Subscriber access provided by Kaohsiung Medical University
Characterization of Natural and Affected Environments
Real-time Quantitative PCR Measurements of Fecal Indicator Bacteria and Human-associated Source Tracking Markers in a Texas River following Hurricane Harvey Vikram Kapoor, Indrani Gupta, ABM Tanvir Pasha, and Duc Phan Environ. Sci. Technol. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.estlett.8b00237 • Publication Date (Web): 10 May 2018 Downloaded from http://pubs.acs.org on May 11, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 26
Environmental Science & Technology Letters
1
Real-time Quantitative PCR Measurements of Fecal
2
Indicator Bacteria and Human-associated Source
3
Tracking Markers in a Texas River following Hurricane
4
Harvey
5
Vikram Kapoor*, Indrani Gupta, A.B.M. Tanvir Pasha, and Duc Phan
6
Department of Civil and Environmental Engineering, University of Texas at San Antonio, San
7
Antonio, TX 78249, USA
8 9 10 11
Keywords. Fecal source tracking, human-associated Bacteroidales, quantitative PCR, genetic
12
markers, water quality, Hurricane Harvey
13 14 15 16 17
1 ACS Paragon Plus Environment
Environmental Science & Technology Letters
18
ABSTRACT
19
Hurricane Harvey has caused unprecedented devastation to huge parts of southeastern Texas,
20
particularly damaging the wastewater infrastructure resulting in release of sewage contamination
21
into environmental waters. The purpose of this study was to conduct a preliminary assessment of
22
fecal indicator bacteria (E. coli and enterococci) and human-associated fecal genetic markers
23
(Human-associated Bacteroidales), measured using qPCR assays, across a Texas river impacted
24
by Hurricane Harvey. Water samples were collected along the Guadalupe River during
25
September – December 2017. The most heavily flooded sites showed highest abundance of fecal
26
indicator bacteria and human-associated Bacteroidales markers indicating that a large number of
27
sewage overflows and stormwater runoff occurred during Harvey flooding. These findings
28
suggest that high levels of human fecal contamination were introduced into waterways draining
29
into the Gulf of Mexico and impaired surface water quality. The human-associated Bacteroidales
30
markers exhibited low to slightly strong correlation with conventional fecal indicators suggesting
31
the variable occurrence of different markers and uncertainty of enterococci and E. coli for
32
detection of human fecal pollution. In general, results of this initial microbiological contaminant
33
assessment will serve as baseline information for follow-on studies to monitor existing and
34
emerging public health risks to residents of Texas and potential long-term environmental impacts
35
upon the water resources in the impacted regions.
36
INTRODUCTION
37
Hurricane Harvey originated as a tropical storm over the Atlantic Ocean, becoming a Category 4
38
hurricane on August 25, 2017 as it approached the coast of Texas.1 After a week of storms and
39
heavy flooding, Hurricane Harvey devastated huge parts of southeastern Texas, particularly
40
damaging the wastewater infrastructure resulting in release of sewage contamination into 2 ACS Paragon Plus Environment
Page 2 of 26
Page 3 of 26
Environmental Science & Technology Letters
41
environmental waters.2,3 Immediately following Harvey, both the Texas Commission on
42
Environmental Quality (TCEQ) and the U.S. Environmental Protection Agency (EPA) reported
43
the release of wastewater from sanitary sewers due to the historic flooding and more than 800
44
wastewater treatment plants reported spills.4,5 The widespread flooding resulted in sewage
45
overflows and contamination of floodwaters which eventually make their way into surface
46
waters, before finally discharging into the Gulf of Mexico. These sewage overflows contain high
47
levels of fecal bacteria and potentially pathogenic organisms posing a serious risk to human and
48
environmental health via waterborne disease outbreaks, deterioration of recreational and drinking
49
water quality, and degradation of aquatic ecology.6 Initial water samples collected from
50
floodwaters in affected areas showed E. coli levels greater than 100 times the EPA criteria for
51
recreation water quality standard,7 strongly suggesting mixture of sewage in water.2,8 Water
52
quality impacts have been reported for previous hurricanes but have primarily been limited to
53
chemical contaminants, and transient increases in nutrient loading.9-12 Moreover, these studies
54
have focused on contamination of localized water bodies such as the contamination of Lake
55
Pontchartrain (Louisiana) during Hurricane Katrina12,13 and limited data is available on the
56
distribution of microbial contaminants in impacted rivers and streams.
57
Hurricanes and large storms play a significant role in the transport of water contaminants across
58
environmental waters.10,11 Among these contaminants, the mobilization and distribution of fecal
59
indicator bacteria and human-associated fecal markers due to sewage overflows and stormwater
60
runoff resulting from flooding is poorly understood. There is a critical need to understand the
61
distribution and transport of fecal bacteria in contaminated surface waters due to ecological as
62
well as human health concerns. Emerging real-time quantitative PCR (qPCR) methods designed
63
to estimate the concentration of fecal pollution by targeting genomic DNA from fecal indicator
3 ACS Paragon Plus Environment
Environmental Science & Technology Letters
64
bacteria such as E. coli, Enterococcus spp., and Bacteroidales are now available and can
65
generate test results in just a few hours after sampling, which is critical during extreme events
66
such as hurricanes. These genetic markers may also detect viable but nonculturable (VBNC)
67
cells that are not detected by conventional cultivation approaches but may still pose a public
68
health risk. In this study, we have documented the levels of fecal indicator bacteria (E. coli and
69
enterococci) including human-associated fecal genetic markers (Human-associated
70
Bacteroidales), measured using qPCR assays, across a Texas river impacted by Hurricane
71
Harvey. In addition, we studied the correlation between human-associated Bacteroidales markers
72
and traditional fecal indicators to better evaluate the efficacy of using these markers for the
73
detection of human fecal pollution.
74
The primary objective of this study was to investigate the extent to which the flooded
75
waterways in Texas have been contaminated with microbial contaminants following Hurricane
76
Harvey. To accomplish this, we have evaluated the presence and abundance of fecal indicator
77
bacteria and selected human-associated markers in flood-impacted environmental waters.
78
Surface water samples were collected along the Guadalupe River, ranging from sites in Victoria,
79
TX (which recorded devastating flooding; see Figure S1)14 to its outfall in the Texas coast.
80
Samples were also collected from the San Antonio River as a control site since it is in close
81
proximity to Guadalupe River and was not severely impacted due to flooding.14 We applied a
82
suite of qPCR assays targeting E. coli, enterococci, and human-associated Bacteroidales markers
83
(HF183 and BacHum) to identify and quantify the fecal contamination in water. The results
84
presented here provide the much needed and critical data on the effects of human fecal
85
contamination sources on the surface water quality following Hurricane Harvey.
86 4 ACS Paragon Plus Environment
Page 4 of 26
Page 5 of 26
Environmental Science & Technology Letters
87
MATERIALS AND METHODS
88
Sampling sites and sample collection. The Guadalupe River runs from Kerr County, Texas to
89
the San Antonio Bay in the Gulf of Mexico. The excessive rainfall from Hurricane Harvey
90
resulted in major flooding over the Guadalupe River at Victoria, TX during which the river
91
crested at about 32 feet (i.e. around 10 feet above the flood stage; see Figure S1). The water
92
quality of the river is a major concern as it is an important drinking water source for several
93
cities and some sections are used in recreational activities. Thus, fecal contamination of the river
94
is a significant public health concern and has a negative economic impact.
95
An a priori identification of representative sampling locations was not possible due to the
96
considerable logistical complications involving access to the sites and safety concerns. An initial
97
rapid assessment was performed on September 8, 2017 where multiple water samples were
98
collected from the Guadalupe River in flooded regions in Victoria, TX, and potential sampling
99
sites were identified. A more intensive sampling campaign was conducted two weeks later on
100
September 22, 2017 followed by three more sampling events each in October, November and
101
December 2017. During the intensive sampling campaign, water samples were collected from six
102
different sites along the Guadalupe River (G1, G2, G3, G4, G5 and G6) and two sites on the San
103
Antonio River (SA1 and SA2) over a period of 10 weeks from September 2017 to December
104
2017 (Figure 1). The sampling locations were chosen and categorized into flooded or non-
105
flooded sites based on the severity of impact due to rainfall and flooding (see Figure S2). The
106
description and coordinates of the sites has been listed in Table 1. Additional description of the
107
sampling sites has been given in Supporting Information.
108 109
Water samples were collected in duplicate from each site using sterilized 1 L bottles (Nalgene, Rochester, NY) and transported on ice to the laboratory at the University of Texas at 5 ACS Paragon Plus Environment
Environmental Science & Technology Letters
110
San Antonio (San Antonio, TX) within 6 h of collection. The water samples (200 to 1,000 mL)
111
were filtered onto 0.45 μm pore-size, 47 mm diameter polycarbonate membranes (Pall Life
112
Sciences, Ann Arbor, Michigan) and stored at −20 °C until DNA extraction.
113
qPCR analyses. The occurrence and relative abundance of four different fecal bacterial markers
114
in environmental water samples was measured using TaqMan qPCR assays (Table S1) and DNA
115
extracts as the templates. DNA was extracted from filter samples using the DNeasy PowerLyzer
116
PowerSoil Kit (Qiagen, Germantown, MD) according to the manufacturer’s protocol. The
117
targeted fecal bacterial groups were E. coli (EC23S857 assay),15 Enterococcus spp. (Entero1
118
assay),16 and human-associated Bacteroidales (HF183 and BacHum assays).17,18 The qPCR
119
assays were performed as previously described.19 Additional details regarding DNA extraction
120
and qPCR assays have been provided in Supporting Information. The genetic markers used in
121
our study for the detection of E. coli and enterococci, have been correlated to public health risk20
122
and Entero1 assay has been incorporated by the EPA into water quality standards in the United
123
States.7 The two human-associated Bacteroidales qPCR markers have been evaluated in a
124
number of studies and been shown to be highly specific to human feces, with little to no cross
125
reactivity with other host feces.21,22 In addition, all the four qPCR markers have been validated
126
and tested in several studies for measuring human fecal contamination in environmental
127
waters.19,21-23
128
Data analyses. The marker copy number per 100 mL of water was calculated for those samples
129
with values above the limit of quantification for each assay, and all data were log10 transformed
130
before statistical analysis. The qPCR results for initial grab samples were excluded from any
131
subsequent analyses as these samples were obtained from slightly different locations than the
132
sampling sites established in this study. Differences in marker concentrations were analyzed 6 ACS Paragon Plus Environment
Page 6 of 26
Page 7 of 26
Environmental Science & Technology Letters
133
using non-parametric Kruskal-Wallis one-way analysis of variance. The correlation between
134
human-associated Bacteroidales and conventional fecal indicators was analyzed using the
135
logistic regression analysis. All analyses were performed using Microsoft Excel (2016) and
136
correlation strength was interpreted according to an accepted scale for biological statistics.24 All
137
statistical test outcomes were regarded as significant at p < 0.001.
138
RESULTS AND DISCUSSION
139
Performance of qPCR assays. Standard curves were generated using serial dilutions of known
140
copy numbers to determine the amplification efficiencies and linear range of the qPCR assays.
141
The qPCR amplification efficiencies for all the assays ranged from 92.5 to 105.6%, with r2
142
values between 0.986 and 0.999. The linear range of quantification for qPCR assay of human-
143
associated Bacteroidales (HF183 and BacHum) and E. coli markers were between 10 and 106
144
copies, while the linear range for qPCR assay of Entero1 was 20 – 2 × 106 copies. PCR
145
inhibition tests were performed with 10-fold dilutions of each DNA extract as described in
146
previous study.23 In these tests, a Ct value proportional to a 10-fold dilution relative to the
147
undiluted DNA templates resulted, suggesting that PCR inhibition did not interfere with the
148
amplification efficiency. No template controls (two per PCR plate) and DNA extraction controls
149
indicated the absence of contamination in the qPCR experiments.
150
Distribution of fecal indicator bacteria and human-associated marker levels. Initial grab
151
water samples (n = 4) were collected on September 8, 2017 near sites G4 and G5. The average
152
concentration of fecal bacterial indicators in these samples were 1.31 × 102 copies and 5.05 × 105
153
copies per 100 mL water for E. coli and enterococci, respectively. The average concentration of
154
human-associated Bacteroidales in these samples were 6.74 × 101 copies and 5.67 × 101 copies
7 ACS Paragon Plus Environment
Environmental Science & Technology Letters
155
per 100 mL water for BacHum and HF183, respectively. This was followed by a more intensive
156
sampling campaign from September to December 2017.
157
The targeted fecal bacterial groups were frequently detected in the surface water samples
158
(n = 32) (40–100%). E. coli and enterococci were present in all the samples using the EC23S857
159
and Entero1 assays, respectively while the human-associated Bacteroidales were detected less
160
frequently. Enterococci exhibited the highest levels across all the sampling sites with mean
161
marker abundance greater than 103 copies per 100 mL water. The spatial distribution of the
162
levels of markers across the study sites is represented in Figure 2. The water samples yielded
163
significantly different amplification results across study sites with both the EC23S857 and
164
Entero1 assays, (Kruskal-Wallis test, p < 0.001). Site G4 had the highest concentrations of E.
165
coli and enterococci among all the study sites. The two human-associated Bacteroidales markers,
166
HF183 and BacHum, exhibited a similar spatial distribution pattern across the sampling sites,
167
although the level of BacHum marker was slightly higher for all the samples. Both of the
168
markers tested positive at most of the sites, with the exception of sites G1 and SA2 where HF183
169
was not detected in any of the samples. The levels of the human-associated Bacteroidales
170
markers were statistically different (Kruskal-Wallis test, p < 0.001) from each other among the
171
study sites. Site G4 had the highest mean copy number for the human-associated Bacteroidales
172
markers which is consistent with the excessive flooding at this site, and also because it was
173
located right after a WWTP and the discharge from the WWTP was fed into this region. Site G5
174
also had a relatively high concentration of human-associated Bacteroidales compared to other
175
sites. However, for site G6, which is located right after the confluence of Guadalupe River and
176
San Antonio River, the concentrations of both human-associated Bacteroidales markers
177
decreased, indicating that San Antonio River was not severely impacted by human fecal
8 ACS Paragon Plus Environment
Page 8 of 26
Page 9 of 26
Environmental Science & Technology Letters
178
contamination. This was also true for sites SA1 and SA2 (control sites on San Antonio River)
179
which had considerably lower levels of human-associated Bacteroidales.
180
The temporal distribution of the average levels of fecal indicator bacteria and human-
181
associated markers is presented in Figure 3. Sites G4 and G5 were classified as flooded regions
182
of the Guadalupe River, while sites G1, G2, G3 and G6 were located in the non-flooded areas.
183
Sites SA1 and SA2 were located in the non-flooded segments of the San Antonio River. On
184
average, all the markers had the highest concentrations for the flooded regions of the Guadalupe
185
River. Although, the concentrations of human-associated Bacteroidales was significantly higher
186
for the flooded sites than the non-flooded sites throughout the collection period, the observed
187
levels of human-associated fecal markers were considered relatively low for an impacted site.25
188
However, the concentrations for human-associated Bacteroidales remained elevated for flooded
189
sites even 2-3 months after Harvey suggesting that human fecal pollution may be a chronic and
190
relatively constant source of fecal contamination for the flood-impacted sites, both before and
191
after the hurricane. Samples collected on September 22, 2017, ~3 weeks after flooding,
192
demonstrated highest levels for E. coli, BacHum and HF183, both in the flooded and non-
193
flooded segments; however, no significant variation occurred over time (Kruskal-Wallis test, p >
194
0.5). Initial grab samples collected on September 8, 2017 (during recovery period from flooding)
195
from the flooded segment of Guadalupe River (near sites G4 and G5) had higher concentrations
196
of Entero1 marker than the samples collected on September 22, 2017 (3 weeks after flooding),
197
however the concentrations of other markers was slightly lower. This may be due to the
198
differences in the locations of the initial grab samples and the actual sampling sites, which were
199
not accessible earlier due to safety concerns.
9 ACS Paragon Plus Environment
Environmental Science & Technology Letters
200
The presence of high concentrations of bacterial indicators in surface water samples
201
collected from flooded regions of the river were indicative of the fecal contamination in the
202
environmental waters of the impacted areas. The most heavily flooded sites showed the highest
203
abundance of fecal indicator bacteria and human-associated Bacteroidales indicating that the
204
large number of sewage overflows and stormwater runoff which occurred during Harvey
205
flooding introduced high levels of fecal bacteria into waterways draining into the Gulf of Mexico
206
and thereby impairing surface and coastal water quality. The concentrations of fecal indicator
207
bacteria, such as Enterococcus spp., in recreational waters impacted by sewage have been linked
208
to adverse health outcomes in swimmers through epidemiology studies.20 The U.S. EPA recently
209
published revised recreational water quality criteria for enterococci that aim to keep the risk of
210
gastrointestinal illness (GI) in swimmers below approximately 30 illnesses per 1000 primary
211
contact recreators.7 All initial water samples collected from the flooded regions of the Guadalupe
212
River had E. coli and enterococci concentrations above the regulatory level for contact
213
recreational waters,7 although it should be noted that our results were measured by qPCR
214
methods while the Recreational Water Quality Criteria is based on culturable E. coli or
215
enterococci. A recent study used quantitative microbial risk assessment (QMRA) to simulate the
216
risk of GI illness associated with swimming in waters containing different concentrations of the
217
HF183 qPCR marker, and estimated there was a linear relationship between HF183 median
218
concentrations (in units of copies per 100 mL) and simulated GI illness rates [log10(GI risk) =
219
−4.93 + 0.94 × log10HF183; RSQ = 0.99].25 On the basis of this regression relationship, the
220
simulated GI illness rate per 1000 swimmers varied between 5.2 × 10-4 and 0.005 for the flooded
221
sites in Guadalupe River. This level of predicted human health risk based on HF183 marker is
222
relatively low as compared to the EPA-adopted health target of ∼30 GI illnesses per 1000
10 ACS Paragon Plus Environment
Page 10 of 26
Page 11 of 26
Environmental Science & Technology Letters
223
swimmers. Although the QMRA model used here relies on a scenario in which raw sewage
224
contaminates recreational waters, there are many other potential sources of human fecal pollution
225
such as leaking septic systems, open defecation, and treated wastewater. Future studies to
226
explore the relationships between human health and human-associated qPCR marker levels in
227
recreational waters impacted by sewage, treated effluents, and human feces are needed.
228
Stormwater runoff resulting from flooding can cause an influx of indicator bacteria to
229
receiving waters. Previous studies have demonstrated increased indicator bacteria levels in
230
coastal waters influenced by stormwater runoff.26-28 However, there are limited studies relating
231
the impact of flooding caused by hurricanes to the sources of fecal pollution in inland waterways.
232
Fecal bacteria may be introduced into surface waters during flooding through numerous sources
233
such as municipal waste from household sewage treatment systems, combined sewer overflows
234
(CSOs), sanitary sewer overflows (SSOs), damaged WWTPs, leaky septic tanks, and stormwater
235
and urban runoff.6,19 In this study we quantified the extent of human fecal contamination in
236
environmental waters by an integrated analysis of host-specific markers and fecal bacteria. The
237
assessment of study sites for human-associated Bacteroidales demonstrates that sewage sources
238
of fecal pollution are major contributors to water quality deterioration within our study area.
239
Because the Guadalupe River does not encompass agriculture or farming runoff, and there are no
240
CSOs in the sampling region, the primary source of human-associated Bacteroidales can be
241
attributed to the waste influx from damaged wastewater infrastructure and stormwater runoff. As
242
the river flow into the Gulf of Mexico, these sources appear to represent a chronic and relatively
243
constant source of human contamination for inland and near shore communities.
244
Correlation between human-associated Bacteroidales and fecal indicator bacteria. The
245
numbers of markers per 100 mL water sample were used to study the correlation between 11 ACS Paragon Plus Environment
Environmental Science & Technology Letters
246
human-associated Bacteroidales and conventional fecal indicators (enterococci and E. coli). The
247
human-associated Bacteroidales markers showed low to slightly strong correlation with
248
enterococci and E. coli markers, which are the most frequently used water quality indicators
249
(Figure S3). There was a weak correlation between the BacHum and Entero1 markers in the
250
samples (correlation coefficient, r2 = 0.24, p < 0.001), although E. coli exhibited slightly stronger
251
correlation with the BacHum marker (r2 = 0.72, p < 0.001). HF183 also showed similar
252
correlation as BacHum with both Entero1 (r2 = 0.32, p < 0.001) and E. coli (r2 = 0.75, p