Detection of Pathogenic and Non-pathogenic 
Bacteria in Drinking Water and Associated 
Biofilms on the Crow Reservation, Montana, 
USA
Authors: Crystal L. Richards, Susan C. Broadaway, 
Margaret J. Eggers, John Doyle, Barry H. Pyle, Anne 
K. Camper, and Timothy E. Ford
The final publication is available at Springer via http://dx.doi.org/10.1007/s00248-015-0595-6.
Richards, Crystal L. , Susan C. Broadaway, Margaret J. Eggers, John Doyle, Barry H. Pyle, Anne 
K. Camper, and Timothy E. Ford. "Detection of Pathogenic and Non-pathogenic Bacteria in 
Drinking Water and Associated Biofilms on the Crow Reservation, Montana, USA." Microbial 
Ecology 76, no. 1 (July 2018): 52-63. DOI:10.1007/s00248-015-0595-6.
Made available through Montana State University’s ScholarWorks 
scholarworks.montana.edu 
1 
Detection of Mycobacteria, Legionella, and Helicobacter in Drinking Water and Associated 
Biofilms on the Crow Reservation, Montana, USA 
Crystal L. Richards1, Susan C. Broadaway1, Margaret J. Eggers1,2, Emily Colgate3, John
Doyle4,5,6, Barry H. Pyle1, Anne K. Camper7*, and Timothy E. Ford8
1Department of Microbiology and Center for Biofilm Engineering 
Montana State University 
Bozeman, MT 59717 
2Little Big Horn College 
Crow Agency, MT 59022 
3University of Vermont 
College of Medicine 
Burlington, VT 05405 
4Apsaalooke Water and Wastewater Authority 
Hardin, Montana 59034 
5Crow Tribal Member 
Crow Agency, MT 59022 
6Big Horn County Commissioner 
Hardin, Montana 59034 
7Department of Civil Engineering and Center for Biofilm Engineering 
Montana State University 
Bozeman, MT 59717 
8Department of Research and Graduate Studies 
University of New England 
Biddeford, ME 04005 
Journal: Applied and Environmental Microbiology 
Journal Section: Public Health 
_____________________ 
(*)Corresponding author.  Mailing address: 
Center for Biofilm Engineering,  
Montana State University, EPS 366, Bozeman, MT 59717. 
Phone: (406) 994-4906.  Fax: (406) 994-6098.  E-mail: anne_c@biofilm.montana.edu 
2 
 
ABSTRACT 1 
Public health has been shown to be directly related to water quality, and although drinking water 2 
quality has improved in much of the United States, rural areas typically have underserved water 3 
systems.  Private residences in rural areas with water systems that are not adequately regulated, 4 
monitored, and updated could have drinking water that poses a health risk.  To investigate water 5 
quality on the Crow Reservation in Montana, water and biofilm samples were collected from 57 6 
public buildings and private residences served by both treated municipal and individual 7 
groundwater well systems.  Three bacterial genera, with members that are potential drinking 8 
water pathogens, were chosen for investigation.  Mycobacteria, Legionella, and Helicobacter 9 
were detected by PCR and/or standard culture techniques.  Free and total chlorine, temperature, 10 
and pH were recorded at the time of sampling.  Fecal coliform bacteria and heterotrophic plate 11 
count (HPC) bacteria were enumerated using m-Coliblue24® and R2A agar, respectively.  All 12 
three target genera were detected in drinking water systems on the Crow Reservation.  Species 13 
detected included the opportunistic and frank pathogens Mycobacterium avium, M. gordonae, M. 14 
flavescens, Legionella pneumophila, and H. pylori.  There was no correlation between the 15 
presence of any genera and chlorine (free and total), temperature, pH or fecal coliforms. 16 
However, there was an association between HPC bacteria and the presence of Mycobacteria and 17 
Legionella but not the presence of Helicobacter.  This research has shown that groundwater and 18 
municipal drinking water systems and associated biofilms may be reservoirs for Mycobacteria, 19 
Legionella, and Helicobacter. 20 
 21 
 22 
 23 
3 
 
INTRODUCTION 24 
In the United States over 15 million households rely on private ground water wells for 25 
their primary drinking water source (82), and in many rural areas private and community 26 
groundwater wells provide a major source of drinking water (12).  Generally, most water 27 
obtained from private groundwater systems is considered safe to drink (23).  However, in the 28 
United States from 1999-2002, 22% of water-borne illnesses were attributed to individual water 29 
systems and 36% were attributed to community systems (24).  Private water systems are not 30 
routinely monitored for bacteriological water quality, thus little is known about the presence of 31 
bacterial pathogens in these systems.  Information regarding water quality on Indian 32 
Reservations in the United States is equally scant.  However, it is known that American Indian 33 
populations have disproportionately high disease burdens compared to the overall population of 34 
the United States (59).  This is due to many factors which include economics, geographic 35 
isolation, cultural barriers, and inadequate sewage disposal (5).  36 
The United States Centers for Disease Control report that chronic lower respiratory 37 
disease, influenza and pneumonia are among the top ten causes of death among American Indian 38 
and Alaska Native populations (82).  In Montana, cancer is included as a major cause of death 39 
for American Indian populations (21).  Although it has been observed that the disease burden of 40 
these populations is greater than the overall population of the United States (59), very little 41 
research has been done to identify causes and potential routes of exposure to infectious agents 42 
and environmental carcinogens.  In the present study, three bacterial genera with members that 43 
are potential drinking water pathogens, Mycobacteria, Legionella, and Helicobacter, were 44 
chosen for investigation due to concerns expressed by Crow Tribal community members about 45 
4 
 
poor drinking water quality and the relationship of these organisms to respiratory disease and 46 
stomach cancer (28).    47 
Mycobacteria are common inhabitants of drinking water systems and are known to 48 
survive and proliferate in biofilms (29, 49).  Several species of this genus cause respiratory 49 
disease in mainly immunocompromised humans. Species include members of the 50 
Mycobacterium avium complex, M. gordonae, M. flavescens, and others (18, 36, 54, 55).  51 
Legionella are ubiquitous throughout aquatic environments including ground and surface water, 52 
and manmade water reservoirs such as potable water systems and cooling towers. (20, 53, 81).  53 
Legionella pneumophila is the main causative agent for respiratory disease in that genus, causing 54 
Legionellosis in the form of Legionnaire’s disease and Pontiac fever (30).  Legionellosis is 55 
thought to occur when Legionella are aerosolized and inhaled (62).  However, it has been 56 
suggested that transmission of the different forms of Legionellosis, and the resultant severity of  57 
disease, may be related to an association with biofilms (42).  Helicobacter are pathogens of the 58 
gastrointestinal tract of mammals but have been found in many environments such as well, river, 59 
and pond water, in addition to house flies, and cattle feces (68).  Helicobacter pylori are the 60 
primary bacterial cause of gastritis, as well as peptic and duodenal ulcers in people around the 61 
world (63).  Infection is known to increase the risk of the development of gastric mucosa-62 
associated lymphoma and adenocarcinoma (62). Water is a short term reservoir, with the 63 
pathogen often occurring sporadically in drinking water supplies that have been exposed to 64 
sewage, or have been contaminated by infected animals (10).  65 
Drinking water samples and their associated biofilms were tested for heterotrophic and 66 
coliform bacteria by traditional culture methods.  Mycobacteria, Legionella, and Helicobacter 67 
species were detected by culture and PCR.  The aim of this study was to investigate whether 68 
5 
 
these organisms are common inhabitants of drinking water systems on the Crow Reservation in 69 
southeast Montana.  70 
 71 
MATERIALS AND METHODS 72 
Study Area.  The Crow Indian Reservation, Montana, USA was the primary location for sample 73 
collection and analysis.  Fifty-seven locations were sampled across the Crow Reservation (41 74 
private residences and 14 public buildings) from March 2007 through July 2009.  The Crow 75 
Reservation, the largest reservation in Montana, is rural with an average population density of 76 
1.9 individuals per square mile (82).  The Crow tribe has an enrolled membership of 11,357 and 77 
approximately 72% of members live on or near the Reservation (65).  This Reservation has a 78 
diverse landscape spanning the Wolf, Big Horn and Pryor Mountain ranges, as well as the Big 79 
Horn and Little Big Horn River valleys.  Land use is typical of rural areas in Montana with 80 
approximately 68% grazing rangeland, 12% dry cropland, 3% irrigated cropland, 15% forested 81 
areas, 1% wild land, and 1% developed areas (4).  The Crow Reservation area receives 82 
approximately 12-18 total inches annual precipitation (4).  The surface water in the area is 83 
dependent on precipitation, snowpack and groundwater for recharge while the aquifers on the 84 
reservation rely on infiltration from rivers, streams, precipitation, stock ponds and reservoirs 85 
(38). The major township on the reservation, Crow Agency, has drinking water provided by 86 
treated surface water, while other townships utilize community and private groundwater wells 87 
and springs (33).  The Crow Agency treatment facility performs reliably and adequately; 88 
however the distribution system in Crow Agency is nearly 100 years old and is vulnerable to 89 
cracks and leaks (27).  Most of the residents outside of designated townships have privately 90 
maintained groundwater wells, often only drilled to first water.  91 
6 
 
Sample Collection and Processing.  Samples were primarily collected from kitchen sinks in private 92 
residences and kitchen or restroom sinks in public buildings.  Biofilm samples were collected first, 93 
before any flushing or sterilization of the tap. Biofilm samples were collected by systematically 94 
wiping the inside of the drinking water faucet with a sterile cotton swab before any flushing or 95 
sterilization of the tap.  Three swabs were collected from one faucet at each residence or building 96 
and were placed in individual tubes containing sterile water for transport.  To calculate surface area 97 
of the biofilm, the faucet dimensions (depth and width) were measured and recorded. After biofilm 98 
collection, the faucet was wiped with 95% ethanol to sanitize it before bulk water collection. After 99 
sanitization of the tap, one liter of water was collected without flushing and is denoted as “first 100 
flush”.  First flush sample collection was added in 2008 (n=20) and thus this fraction was only 101 
analyzed for groundwater wells.  After first flush collection, the water was run from the tap for two 102 
minutes minimum or until water temperature stabilized prior to parameter measurement.   Physical 103 
and chemical characteristics were measured using standard methodologies.  The presence and 104 
quantity of free and total chlorine was measured using a colorimetric method (Hach kit model CN-105 
70 chlorine test kit, Hach Co., Loveland, CO).  The temperature and pH were measured using a 106 
multi-parameter probe (Oakton, Vernon Hills, IL).  After a minimum of two minutes of flushing, 107 
“post flush” water was collected in three separate one liter sterile plastic bottles.  All samples were 108 
placed on ice and transported to the laboratory and processed within 24 h.  In the laboratory, tubes 109 
containing biofilm samples were vortexed for one minute and the cell suspensions from each swab 110 
were pooled and mixed. Pooled cell suspensions from the same source were used for all biofilm 111 
analysis.  To concentrate water samples, 900 ml from each liter sample was filtered through a 0.45 112 
µm 25mm diameter mixed cellulose ester filter (Pall Corp., Ann Arbor, MI).  The filter was vortexed 113 
in 1.5 ml of PBS at maximum speed for one minute, and then removed and the cell suspension 114 
7 
 
centrifuged at 13,000 x g for 10 minutes.  The supernatant was removed and the cell pellet was 115 
resuspended in 100 µl of sterile water. Pooled biofilm and both concentrated and un-concentrated 116 
water samples were used in DNA extractions. Pooled biofilm and un-concentrated water samples 117 
were used for genus specific culture methods. 118 
Quantification of Fecal Indicator Bacteria and Heterotrophic Bacteria.  Fecal indicator bacteria 119 
were quantified for first flush and post flush water samples using standard methodologies.  The 120 
presence of fecal contamination was determined by growth on the selective and differential m-121 
Coliblue24® broth (Hach Co., Loveland, CO).  This medium was used to culture coliform bacteria 122 
and differentiates Escherichia coli by using an enzymatic indicator.  One hundred ml of each water 123 
sample and appropriate dilutions were filtered and the filter placed on a pad soaked with 2 ml of the 124 
Coliblue24® broth.  The filters were incubated at 37˚C and growth was observed at 24 h.  Post-flush 125 
water samples were collected in triplicate and each replicate was analyzed separately.  Sterile water 126 
was filtered as a negative control, and sterile water was inoculated with environmental isolates of 127 
Escherichia coli and Klebsiella pneumoniae obtained from drinking water samples from New 128 
Haven, CT as a positive control.  Heterotrophic bacteria were enumerated for biofilm, first flush and 129 
post flush samples.  Each sample was diluted and plated on R2A agar followed by incubation at 130 
30ºC for 2 weeks. 131 
Control Bacterial Strains and Growth Conditions.  Representative species from each genus of 132 
interest were kept as frozen stocks at -70˚C and used as positive controls in both PCR and culture 133 
methods. The M. avium W2001 strain used in this study was originally isolated from drinking water 134 
in the Boston area and has been classified as M. avium subsp. hominisuis based on the hsp65 gene 135 
(80). The M. avium strain was cultured on Middlebrook 7H10 (Difco) and incubated at 37˚C for 10 136 
days.  L.  pneumophila strain 33153 was obtained from the American Type Culture Collection, 137 
8 
 
cultured on Legionella agar (Difco) enriched with 0.7% L-cysteine and 0.3% ferric pyrophosphate 138 
(Difco) and incubated at 37˚C for 7 days.  The H. pylori strain 43504 was obtained from the 139 
American Type Culture Collection and was cultured on H. pylori specific HP medium (25).  The 140 
Helicobacter cultures were placed in a BBL anaerobe jar with a BBL CampyPak PlusTM sachet, 141 
which creates a microaerophilic atmosphere of 5-10% oxygen and 10% carbon dioxide, and 142 
incubated for one week at 37˚C.  All control strains were grown and sequentially transferred twice 143 
prior to use as positive controls.   144 
Culture of Drinking Water and Biofilm Samples.  The drinking water and biofilm samples were 145 
analyzed for the presence of Mycobacteria, Legionella, and Helicobacter, by organism appropriate 146 
culture techniques.  Due to overgrowth by other micoorganisms, specific selection methods were 147 
employed to target the organisms of interest.  To select for members of the genus Mycobacteria, two 148 
hundred microliters of each unconcentrated water sample as well as the pooled biofilm suspension 149 
were treated with a final concentration of 0.005% cetyl pyridinium chloride (CPC) (Sigma, St. 150 
Louis, MO) for 30 minutes as previously described (70).   Sterile tap water was inoculated with M. 151 
avium W2001 and treated with CPC as a positive control.  The CPC treated cells were washed with 152 
phosphate buffered saline twice by centrifuging at 10,000 x g for 5 minutes.  Subsequently, one 153 
hundred microliters were plated onto M7H10 agar (Difco), two replicates were plated for each 154 
sample and were incubated at 37˚C for up to three weeks.  To select for Legionella species, each 155 
sample was heated to 50˚C for 30 minutes in a water bath (9).  Subsequently, one hundred 156 
microliters were plated onto enriched Legionella agar (Difco), two replicates were plated for each 157 
sample and incubated at 37˚C for one week.  Sterile tap water was inoculated with L. pneumophila 158 
ATCC 33153 and treated with heat as a positive control.  The samples were also cultured on H. 159 
pylori specific HP medium (25).  H.  pylori ATCC 43504 was inoculated into sterile water and 160 
9 
 
plated as a positive control.  The plates were placed in a BBL anaerobe jar with a BBL CampyPak 161 
PlusTM sachet, and incubated for one week at 37˚C.  All presumptive isolates were subcultured and 162 
subsequently identified by PCR and phylogenetic analysis. 163 
DNA Extraction from Biofilm and Water Samples.  Nucleic acids were extracted from the pooled 164 
biofilm suspensions and from concentrated and unconcentrated water samples.  DNA was extracted 165 
within 48 h of sampling and the extracts were immediately frozen at -20˚C.  Two ml of each biofilm 166 
sample was centrifuged at 12,000 x g for 15 minutes.  Subsequently, all but 100μl of the supernatant 167 
was removed, the pellet was mixed thoroughly into the liquid and the suspension was added to 2 ml 168 
plastic screw cap tubes with o-rings (Fisher) containing 0.4g of 0.1 mm sterile glass beads.  169 
Similarly, 200μl of each concentrated and unconcentrated water sample was added to individual 170 
sterile bead tubes for DNA extraction.  Two hundred microliters of lysis buffer consisting of 20 mM 171 
sodium acetate (Fisher Scientific, Fair Lawn, New Jersey), 0.5% sodium dodecyl sulfate (Fisher), 172 
and 1mM ethylenediamine-tetraacetic acid (Fisher) and 500 μl phenol (pH 8.1) (Fisher) was also 173 
added to each 2 ml tube and the mixture was homogenized in a Fastprep® FP120 cell disrupter at 174 
speed 5.0 for 40 seconds.  After homogenization, samples were placed on ice and allowed to rest for 175 
10 minutes.  The samples were then centrifuged at 12,000 x g for 10 minutes.  The DNA was 176 
precipitated by transferring the supernatant to a fresh 2 ml tube containing an equal volume of 177 
chloroform: isoamylalcohol (24:1).  The samples were vortexed for 30 seconds and then centrifuged 178 
at 12,000 x g for 5 minutes.  The supernatant was transferred to another fresh tube containing an 179 
equal volume of isopropanol and 1/10 volume of 3M sodium acetate and held at -20˚C for 24 h.  The 180 
nucleic acids were subsequently pelleted by centrifugation, washed once with 70% ethanol, air-181 
dried, and finally resuspended in 100μl of Tris-EDTA buffer (TE) consisting of 10mM Tris and 182 
1mM EDTA (Fisher).     183 
10 
 
PCR Amplification, Sequencing, and Phylogenetic Analysis. The detection limit of each primer 184 
set was determined by amplification of a 10-fold dilution series of purified genomic DNA (10ng-185 
0.0001pg).  The target genes, sequences, product sizes, and PCR conditions are listed in Table 1.  To 186 
ensure that the PCR reaction was not inhibited by environmental contaminants, amplification of each 187 
sample was performed using eubacterial 16S rRNA primers as described by Voytek et al. (84). 188 
Amplification of the PCR products was done in 25µl PCR mixture containing 1x PCR buffer II, 50-189 
200ng template DNA, 200µM (each) deoxynucleoside triphosphates (Takara Bio Inc., Japan), 0.1 190 
µM (each) of primer (Integrated DNA Technologies, Coralville, IA), and 1U LA Taq polymerase 191 
(Takara).  Aliquots of each PCR product were separated by electrophoresis in a 0.8% (w/v) agarose 192 
gel (Fisher) in TBE buffer consisting of 90mM Tris-HCl (Fisher), 80mM boric acid (Fisher), 2.5mM 193 
EDTA (Fisher) and stained with ethidium bromide (0.5µg/ml).  PCR products were purified using 194 
the Qiaquick PCR purification kit (Qiagen, Valencia, CA) according to the manufacturer’s 195 
instructions.  Automated sequencing from both strands of PCR products of positive samples was 196 
performed by the Molecular Research Core Facility at Idaho State University.  DNA sequences were 197 
assessed for their similarity to published DNA sequences using the BLAST database 198 
(http://www.ncbi.nlm.nih.gov/BLAST/).  Sequences were aligned with ClustalW (77).  Phylogenetic 199 
trees were constructed with the neighbor-joining method (67) and the Jukes-Cantor distance model 200 
(41) with bootstrap values of 1,000 replicates within MEGA v4.0 (75). All sequences were deposited 201 
in GenBank, accession numbers HQ018935-HQ018989. 202 
Statistical Analysis.  All data were compiled and for all instances where plate count values had a 203 
value of zero indicating none detected, a substitution rule was used (83).  An arbitrary value (0.25) 204 
was chosen to replace all zero plate counts so that log transformations could be performed.  Multiple 205 
linear regression and logistic regression tests were performed in Minitab® to determine correlations 206 
11 
 
between pH, temperature, drinking water and biofilm heterotrophic bacteria, total coliform bacteria, 207 
Helicobacter, Legionella and, Mycobacteria (44).  Additionally, paired and Welch two sample t-tests 208 
were performed on heterotrophic and total coliform bacteria to determine if there were significant 209 
differences between first flush, post flush, and biofilm samples (44).  Fisher’s exact tests were 210 
performed to determine if the presence of Mycobacteria, Legionella, and/or Helicobacter had a 211 
relationship with each other (19).   Fisher’s exact tests were also done to determine if there was a 212 
relationship between Mycobacteria, Legionella, and Helicobacter and the source of the drinking 213 
water (treated municipal or groundwater well) (19).  A Benjamini-Hochberg correction (10%) was 214 
applied to all analyses to minimize false discovery due to multiple comparisons (11).  215 
 216 
RESULTS 217 
Physical Characteristics of Sampled Drinking Water.   A total of 57 sites were sampled during 218 
this study.  Sixteen samples were collected from public buildings and private residences that had 219 
drinking water supplied by treated municipal systems, while the 41 remaining systems were 220 
community and private groundwater wells.  Total and free chlorine were quantified in drinking water 221 
sampled from municipal systems and ranged from none detected - 2.5 mg/L and none detected - 1.3 222 
mg/L with means of 0.34 mg/L and 0.27 mg/L, respectively.  The pH of the drinking water ranged 223 
from 5.82-9.56 with a mean of 7.42.  The temperature of the bulk water was recorded after flushing 224 
the tap and ranged from 8 - 33ºC with a mean of 15.7ºC and one outlier at 46ºC. Treated municipal 225 
systems had a mean temperature of 21.5ºC, while groundwater well systems had a mean temperature 226 
of 14.2ºC.  Simple linear regression and logistic regression were used to analyze relationships 227 
between the measured physical characteristics and HPC bacteria, total coliforms, Legionella, 228 
12 
 
Mycobacteria, and Helicobacter.  No significant statistical correlation was detected between the 229 
measured physical characteristics and any of the bacteria identified. 230 
Detection of Heterotrophic Bacteria and Fecal Indicator Bacteria. Heterotrophic bacteria were 231 
enumerated to assess whether these organisms were associated with the presence of potential 232 
pathogens.  Table 2 shows the range and arithmetic mean of HPC bacteria in first flush, biofilm and 233 
post flush drinking water samples.  The significance from the statistical analyses of the interactions 234 
between HPC bacteria counts and response variables are shown in Tables 3 and 4. Differences in the 235 
mean HPC populations between the first flush and post flush fractions collected were evaluated 236 
using Minitab® (Table 3).  There was a significant difference in mean HPC bacteria when first flush 237 
and bulk water samples were compared (P = 0.025, significant with Benjamini-Hochberg correction) 238 
with first flush samples having higher numbers of HPC bacteria on average.  Differences in HPC 239 
bacteria between treated municipal and groundwater wells were also evaluated (Table 5).  There was 240 
a difference between the mean HPC bacteria in biofilm samples and the drinking water source (P = 241 
0.049), however after applying the Benjamini-Hochberg correction this relationship was not 242 
significant.  Overall, biofilm samples collected from groundwater wells had higher HPC bacteria 243 
counts than biofilm samples collected from treated municipal systems.  There was not a significant 244 
difference in HPC bacteria numbers between the source water types for the post flush water samples.  245 
 Total coliform bacteria were enumerated in first flush and post flush drinking water samples.  246 
Coliform bacteria were found in both treated municipal (37.5% or 6/16) and untreated groundwater 247 
wells (40% or 16/40) in post flush water samples.  Escherichia coli was not observed in treated 248 
municipal samples but was found in 10% or 4/40 of post flush groundwater well samples.  Table 2 249 
shows the range and arithmetic mean of coliform bacteria and E. coli in drinking water samples.  250 
Although there was a significant difference in HPC bacteria in first flush and post flush fractions, 251 
13 
 
there was not a significant difference between mean coliform counts in first flush and post flush 252 
fractions (P > 0.13).  E. coli had a positive association with post flush HPC bacteria (P = 0.026, 253 
significant with correction) (Table 3).  254 
Presence of Mycobacteria, Legionella, and Helicobacter.  Mycobacterium species were detected in 255 
35.1% or 20/57 of the locations sampled, with 15 found in the biofilm fraction, and 8 in the drinking 256 
water fraction. Three of these occurrences of Mycobacteria were found in both the drinking water 257 
and biofilm fractions.  From the biofilm fractions, 7 of the 15 positive samples were identified by 258 
PCR alone and 5 were identified by culture alone while 3 were identified by both PCR and culture.  259 
From the drinking water fractions, 2 were identified by both PCR and culture while the remaining 6 260 
were identified by PCR only.  Fig. 1 shows the phylogenetic relatedness of the PCR and culture 261 
isolates.  The Mycobacterium species sequences detected were closely related to known species 262 
including M. gilvum, M. mucogenicum, M. murale, M. flavescens, M. gordonae, M. manitobense and 263 
members of the Mycobacterium avium complex (MAC) (>95% similarity).   264 
 To assess whether total coliforms or HPC bacteria influence the likelihood that Mycobacteria 265 
may be present, logistic regression was applied in Minitab®.  Identical analyses were performed for 266 
all three genera tested in this study.  The analysis showed a relationship between Mycobacteria and 267 
both post flush (P = 0.044, significant after correcting for multiple comparisons) and biofilm HPC 268 
bacteria (P = 0.01, not significant after corrections) (Tables 3 and 4 respectively).  This showed that, 269 
in general, as HPC bacteria increased, the odds of encountering Mycobacteria increased as well.  270 
Mycobacteria were detected when Coliforms were present in 50% or 10/20 of the locations sampled.  271 
Of the 20 locations that tested positive for Mycobacteria, 8 were treated municipal systems and 12 272 
were groundwater well systems.  There were no significant relationships between the presence of 273 
14 
 
Mycobacteria and total coliforms (logistic regression) or the source type of the drinking water 274 
system (Fisher’s exact test).  275 
 Legionella species were detected in 21% or 12 of the 57 locations sampled with 5 of those in 276 
the biofilm fraction, 8 in the drinking water fraction and only one occurrence of Legionella in both 277 
the biofilm and drinking water.  Of the 5 positive biofilm samples, 3 were identified by PCR alone, 1 278 
was identified by culture alone, and 1 was identified by both PCR and culture.  From the 8 drinking 279 
water samples, 2 were identified by both PCR and culture while the remaining 6 were identified by 280 
PCR alone.  Fig. 2 shows the phylogenetic relatedness of Legionella detected by PCR directly and 281 
from culture isolates.  The Legionella species detected include uncultured Legionella sp., L. 282 
pneumophila, Legionella sp., L. fairfieldensis, and L. dresdeniensis (sequence similarity >95%).  283 
The results of the logistic regression showed a positive relationship between post flush HPC 284 
bacteria counts and Legionella in the system (P = 0.003, significant after correcting for multiple 285 
comparisons) (Table 3).  In general, as post flush HPC bacteria increased, the odds of encountering 286 
Legionella increase as well.  The greatest interaction occurred between Legionella detected in the 287 
biofilm fraction and post flush HPC bacteria (P = 0.001).  Legionella detected in the drinking water 288 
fraction did not have a significant interaction (P = 0.068) with post flush fractions (Table 3).  There 289 
was no significant relationship between the presence of Legionella (in either the biofilm or drinking 290 
water fractions) and biofilm HPC bacteria (Table 4).  Coliforms were present in 6 of the 12 samples 291 
where Legionella were detected, but there was no significant relationship between the presence of 292 
Legionella and total coliforms (P = 0.679).  However, there was a significant association between 293 
the presence of Legionella and E. coli (P = 0.018).  Of the 12 samples positive for Legionella, 8 294 
were at treated municipal sites and 4 were in groundwater.  Unlike Mycobacteria, the source type of 295 
15 
 
the drinking water system did have a relationship with the presence of Legionella (P = 0.002) (Table 296 
5).  297 
Helicobacter species were detected in 7% or 4/57 of locations sampled, with 2 of those in the 298 
biofilm and 2 in the drinking water.  There were no occurrences of Helicobacter in the drinking 299 
water and biofilm concurrently.  All of the positive samples were identified by PCR alone.  Fig. 3 300 
shows the phylogenetic relatedness of the Helicobacter sequences detected by PCR directly.  The 301 
only Helicobacter species detected was H. pylori.  Coliforms were found in 2 of the 4 samples where 302 
Helicobacter were detected.  Logistic regression did not demonstrate any significant correlation 303 
between the presence of Helicobacter and any of the biological or physical parameters collected or 304 
the source type of the drinking water.  305 
Interactions Between Potentially Pathogenic Genera. To determine whether there was a 306 
relationship between the presence of the three genera of interest a Fisher’s exact test was performed.  307 
There was no statistically significant relationship between the three genera.  Interestingly, 50% or 308 
6/12 of locations positive for Legionella were also positive for Mycobacteria while only one location 309 
had both Legionella and Helicobacter.  Conversely, there were 20 occurrences of Mycobacteria with 310 
six of these samples also positive for Legionella (28.5%) and two samples positive for Helicobacter 311 
(9.5%). Helicobacter was found alone in one location (25% or 1/4). 312 
 313 
DISCUSSION 314 
 The results of our study show that Mycobacteria, Legionella, and Helicobacter can be found 315 
in drinking water and associated biofilms on the Crow Reservation, in both treated municipal water 316 
and untreated well water.  The data also indicated that the number of HPC bacteria correlated with 317 
the presence of Mycobacteria or Legionella.   318 
16 
 
Fecal Coliforms and HPC Bacteria in Drinking Water.  Although the presence of coliform 319 
bacteria in drinking water is a potential indicator of fecal contamination and may indicate the 320 
possible presence of harmful pathogens in drinking water (48), members of the coliform group  321 
are also common inhabitants of rural drinking water systems (45).  This study found that 40% of 322 
community and private groundwater wells contained coliform bacteria while 37.5% of treated 323 
municipal samples were positive.  This is in agreement with an Iowa statewide rural well water 324 
survey that found that 44% of private groundwater systems were contaminated with coliforms 325 
(37).  In our study area, surface water municipal and groundwater systems are vulnerable to 326 
contamination, particularly during wet seasons that result in flooding events.  These flooding 327 
events can drastically increase the turbidity of surface waters and hinder water treatment, 328 
potentially allowing coliform contamination of finished water.  During March 2007, the largest 329 
treatment facility on the Reservation was required to shut down due to mud and debris that 330 
clogged the intake pipe after a flood (16). This particular event accounts for all of the coliform 331 
positive municipal system samples except one, which occurred shortly after this flood.  The 332 
temporary closure of the treatment facility required a town-wide boil order and resulted in turbid 333 
water at the tap. Groundwater wells in rural areas are vulnerable to flooding, but are also 334 
susceptible to contamination from septic systems and inappropriate disposal of sewage effluents 335 
and sludges (12).  During sample collection, we occasionally observed instances where well 336 
heads were completely inundated after precipitation, and water at the tap was turbid and/or 337 
odiferous.  338 
 Coliform detection has inherent limitations and high levels of background bacteria can 339 
interfere with the assays (17, 32, 48).  Coliform bacteria often do not adequately predict the 340 
presence of pathogens, as has been demonstrated in waterborne outbreaks of Cryptosporidia, 341 
17 
 
Giardia, and Salmonella (43).  The lack of concurrence between the detection of Mycobacteria, 342 
Legionella, and Helicobacter and coliforms indicates that fecal indicator bacteria have limited 343 
use in predicting the presence of these environmental pathogens.  Our finding agrees with that of 344 
others who found no correlation between these organisms and fecal coliform bacteria (61, 72, 84, 345 
86).   346 
 Heterotrophic plate count bacteria are the normal flora of drinking water and include a 347 
wide range of organisms including Acinetobacter, Aeromonas, Bacillus, Corynebacterium, 348 
Pseudomonas, Mycobacteria, and Legionella (2).  Helicobacter also utilize organic nutrients for 349 
growth and thus fit the general definition of a heterotroph, but their microaerophilic lifestyle 350 
make them less suited for growth in the drinking water environment (1, 35).  This study has 351 
shown that HPC bacteria can occur in numbers >106 CFU/ml and that water that was stagnant in 352 
plumbing (first flush) had significantly greater numbers of HPC bacteria than water that has been 353 
collected after flushing (P = 0.025).  Water stagnation in drinking water pipes promotes bacterial 354 
accumulation and may compromise microbiological quality of drinking water when those 355 
organisms are flushed out (52).  In this study, groundwater wells generally had higher levels of 356 
HPC bacteria than treated municipal water, which can be at least partially explained by the 357 
presence of chlorine residuals in municipal systems.  358 
 Heterotrophic plate count bacteria in biofilm and post flush drinking water fractions had 359 
significant relationships with the presence of both Mycobacteria and Legionella.  Logistic 360 
regression showed that as the number of HPC bacteria increase, the odds of encountering 361 
Mycobacteria or Legionella increase as well.  The presence of Mycobacteria in the system had a 362 
stronger relationship with HPC bacteria in post flush drinking water fractions (odds ratio 1.68, P 363 
= 0.044) than in biofilm fractions (odds ratio 0.63, P = 0.010).  The data showed a relationship 364 
18 
 
between Mycobacteria identified in different fractions (biofilm and drinking water) and the 365 
number of HPC bacteria in the different fractions.  The most significant relationship was the 366 
interaction between the presence of Mycobacteria in the drinking water and elevated HPC 367 
bacteria in post flush water (odds ratio 2.05, P = 0.030).  September et al. (72) found that water 368 
quality parameters do not provide any indication of the possible presence of Mycobacteria in 369 
drinking water biofilms, while another group found a relationship between elevated HPC counts 370 
and Mycobacteria in surface waters (39).  The relationship between Mycobacteria and HPC in 371 
drinking water systems remains unclear and more research is required to elucidate all of the 372 
factors involved.  The presence of Legionella in the system had a stronger relationship with HPC 373 
bacteria in post flush drinking water (odds ratio 2.75, P = 0.003) than in biofilm samples (odds 374 
ratio 0.77, P = 0.185).  The relationship between Legionella identified in biofilm and drinking 375 
water and the number of HPC bacteria in the corresponding fractions was analyzed.  Although 376 
the minority of Legionella sequences were found in biofilm samples (41.6%), they accounted for 377 
the significant interaction with elevated HPC bacteria in post flush water. Finding elevated levels 378 
of HPC bacteria in post flush samples significantly increased the odds of encountering 379 
Legionella in a biofilm (odds ratio 31.66, P = 0.001).  There is very little data regarding the 380 
usefulness of HPC counts for predicting the presence of Legionella. It has been shown that 381 
certain common HPC bacteria inhibit the growth of Legionella while others stimulate it (79).  382 
Our data is in agreement with LeChevallier et al. (47) who concluded that HPC bacteria were 383 
useful for predicting the presence of opportunistic pathogens and provide insight into the overall 384 
quality of drinking water.  There was no relationship between HPC bacteria and the presence of 385 
Helicobacter in any fraction.   386 
19 
 
Sampling Strategy Influences Detection of Mycobacteria, Legionella, and Helicobacter. This 387 
research has attempted to identify the presence of potential pathogens and identify factors that 388 
may play a role in where and when these organisms may be present.  Because the residents on 389 
the Crow Reservation were concerned with overall drinking water quality, samples of drinking 390 
water and associated biofilms were taken at public buildings and private residences.  This 391 
resulted in samples being collected from treated municipal and untreated groundwater systems.  392 
In this study, biofilm samples were collected in addition to drinking water samples according to 393 
the recommendations for Legionella (VAMC; Pittsburgh, Pa., CDC; Atlanta, Ga.).  Although this 394 
recommendation specifically addresses Legionella detection, it is in agreement with many other 395 
findings that Mycobacteria, Legionella, and Helicobacter can be harbored and detected in 396 
drinking water biofilms (35, 69, 74).  Finally, two methods for detecting the organisms of 397 
interest, PCR and culturing, were chosen.  It is well documented that traditional culturing 398 
techniques underestimate the quantity and diversity of microorganisms in the environment (60).  399 
However, when looking at issues of public health it is also important to identify whether these 400 
organisms are viable and perhaps capable of infection.  By combining molecular detection with 401 
traditional culturing methods, it is possible to increase the likelihood of detecting an organism of 402 
interest.   403 
Mycobacteria, Legionella, and Helicobacter were found in both treated municipal water 404 
and untreated well water systems.  Mycobacteria were found more often in groundwater systems 405 
than in treated municipal systems (61.9% and 38.1% of the 20 samples positive for 406 
Mycobacteria, respectively).  Reports of the detection of Mycobacteria in groundwater have 407 
been sporadic, but generally have shown relative frequencies from not detected to up to 68% of 408 
locations testing positive (46, 71).  Mycobacteria have been detected in treated systems with 409 
20 
 
varying results as well (22, 46, 78).  Overall, our results are consistent with other reports of 410 
Mycobacteria in treated municipal and groundwater systems.  Unlike Mycobacteria, Legionella 411 
had a statistically significant relationship with the source of the drinking water.  Legionella were 412 
found more often in treated municipal systems (66.7%) than in groundwater systems (33.3%).  In 413 
other studies, Legionella has been frequently found in municipal water systems and sporadically 414 
in groundwater (14, 20, 53, 87).  It has been shown that the presence and diversity of Legionella 415 
varies spatially in drinking water distribution systems and in groundwater (87).  It is possible that 416 
premise plumbing in buildings with light or sporadic use could promote the planktonic and/or 417 
necrotrophic growth of Legionella as described by others (51, 76).  It is also likely that the 418 
overall warmer temperature of the treated municipal system is supportive for Legionella survival 419 
and growth.  Helicobacter were detected in 4 locations of our study area with 50% in treated 420 
municipal systems and 50% in untreated groundwater systems.  One of the instances of 421 
Helicobacter occurred during the flood event that closed the water treatment facility for a short 422 
period of time.  Reports of Helicobacter detection in drinking water have been intermittent with 423 
most reports finding infrequent positive samples (15, 40, 85). Although the environmental 424 
reservoir of Helicobacter is unknown, it is possible that water distribution systems may be 425 
vulnerable to contamination through breaks or leaks in distribution pipes.  Groundwater systems 426 
that are too shallow may be under the influence of surface water which could be contaminated by 427 
agricultural practices and inadequate sewage disposal (12).   428 
 Mycobacteria, Legionella, and Helicobacter were detected in both biofilm and drinking 429 
water samples.  Mycobacterium gilvum and M. avium complex were most frequently identified 430 
(>95% sequence similarity) and were both found in biofilms more often than drinking water.   431 
Legionella pneumophila occurred more often in biofilm samples while sequences identified as 432 
21 
 
Legionella sp. were more often identified in drinking water samples.  Interestingly, both 433 
Mycobacteria and Legionella had greater rates of culture positive tests in biofilm samples. This 434 
could indicate that tap water biofilms are protective and supportive for these organisms. H. pylori 435 
sequences occurred in two drinking water samples and two biofilm samples.  Although H. pylori 436 
were not detected in a large number of locations, biofilm sampling doubled the detection of this 437 
organism.  These data are consistent with reports of others that indicate that all three of these 438 
genera can be found in both drinking water and associated biofilm samples (50, 53). 439 
 Consistent with other reports, molecular detection of all three genera was more successful 440 
than traditional culture methods (3, 49, 73, 85).  The majority of detections were achieved by PCR, 441 
with only a small fraction of the samples positive for Mycobacteria and Legionella culture isolates.  442 
While it is known that molecular techniques are important for detecting organisms that are injured, 443 
or viable but not culturable (1, 7), culture techniques provide valuable information as well.  In this 444 
study, both Mycobacteria and Legionella were detected by culture methods and PCR.  However, in a 445 
minority of cases, culture positive locations could not be identified by PCR performed directly on 446 
the samples.  This has been documented by others as well and could be due to PCR inhibitors, such 447 
as heavy metals, intrinsic to the drinking water system (31, 66).   448 
Health Consequences of Mycobacteria, Legionella, and Helicobacter in Drinking Water.  All of 449 
the Mycobacteria sequences detected in this study were of the nontuberculous Mycobacteria group 450 
(NTM).  One important detected NTM are the slow-growing opportunistic pathogens in the M. 451 
avium complex (MAC), which includes the M. avium subsp. avium, M. avium subsp. intracellulare, 452 
M. avium subsp. hominisuis, and M. avium subsp. paratuberculosis (64).  MAC accounts for over 70 453 
percent of nontuberculous mycobacterial disease in the United States and for more than 95 percent 454 
of nontuberculous disease among persons infected with human immunodeficiency virus (HIV) (64).  455 
22 
 
It has been shown that MAC isolates recovered from hospital water had a close relationship (large-456 
restriction-fragment pattern analysis) with clinical isolates recovered from patients indicating that 457 
water could be the reservoir for infection (6).  Other sequences identified in this study were of >95% 458 
similarity to M. gordonae, M. flavescens, and M. mucogenicum. These species are all known to be 459 
inhabitants of drinking water systems and have been implicated in adverse health consequences (36, 460 
46, 54, 55).  One interesting fast-growing Mycobacterium species that we encountered fairly 461 
frequently was M. gilvum.  This bacterium is an environmental mycobacterium that has been isolated 462 
from soils in Montana (57), which is known to degrade polyaromatic hydrocarbons and has not been 463 
implicated in any health effects.   464 
 The Legionella species sequences detected in this study were mainly L. pneumophila and 465 
Legionella sp. but also included sequences similar to L. fairfieldensis, L. dresdeniensis, and L. 466 
birminghamiensis.  L. pneumophila is a well-documented opportunistic pathogen that has a low 467 
infection rate (1-6%), but a mortality rate of 10-15%, and accounts for 1-4% of all pneumonia cases 468 
in the United States’ general population (62).  L.  pneumophila is the primary disease causing species 469 
of this genus but there have been occasional cases of disease caused by other Legionella sp. (58).  470 
Other Legionella species such as L. fairfieldensis and L. birminghamiensis have been documented in 471 
drinking water systems (26), but are not implicated in health effects.   472 
This study detected H. pylori sequences at four of the locations sampled.  Recent research 473 
using the 13C urea breath test has shown that the prevalence of H. pylori in one rural community of 474 
Montana is greater than 50% (56).  Their research also indicated that the presence of H. pylori 475 
infection was associated with regular consumption of city water as indicated by questionnaire results 476 
(Unpublished data, USEPA) (56).  Untreated well water has also been implicated in clinical 477 
23 
 
infections in the United States (8, 10).  Areas with poor water quality may be more likely to have 478 
higher rates of water-borne transmission of disease, especially in children (15, 85).   479 
In conclusion, microbes such as M. avium, L. pneumophila, and H. pylori can be found in 480 
drinking water systems in rural underserved areas. Coliforms were shown to be inadequate indicators 481 
for all of these organisms, while HPC bacterial levels did have a relationship with the presence of 482 
Mycobacteria and Legionella. Both treated municipal water and groundwater fed systems can harbor 483 
these organisms, which can be found in both bulk water and associated biofilms. Consequently, it is 484 
impossible to rule out drinking water as a route of infection for pathogenic bacteria such as M. 485 
avium, L. pneumophila, and H. pylori.  To address health disparities in underserved communities 486 
such as American Indian reservations it is important to determine potential reservoirs of infection.  487 
These results are pertinent to water utility managers, regulatory agencies, as well as epidemiologists 488 
interested in identifying disease causing agents in rural drinking water systems. 489 
 490 
ACKNOWLEDGEMENTS 491 
The authors would like to thank the entire Crow Environmental Health Steering 492 
Committee, Ada Bends, Urban Bear Don't Walk, John Doyle, Brandon Goodluck, Vernon Hill, 493 
Larry Kindness, Myra Lefthand, Henry Pretty On Top, Ronald Stewart, and Sara Young, for 494 
providing insight into research in reservation communities and community support for this work. 495 
Thanks to Crow community coordinators, Crescentia Cummins and Gail Whiteman, without 496 
whom sample collection would not have been possible. Additional thanks to Al Parker for 497 
statistical support. 498 
This work was supported by the National Institute for Health, Center for Native Health 499 
Partnerships, (M201-10-W2724).  Dr. Ford is supported in part by a grant from the U.S. 500 
24 
 
Environmental Protection Agency’s Science to Achieve Results (US-EPA STAR) program and 501 
Crystal Richards is supported by a US-EPA STAR Fellowship.  Although the research described 502 
in the article has been funded in part by the U.S. Environmental Protection Agency's STAR 503 
program through grant numbers RD833706 and FP916936, it has not been subjected to any EPA 504 
review and therefore does not necessarily reflect the views of the Agency, and no official 505 
endorsement should be inferred.  506 
 507 
REFERENCES 508 
1. Adams, B. L., T. C. Bates, and J. D. Oliver. 2003. Survival of Helicobacter pylori in a 509 
natural freshwater environment. Appl. Environ. Microbiol. 69:7462-7466. 510 
2. Allen, M. J., S. C. Edberg, and D. J. Reasoner. 2004. Heterotrophic plate count 511 
bacteria - what is their significance in drinking water? Int. J. Food Microbiol. 92:265-512 
274. 513 
3. Angenent, L. T., S. T. Kelley, A. St Amand, N. R. Pace, and M. T. Hernandez. 2005. 514 
Molecular identification of potential pathogens in water and air of a hospital therapy 515 
pool. Proc. Natl. Acad. Sci. U.S.A. 102:4860-4865. 516 
4. Anonymous. 2002. Crow Indian Reservation Natural, Socio-Economic, and Cultural 517 
Resources Assessment and Conditions Report, Crow Tribe of Indians, Montana. 518 
5. Anonymous. 1999. The Health Care Challenge: Acknowledging Disparity, Confronting 519 
Discrimintaion, and Ensuring Equality. United States National Archives and Records 520 
Administration, Washington D.C. 20402. 521 
6. Aronson, T., A. Holtzman, N. Glover, M. Boian, S. Froman, O. G. W. Berlin, H. Hill, 522 
and G. Stelma. 1999. Comparison of large restriction fragments of Mycobacterium 523 
25 
 
avium isolates recovered from AIDS and non-AIDS patients with those of isolates from 524 
potable water. J. Clin. Microbiol. 37:1008-1012. 525 
7. Atlas, R. M. 1999. Legionella: from environmental habitats to disease pathology, 526 
detection and control. Environ. Microbiol. 1:283-293. 527 
8. Baker, K. H., and J. P. Hegarty. 2001. Presence of Helicobacter pylori in drinking 528 
water is associated with clinical infection. Scand. J. Infect. Dis. 33:744-746. 529 
9. Bartie, C., S. N. Venter, and L. H. Nel. 2003. Identification methods for Legionella 530 
from environmental samples. Water Res. 37:1362-1370. 531 
10. Bellack, N. R., M. W. Koehoorn, Y. C. MacNab, and M. G. Morshed. 2006. A 532 
conceptual model of water's role as a reservoir in Helicobacter pylori transmission: a 533 
review of the evidence. Epidemiol. Infect. 134:439-449. 534 
11. Benjamini, Y., and Y. Hochberg. 1995. Controlling the false discovery rate - a practical 535 
and powerful approach to multiple testing. J. R. Stat. Soc. Series B Stat. Methodol.  536 
57:289-300. 537 
12. Bitton, G., and C. P. Gerba. 1984. Groundwater Pollution Microbiology: the Emerging 538 
Issue, p. 1-7. In G. Bitton and C. Gerba (ed.), Groundwater pollution microbiology. John 539 
Wiley & Sons, New York. 540 
13. Boddinghaus, B., T. Rogall, T. Flohr, H. Blocker, and E. C. Bottger. 1990. Detection 541 
and Identification of Mycobacteria by Amplification of RNA. J. Clin. Microbiol. 542 
28:1751-1759. 543 
14. Brooks, T., R. A. Osicki, V. S. Springthorpe, S. A. Sattar, L. Filion, D. Abrial, and S. 544 
Riffard. 2004. Detection and identification of Legionella species from groundwaters. J. 545 
Toxicol. Environ. Health. 67:1845-1859. 546 
26 
 
15. Bunn, J. E. G., W. G. MacKay, J. E. Thomas, D. C. Reid, and L. T. Weaver. 2002. 547 
Detection of Helicobacter pylori DNA in drinking water biofilms: implications for 548 
transmission in early life. Lett. Appl. Microbiol. 34:450-454. 549 
16. Burkland, B. 2010. Personal Communication regarding treatment facility at Crow 550 
Agency, US Environmental Protection Agency. Helena, MT. 551 
17. Burlingame, G. A., J. McElhaney, M. Bennett, and W. O. Pipes. 1984. Bacterial 552 
interference with coliform colony sheen production on membrane filters. Appl. Environ. 553 
Microbiol. 47:56-60. 554 
18. Cassidy, P. M., K. Hedberg, A. Saulson, E. McNelly, and K. L. Winthrop. 2009. 555 
Nontuberculous Mycobacterial Disease Prevalence and Risk Factors: A Changing 556 
Epidemiology. Clin. Infect. Dis. 49:E124-E129. 557 
19. Conover, W. J. 1999. Practical nonparametric statistics, Third ed. Elm Street Publishing 558 
Services, Inc., New York. 559 
20. Costa, J., I. Tiago, M. S. da Costa, and A. Verissimo. 2005. Presence and persistence 560 
of Legionella spp. in groundwater. Appl. Environ. Microbiol. 71:663-671. 561 
21. Covers Up, T., R. Turnsplenty, D. Wetzel, and J. Giroux. 2005. 1990-2002 Montana 562 
and Wyoming American Indian Top Ten Causes of Death. Rocky Mountain Tribal 563 
Epidemiology Center. 564 
22. Covert, T. C., M. R. Rodgers, A. L. Reyes, and G. N. Stelma. 1999. Occurrence of 565 
nontuberculous mycobacteria in environmental samples. Appl. Environ. Microbiol.  566 
65:2492-2496. 567 
23. Craun, G. F., R. L. Calderon, and T. J. Wade. 2006. Assessing waterborne risks: an 568 
introduction. J. Water Health 04. Suppl 2:3-18. 569 
27 
 
24. Craun, M. F., G. F. Craun, R. L. Calderon, and M. J. Beach. 2006. Waterborne 570 
outbreaks reported in the United States. J. Water Health 04. Suppl 2:19-30. 571 
25. Degnan, A. J., W. C. Sonzogni, and J. H. Standridge. 2003. Development of a plating 572 
medium for selection of Helicobacter pylori from water samples. Appl. Environ. 573 
Microbiol.  69:2914-2918. 574 
26. Diederen, B. M. W., C. M. A. de Jong, I. Aarts, M. F. Peeters, and A. van der Zee. 575 
2007. Molecular evidence for the ubiquitous presence of Legionella species in Dutch tap 576 
water installations. J. Water Health 5:375-383. 577 
27. Doyle, J. 2010. Personal communication regarding municipal distribution system at 578 
Crow Agency, Crow Environmental Health Steering Committee. Crow Agency, MT. 579 
28. Doyle, J., L. Kindness, U. Bear Don't Walk, S. Young, and M. Lefthand. 2006. 580 
Personal communication regarding water quality on Crow Reservation, Crow 581 
Environmental Health Steering Committee. Crow Agency, MT. 582 
29. Falkinham, J. O., C. D. Norton, and M. W. LeChevallier. 2001. Factors influencing 583 
numbers of Mycobacterium avium, Mycobacterium intracellulare, and other 584 
mycobacteria in drinking water distribution systems. Appl. Environ. Microbiol.  67:1225-585 
1231. 586 
30. Fields, B. S., R. F. Benson, and R. E. Besser. 2002. Legionella and Legionnaires' 587 
disease: 25 years of investigation. Clin. Microbiol. Rev. 15:506-526. 588 
31. Fiume, L., M. A. Bucca Sabattini, and G. Poda. 2005. Detection of Legionella 589 
pneumophila in water samples by species-specific real-time and nested PCR assays. Lett. 590 
Appl. Microbiol. 41:470-475. 591 
28 
 
32. Franzblau, S. G., B. J. Hinnebusch, L. M. Kelley, and N. A. Sinclair. 1984. Effect on 592 
noncoliforms on coliform detection in potable groundwater - improved recovery with an 593 
anaerobic membrane-filter technique. Appl. Environ. Microbiol. 48:142-148. 594 
33. Geach, J. 2007. Source Water Delineation and Assessment Report, p. 1-34. United States 595 
Environmental Protection Agency, Helena, MT. 596 
34. Germani, Y., C. Dauga, P. Duval, M. Huerre, M. Levy, G. Pialoux, P. Sansonetti, 597 
and P. A. D. Grimont. 1997. Strategy for the detection of Helicobacter species by 598 
amplification of 16S rRNA genes and identification of H. felis in a human gastric biopsy. 599 
Res. Microbiol. 148:315-326. 600 
35. Giao, M. S., N. F. Azevedo, S. A. Wilks, M. J. Vieira, and C. W. Keevil. 2008. 601 
Persistence of Helicobacter pylori in heterotrophic drinking-water biofilms. Appl. 602 
Environ. Microbiol.  74:5898-5904. 603 
36. Guillen, S. M., J. S. Hospital, E. G. Mampaso, A. G. Espejo, C. E. Baquedano, and 604 
A. O. Calderon. 1986. Gluteal abscess caused by Mycobacterium flavescens. Tubercle 605 
67:151-153. 606 
37. Hallberg, G. R., and B. C. Kross. 1990. Iowa Statewide Rural Well Water Survey - 607 
Summary of Results. Iowa Department of Natural Resources Geologocal Survey Bureau 608 
and University of Iowa Center for Health Effects of Environmental Contamination, Iowa 609 
City, IA 52242. 610 
38. Heath, R. C. 1984. Ground-Water Regions of the United States, p. 1-78. U.S. Geological 611 
Survey Supply Paper 2242. 612 
29 
 
39. Iivanainen, E. K., P. J. Martikainen, P. K. Vaananen, and M. L. Katila. 1993. 613 
Environmental factors affecting the occurrence of mycobacteria in brook waters. Appl. 614 
Environ. Microbiol.  59:398-404. 615 
40. Janzon, A., A. Sjoling, A. Lothigius, D. Ahmed, F. Qadri, and A. M. Svennerholm. 616 
2009. Failure To Detect Helicobacter pylori DNA in Drinking and Environmental Water 617 
in Dhaka, Bangladesh, Using Highly Sensitive Real-Time PCR Assays. Appl. Environ. 618 
Microbiol.  75:3039-3044. 619 
41. Jukes, T. H., and C. R. Cantor. 1969. Evolution of protein molecules, p. 21-132. In H. 620 
N. Munro (ed.), Mammalian protein metabolism. Academic Press, New York. 621 
42. Kramer, M. H., and T. E. Ford. 1994. Legionellosis: ecological factors of an 622 
environmentally 'new' disease. Zentralbl Hyg Umweltmed 195:470-82. 623 
43. Kramer, M. H., B. L. Herwaldt, R. L. Calderon, and D. D. Juranek. 1996. 624 
Surveillance for Waterborne-Disease Outbreaks -- United States, 1993-1994, p. 1-33, vol. 625 
45. Centers for Disease Control, Washington D.C. 626 
44. Kutner, M., C. Nachtsheim, J. Neter, and W. Li. 2004. Applied linear statistical 627 
models, Fifth ed. McGraw-Hill/Irwin, New York. 628 
45. Lamka, K. G., M. W. LeChevallier, and R. J. Seidler. 1980. Bacterial contamination 629 
of drinking water supplies in a modern rural neighborhood. Appl. Environ. Microbiol.  630 
39:734-738. 631 
46. Le Dantec, C., J. P. Duguet, A. Montiel, N. Dumoutier, S. Dubrou, and V. Vincent. 632 
2002. Occurrence of mycobacteria in water treatment lines and in water distribution 633 
systems. Appl. Environ. Microbiol. 68:5318-5325. 634 
30 
 
47. LeChevallier, M. W., R. J. Seidler, and T. M. Evans. 1980. Enumeration and 635 
characterization of standard plate-count bacteria in chlorinated and raw water supplies. 636 
Appl. Environ. Microbiol. 40:922-930. 637 
48. Leclerc, H., D. A. A. Mossel, S. C. Edberg, and C. B. Struijk. 2001. Advances in the 638 
bacteriology of the coliform group: their suitability as markers of microbial water safety. 639 
Annu. Rev. Microbiol. 55:201-234. 640 
49. Lehtola, M. J., E. Torvinen, J. Kusnetsov, T. Pitkanen, L. Maunula, C. H. von 641 
Bonsdorff, P. J. Martikainen, S. A. Wilks, C. W. Keevil, and I. T. Miettinen. 2007. 642 
Survival of Mycobacterium avium, Legionella pneumophila, Escherichia coli, and 643 
caliciviruses in drinking water-associated biofilms grown under high-shear turbulent 644 
flow. Appl. Environ. Microbiol. 73:2854-2859. 645 
50. Mackay, W. G., L. T. Gribbon, M. R. Barer, and D. C. Reid. 1998. Biofilms in 646 
drinking water systems - A possible reservoir for Helicobacter pylori. Water Sci. 647 
Technol. 38:181-185. 648 
51. Mampel, J., T. Spirig, S. S. Weber, J. A. J. Haagensen, S. Molin, and H. Hilbi. 2006. 649 
Planktonic replication is essential for biofilm formation by Legionella pneumophila in a 650 
complex medium under static and dynamic flow conditions. Appl. Environ. Microbiol. 651 
72:2885-2895. 652 
52. Manuel, C. M., O. C. Nunes, and L. F. Melo. 2010. Unsteady state flow and stagnation 653 
in distribution systems affect the biological stability of drinking water. Biofouling 654 
26:129-139. 655 
31 
 
53. Marciano-Cabral, F., M. Jamerson, and E. S. Kaneshiro. 2010. Free-living amoebae, 656 
Legionella and Mycobacterium in tap water supplied by a municipal drinking water 657 
facility. J. Water Health 08:71-82. 658 
54. Marshall, C., J. Samuel, A. Galloway, and S. Pedler. 2008. Mycobacterium 659 
mucogenicum from the Hickman line of an immunocompromised patient. J. Clin. Pathol. 660 
61:140-141. 661 
55. Mazumder, S. A., A. Hicks, and J. Norwood. 2010. Mycobacterium gordonae 662 
pulmonary infection in an immunoocompetent adult. N. A. J. Med. Sci. 2:205-207. 663 
56. Melius, E., R. Wierzba, S. Davis, J. Sobel, B. Gold, A. Henderson, and J. Cheek. 664 
2005. Risk factors for Helicobacter pylori in a rural community.  665 
57. Miller, C. D., R. Child, J. E. Hughes, M. Benscai, J. P. Der, R. C. Sims, and A. J. 666 
Anderson. 2007. Diversity of soil mycobacterium isolates from three sites that degrade 667 
polycyclic aromatic hydrocarbons. J. Appl. Microbiol. 102:1612-1624. 668 
58. Muder, R. R., and V. L. Yu. 2002. Infection due to Legionella species other than L. 669 
pneumophila. Clin. Infect. Dis. 35:990-998. 670 
59. National Center for  Health Statistics. 2010. Health, Unites States, 2009: With Special 671 
Feature on Medical Technology., Hyattsville, MD. 672 
60. Pace, N. R. 1997. A molecular view of microbial diversity and the biosphere. Science 673 
276:734-740. 674 
61. Palmer, C. J., G. F. Bonilla, B. Roll, C. Paszkokolva, L. R. Sangermano, and R. S. 675 
Fujioka. 1995. Detection of Legionella species in reclaimed water and air with the 676 
ENVIROAMP Legionella PCR kit and direct fluorescent-antibody staining. Appl. 677 
Environ. Microbiol. 61:407-412. 678 
32 
 
62. Percival, S., R. Chalmers, M. Embrey, P. Hunter, J. Sellwood, and P. Wyn-Jones. 679 
2004. Microbiology of Waterborne Diseases. Elsevier Academic Press, California. 680 
63. Perez-Perez, G. I., D. Rothenbacher, and H. Brenner. 2004. Epidemiology of 681 
Helicobacter pylori infection. Helicobacter 9:1-6. 682 
64. Reed, C., C. F. von Reyn, S. Chamblee, T. V. Ellerbrock, J. W. Johnson, B. J. 683 
Marsh, L. S. Johnson, R. J. Trenschel, and C. R. Horsburgh. 2006. Environmental 684 
risk factors for infection with Mycobacterium avium complex. Am. J. Epidemiol. 164:32-685 
40. 686 
65. Research and Analysis Bureau. 2008. Demographic and Economic Information for the 687 
Crow Reservation, p. 1-8. Montana Department of Labor and Industry, Helena, MT 688 
59624. 689 
66. Riffard, S., S. Douglass, T. Brooks, S. Springthorpe, L. G. Filion, and S. A. Sattar. 690 
2001. Occurrence of Legionella in groundwater: an ecological study. Water Sci. Technol.  691 
43:99-102. 692 
67. Saitou, N., and M. Nei. 1987. The neighbor-joining method - a new method for 693 
reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425. 694 
68. Sasaki, K., Y. Tajiri, M. Sata, Y. Fujii, F. Matsubara, M. G. Zhao, S. Shimizu, A. 695 
Toyonaga, and K. Tanikawa. 1999. Helicobacter pylori in the natural environment. 696 
Scand. J. Infect. Dis. 31:275-280. 697 
69. Schulze-Robbecke, R., B. Janning, and R. Fischeder. 1992. Occurrence of 698 
mycobacteria in biofilm samples. Tuber. Lung Dis. 73:141-144. 699 
33 
 
70. Schulze-Robbecke, R., A. Weber, and R. Fischeder. 1991. Comparison of 700 
decontamination methods for the isolation of mycobacteria from drinking water samples. 701 
J. Microbiol. Methods 14:177-183. 702 
71. Schwartz, T., S. Kalmbach, S. Hoffmann, U. Szewzyk, and U. Obst. 1998. PCR-based 703 
detection of mycobacteria in biofilms from a drinking water distribution system. J. 704 
Microbiol. Methods 34:113-123. 705 
72. September, S. M., V. S. Brozel, and S. N. Venter. 2004. Diversity of nontuberculoid 706 
Mycobacterium species in biofilms of urban and semiurban drinking water distribution 707 
systems. Appl. Environ. Microbiol. 70:7571-7573. 708 
73. Springer, B., L. Stockman, K. Teschner, G. D. Roberts, and E. C. Bottger. 1996. 709 
Two-laboratory collaborative study on identification of mycobacteria: Molecular versus 710 
phenotypic methods. J. Clin. Microbiol. 34:296-303. 711 
74. Storey, M. V., J. Langmark, N. J. Ashbolt, and T. A. Stenstrom. 2004. The fate of 712 
legionellae within distribution pipe biofilms: measurement of their persistence, 713 
inactivation and detachment. Water Sci. Technol. 49:269-275. 714 
75. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4: Molecular evolutionary 715 
genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24:1596-1599. 716 
76. Temmerman, R., H. Vervaeren, B. Noseda, N. Boon, and W. Verstraete. 2006. 717 
Necrotrophic growth of Legionella pneumophila. Appl. Environ. Microbiol. 72:4323-718 
4328. 719 
77. Thompson, J. D., T. J. Gibson, and D. G. Higgins. 2002. Multiple sequence alignment 720 
using ClustalW and ClustalX. Curr Protoc Bioinformatics Chapter 2:Unit 2.3. 721 
34 
 
78. Torvinen, E., S. Suomalainen, M. J. Lehtola, I. T. Miettinen, O. Zacheus, L. Paulin, 722 
M. L. Katila, and P. J. Martikainen. 2004. Mycobacteria in water and loose deposits of 723 
drinking water distribution systems in Finland. Appl. Environ. Microbiol. 70:1973-1981. 724 
79. Toze, S., L. I. Sly, I. C. Macrae, and J. A. Fuerst. 1990. Inhibition of growth of 725 
Legionella species by heterotrophic plate-count bacteria isolated from chlorinated 726 
drinking water. Curr. Microbiol. 21:139-143. 727 
80. Turenne, C. Y., M. Semret, D. V. Cousins, D. M. Collins, and M. A. Behr. 2006. 728 
Sequencing of hsp65 distinguishes among subsets of the Mycobacterium avium complex. 729 
J. Clin. Microbiol. 44:433-440. 730 
81. Turetgen, I., E. I. Sungur, and A. Cotuk. 2005. Enumeration of Legionella 731 
pneumophila in cooling tower water systems. Environ. Monit. Assess. 100:53-58. 732 
82. United States Census Bureau. 2007. Current Housing Reports, Series H150/07, p. 1-733 
642, American Housing Survey for the United States: 2007. U.S. Government Printing 734 
Office, Washington D.C. 735 
83. United States Environmental Protection Agency. 1998. Guidance for data quality 736 
assessment-Practical methods for data analysis. Office of Research and Development, 737 
Washington, D.C. 738 
84. Voytek, M. A., J. B. Ashen, L. R. Fogerty, J. D. Kirshtein, and E. R. Landa. 2005. 739 
Detection of Helicobacter pylori and fecal indicator bacteria in five North American 740 
rivers. J. Water Health 03:405-422. 741 
85. Watson, C. L., R. J. Owen, B. Said, S. Lai, J. V. Lee, S. Surman-Lee, and G. Nichols. 742 
2004. Detection of Helicobacter pylori by PCR but not culture in water and biofilm 743 
35 
 
samples from drinking water distribution systems in England. J. Appl. Microbiol.  744 
97:690-698. 745 
86. Whan, L., H. J. Ball, I. R. Grant, and M. T. Rowe. 2005. Occurrence of 746 
Mycobacterium avium subsp paratuberculosis in untreated water in Northern Ireland. 747 
Appl. Environ. Microbiol. 71:7107-7112. 748 
87. Wullings, B. A., and D. van der Kooij. 2006. Occurrence and genetic diversity of 749 
uncultured Legionella spp. in drinking water treated at temperatures below 15 degrees C. 750 
Appl. Environ. Microbiol. 72:157-166. 751 
36 
 
Table 1. Primer Sequences, References and PCR Conditions. 
Target 
(Reference) 
Sequence 
Product 
size 
PCR Conditions 
16S RNA gene 
Legionella spp. 
(87) 
LEG-225 5’AAGATTAGCCTGCG 
TCCGAT;  LEG-858 5’GTCAACT 
TATCGCGTTTGCT 
 
 
656 bp 
94˚C 2 min (1 cycle); 94˚C 
20 sec, 60˚C 30 sec, 72˚C 
40 sec (40 cycles); 72˚C 5 
min (1 cycle) 
 
16S RNA gene 
Mycobacterium 
spp. (13) 
MycgenF 5’ AGAGTTTGATCCT 
GGCTCAG;  MycgenR 5’ TGCAC 
ACAGGCCACAAGGGA 
 
 
1,030 bp 
95˚C 2 min (1 cycle); 93˚C 
1 min, 60˚C 1 min, 72˚C 1 
min (35 cycles); 72˚C 5 
min (1 cycle) 
 
16S RNA gene 
Helicobacter 
spp. (34) 
HS1 5’ AACGATGAAGCTTCT 
AGCTTGCTAG;  HS2 5’ GTGCT 
TATTCGTTAGATACCGTCAT 
 
 
400 bp 
94˚C 5 min (1 cycle); 94˚C 
1 min, 65˚C 1 min, 72˚C 1 
min (35 cycles); 72˚C 5 
min (1 cycle) 
 
16S RNA gene 
Eubacteria (84) 
46f 5' GCYTAACACATGCA 
AGTCGA;  519r 5' GTATTACCG 
CGGCKGCTG 
 
490 bp 
95˚C 5 min (1 cycle); 94˚C 
0.5 min, 56˚C 0.5 min, 
72˚C 1.5 min (30 cycles); 
72˚C 7 min (1 cycle) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37 
 
 
Table 2. Range and Arithmetic Mean of HPC bacteria, total coliforms, and E. coli 
Bacteria Measurement 
Source 
Treated Municipal 
(n=16) 
Groundwater Well 
(n=41) 
Heterotrophic plate counts    
In first flush (CFU/ml) Range * 1.5 x 100 - 5.12 x 107 
 Arithmetic Mean * 2.81 x 106 
In water (CFU/ml) Range 3.57 x 102 - 5.15 x 105 2.0 x 100 - 9.23 x 105 
 Arithmetic Mean 9.02 x 104 5.7 x 104 
In biofilm (CFU/mm2) Range <1 - 1.24 x 105 <1 - 3.22 x 105 
 Arithmetic Mean 1.79 x 104 4.29 x 104 
Total Coliforms    
In first flush (CFU/100ml) Range * <1 - 1.19 x 103 
 Arithmetic Mean * 1.17 x 102 
In water (CFU/100ml) Range <1 - 2.63 x 101 <1 - 2.96 x 103 
 Arithmetic Mean 2.71 x 100 1.07 x 102 
Escherichia coli    
In first flush (CFU/100ml) Range * <1 
 Arithmetic Mean * <1 
In water (CFU/100ml) Range <1 <1 - 2.22 x 102 
 Arithmetic Mean <1 5.68 x 100 
* No samples were taken in this category. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
38 
 
Table 3. Statistical Analysis of the Interactions between HPC Bacteria in Drinking Water and 
Response Variables 
    Significance and FDR 
Response Variable Model P P(12) 
Helicobacter (BF) Binary Logistic Regression 0.95 0.100 
Mycobacteria (BF) Binary Logistic Regression 0.778 0.092 
Helicobacter (System) Binary Logistic Regression 0.628 0.083 
Total coliforms (DW) Simple Linear Regression 0.463 0.075 
Helicobacter (DW) Binary Logistic Regression 0.457 0.067 
Legionella (DW) Binary Logistic Regression 0.068 0.058 
Mycobacteria (System) Binary Logistic Regression 0.044* 0.050 
Mycobacteria (DW) Binary Logistic Regression 0.03* 0.042 
Escherichia coli (DW) Simple Linear Regression 0.026* 0.033 
HPC bacteria (FF) Paired t-test 0.025* 0.023 
Legionella (System) Binary Logistic Regression 0.003* 0.017 
Legionella (BF) Binary Logistic Regression 0.001* 0.008 
* Denotes P-value with statistical significance.  
(FDR) false discovery rate, (FF) first flush, (BF) biofilm, (DW) drinking water, (System) 
combines all sample fractions. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39 
 
Table 4. Statistical Analysis of the Interactions between HPC Bacteria in Biofilms and Response 
Variables 
    Significance and FDR 
Response Variable Model P P(10) 
Helicobacter (BF) Binary Logistic Regression 0.872 0.100 
Helicobacter (DW) Binary Logistic Regression 0.746 0.082 
Helicobacter (System) Binary Logistic Regression 0.726 0.073 
Total coliforms (DW) Simple Linear Regression 0.606 0.064 
Mycobacteria (DW) Binary Logistic Regression 0.509 0.055 
Legionella (BF) Binary Logistic Regression 0.204 0.045 
Legionella (DW) Binary Logistic Regression 0.208 0.036 
Legionella (System) Binary Logistic Regression 0.185 0.027 
Mycobacteria (BF) Binary Logistic Regression 0.051 0.018 
Mycobacteria (System) Binary Logistic Regression 0.01 0.009 
(FDR) false discovery rate, (BF) biofilm, (DW) drinking water, (System) combines all sample 
fractions. 
 
 
 
 
 
 
 
Table 5. Statistical Analysis of the Interactions between Drinking Water Source (Treated 
Municipal or Groundwater Well) and Response Variables 
    Significance and FDR 
Variable Model P P(7) 
Escherichia coli (DW) Two sample t-test 0.418 0.100 
Helicobacter (System) Fisher's exact test 0.393 0.086 
Mycobacteria (System) Fisher's exact test 0.216 0.071 
HPC bacteria (DW) Two sample t-test 0.13 0.057 
Total coliforms (DW) Two sample t-test 0.128 0.043 
HPC bacteria (BF) Two sample t-test 0.049 0.029 
Legionella (System) Fisher's exact test 0.003* 0.014 
* Denotes P-value with statistical significance.  
(FDR) false discovery rate, (BF) biofilm, (DW) drinking water, (System) combines all sample 
fractions. 
40 
 
 BFc02.4(m) 
 Mycobacterium sp. (AF4805801) 
 DWp03.2 (g) 
 BFc02.3(m) 
 BFp02.1(m) 
 Mycobacterium sp. (GQ924943) 
 Mycobacterium sp. (FJ655010) 
 BFp01.1(m) 
 DWc42.1(g) 
 BFc11.1(m) 
 BFc13.1(g) 
 BFc31.1(g) 
 DWp38.1(g) 
 BFp11.1(m) 
 BFp13.1(g) 
 M. gilvum (CP000656) 
 M. fortuitum (AF480580) 
 M. septicum (AY457070) 
 BFc02.1(m) 
 M. mucogenicum (AM884316) 
 M. flavescens (AF480578) 
 BFc04.1(g) 
 M. vaccae (AF544639) 
 BFc05.1(g) 
 BFc02.2(m) 
 M. tokaiense (NR 025236) 
 M. murale (AM056053) 
 BFp36.1(m) 
 BFp36.2(m) 
 DWp36.1(m) 
 DWc36.1(m) 
 DWp21.1(m) 
 M. gordonae (GQ924935) 
 M. chimaera (GQ153272) 
 M. intracellulare (GQ153276) 
 BFc50.1(g) 
 M. avium (CP000479) 
 DWp13.1(g) 
 M. avium sp. paratuberculosis (EF521896) 
 BFp22.1(m) 
 DWp17.1(g) 
 BFp17.1(g) 
 M. avium (GQ153272) 
 DWp42.1(g) 
 BFp20.1(m) 
 DWp45.1(g) 
 M. avium sp. silvaticum (EF521891) 
 BFp07.1(g) 
 M. avium sp. hominissuis (EF521892) 
 BFc28.1(g) 
 M. manitobense (AY082001) 
 BFp12.1(m) 
 Corynebacterium glutamicum (NC 003450) 
100 
100 
92 
95 
100 
88 
61 
64 
62 
52 
73 
96 
95 
67 
62 
54 
96 
52 
98 
0.05 
41 
 
Figure 1. Phylogenetic relationship of 16s rRNA gene amplified with Mycobacterium genus-
specific primers with Corynebacterium glutamicum as the out-group. Reference sequences are in 
bold with accession numbers in parentheses. Sequences from this study are indicated by code as 
follows. (DW) drinking water, (BF) biofilm, (p) PCR, (c) culture, (g) groundwater, and (m) 
municipal. 
 
 
42 
 
 
 DWp33.1(m) 
 BFp07.1(g) 
 BFc06.1(m) 
 BFc07.1(g) 
 BFc07.2(g) 
 L. pneumophila (CP000675) 
 L. pneumophila (AE017354) 
 L. pneumophila (CR628336) 
 L. pneumophila (EU054324) 
 L. pneumophila (CR628337) 
 DWp35.1(m) 
 BFp01.1(m) 
 Legionella sp. (X97361) 
 BFp02.1(m) 
 Uncultured Legionella sp. (AY924167) 
 DWc35.1(m) 
 Uncultured bacterium clone (DQ336999) 
 DWp01.1(m) 
 DWp34.1(m) 
 DWp03.1(g) 
 DWp12.1(m) 
 L. dresdeniensis (AM747393) 
 L.fairfieldensis (Z497722) 
 L.birminghamiensis (Z49717) 
 DWp21.1(m) 
 DWc12.1(m) 
 DWp12.2(m) 
 DWp19.1(g) 
 Uncultured Legionella sp. (AY924046) 
 L. anisa UCSC 23 (AJ969024) 
 L. longbeachae (AY444741) 
 BFp26.1(g) 
 L. parisiensis (U59697) 
 Coxiella burnetii (AE016828) 
74 
89 
59 
78 
97 
86 
70 
83 
93 
85 
63 
63 
77 
0.02 
43 
 
 
Figure 2. Phylogenetic relationship of 16s rRNA gene amplified with Legionella genus-specific 
primers with Coxiella burnetii as the out-group. Reference sequences are in bold with accession 
numbers in parentheses. Sequences from this study are indicated by code as follows. (DW) 
drinking water, (BF) biofilm, (p) PCR, (c) culture, (g) groundwater, and (m) municipal. 
 
 
 
 
Figure 3. Phylogenetic relationship of 16s rRNA gene amplified with Helicobacter genus-
specific primers with Campylobacter jejuni as the out-group. Reference sequences are in bold 
with accession numbers in parentheses. Sequences from this study are indicated by code as 
follows. (DW) drinking water, (BF) biofilm, (p) PCR, (c) culture, (g) groundwater, and (m) 
municipal. 
 
 
 
 
 H. pylori (FM991728) 
 DWp11.1(m) 
 H. pylori (CP001173) 
 BFp05.1(g) 
 H. pylori (EU035396) 
 DWp32.1(m) 
 H. pylori (CP001680) 
 BFp19.1(g) 
 H. nemestrinae (AF363064) 
 H. pylori (FJ788640) 
 H. pylori (EU544199) 
 H. pylori (FJ788642) 
 H. pylori (EU033951) 
 H. pylori (U01330) 
 H. heilmannii (AF506786) 
 H. mustelae (NR 029169) 
 H. canis (U04344) 
 Campylobacter jejuni (CP001876) 
95 
87 
99 
0.02