Subscriber access provided by SELCUK UNIV
Article
Metagenomics shows low energy anaerobic-aerobic treatment reactors reduce antibiotic resistance gene dissemination from domestic wastewater Beate Christgen, Ying Yang, Ziauddin Ahamed, Bing Li, D Catalina Rodriguez, Tong Zhang, and David W. Graham Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/es505521w • Publication Date (Web): 20 Jan 2015 Downloaded from http://pubs.acs.org on January 30, 2015
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 free 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 accessible to all readers and 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.
Environmental Science & Technology 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 29
Environmental Science & Technology
1 2 3 4
Metagenomics shows low-energy anaerobic-aerobic treatment reactors
5
reduce antibiotic resistance gene dissemination from domestic
6
wastewater
7 8 9
Beate Christgen1#, Ying Yang2#, S. Ziauddin Ahammad1,3, Bing Li2, D. Catalina Rodriquez1,4,
10
Tong Zhang2* & David W. Graham1*
11 12
1- School of Civil Engineering & Geosciences, Newcastle University, Newcastle upon
13
Tyne, United Kingdom
14
2- Environmental Biotechnology Laboratory, Department of Civil Engineering, University
15
of Hong Kong, Hong Kong SAR, China
16
3- Department of Biochemical Engineering and Biotechnology, Indian Institute of
17
Technology Delhi, Hauz Khas, New Delhi, India
18
4- University of Antioquia, Laboratory Diagnostics and Pollution Control (GDCON),
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
Medellin, Colombia
#
*
These authors contributed equally to the manuscript.
Corresponding Authors:
Prof David Graham School of Civil Engineering and Geosciences Newcastle University Newcastle upon Tyne NE1 7RU United Kingdom Phone: (44)-0-191-222-7930 Fax: (44)-0-191-222-6502 E-mail:
[email protected] Dr Tong Zhang Environmental Biotechnology Laboratory Department of Civil Engineering University of Hong Kong, Hong Kong SAR, China Phone: 0-852-2857-8551 Fax: 0-852-2559-5337 E-mail:
[email protected] ACS Paragon Plus Environment
Environmental Science & Technology
40
Graphic Abstract
41
ACS Paragon Plus Environment
Page 2 of 29
Page 3 of 29
Environmental Science & Technology
42
Abstract
43
Effective domestic wastewater treatment is among our primary defences against the
44
dissemination of infectious waterborne disease. However, reducing the amount of energy
45
used in treatment processes has become essential for the future. One low-energy treatment
46
option is anaerobic-aerobic sequence (AAS) bioreactors, which use an anaerobic pre-
47
treatment step (e.g., anaerobic hybrid reactors) to reduce carbon levels followed by some
48
form of aerobic treatment. Although AAS is common in warm climates, it is not known how
49
its compares with other treatment options relative to disease transmission, including its
50
influence on antibiotic resistance (AR) in treated effluents. Here we used metagenomic
51
approaches to contrast the fate of AR genes (ARG) in anaerobic, aerobic and AAS
52
bioreactors treating domestic wastewater. Five reactor configurations were monitored for six
53
months, and treatment performance, energy use and ARG abundance and diversity were
54
compared in influents and effluents. AAS and aerobic reactors were superior to anaerobic
55
units in reducing ARG-like sequence abundances, with effluent ARG levels of 29, 34 and 74
56
ppm (198 ppm influent), respectively. AAS and aerobic systems especially reduced
57
aminoglycoside, tetracycline, and β-lactam ARG levels relative to anaerobic units, although
58
63 persistent ARG subtypes were detected in effluents from all systems (of 234 assessed).
59
Sulfonamide and chloramphenicol ARG levels were largely unaffected by treatment,
60
whereas a broad shift from target-specific ARGs to ARGs associated with multidrug-
61
resistance was seen across influents and effluents. AAS reactors show promise for future
62
applications because they can reduce more ARGs for less energy (32% less energy here),
63
but all three treatment options have limitations and need further study.
64 65
Keywords:
66
minimization
Wastewater
treatment,
metagenomes,
antibiotic
67 68
ACS Paragon Plus Environment
resistance,
energy
Environmental Science & Technology
69
Introduction
70
Domestic biological wastewater treatment is among our most effective defences against the
71
transmission of infectious waterborne disease and poor water quality. Contemporary
72
wastewater treatment plants (WWTPs) effectively reduce waste organic matter (as COD),
73
nitrogen (N) and fecal bacterial levels,1 often employing aerobic microbiological processes
74
with active aeration (e.g., activated sludge; AS). However, AS processes often have high
75
operating costs due to energy-consuming aeration, which will almost certainly increase if
76
energy prices rise in the future. This is a worldwide concern, but it is especially problematic
77
in emerging countries like China and India, which have rapidly growing economies,
78
increased urbanization, chronic shortages of energy, and higher baseline levels of endemic
79
disease.
80 81
One possible energy-saving treatment option is anaerobic-aerobic sequence (AAS) reactors,
82
which pre-treat the wastewater anaerobically to reduce COD loads and potentially produce
83
biogas, followed by aerobic processes that polish the anaerobic effluents to meet effluent
84
discharge standards. Depending on the design, AAS systems can have lower capital costs
85
(relative to conventional treatment systems), smaller footprints, lower sludge production,
86
possible biogas production, and reduced energy use because of lower oxygen demand.1-4 In
87
fact, AAS treatment systems are now quite common for domestic waste treatment in warmer
88
climate regions (e.g. Brazil, Egypt, India, and Israel),5-7 although AAS systems are not typical
89
in temperate and colder parts of the world.
90 91
Although a shift towards low-energy waste treatment has become critical, treatment
92
technologies must still protect against disease transmission, such as reducing pathogen
93
levels in treated effluents. However, there is strong evidence that WWTPs are reservoirs for
94
and can promote antibiotic resistant bacteria (ARB) and genes (ARG) in their microbial
95
communities,8-11 which then can be released to the environment via liquid effluents and
96
biosolids.8,12-15 The increased potential for ARBs/ARGs in WWTPs has implications to all
ACS Paragon Plus Environment
Page 4 of 29
Page 5 of 29
Environmental Science & Technology
97
plants and their ability to protect against disease because acquired AR in pathogenic and
98
other bacteria is increasing on a global scale.16 Therefore, for low-energy waste treatment
99
options to be implemented, they must be at least as effective as current treatment
100
approaches (ideally better) in terms of ARG levels or they will not effectively protect against
101
resistant disease dissemination.
102 103
Here we assessed ARG dissemination in the treated effluents of three different domestic
104
wastewater treatment options that use different amounts of energy. Specifically, anaerobic,
105
aerobic, and AAS reactors treating domestic wastes were compared in terms of energy use,
106
treatment performance, and ARG abundance and diversity over six months. High-throughput
107
sequencing and metagenomics were employed to characterise ARG occurrence and
108
diversity levels between influents and effluents, which were compared with relative energy
109
consumption levels during treatment.
110 111
Methods
112
Reactor Set-up and Operations: Reactor configurations are shown in Figure S1 (see
113
Supporting Information; SI). Fresh settled wastewater was collected weekly from a local
114
domestic WWTP in Northern England and stored in 18-L carboys at 4oC for use as reactor
115
feeds. Mean waste influent characteristics were as follows (± standard error): soluble COD
116
(sCOD) = 265.9 ± 14.9 mg/L, Total Kejdahl Nitrogen (TKN) = 82.7 ± 8.41 mg-N/L, ammonia
117
(NH3) = 47.2 ± 3.57 mg-N/L, and pH = 7.2 ± 0.02. Wastewater was fed using peristaltic
118
pumps (Watson Marlow 520S, UK) directly into three initial reactors, which included an
119
upflow anaerobic sludge blanket (UASB) reactor, an anaerobic hybrid reactor (AHR), and a
120
completely-mixed aerobic reactor (AER1); all reactor designs used in domestic wastewater
121
treatment.17 Effluents from the AHR and UASB units, respectively, were pumped into
122
second-stage aerobic units (AER2 and AER3) for further treatment (i.e., the two AAS
123
systems were AHR-AER2 and UASB-AER3). Two anaerobic reactor types were tested
124
because a sub-goal was to contrast UASB vs AHR designs relative to ARG/ARB mitigation.
ACS Paragon Plus Environment
Environmental Science & Technology
125 126
UASB and AHR reactors had working volumes of 1.5 L and height-diameter ratios of 4.0
127
(H/D). Wastewater influent was introduced at the bottom of the units and both units were
128
operated at three-day hydraulic retention times (HRT). Recirculation flow rates for the UASB
129
and AHR reactors were 133 mL/min and 155 mL/min, respectively. A three-phase separator
130
(gas–liquid–solid) was located at the top of each reactor to reduce washout of suspended
131
solids and also to trap biogas in attached Tedlar sampling bags. The anaerobic units were
132
maintained at 35 ± 1°C by circulating warm water around their cores. pH was maintained
133
between 6.8 and 7.2.
134 135
The three aerobic reactors had 1.04 L operating volumes. Wastewater was fed by peristaltic
136
pump in the aerobic units, either directly from the wastewater source (AER1) or from AHR
137
(AER2) or UASB (AER3) reactor effluent lines. The aerobic reactors were operated at two-
138
day HRTs, which were chosen based on preliminary system testing. No recycle was
139
provided because it was desired to keep operations simple to minimise the chance of
140
mechanical or other failures disturbing the systems. Air supply was tightly regulated and
141
monitored, which was essential for energy calculations. Mixing was provided using air
142
diffusers, stir plates and magnetic stir bars. Aerobic reactors were maintained at room
143
temperature (20 to 22 oC).
144 145
Sample Collection and Routine Sample Analysis: Wastewater influent and effluent
146
samples from the UASB, AHR and AER1 units, and effluents from the two AAS systems
147
(UASB-AER2 and AHR-AER3) were collected and analysed weekly. These reactors were
148
actually operated for over six months; however, sample collection for metagenomic analysis
149
was only performed over a six-week window of time after the reactors had been very stable
150
for six weeks. Our goal was to quantify ARG abundances and diversity under pseudo-steady
151
state operating conditions. Routine monitoring included sCOD, TKN, NH3, and pH for all
152
units; total suspended solids (TSS) and volatile suspended solids (VSS) for aerobic units;
ACS Paragon Plus Environment
Page 6 of 29
Page 7 of 29
Environmental Science & Technology
153
and methane (CH4) production in the anaerobic units. Influent and effluent samples for DNA
154
extraction and metagenomic analyses were collected weekly and frozen immediately.
155
All chemical analyses were performed according to standard methods for wastewater
156
characterisation,18 except biogas and CH4 analysis. Gas bag volumes were recorded
157
regularly and CH4 levels were quantified using direct injection into a Carlo Erba 5160 (UK)
158
Mega Gas Chromatograph (GC) equipped with a Flame Ionisation Detector (FID) and a HP-
159
PLOT Q capillary column (30m X 0.32 mm internal diameter) packed with 20µm Q phase.
160
The GC oven temperature was maintained at 35°C, and the injector and detector were kept
161
at 300°C. Hydrogen was used as the carrier gas (1ml/min, 65kPa). Methane analysis
162
standards were prepared using a 70:30% analytical grade CH4:CO2 standard gas mixture
163
(BOC Gases, UK). All GC injections were performed in duplicate.
164 165
DNA extraction, sequencing and bioinformatics analysis: DNA was extracted from the
166
weekly samples after completion of all sampling, using the Fast Soil DNA extraction kit (MP
167
Biomedicals, USA) and a Ribolyzer (MP Biomedicals, USA) according manufacturer’s
168
instructions. After extraction, DNA were sorted and combined into two tri-weekly groups for
169
sequencing (i.e., the first three weeks were combined and the second three weeks were
170
combined). Combining DNA from sequential sampling windows provided two independent
171
samples for influent and effluent microbial meta-communities in each reactor.
172 173
To assess the diversity and relative abundances of ARGs in the combined samples, DNA
174
extracts were provided (in duplicate) to the Beijing Genomics Institute (BGI) for shot-gun
175
libraries construction (insert size 170 bp) and high-throughput sequencing using the Illumina
176
HiSeq 2000 platform. Approximately 3 Gb (giga base pairs) of data was generated for each
177
DNA sample. Quality filtering was conducted for all metagenomic data to ensure validity in
178
downstream analysis. Raw reads that contained three or more ambiguous nucleotides,
179
quality scores below 20 for more than 36 bases, or with adapter contamination were
180
removed prior to subsequent data processing. Quality-filtered metagenomic data was
ACS Paragon Plus Environment
Environmental Science & Technology
181
searched for putative ARGs against the clean Antibiotic Resistance Database (ARDB),
182
which had non-redundant sequences using BLASTX with E-value ≤ 10-5.19,20 A read was
183
identified as an ARG-like sequence according to its best BLASTX hit with amino acid identity
184
≥ 90% and alignment length ≥ 25 amino acids.21,22
185 186
Identified ARG-like sequences were sorted into ARG types and sub-types, using a structured
187
ARDB and customized python script.20 ARG definitions, which included presumed resistance
188
mechanisms for each ARG, were based on previous work and are simplifications because
189
ARGs often code for proteins with multiple effects.20,22 However, simplification is reasonable
190
because over 230 ARGs were assessed and overarching trends were more important than
191
individual details. To compare ARG levels among samples, the number of ARG-like
192
sequences was normalized to the number of metagenome and ARG-sequences in each
193
sample. As such, ARG levels are reported either as “total metagenome sequences” (in ppm;
194
one read in one million reads) or as “total ARG-like sequences” as percentages (%).
195 196
Energy Calculations: Energy use was calculated for all systems, and included energy used
197
for aeration, mixing, heating, and pumping, and potential energy gained from biogas
198
production.2 To produce realistic energy estimates, calculations for each case were based
199
on extrapolated energy use in geometrically identical, scaled-up treatment units treating
200
1000 L/d wastewater. However, air supply rate, mixing and pumping rate data were based
201
upon measured lab reactor data.
202 203
Energy used in the aerated units was dominated by aeration and mixing needs, which was
204
calculated using measured air flow rate data and equivalent paddle-mixing requirements for
205
scaled-up aerobic reactors. Power ratings for characteristic industrial-scale centrifugal
206
pumps were used to estimate feed and effluent recycle pump energy needs.2 Energy
207
estimates for anaerobic units were based on required mixing energy as well as energy
208
required for heating anaerobic units to 35°C. Potential energy recovery from biogas
ACS Paragon Plus Environment
Page 8 of 29
Page 9 of 29
Environmental Science & Technology
209
production was estimated based on previous CH4 data and use of industrial-scale gas
210
turbines. Net energy used by the anaerobic units was the difference between energy used
211
for pumping, heating and mixing minus the potential energy gained from biogas production.
212
Energy calculations for AAS systems included the energy used for mixing and aeration in the
213
aerobic units (less air was used due to lower sCODs), and mixing, heating and biogas
214
production in the anaerobic units, which varied slightly between the UASB and AHR
215
reactors.
216 217
Other Data Analysis: This study contrasted aerobic, anaerobic and AAS treatment systems
218
relative to ARG fate and energy use using two independent influent and effluent
219
metagenomes per system (except AHR where one metagenome was available; see SI
220
Tables S1-S3). However, to allow statistical comparisons among the three treatment options,
221
it was first necessary to determine whether ARG characteristics in each duplicate-pair were
222
statistically the same in paired DNA samples. One-way ANOVA and t-tests were performed
223
on relative ARG levels in each duplicate for the 19 AR groups (identified in the metagenomic
224
analysis; see Table S1). If the duplicate-pairs were statistically the same (i.e., p < 0.05), data
225
from the duplicates were clumped as averages to allow more rigorous statistical
226
comparisons among treatment opens. Follow-on tests included two-sample testing, such as
227
ANOVA among reactors under each treatment, and statistical comparisons between
228
performance data for each reactor system. All analyses were conducted using SPSS
229
(Chicago,
230
(http://bioinfogp.cnb.csic.es/tools/venny/), and the relatedness of influents and treated
231
effluents was visualized using Circos.23
IL;
v.
17.0).
Venn
Diagram
was
plotted
using
Venny
232 233
Results and Discussion
234
Bioreactor Performance, Operational Data and Energy Used per Design. Two anaerobic
235
(UASB and AHR), one aerobic (AER1), and two AAS (AHR-AER2 and UASB-AER3)
236
treatment systems (see Figure S1) were monitored to compare treatment performance and
ACS Paragon Plus Environment
Environmental Science & Technology
237
ARG metagenomes before and after treatment of domestic wastes. Relative to treatment
238
performance, Table 1 data show effluent sCOD, TKN and NH3 levels did not significantly
239
differ between the two anaerobic reactors or between the two AAS systems (p > 0.05).
240
Therefore, these data were grouped under ‘anaerobic’ (UASB and AHR combined) and
241
‘AAS’ (AHR-AER2 and UASB-AER3 combined), which were then statistically compared with
242
each other and ‘aerobic’ treatment unit (AER1), relative to effluent sCOD, TKN and NH3
243
levels.
244 245
The AAS systems had significantly lower effluent sCOD levels than aerobic or anaerobic
246
units alone (p < 0.027 and p < 0.002, respectively). In contrast, aerobic units had
247
significantly lower effluent TKN and NH3 levels than the AAS systems (p < 0.024 and p
