited. Microbiol. Mol. Biol. Rev. 67, 593-656.

Nikaido, H., Kim, S., Rosenberg, E. 1993. Physical organization of lipids in the cell wall of Mycobacterium chelonae. Mol. Microbiol. 8(6), 1025-1030.

Nolan, B., Ruddy, B., Hitt, K., Helsel, D. 1997. Risk of Nitrate in Groundwaters of the United States, A National Perspective. Environ Sci Tech. 31(8), 2229-2236.

Nonnenmacher, C., Dalpke, A., Mutters, R., Heeg, K. 2004. Quantitative detection of periodontopathogens by real-time PCR. J. Microbiol. Methods. 59, 117– 125.

215

Norton, C., LeChevallier, M., Falkingham, J. 2004. Survival of Mycobacterium avium in a model distribution system. Water Res. 38, 1457-1466.

Nunn, P., McAdam, K. 1988. Mycobacterial infections and AIDS. Br. Med. Bull. 44(3), 801-813.

O‘Brien, R., Geiter, L., Snider, D. 1987. The epidemiology of nontuberculous mycobacterial diseases in the United States: results from a national survey. Am. Rev. Respir. Dis. 135, 1007-1014.

Odiawo, G., Mukurira, J. 1988. Avian cerebral tuberculosis. Vet. Recom. 122, 279-280.

Odoi, A., Martin, S., Michel, P., Holt, J., Middleton, D., Wilson, J. 2003. Geographical and temporal distribution of human giardiasis in Ontario, Canada. Int. J. Health Geogr. 2, 5.

Odumeru, J., Gao, A., Chen, S., Raymond, M., Mutharia, L. 2001. Use of bead beater for preparation of Mycobacterium paratuberculosis template DNA in milk. Canad. J. Vet. Res. 65, 201-205.

O‘Grady, D., Flynn, O., Costello, E., Quigley, F., Gogarty, A., McGuirk, J., O‘Rourke, J., Gibbons, N. 2000. Restriction fragment length polymorphism analysis of Mycobacterium avium isolates from animal and human sources. Intern. J. Tuberculosis Lung Dis. 4, 278-281.

Olivier, K., Weber, D., Wallace, R., Faiz, A., Lee, J., Zhang, Y., Brown-Elliot, B., Handler, A., Wilson, R., Schechter, M., Edwards, L., Chakraborti, S., Knowles, M. 2002. Nontuberculosis Mycobacteria: I. Multicenter prevalence study in cystic fibrosis. Am. J. Respir. Critical Care Med. 14, 14.

Ollar, R., Dale, J., Felder, M., Favate, A. 1990. The use of paraffin wax metabolism in the speciation of Mycobacterium avium-intracellulare. Tubercle. 71, 23-28.

Orholm, M., Munkholm, P., Langholz, E., Nielsen, O., Sorensen, T., Binder, V. 1991. Familial occurrence of inflammatory bowel disease. New England J. Med. 324, 84- 88.

Orholm, K., Binder, V., Sorensen, T., Rasmussen, L., Kyvik, K. 2000. Concordance of inflammatory bowel disease among Danish twins. Results of a nationwide study. Scandinavian J. Gastroenterology. 35, 1075-1081.

O‘Sullivan, D., Unwin, D. 2010. Geographic Information Analysis. John Wiley and Sons, Inc., Hoboken.

Owusu-Edusei, K., Owens, C. 2009. Monitoring county-level chlamydia incidence in Texas, 2004 - 2005: application of empirical Bayesian smoothing and Exploratory Spatial Data Analysis (ESDA) methods. Int. J. Health Geogr. 8, 12.

216

Palomino, J., Obiang, A., Realini, L., Meyers, W., Portaels, F. 1998. Effect of Oxygen on Growth of Mycobacterium ulcerans in the BACTEC System. J. Clin. Microbiol. 36(11), 3420-3422.

Panek, F., Bobo, T. 2006. Striped Bass Mycobacteriosis: A Zoonotic Disease of Concern in Chesapeake Bay, pp. 5214. USGS/NOAA Workshop on Mycobacteriosis in Striped Bass. USGS Scientific Investigations Report, Annapolis.

Paramasivan C., Govindan, D., Prabhakar, R., Somasundaram, P., Subbammal, S., Tripathy, S. 1985. Species level identification of non-tubercular mycobacteria from South Indian BCG trial area during 1981. Tubercle. 66, 9–15.

Parashar, D., Chauhan, D., Sharma, V., Chauhan, A., Chauhan, S., Katoch, V. 2004. Optimization of Procedures for Isolation of Mycobacteria from Soil and Water Samples Obtained in Northern India. Appl. Environ. Microbiol. 70(6), 3751-3753.

Parker D., Munroe, D. 2004. Spatial Tests for Edge-effect Externalities and External Scale Economies in California Certified Organic Agriculture. Am. Ag. Econ. Assoc. Ann. Meeting. Denver.

Parker, D., Munroe, D. 2007. The geography of market failure: Edge-effect externalities and the location and production patterns of organic farming. Ecol. Econ. 60(4), 821-833.

Paustian, M., Zhu, X., Sreevatsan, S., Robbe-Austerman, S., Kapur, V., Bannantine, J. 2008. Comparative genomic analysis of Mycobacterium avium subspecies obtained from multiple host species. BMC Genomics. 9, 135.

Pavlik, I., Bartl, J., Dvorska, L., Svastova, P., du Maine, R., Machackova, M., Yayo Ayele, W., Horvathova, A. 2000. Epidemiology of paratuberculosis in wild ruminants studied by restriction fragment length polymorphism in the Czech Republic during the period 1995-1998. Vet. Microbiol. 77, 231-251.

Pavlik, I., Svastova, P., Bartl, J., Dvorska, L., Rychlik, M. 2000. Relationship between IS901 in the Mycobacterium avium complex strains isolated from birds, animals, humans, and the environment and virulence for poultry. Clin. Diag. Lab Immunol. 7, 212-217.

Pedley S., Bartram, J., Rees, G., Dufour, A., Cotruvo, J. 2004. Pathogenic Mycobacteria in Water: A Guide to Public Health Consequences, Monitoring, and Management. IWA Publishing, Colchester.

Peeters, M., Nevens, H., Baert, F., Hiele, M., DeMeyer, A., Vlietinck, R., Rutgeerts, P. 1996. Familial Aggregation in Crohn‘s disease: increased age-adjusted risk and concordance in clinical characteristics. Gastroenterol. 111, 597-603.

217

Pfeiffer, D., Robinson, T., Stevenson, M., Stevens, K., Rogers, D., Clements, A. 2008. Spatial Analysis in Epidemiology. Oxford University Press, Oxford.

Pelletier, P., du Moulin, G., Stottmeier, K. 1988. Mycobacteria in public water supplies: comparative resistance to chlorine. Microbial Sci. 5, 147-148.

Perazella, M., Eisen, T., Brown, E. 1993. Peritonitis associated with disseminated Mycobacterium avium complex in an acquired immunodeficiency syndrome patient on chronic ambulatory peritoneal dialysis. Am. J. Kidney Dis. 21, 319-321.

Peres, R., Maciel, E., Morais, C., Ribeiro, F., Vinhas, S., Pinheiro, C., Dietze, R., Johnson, J., Eisenach, K., Palaci, M. 2009. Comparison of two concentrations of NALC-NaOH for decontamination of sputum for mycobacterial culture. Intern. J. Tuberculosis Lung Dis. 13(12), 1572-1575.

Peters, M., Muller, C., Rusch-Gerdes, S., Seidel, C., Gobel, U., Pohle, H., 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. J. Infect. 31, 39-44.

Phillips, M., von Reyn, C. 2001. Nosocomial infections due to nontuberculous mycobacteria. Clin. Infect. Dis. 33, 1363-1374.

Philpott J., Woodburne, A., Philpott, O., Schaefer, W., Mollohan, C. 1963. Swimming pool granuloma. Arch. Dermatol. 88,158-162.

Pierce, E. 2009. Possible transmission of Mycobacterium avium subspecies paratuberculosis through potable water: lesson from an urban cluster of Crohn's disease. Gut Pathog. 1, 17.

Pillai, S., Jayarao, B., Gummo, J., Hue, E., Tiwari, D., Stabel, J., Whitlock, R. 2001. Identification and sub-typing of Mycobacterium avium subspecies paratuberculosis and Mycobacterium avium subspecies avium by randomly amplified polymorphic DNA. Vet. Microbiol. 79, 275-284.

Plikaytis, B., Plikaytis, B., Yakrus, M., Butler, W., Woodley, C., Silcox, V., Shinnick, T. 1992. Differentiation of slowly growing Mycobacterium species, including Mycobacterium tuberculosis, by gene amplification and restriction fragment length polymorphism analysis. J. Clin. Microbiol. 30, 1815-1822.

Pond, C., Rush, H. 1981. Infection of white carneaux pigeons (Columbia livia) with Mycobacterium avium. Lab. Animal Sci. 31, 196-199.

Portaels, F., Elsen, P., Guimares-Peres, A., Fonteyne, P., Meyer, W. 1999. Insects in the Transmission of Mycobacterium ulcerans Infection. Lancet. 353, 986.

Portaels, F., Larsson, L., Smeets, P. 1988. Isolation of mycobacteria from healthy persons stools. Int. J. Lepr. 56, 468– 471.

218

Portaels, F., Pattyn, S. 1982. Growth of mycobacteria in relation to the pH of the medium. Ann. Microbiol. 133, 213-22.

Pradhan, M., Sahn, D., Younger, S. 2003. Decomposing world health inequality. J. Health Econ. 22(2), 271-293.

Preacher, K. J. 2001, April. Calculation for the chi-square test: An interactive calculation tool for chi-square tests of goodness of fit and independence [Computer software]. Available at: http://quantpsy.org.

Primm, T., Lucero, C., Falkinham, J. 2004. Health Impacts of Environmental Mycobacteria. Clin. Microbiol. Rev. 17(1), 98-106.

Prince, D., Peterson, D., Steiner, R., Gottleib, J., Scott, R., Israel, H. 1989. Infection with Mycobacterium avium complex in patients without predisposing conditions. New England J. Med. 321, 863-868.

Prissick, F., Masson, A. 1957. Yellow-pigmented pathogenic mycobacteria from cervical lymphadenitis. Canad. J. Microbiol. 3, 91-100.

Radomski, N., Cambau, E., Moulin, L., Haenn, S., Moilleron, R., Lucas, F. 2010. Comparison of culture methods for isolation of nontuberculous mycobacteria from surface waters. Appl Environ Microbiol. 76(11), 3514-3520.

Rafii, F., Selby, A., Newton, R., Cerniglia, C. 1994. Reduction and mutagenic activation of nitroaromatic compounds by Mycobacterium spp. Appl. Environ. Microbiol. 60, 4263-4267.

Ragunathan, P., Whitney, E., Tappero, J., Ashford, D., Bugri, S., Amofah, G., Asomoa, K., Stienestra, E., van der Graaf, W., van der Werf, K., Dobos, S., Kihlstrom, L., King, C. 2001. Burden of Buruli ulcer in Uper Denkyira District, Ghana 1994-2000. In: World Health Organization (ed.). Report of 4th WHO Advisory Group Meeting on Buruli ulcer. World Health Organization, Geneva. Available at: http://www.who.int/gtb-buruli/publications/index.html.

Ramasoota, P., Chansiripornchai, N., Kallenius, G., Hoffner, S., Svenson, S. 2001. Comparison of Mycobacterium avium complex (MAC) from pigs and humans in Sweden by random amplified polymorphic DNA (RAPD) using standardized reagents. Vet. Microbiol. 78, 251-259.

Rappaport, J. 2007. Moving to nice weather. Regional Sci. Urban Econ. 37(3), 375-398.

Rastogi, N., Frehel, C., Ryter, A., Ohayon, H., Lesourd, M., David, H. 1981. Multiple drug resistance in Mycobacterium avium: is the wall architecture responsible for the exclusion of antimicrobial agents? Antimicro. Agents Chemo. 20, 666-677.

Rastogi, N., Legrand, E., Sola, C. 2001. The mycobacteria: an introduction to nomenclature and pathogenesis. Rev. Sci. Tech. 20(1), 21-54.

219

Ravva, S., Stanker, L. 2005. Real-time quantitative PCR detetction of Mycobacterium avium subsp. paratuberculosis and differentiation from other mycobacteria using SYBR Green and TaqMan assays. J. Microbiol. Methods. 63(3), 305-317.

Razvi, S., Quittell, L., Sewall, A., Quinton, H., Marshall, B., Saiman, L. 2009. Respiratory Microbiology of Patients With Cystic Fibrosis in the United States, 1995-2005. Chest. 136, 1554-1560.

Read, J., Eames, K., Edmunds, W. 2008. Dynamic social networks and the implications for the spread of infectious disease. J. R. Soc. Interface. 5, 1001-1007.

Reddacliff, L., Vadali, A., Whittington, A. 2003. The effect of decontamination protocols on the numbers of sheep strain Mycobacterium avium subsp. paratuberculosis isolated from tissues and faeces. Vet. Microbiol. 95(4), 271–282.

Reddacliff, L., Marsh, I., Fell, S., Austin, S., Whittington, R. 2010. Isolation of Mycobacterium avium subspecies paratuberculosis from muscle and peripheral lymph nodes using acid-pepsin digest prior to BACTEC culture. Vet Microbiol. 145, 122-128.

Redmond, W., Ward, D. 1966. Media and methods for phage-typing mycobacteria. Bull World Health Organ. 35(4), 563-568.

Reed, C., von Reyn, C., Chamblee, S., Ellerbrock, T., Johnson, J., Marsh, B., Johnson, B., Trenschel, R., Horsburgh, C. 2006. Environmental Risk Factors for Infection with Mycobacterium avium Complex. Am. J. Epidemiol. 164(1), 32-40.

Rehm, S., Farley, M., File, T., Hall, W., Hopkins, R., Levine, O., Nichol, K., Nuorti, P., Zimmerman, R., Schaffner, W. 2009. Higher pneumococcal disease vaccination rates needed to protect more at-risk US adults. Postgrad Med. 121(6), 101-105.

Reviriego, F., Moreno, M., Dominguez, L. 2000. Soil type as a putative risks factor of ovine and caprine paratuberculosis seropositivity in Spain. Prev. Vet. Med. 43, 43- 51.

Reynolds, K., Mena, K., Gerba, C. 2008. Risk of Waterborne Illness Via Drinking Water in the United States. Rev. Environ. Contamin. Toxicol. 192, 117-158.

Rickman, O., Ryu, J., Fidler, M., Kalra, S. 2002. Hypersensitivity pneumonitis associated with Mycobacterium avium complex and hot tub use. Mayo. Clin. Proc. 77, 1233-1237.

Ridgeway, H., Rigby, M., Argo, D. 1984. Adhesion of a Mycobacterium sp. to cellulose diacetate membranes used in reverse osmosis. Appl. Environ. Microbiol. 47, 61- 67.

Riemann, H., Abbas, B. 1983. Diagnosis and Control of Bovine Paratuberculosis (Johne‘s Disease). Adv. Vet. Sci. and Comp. Medicine. 27, 481-506.

220

Ringuet, H., Akoua-Koffi, C., Honore, S., Varnerot, A., Vincent, V., Berche, P., Gaillard, J., Pierre-Audigier, C. 1999. Hsp65 Sequencing for Identification of Rapidly Growing Mycobacteria. J. Clin. Microbiol. 37(3), 852-857.

Ristola, M., von Reyn, C., Arbeit, R., Soini, H., Lumio, J., Ranki, A., Buhler, S., Waddell, R., Tosteson, A., Falkingham, J., Sox, C. 1999. High rates of disseminated infection due to non-tuberculous mycobacteria among AIDS patients in Finland. J. Infect. 39, 61-67.

Robicsek, F., Hoffman, P., Masters, T., Daugherty, H., Cook, J., Selle, J., Mauney, C., Hinson, P. 1988. Rapidly growing nontuberculous mycobacteria: a new enemy of the cardiac surgeon. Ann. Thoracic Surg. 46, 703-710.

Robinson, R., Phillips, P., Stevens, G., Storm, P. 1989. An outbreak of Mycobacterium bovis infection in fallow deer (Dama dama). Aust. Vet. J. 66, 195-197.

Rodgers, M., Backstone, B., Reyers, A., Covert, T. 1999. Colonization of point-of-use water filters by silver resistant non-tuberculous mycobacteria. J. Clin. Pathol. 52, 629.

Rogers, D. 2006. Models for vectors and vector-borne diseases. In: Hay, S., Graham, A., Rogers, D. (ed.) Global mapping of infectious diseases: methods, examples, and emerging application. Academic Press, London.

Romanus, V., Hallander, H., Wahlen, P., Olinder-Nielsen, A., Magnusson, P., Juhlin, I. 1995. Atypical mycobacteria in extrapulmonary disease among children. Incidence in Sweden from 1969 to 1990, related to changing BCG-vaccination coverage. Tuberculous Lung Dis. 76, 300-310.

Rose, C., Martyny, J., Newman, L., Milton, D., King, T., Beebe, J., McCammon, J., Hoffman, R., Kreiss, K. 1998. ―Lifeguard Lung‖: endemic granulomatous pneumonitis in an indoor swimming pool. Am. J. Public Health. 88, 1795-1800.

Roth, A., Fischer, M., Hamid, M., Michalke, S., Ludwig, W., Mauch, H. 1998. Differentiation of phylogenetically related slowly growing mycobacteria based on 16S-23S rRNA gene internal transcribed spacer sequences. J. Clin. Microbiol. 36, 139-147.

Roy, V., Weisdorf, D. 1997. Mycobacterial infections following bone marrow transplantation: a 20 year retrospective review. Bone Marrow Transplantation. 19, 467-470.

Rushton, S., Goodfellow, M., O‘Donnell, A., Magee, J. 2007. The epidemiology of atypical mycobacterial diseases in northern England: a space-time clustering and generalized linear modeling approach. Epi. Infect. 135, 765-774.

Sadiq, R., Kleiner, Y., Rajani, B. 2003. Forensics of water quality failure in distribution systems- a conceptual framework. J. Indian Water Works Assoc. 35(4), 1-23.

221

Safranek, T., Jarvis, W., Carlson, L., Cusick, L., Bland, L., Swenson, J., Silcox., V. 1987. Mycobacterium chelonae wound infections after plastic surgery employing contaminated gentian violet skin-marking solution. New England J. Med. 317, 197- 201.

Saleeb, P., Olivier, K. 2010. Pulmonary Nontuberculous Mycobacterial Disease: New Insights into Risk Factors for Susceptibility, Epidemiology, and Approaches to Management in Immunocompetent and Immunocompromised Patients. Curr. Infect. Dis. Reports. 12(3), 198-203.

Sanford, S., Rehmtulla, A., Josephson, G. 1994. Tuberculosis in farmed rheas (Rhea americana). Avian Dis. 38, 193-196.

Sangari, F., Goodman, J., Bermudez, L. 2000. Mycobacterium avium enters intestinal epithelial cells through the apical membrane, but not by the basolateral surface, activates small GTPase Rho and, once within epithelial cells, expresses an invasive phenotype. Cell. Microbiology. 2, 561–568.

Sangari, F., Goodman, J., Petrofsky, M., Kolonoski, P., Bermudez, L. 2001. Mycobacterium avium invades the intestinal mucosa primarily by interacting with enterocytes. Infect. Immunol. 69, 1515–1520.

Schaad, U., Votteler, T., McCracken, G., Nelson, J. 1979. Management of atypical mycobacterial lymphadenitis in childhood: A review based on 380 cases. J. Ped. 95, 356-360.

Scheibl P., Gerlach, G. 1997. Differentiation of Mycobacterium paratuberculosis isolates by rDNA-spacer analysis and random amplified polymorphic DNA patterns. Vet. Microbiol. 51, 151-158.

Schelonka, R., Ascher, D., McMahon, D., Drehner, D., Kuskie, M. 1994. Catheter- related sepsis caused by Mycobacterium avium complex. Ped. Infect. Dis. J. 13, 236-238.

Schlesinger, L., Bellinger-Kawahara, C., Payne, N., Horwitz, M. 1990. Phagocytosis of Mycobacterium tuberculosis is mediated by human monocyte complement receptors and complement component C3. J. Immunol. 144, 2771-2780.

Schlesinger, L., Horwitz, M. 1991. Phagocytosis of Mycobacterium leprae by human monocyte-derived macrophages is mediated by complement receptors CR1 (CD35), CR3 (CD11b/CD18) and CR4 (CD11c/CD18) and interferon-gamma inactivation inhibits complement receptor function and phagocytosis of this bacterium. J. Immunol. 147, 1983-1994.

Schochetman, G., Ou, C., Jones, W. 1988. Polymerase Chain Reaction. J. Infect. Dis. 158(6), 1154-1157.

222

Schoon, H., Schoon, D., Kirpal, G., Richter, E., Gerdes, J., Weiss, R., Dressen, W. 1993. Enzootic Mycobacterium avium infection in captive Parma wallabies (Macropus parma) with unusual spinal cord manifestations. J. Compar. Pathol. 108, 311-316.

Schulze-Robbecke, R., Buchholtz, K. 1992. Heat susceptibility of aquatic mycobacteria. Appl. Environ. Microbiol. 58, 1869-1873.

Schulze-Robbecke, R., Feldman, C., Fischeder, R., Janning, B., Exner, M., Wahl, G. 1995. Dental units: an environmental study of sources of potentially pathogenic mycobacteria. Tuberculosis Lung Dis. 76, 318-323.

Schulze-Robbecke, R., Fischeder, R. 1989. Mycobacteria in biofilms. Zentralbl. Hyg. Umweltmed. 188, 385-390.

Schulze-Robbecke, R., Weber, A., Fischeder, R. 1991. Comparison of decontamination methods for the isolation of mycobacteria from drinking water samples. J. Microbiol. Methods. 14, 177-183.

Semenza, J., Roberts, L., Henderson, A., Bogan J., Rubin, C. 1998. Water distribution system and diarrheal disease transmission: a case study in Uzbekistan. Am. J. Trop. Med. Hyg. 59(6), 941–946.

Shackelford, C., Reed, W. 1989. Disseminated Mycobacterium avium infection in a dog. J. Vet. Diagn. Invest. 1, 273-275.

Shane, S., Camus, A., Strain, M., Thoen, C., Tully, T. 1993. Tuberculosis in commercial emus (Dromaius novaehollandiae). Avian Dis. 37, 1172-1176.

Shelton, B., Flanders, W., Morris, G. 1999. Mycobacterium sp. as a Possible Cause of Hypersensitivity Pneumonitis in Machine Workers. Emerg. Infect. Dis. 5, 270-273.

Sheng, W., Huang, Y., Chang, S., Hsueh, P. 2009. Brain Abscess Caused by Tsukamurella tyrosinosolvens in an Immunocompetent Patient. J. Clin. Microbiol. 47(5), 1602-1604.

Shepard, C., McRae, D. 1965. Mycobacterium leprae in Mice: Minimal Infectious Dose, Relationship Between Staining Quality and Infectivity, and Effect of Cortisone. J. Bacteriol. 89(2), 365-372.

Shepherd, B., Jenkins, C., Rebeiro, P., Stinnette, S., Bebawy, S., McGowan, C., Hulgan, T., Sterling, T. 2010. Estimating the Optimal CD4 Count for HIV-Infected Persons to Start Antiretroviral Therapy. Epidemiol. 21(5), 698-705.

Shinnick, T. 1987. The 65-kilodalton Antigen of Mycobacterium tuberculosis. J. Bacteriol. 169(3), 1080-1088.

223

Sills, R., Mullaney, T., Stickle, R., Darien, B., Brown, C. 1990. Bilateral granulomatous guttural pouch infection due to Mycobacterium avium complex in a horse. Vet. Pathol. 27, 133-135.

Silva, M., Portaels, F., Macedo, P. 1989. New data on the ultrastructure of the membrane of Mycobacterium leprae. Intern. J. Leprosy Other Mycobacterial Dis. 57, 54-64.

Simmonds, N., MacNeil, S., Cullinan, P., Hodson, M. 2010. Cystic fibrosis and survival to 40 years: a case-control study. Eur. Respir. J. 36, 1277-1283.

Simoes, L., Simoes, M., Vieira, M. 2010. Influence of the Diversity of Bacterial Isolates from Drinking Water on Resistance of Biofilms to Disinfection. Appl. Environ. Microbiol. 76(19), 6673-6679.

Skriwan, C., Fajardo, M., Hagele, S., Horn, M., Wagner, M., Michel, R., Krohne, G., Schleicher, M., Hacker, J., Steinert, J. 2002. Various Bacterial Pathogens and Symbionts Infect the Amoeba Dictyostelium discoideum. Intern. J. Med. Microbiol. 291, 615-624.

Slutsky, A., Arbeit, R., Barber, T., Rich, J., von Reyn, C., Pieciak, W., Barlow, M., Maslow, J. 1994. Polyclonal infections due to Mycobacterium avium complex in patients with AIDS detected by pulse-field gel electrophoresis of sequential clinical isolates. J. Clin. Microbiol. 32, 1773-1778.

Small, P., McClenny, N., Singh, S., Schoolnik, G., Tompkins, L., Mickelsen, P. 1993. Molecular strain typing of Mycobacterium tuberculosis to confirm cross- contamination in the mycobacteriology laboratory and modification of procedures to minimize occurrence of false-positive cultures. J. Clin. Microbiol. 31, 1677-1682.

Smith D. 1967. Diagnostic and Prognostic Significance of the Quantitative Tuberculin Tests. The Influence of Subclinical Infections with Atypical Mycobacteria. Ann. Internal Med. 67, 919–46.

Smith, H., Rose, J. 1990. Waterborne cryptospoidiosis. Parasitol Today. 6(1), 8-12.

Smith, K., De Vos, V., Price, L., Hugh-Jones, M., Keim, P. 2000. Bacillus anthracis diversity in Kruger National Park. J. Clin. Microbiol. 38, 3780-3784.

Smythe, L., Smith, I., Smith, G., Dohnt, M., Symonds, M., Barnett, L., McKay, D. 2002. A quantitative PCR (TaqMan) assay for pathogenic Leptospira spp. BMC Infect. Dis. 2, 1-7.

Sniadack, D., Ostroff, S., Karlix, M., Southwick, R., Schwartz, B., Sprauer, M., Silcox, V., Good, R. 1993. A Nosocomial Pseudo-Outbreak of Mycobacterium xenopi Due to a Contaminated Potable Water Supply: Lessons in Prevention. Infect. Control Hospital Epidemiol. 14, 636-641.

224

Song, H., Sandie, R., Wang, Y., Andrade-Navarro, M., Niederweis, M. 2008. Identification of outer membrane proteins of Mycobacterium tuberculosis. Tuberculosis. 88(6), 526-544.

Sood, A., Sreedhar, R., Kulkarnl, P., Nawoor, A. 2007. Hypersensitivity Pneumonitis-like Granulomatous Lung Disease with Nontuberculous Mycobacteria from exposure to Hot Water Aerosols. Environ. Health Perspectives. 115(2), 262-266.

Stall, R., Duran, L., Wisniewski, S., Friedman, M., Marshal, M., McFarland, W., Guadamuz, T., Mills, T. 2009. Running in Place: Implications of HIV Incidence Estimates among Urban Men Who Have Sex with Men in the United States and Other Industrialized Countries. AIDS Behavior. 13(4), 615-629.

Starke, J., Correa, A. 1995. Management of mycobacterial infection and disease in children. Ped. Infect. Dis. J. 14, 455-470.

Stead, W., Lofgren, J., Warren, J., Thomas, C. 1985. Tuberculosis as an endemic and nosocomial infection among the elderly in nursing homes. N. Engl. J. Med. 312, 1483-1487.

Steed, K., Falkinham, J. 2006. Effect of growth in biofilms on chlorine susceptibility of Mycobacterium avium and Mycobacterium intracellulare. Appl. Environ. Microbiol. 72, 4007–4100.

Steinert, M., Birkness, K., White, E., Fields, B., Quinn, F. 1998. Mycobacterium avium bacilli Grow Saprozoically in Coculture with Acanthamoeba polyphaga and Survive within Cyst Walls. Appl. Environ. Microbiol. 64, 2256–2261.

Steingrube V., Wilson, R., Brown, B., Jost, K., Blacklock, Z., Gibson, J., Wallace, R. 1997. Rapid Identification of Clinically Significant Species and Taxa of Aerobic Actinomycetes, Including Actinomadura, Gordona, Nocardia, Rhodococcus, Streptomyces, and Tsukamurella Isolates, by DNA Amplification and Restriction Endonuclease Analysis. J. Clin. Microbiol. 35(4), 817-822.

Steinhauer, K., Eschenbacher, I., Radischat, N., Detsch, C., Niederweis, M., Goroncy- Bermes, P. 2010. Rapid Evaluation of the Mycobactericidal Efficacy of Disinfectants in the Quantitative Carrier Test EN 14563 by Using Fluorescent Mycobacterium terrae. Appl. Environ. Microbiol. 76, 546-554.

Stine, C., Kane, A., Baya, A. 2010. Mycobacteria isolated from Chesapeake Bay Fishes. J. Fish. Dis. 33, 39-46.

Stine, T., Harris, A., Levin, S., Rivera, N., Kaplan, R. 1987. A Pseudoepidemic Due To Atypical Mycobacteria in a Hospital Water Supply. JAMA. 258, 809-811.

Stinear, T., Jenkins, G., Johnson, P., Davies, J. 2000. Comparative genetic analysis of Mycobacterium ulcerans and Mycobacterium marinum reveals evidence of recent divergence. J. Bacteriol. 182, 6322-6330.

225

Stinear, T., Johnson, P., Davies, J. 2001. Direct Detection of Mycobacteria in the Environment with Sequence Capture PCR, pp. 55-62. In: Rochelle, P. (ed.). Environmental Molecular Microbiology: Protocols and Applications. Horizon Press, Wymondham.

Stinear, T., Mye-Obiang, A., Small, P., Frigui, W., Pryor, M., Brosch, R., Jenkins, G., Johnson, P., Davies, J., Lee, R., Adusumilli, S., Garnier, T., Haydock, S., Leadlay, P., Cole, S. 2004. Giant plasmid-encoded polyketide synthases produce the macrolide toxin of Mycobacterium ulcerans. PNAS. 101(5), 1345-1349.

Stinear, T., Seemann, T., Harrison, P., Jenkin, G., Davies, J., Johnson, P., Abdellah, Z., Arrowsmith, C., Chillingworth, T., Churcher, C., Clarke, K., Cronin, A., Davis, P., Goodhead, I., Holroyd, N., Jagels, K., Lord, A., Moule, S., Mungall, K., Norbertczak, H., Quail, M., Rabbinowitsch, E., Walker, D., White, B., Whitehead, S., Small, P., Brosch, R., Ramakrishnan, L., Fischbach, M., Parkhill, J., Cole, S. 2008. Insights from the complete genome sequence of Mycobacterium marinum on the evolution of Mycobacterium tuberculosis. Genome Res. 18(5), 729-741.

Stoodley, P., Dodds I., Boyle, J., Lappin-Scott, H. 1998. Influence of hydrodynamics and nutrients on biofilm structure. J. Appl. Microbiol. 85, 19-28.

Stoodley, P., Hall-Stoodley, L., Lappin-Scott, H. 2001. Detachment, surface migration, and other dynamic behavior in bacterial biofilms revealed by digital time-lapse imaging. Methods Enzymol. 337, 306-319.

Stoodley, P., Wilson, S., Hall-Stoodley, L., Boyle, J., Lappin-Scott, H., Costerton, J. 2001. Growth and detachment of cell clusters from mature mixed species biofilms. Appl. Environ. Microbiol. 67, 5608-5613.

Stormer, R., Falkinham, J. 1989. Differences in antimicrobial susceptibility of pigmented and unpigmented colonial variants of Mycobacterium avium. J. Clin. Microbiol. 27, 2459–2465.

Strahl, E., Gillaspy, G., Falkingham, J. 2001. Fluorescent acid-fast microscopy for measuring phagocytosis of Mycobacterium avium, Mycobacterium intracellulare, and Mycobacterium scrofulaceum by Tetrahymena pyriformis and their intracellular growth. Appl. Environ. Microbiol. 67, 4432-4439.

Streeter, R., Hoffsis, G., Bech-Nielsen, S., Shulaw, W., Rings, D. 1995. Isolation of Mycobacterium paratuberculosis from colostrum and milk of subclinically infected cows. Am. J. Vet. Res. 56, 1322-1324.

Streit, M., Bregenzer, T., Heinzer, I. 2008. Cutaneous Infections due to Atypical Mycobacteria. Der. Hautarzt. 59(1), 59-70.

226

Sturgill-Koszycki, S., Schlesinger, P., Chakraborty, P., Haddix, P., Collins, H., Fok, A., Allen, R., Gluck, S., Heuser, J., Russell, D. 1994. Lack of acidification in Mycobacterium phagosomes produced by exclusion of the vesicular proton ATPase. Sci. 263, 678-681.

Sudha, G., Nirupa, C., Rajasakthivel, M., Sivasusbramanian, S., Sundaram, V., Bhatt, S., Subramaniam, K., Thiruvalluvan, E., Mathew, R., Renu, G., Santha, T. 2003. Factors influencing the care-seeking behavior of chest symptomatics: a community-based study involving rural and urban populations. Trop. Med. Intern. Health. 8(4), 336-341.

Sung, N., Collins, M. 2003. Variation in resistance of Mycobacterium paratuberculosis to acid environments as a function of culture medium. Appl. Environ. Microbiol. 69, 6833-6840.

Svetlikova, Z., Skovierova, H., Niederweis, M., Gaillard, J., McDonnell, G., Jackson, M. 2009. Role of Porins in the Susceptibility of Mycobacterium smegmatis and Mycobacterium chelonae to Aldehyde-Based Disinfectants and Drugs. Antimicro. Agents Chemo. 53, 4015-4018.

Swarz, R., Naai, D., Vogel, C., Yeager, H. 1988. Differences in uptake of mycobacteria by human monocytes: a role for complement. Infect. Immun. 56, 2223-2227.

Swerdlow, D., Mintz, E., Rodriguez, M., Tejada, E., Ocampo, C., Espejo, L., Greene, K., Saldana, W., Seminario, L., Tauxe, R., Wells, J., Bean, N., Ries, A., Pollack, M., Vertiz, B., Blake, P. 1992. Waterborne transmission of epidemic cholera in Trujillo, Peru: lessons for a continent at risk, Lancet. 340(8810), 28-33.

Takai, K., Horikoshi, K. 2000. Rapid detection and quantification of members of the archaeal community by quantitative PCR using fluorogenic probes. Appl. Environ. Microbiol. 66, 5066-5072.

Tan, C., Lai, C., Liao, C., Chou, C., Hsu, H., Huang, Y., Hsueh, P. 2010. Mycobacterial bacteraemia in patients infected and not infected with human immunodeficiency virus, Taiwan. Clin. Microbiol Inf. 16, 627–630.

Taylor, G., Worth, D., Palmer, S., Jahans, K., Hewinson, R. 2007. Rapid detetction of Mycobacterium bovis DNA in cattle lymph nodes with visible lesions using PCR. BMC Vet. Res. 3, 12.

Taylor, R., Falkingham, J., Norton, C., LeChevallier, M. 2000. Chlorine, Chloramine, Chlorine Dioxide, and Ozone Susceptibility of Mycobacterium avium. Appl. Environ. Microbiol. 66, 1702-1705.

Taylor, T., Patterson, C., Hale, Y., Safranek, W. 1997. Routine use of PCR-restriction fragment length polymorphism analysis for identification of mycobacteria growing in liquid media. J. Clin. Microbiol. 35, 79-85.

227

Taylor, T., Patterson, C., Hale, Y., Safranek, W. 1996. Routine Use of PCR-Restriction Fragment Length Polymorphism Analysis for Identification of Mycobacteria Growing in Liquid Media. J. Clin. Microbiol. 35(1), 79-85.

Tebbe, C., Vahjen, W. 1993. Interference of humic acids and DNA extracted directly from soil in detection and transformation of recombinant DNA from bacteria and a yeast. Appl. Environ. Microbiol. 59, 2657-2665.

Teunis, P., Xu, M., Fleming, K., Yang, J., Moe, C., LeChevallier, M. 2010. Enteric Virus Infection Risk from Intrusion of Sewage into a Drinking Water Distribution Network. Environ. Sci Tech. 44(22), 8561-8566.

Thegerstrom, J., Romanus, V., Friman, V., Brudin, L., Haemig, P., Olsen, B. 2008. Mycobacterium avium Lymphadenopathy among Children, Sweden. Emerg. Infect. Dis. 14(4), 661-663.

Thoen, C., Himes, E., Barrett, R. 1977. Mycobacterium avium serotype 1 infection in a sandhill crane (Grus canadensis). J. Wildlife Dis. 13, 40-42.

Thofern, E., Schoenen, D., Tuschewitzki, G. 1987. Microbial surface colonization and disinfection problems. Offentl. Gesundh.-wes. 49, 14-20.

Thomas, G., Swift, G., Green, J., Newcombe, R., Braniff-Mathews, C., Rhodes, J., Wilkinson, S., Strohmeyer, G., Kreuzpainter, G. 1998. Controlled trial of antituberculous chemotherapy in Crohn‘s disease: a five year follow up study. Gut. 42, 497-500.

Thomson, R. 2010. Changing Epidemiology of Pulmonary Nontuberculous Mycobacteria Infections. EID. 16(10), 1576-1583.

Thomson, R.,Yew, W. 2009. When and how to treat pulmonary non-tuberculous mycobacterial diseases. Respirol. 14, 12–26.

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. Appl. Environ. Microbiol. 74, 3094-3098.

Thorel, M., Moreau, R., Charvin, M., Ebiou, D. 1991. Debusquement enzymatique des mycobacteries dans les millieaux naturels. C. R. Soc. Biol. 185, 331-337.

Timms, V., Gehringer, M., Mitchell, H., Daskalopoulos, G., Neilan, B. 2011. How accurately can we detect Mycobacterium avium subsp. paratuberculosis infection? J. Microbiol. Methods. In Press. Available online 2011 March 1.

Torriani, F., Behling, C., McCutchan, J., Haubrich, R., Havlir, D. 1996. Disseminated Mycobacterium avium complex: correlation between blood and tissue burden. J. Infect. Dis. 173, 942-949.

228

Torvinen, E., Lehtola, M., Martikainen, P., Miettinen, I. 2007. Survival of Mycobacterium avium in Drinking Water Biofilms as Affected by Water Flow Velocity, Availability of Phosphorus, and temperature. Appl. Environ. Microbiol. 73(19), 6201-6207.

Trias, J., Jarlier, V., Benz, R. 1992. Porins in the cell wall of mycobacteria. Sci. 258(5087), 1479-1481.

Tsang, A., Denner, J., Brennan, P., McClatchy, K. 1992. Clinical and epidemiological importance of typing of Mycobacterium avium complex isolates. J. Clin. Microbiol. 30, 479-484.

Tu, H., Chen, C., Huang, T., Huang, W., Chen, Y., Liu, Y., Lin, Y. 2007. Use of a Disposable Water Filter for Prevention of False-Positive Results due to Nontuberculous Mycobacteria in a Clinical Laboratory Performing Routine Acid- fast Staining for Tuberculosis. Appl. Environ. Microbiol. 73(19), 6296-6298.

Tuttle, J., Gomez, T., Doyle, M., Wells, J., Zhao, T., Tauxe, R., Griffin, P. 1999. Lessons from a large outbreak of Escherichia coli O157:H7 Infections: insights into the infectious dose and method of widespread contamination of hamburger patties. Epidemiol. Infect. 122(2), 185-192.

Ucko, M., Colomi, A., Kvitt, H., Diamant, A., Zlotkin, A., Knibb, W. 2002. Strain variation in Mycobacterium marnium fish isolates. Appl. Environ. Microbiol. 68, 5281-5287.

Uthman, O. 2008. Geogrpahical variations and contextual effects on age of initiation of sexual intercourse among women in Nigeria: a multilevel and spatial analysis. Int. J. Health Geogr. 7, 27.

Vaerewijck, M., Huys, J., Palomino, J., Swings, J., Portaels, F. 2005. Mycobacteria in Drinking Wter Distribution Systems: Ecology and Significance for Human Health. FEMS Microbiol. Rev. 29, 911-934. van der Kooij, D. 2003. Assimilable organic carbon as an indicator of bacterial regrowth. J. Am. Water Works Assoc. 84, 57-65. van der Reijden, H., Schipper, M., Danner, S., Arisz, L. 1989. Glomerular lesions and opportunistic infections of the kidney in AIDS: an autopsy study of 47 cases. Adv. Exp. Med. Biol. 252, 181-188. van Ingen, J., Blaak, H., de Beer, J., de Roda, A., van Soolingen, D. 2010. Rapidly growing nontuberculous mycobacteria cultured from home tap and shower water. Appl Environ. Microbiol. 76(17), 6017-6019.

Van Kruningen, H., Chiodini, R., Thayer, W., Coutu, J., Merkal, R., Runnels, P. 1986. Experimental disease in infant goats induced by a Mycobacterium isolated from a patient with Crohn‘s disease. Dig. Dis. Sci. 31, 1351-1360.

229

Van Kruningen, H., Freda, B. 2001. A clustering of Crohn‘s disease in Mankato, Minnesota. Inflamm. Bowel Dis. 7, 27-33. van Oss, C., Gillman, C., Neumann, A. 1975. Phagocytic Engulfment and Cell Adhesiveness. Marcel Dekker, New York.

Verghese, S., Mullaseri, A., Padmaja, P., Subhadra, A., Cherian, K. 1998. Pacemaker implant site infection caused by atypical mycobacteria. Indian Heart J. 50, 201- 202.

Vincenzi C., Bardazzi, F., Tosti, A., Varotti, C., Morganti, L. 1992. Fish tank granuloma: report of a case. Cutis. 49, 275-276.

Volk, C., LeChavallier, M. 2000. Assessing biodegradable organic matter. J. Am. Water Works Assoc. 92, 64-76. von Reyn, C., Arbeit, R., Horsburgh, C., Ristola, M., Waddell, R., Tvaroha, S., Samore, L. Hirschhorn, M., Lumio, J., Lein, A., Grove, M., Tosteson, A. 2002. Sources of disseminated Mycobacterium avium infection in AIDS. J. Infect. 44, 166-170. von Reyn, C., Arbeit, R., Tosteson, A., Ristola, M., Barber, T., Waddell, R., Sox, C., Brindle, R., Gilks, C., Ranki, A., Bartholomew, C., Edwards, J., Falkingham, J., O‘Connor, G. 1996. The International Epidemiology of Disseminated Mycobacterium avium Complex Infection in AIDS. International MAC Study Group. AIDS. 10, 1025-1032. von Reyn, C., Barber, T., Arbeit, R., Sox, C., O‘Connor, G., Brindle, R., Gilks, C., Hakkarainen, R., Ranki, A., Bartholomew, C., Edwards, J., Tosteson, A., Magnussom, M. 1993. Evidence of previous infection with Mycobacterium avium- Mycobacterium intracellulare complex among healthy subjects: an internatinal study of dominant mycobacterial skin test reactions. J. Infect. Dis. 168(6), 1553- 1558. von Reyn, C., Horsburgh, C., Olivier, K., Barnes, P., Waddell, R., Warren, C., Tvaroha, S., Jaeger, A., Lein, A., Alexander, L., Weber, D., Tosteson, A. 2001. Skin test reactions to Mycobacterium tuberculosis purified protein derivative and Mycobacterium avium sensitin among health care workers and medical students in the United States. Intern. J. Tuberculosis Lung Dis. 5, 1122–1128. von Reyn, C., Jacobs, N., Arbeit, R., Maslow, J., Niemczyk, S. 1995. Polyclonal Mycobacterium avium infections in patients with AIDS: variations in antimicrobial susceptibilities of different strains of M. avium isolated from the same patient. J. Clin. Microbiol. 33, 1008-1010. von Reyn, C., Maslow, J., Barber, T., Falkingham, J., Arbeit, R. 1994. Persistent colonization of potable water as a source of Mycobacterium avium infection in patients with AIDS. Lancet. 343, 1137-141.

230

von Reyn, C., Waddell, R., Eaton, T., Arbeit, R., Maslow, J., Barber, T., Brindle, R., Gilks, C., Lumio, J., Lahdevirta, J., Ranki, A., Dawson, D., Falkingham, J. 1993. Isolation of Mycobacterium avium complex from water in the Unites States, Finland, Zaire, and Kenya. J. Clin. Microbiol. 31, 3227-3230.

Wagner D., Young, L. 2003. Nontuberculous Mycobacterial Infections. Inf. 31(5), 257- 270.

Walker, J., Hughes, D., Hedges, D., Anders, B., Laborde, M., Shewale, J., Sinha, S., Batzer, M. 2004. Quantitative PCR for DNA identification based on genome- specific interspersed repetitive elements. Genomics. 83(3), 518-527.

Wallace, J., Hannah, J. 1988. Mycobacterium avium complex Infection in Patients with Immunodeficiency Syndrome. Chest. 93, 926-932.

Wallace, R., Brown, B., Griffith, D. 1998. Nosocomial Outbreaks/Pseudo-outbreaks Caused by Nontuberculous Mycobacteria. Ann. Rev. Microbiol. 52, 453-490.

Wallace, R., O‘Brien, R., Glassroth, J., Raleigh, J., Dutt, A. 1990. Diagnosis and treatment of disease caused by nontuberculous mycobacteria. Am. Rev. Respir. Dis. 142, 940-953.

Wallace, R., Zhang, Y., Wilson, R., Mann, L., Rossmoore, H. 2002. Presence of a single genotype of the newly described species Mycobacterium immunogenum in industrial metalworking fluids associated with hypersensitivity pneumonitis. Appl. Environ. Microbiol. 68, 5580-5584.

Walters, S., Yamahara, K., Boehm, A. 2009. Persistence of Nucleic Acid Markers of Health-Relevant Organisms in Seawater Microcosms: implications for their use in accessing risk in recreational waters. Water Res. 43(19), 4929-4939.

Wang, R., Luneau, A., Cao, W., Cerniglia, C. 1996. PCR detection of polycyclic aromatic hydrocarbon-degrading mycobacteria. Environ. Sci. Tech. 30, 307-311.

Wang, S., Pancholi, P., Stevenson, K., Yakrus, M., Butler, R., Schlesinger, L., Mangino, J. 2009. Pseudo-Outbreak of Mycobacterium paraffinicum Infection and/or Colonization in a Tertiary Care Medical Center. Inf. control hospital epidemiol. 30(9), 848-853.

Wayne, L., Good, R., Tsang, A., Butler, R., Dawson, D., Goothuis, D., Gross, W., Hawkins, J., Kilburn, J., Kubin, M. 1993. Serovar determination and molecular taxonomic correlation in Mycobacterium avium, Mycobacterium intracellulare, and Mycobacterium scrofulaceum: a cooperative study of the international working group on mycobacterial taxonomy. Intern. J. Syst. Bacteriol. 43, 482-489.

Weber, M., Blanchard, D., Syzdek, L. 1983. The mechanism of scavenging of waterborne bacteria by a rising bubble. Limnol. Oceanography. 28, 101-105.

231

Weinstein, R., Golomb, H., Grumet, G., Gelman, E., Schechte, G. 1981. Hairy cell leukemia: associated with disseminated atypical mycobacterial infection. Cancer. 48, 380-383.

Weitzul, S., Eichhorn, P., Pandya, A. 2000. Nontuberculous mycobacterial infections of the skin. Dermatol. Clinic. 18, 359-377.

Wendt, S., George, K., Parker, B., Gruft, H., Falkingham, J. 1980. Epidemiology of infection by nontuberculous Mycobacteria. III. Isolation of potentially pathogenic mycobacteria from aerosols. Am. Rev. Respir. Dis. 122, 259-263.

Whan, L., Grant, I., Ball, H., Scott, R., Rowe, M. 2001. Bactericidal effect of chlorine on Mycobacterium paratuberculosis in drinking water. Letters Appl. Microbiol. 33, 227-231.

Whittington, R., Fell, S., Walker, D., McAllister, S., Marsh, I., Sergeant, E., Taragel, C., Marshall, D., Links, I. 2000. Use of pooled fecal culture for sensitive and economic detection of Mycobacterium avium subsp. paratuberculosis infection in flocks of sheep. J. Clin. Microbiol. 38, 2550-2556.

Whittington, R., Lloyd, J., Reddacliff, L. 2001. Recovery of Mycobacterium avium subspecies paratuberculosis from nematode larvae cultured from the faeces of sheep with Johne‘s disease. Vet. Microbiol. 79, 311-322.

World Health Organization. 2000. Guidelines For Safe Recreational-Water Environments. Vol 2: Swimming pools, spas, and similar recreational-water environments. World Health Organization, Geneva.

World Health Organization. 2001. Buruli ulcer: Diagnosis of Mycobacterium ulcerans disease. In: Portaels, F., Johnson, P., Meyers, W. (ed.). A manual for health care providers. World Health Organization, Geneva.

World Health Organization. 2008. Global leprosy situation, beginning of 2008. Wkly. Epidemiol. Rec. 83, 293-300.

World Health Organization. 2010. Tuberculosis Fact Sheet. Number 104. World Health Organization, Geneva.

Wilson, I. 1997. Inhibition and Facilitation of Nucleic Acid Amplification. Appl. Environ. Microbiol. 63(10), 3741-3751.

Winthrop, K. 2010. Pulmonary disease due to nontuberculous mycobacteria: an epidemiologists view. Future Microbiol. 5(3), 343-345.

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 J. Med. 346, 1366-1371.

232

Winthrop, K., Chang, E., Yamashita, S., Iademarco, M., LoBue, P. 2009. Nontuberculous Mycobacteria Infections and Anti-Tumor Necrosis Factor-α Therapy. Emerg. Infect. Dis. 15(10), 1556-1561.

Winthrop, K., McNelley, E., Kendall, B., Marshall-Olson, A., Morris, C., Cassidy, M., Saulson, A., Hedberg, K. 2010. Pulmonary nontuberculous mycobacterial disease prevalence and clinical features: an emerging public health disease. Am. J. Respir. Crit. Care Med. 182(7), 977-982.

Witty, L., Tapson, V., Piantadosi, C. 1994. Isolation of mycobacteria in patients with pulmonary alveolar proteinosis. Med. 73, 103-107.

Woelk, E., Goroncy-Bermes, P., Sand, W. 2003. Influence of storage on monodispersed cells of Mycobacterium terrae used for quantitative carrier test prEN 14563. Appl. Environ. Microbiol. 69, 6932-6934.

Wolffs, P., Norling, B., Radstrom, P. 2005. Risk assessment of false-positive quantitative real-time PCR results in food, due to detection of DNA originating from dead cells. J. Microbiol. Methods. 60(3), 315-323.

Wolinsky, E. 1995. Mycobacterial lymphadenitis in children: a prospective study of 105 nontuberculous cases with long-term follow-up. Clin. Infect. Dis. 20, 954-963.

Wolinsky, E. 1992. Mycobacterial diseases other than tuberculosis. Clin. Infect. Dis. 15, 1-10.

Wolinsky E. 1979. Nontuberculous mycobacteria and associated diseases. Am. Rev. Respir. Dis. 119, 107-59.

Wolinsky, E., Gomez, F., Zimpfer, F. 1972. Sporotrichoid Mycobacterium marinum infection treated with rifampin-ethambutol. Am. Rev. Respir. Dis. 105, 964-947.

Wolschendorfa, F., Ackartb, D., Shresthac, T., Hascall-Doveb, L., Noland, S., Lamichhaned, G., Wanga, Y., Bossmannc, S., Basarabab, R., Niederweisa, M. 2011. Copper resistance is essential for virulence of Mycobacterium tuberculosis. PNAS. 108(4), 1621–1626.

Wong, B., Edwards, F., Kiehn, T., Whimbey, E., Donnelly, H., Bernard, E., Gold, J., Armstrong, D. 1985. Continuous high-grade Mycobacterium avium-intracellulare bacteremia in patients with acquired immunodeficiency syndrome. Am. J. Med. 78, 35-40.

Wood, E., Montaner, J., Bangsberg, D., Tyndall, M., Strathdee, S., O‘Shaughnessy, M., Hogg, R. 2003. Expanding access to HIV antiretroviral therapy among marginalized populations in the developed world. AIDS. 17(17), 2419-2427.

233

Woods, G., Long, T., Witebsky, F. 1996. Mycobacterial testing in clinical laboratories that participate in the College of American Pathologists Mycobacteriology Surveys. Changes in practices based on responses to 1992, 1993, and 1995 questionnaires. Arch. Pathol. Lab. Med. 120(5), 429-435.

Woods, S., Cole, T. 1989. A rapid method fro the detection of potentially viable Mycobacterium leprae in human biopsies: a novel application of PCR. FEMS Microbiol. Letters. 65(3), 305-309.

Wynne, J., Seemann, T., Bulach, D., Coutts, S., Talaat, A., Michalski, W. 2010. Re- sequencing the Mycobacterium avium subsp. paratuberculosis K10 genome: improved annotation and revised genome sequence. J. Bacteriol. 192(23), 6319- 20.

Yajko, D., Chin, D., Gonzalez, P., Nassos, P., Hopewell, P., Reingold, A., Horsburgh, C., Yakrus, M., Ostroff, S., Hadley, W. 1995. Mycobacterium avium complex in water, food, and soil samples collected from the environment of HIV-infected individuals. J. Acquired Immune Defic. Synd. Human Retrovirals. 9, 176-182.

Yamazaki, Y., Danelishvili, L., Wu, M., MacNab, M., Bermudez, L. 2006. Mycobacterium avium genes associated with the ability to form biofilm. Appl. Environ. Microbiol. 72, 819-825.

Yang, M., Ross, B., Dwyer, B. 1993. Isolation of a DNA probe for identification of Mycobacterium kansasii, including the genetic subgroup. J. Clin. Microbiol. 31, 2769-2772.

Yarnell, S., Wellehan, J., Johnson, J., Wallace, R., Kane, A. 2011. Nontuberculous mycobacteria detected using novel qPCR Probes. Appl. Environ. Microbiol. In review.

Yassin, M., Abu Amr, S., Al-Najar, H. 2006. Assessment of microbiological water quality and its relation to human health in Gaza Governorate, Gaza Strip, Public Health. 120, 1177–1187.

Yeboah-Manu, D., Bodmer, T., Mensah-Quainoo, E., Owusu, S., Ofori-Adjei, D., Pluschke, G. 2004. Evaluation of Decontamination Methods and Growth Media for Primary Isolation of Mycobacterium ulcerans from Surgical Specimens. J. Clin. Microbiol. 42(12), 5875-5876.

Yip, M., Porter, J., Fyfe, J., Lavender, C., Portaels, F., Rhodes, M., Kator, H., Colorni, A., Jenkin, G., Stinear, T. 2007. Evolution of Mycobacterium ulcerans and other mycolactone-producing mycobacteria from a common Mycobacterium marinum progenitor. J. Bacteriol. 189(5), 2021-2029.

Yu, F., Callis, G., Stewart, P., Griebe, T., McFeters, G. 1994. Cryosectioning of biofilms for microscopic examination. Biofouling. 8, 85-91.

234

Zakowski, P., Fligiel, S., Berlin, O., Johnson, B. 1982. Disseminated Mycobacterium avium-intracellulare infection in homosexual men dying of acquired immunodeficiency. JAMA. 248, 2980.

Zeligman I. 1972. Mycobacterium marinum granuloma: A disease acquired in the tributaries of Chesapeake Bay. Arch. Dermatol. 106, 26-31.

Zenone, T., Boibieux, A., Tigaud, S., Fredenucci, J., Vincent, V., Chidiac, C., Peyramond, D. 1999. Non-tuberculous mycobacterial tenosynovitis: a review. Scandinavian J. Infect. Dis. 31, 221-228.

Zhang, M., Zhang, S. 2011. An efficient DNA extraction method for polymerase chain reaction-based detection of Mycobacterium avium subspecies paratuberculosis in bovine fecal samples. J. Vet. Diagn. Invest. 23(1), 41-48.

Zolg, J., Philippi-Schulz, S. 1994. The superoxide dismutase gene, a target for detection and identification of mycobacteria by PCR. J. Clin. Microbiol. 32(11), 2801-2812.

Zuber, B., Chami, M., Houssin, C., Dubochet, J., Griffiths, G., Daffe, M. 2008. Direct visualization of the outer membrane of mycobacteria and corynebacteria in their native state. J. Bacteriol. 190(16), 5672-5680.

235

BIOGRAPHICAL SKETCH

Stephanie C. Yarnell is the daughter of Howard and Donna Yarnell of Mansfield,

Texas. She grew up in Atlanta, Georgia, and graduated valedictorian of Pickens County

High School in 2002. Stephanie went on to attend the University of Georgia where she graduated Summa cum Laude with highest honors in 2006 with a B.S. in biology, B.S. in microbiology, and a B.S. in ecology. While in her undergraduate studies she completed a honors thesis entitled, ―Analysis of the Interactions Between Glycosaminoglycans and

Pectins‖ and was awarded the prestigious CURO Scholar Award, as well as a

Biomedical Research Fellowship. In 2006, she entered the Medical College of Georgia pursuing a combined M.D./Ph.D degree program. Upon completing the first two years of medical school, she transferred to the University of Florida in 2008 to begin her

Doctor of Philosophy studies in the Interdisciplinary Program in biomedical sciences within the College of Medicine concentrating in immunology and microbiology. While at the University of Florida, she has served as a member of the admissions committee for the M.D./Ph.D Program in the College of Medicine and as a Physical Exam Teaching

Associate for the College of Medicine. In 2011, she was awarded the prestigious P.E.O

Scholar award. She received her Ph.D. from the University of Florida in summer 2011.

Upon completion of her Doctor of Philosophy studies, she re-matriculated into her third year of medical school at the University of Florida College of Medicine, and is expected to graduate from medical school in 2013.

236