Subscriber access provided by University of Florida | Smathers Libraries
Article
Changes in glucose fermentation pathways as a response to the free ammonia concentration in microbial electrolysis cells Mohamed Mahmoud, César Iván Torres, and Bruce E. Rittmann Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b05620 • Publication Date (Web): 17 Oct 2017 Downloaded from http://pubs.acs.org on October 19, 2017
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 43
1 2 3 4 5
Environmental Science & Technology
Changes in glucose fermentation pathways as a response to the free ammonia concentration in microbial electrolysis cells Mohamed Mahmoud § ¥ £ *, César I. Torres § Φ, and Bruce E. Rittmann § ¥
6 7
§
8
Tyler Road, Tempe, Arizona, USA.
9
¥
Biodesign Swette Center for Environmental Biotechnology, Arizona State University, 727
School of Sustainable Engineering and the Built Environment, Arizona State University,
10
Tempe, Arizona, USA.
11
Φ
12
Tempe, Arizona, USA.
13
£
14
Cairo 12311, Egypt
School for Engineering of Matter, Transport and Energy, Arizona State University,
Water Pollution Research Department, National Research Centre, 33 El-Buhouth St., Dokki,
15 16
* Corresponding author: Mohamed Mahmoud
17
Email:
[email protected] &
[email protected]; tel.: +1 480 727 7596; fax: +1 480 727
18
0889.
1 ACS Paragon Plus Environment
Environmental Science & Technology
19
Abstract
20 21
When a mixed-culture microbial electrolysis cell (MEC) is fed with a fermentable substrate, such
22
as glucose, a significant fraction of the substrate’s electrons ends up as methane (CH4) through
23
hydrogenotrophic methanogenesis, an outcome that is undesired. Here, we show that free
24
ammonia-nitrogen (FAN, which is NH3) altered the glucose fermentation pathways in batch
25
MECs, minimizing the production of H2, the “fuel” for hydrogenotrophic methanogens.
26
Consequently, the Coulombic efficiency (CE) increased: 57% for 0.02 g FAN/L fed-MEC,
27
compared to 76% for 0.18 fed-MECs and 62% for 0.37 g FAN/L fed-MECs. Increasing FAN
28
concentration was associated with the accumulation of higher organic acids (e.g., lactate, iso-
29
butyrate, and propionate), which was accompanied by increasing relative abundances of
30
phylotypes that are most closely related to anode respiration (Geobacteraceae), lactic-acid
31
production (Lactobacillales), and syntrophic acetate oxidation (Clostridiaceae). Thus, the
32
microbial community established syntrophic relationships among glucose fermenters, acetogens,
33
and anode-respiring bacteria (ARB). The archaeal population of the MEC fed 0.02 g FAN/L was
34
dominated by Methanobacterium, but 0.18 and 0.37 g FAN/L led to Methanobrevibacter
35
becoming the most abundant species. Our results provide insight into a way to decrease CH4
36
production and increase CE using FAN to control the fermentation step, instead of inhibiting
37
methanogens using expensive or toxic chemical inhibitors, such as 2-bromoethane sulfonic acid.
38 39
Keywords: Microbial electrolysis Cells, Free ammonia-nitrogen, Syntrophy, Anode-respiring
40
bacteria, Fermenters.
2 ACS Paragon Plus Environment
Page 2 of 43
Page 3 of 43
41
Environmental Science & Technology
TOC Art
42 43
44 45
3 ACS Paragon Plus Environment
Environmental Science & Technology
46
Introduction
47
Microbial electrolysis cells (MECs) represent one of the newest environmental
48
biotechnologies for wastewater treatment coupled to the production of renewable energy in the
49
form of electrical power, hydrogen gas (H2), or valuable chemicals.1 Since the organic
50
compounds – the “fuel” for the MECs – are complex in nature, their biodegradation requires
51
cooperation among different trophic groups: fermenters, homoacetogens, and anode-respiring
52
bacteria (ARB), as well as competition with methanogens.2,3 Generally, an MEC’s microbial
53
community has a high complexity in terms of community structure and diversity, but key
54
microbial populations have been identified.4,5
55
The defining group in an MEC is the ARB, which perform a unique type of respiration.
56
Most respiring anaerobes transfer electrons intracellularly to a terminal electron acceptor, such as
57
sulfate, nitrate, or carbon dioxide.6,7 ARB oxidize organic matter internally, but transfer the
58
resulting electrons outside their membrane to a solid electron acceptor, the anode.2,5 Another
59
feature of majority of ARB in mixed community is that their electron donors are limited to only a
60
few simple substrates, such as acetate and H2.8–10 Thus, fermentation is necessary to generate the
61
mixture of simple fermentation products that the ARB can utilize directly.
62
In environments lacking electron acceptors, fermentable organic compounds, such as
63
glucose, are transformed into a variety of organic acids and H2. Acetate is the most prevalent
64
organic acid, and Table 1 presents a number of reactions in which acetate and H2 are formed or
65
consumed. H2-consumers include hydrogenotrophic methanogens, which use H2 as the main
66
electron donor and produce methane (reaction 3 in Table 1).7,11–13 Homoacetogens also scavenge
67
H2 to yield acetate. Homoacetogenesis and hydrogenotrophic methanogenesis have very low
68
energy yields.14 H2 must be kept at low level to allow fermentation to be thermodynamically
4 ACS Paragon Plus Environment
Page 4 of 43
Page 5 of 43
Environmental Science & Technology
69
possible.3,11,15 When H2 builds up, the fermentation stoichiometry changes such that higher
70
organic acids, such as lactate, butyrate, propionate, and ethanol, are produced rather than acetate
71
and H2.3,15
72
When the MEC’s anode is the only respiratory electron acceptor, ARB can out-compete
73
methanogens for acetate, due to their thermodynamic and kinetic advantages over the acetate-
74
consuming methanogens (reaction 5 in Table 1); hence, acetoclastic methanogens cannot
75
compete with acetate-oxidizing ARB, although H2-oxidizing methanogens compete well with
76
H2-oxidizing ARB.8 However, ARB do not have similar advantages for H2 consumption, and the
77
electrons in H2 often are channeled to methane (CH4) via hydrogenotrophic methanogens, which
78
are able to grow in suspension, as well as in the anode’s biofilm.9,16 Thus, one of the challenges
79
for MEC scale-up and commercialization is to minimize production of CH4 from the H2
80
generated via fermentation in the presence of hydrogenotrophic methanogens.
81 82
Table 1.
83 84
Previous researchers proposed different strategies to inhibit methanogenesis in
85
laboratory-scale MECs, including thermal treatment, periodic exposure to oxygen (O2), pH
86
excursions, alamethicin exposure, and use of chemical inhibitors such as 2-bromoethanesulfonate
87
(BES).8,17, 18 Among the proposed strategies, chemical inhibitors seem to be the most effective
88
approach, due to their selectivity for inhibiting the activity of methyl coenzyme M reductase A
89
(mcrA) gene in acetoclastic and hydrogenotrophic methanogens. For example, Parameswaran et
90
al.8 inhibited methanogens in an ethanol-fed MEC with 50 mM BES, which boosted the
91
Coulombic efficiency (CE) by ~40%, as electrons from H2 were rerouted to anode respiration
5 ACS Paragon Plus Environment
Environmental Science & Technology
92
instead of to CH4 production. Although BES is an effective way to inhibit methanogens,8,9 it is
93
not practical for field applications due to the continuous need to supply the inhibitor, which is
94
environmentally toxic and expensive.
95
We recently revealed that ARB are tolerant to relatively-high free-ammonia nitrogen
96
(FAN) concentrations, up to at least 200 mg/L; however, high total-ammonia nitrogen (TAN)
97
concentrations (i.e., > 2.2 g/L) might cause inhibition (or lowering) of the anode respiration
98
rate.19 The relatively high threshold for FAN inhibition by ARB may be a tool for managing
99
microbial communities to favor ARB and high electron recovery in MECs fed with fermentable
100
substrates, as long as TAN is not too high. Therefore, we hypothesize that high-enough FAN
101
can promote the desired syntrophy in MECs fed with fermentable substrate by inhibiting
102
hydrogenotrophic methanogens. Suppressing H2-oxidizing methanogens provides ideal
103
conditions for homoacetogens to oxidize H2 and grow, leading to the generation of acetate, the
104
ideal substrate for ARB. 8,9 This hypothesis requires that a FAN concentration high enough to
105
suppress methanogens not have a strong inhibitory effect on ARB and fermenters.14,20 This
106
strategy is highly relevant to MECs treating real wastewater that are characterized by high FAN
107
concentrations, including landfill leachate and animal wastewater.21,22
108
A high FAN concentration also might promote a new pathway for acetate consumption to
109
produce CH4 at very low H2 partial pressure: syntrophic acetate oxidation (SAO) (alternatively
110
known as reverse acetogenesis), which has been explored in anaerobic digesters.23,25 SAO is a
111
two-step process in which acetate is utilized by syntrophic acetate-oxidizing bacteria (SAOB)
112
(reaction 7 in Table 1) with the generation of reducing equivalents, often in the form of H2. This
113
step is a highly energy-demanding reaction that requires syntrophy with H2-consuming bacteria
114
(e.g., hydrogenotrophic methanogens or hydrogenotrophic ARB) to maintain a very low level of
6 ACS Paragon Plus Environment
Page 6 of 43
Page 7 of 43
Environmental Science & Technology
115
H2 (reaction 3 or reaction 6 in Table 1, respectively) in order to make the overall reaction
116
thermodynamically favorable.11,24 In the absence of other electron acceptors, such as nitrate and
117
sulfate, the produced H2 is most likely routed to CH4 through hydrogenotrophic methanogenesis,
118
since ARB are known of being relatively poor H2-consumers.8,25 So far, nothing is known about
119
the SAO process in MECs fed with high FAN concentration.
120
Our overarching goal is to investigate the effect of different FAN levels on the
121
interactions among ARB and other members of the communities, particularly the fermenters and
122
methanogens. We document the microbial interactions through a combination of chemical,
123
electrochemical, and genomic tools. In order to study electron-flow and synergies in an MEC’s
124
anode, we use a fixed concentration of glucose (5 mM) as the sole electron-donor substrate, and
125
we vary the influent FAN concentration (i.e., 0.02, 0.18, and 0.37 g FAN/L) going into the anode
126
during batch and semi-continuous (hydraulic retention time of 2 days) MECs. Key is that we
127
establish electron-equivalent mass balances. We also characterize the relative abundance and
128
composition of bacteria and Archaea by Illumina sequencing, and we track homoacetogens and
129
methanogens by targeting the formyltetrahydrofolate synthetase (FTHFS) gene – a conserved
130
gene involved in their CO2 fixation pathway – and mcrA gene, respectively, by quantitative real-
131
time PCR (qPCR).
132
7 ACS Paragon Plus Environment
Environmental Science & Technology
133
Materials and methods
134 135 136
MEC design and operation For our experiments to understand the fundamentals of how the microbial community
137
changes in response to high FAN concentration, we used half-cell MECs with its anode potential
138
controlled using a potentiostat. Each MEC held ~ 320 mL in each chamber and had a square
139
graphite anode having a total surface area of ~ 11.6 cm2, and a 0.8-cm OD graphite rod as
140
cathode. This setup excluded any interference from potential losses at the cathode or due to a
141
large distance between the anode and the cathode; fixing the anode potential allowed us to focus
142
only on the anodic reactions. In order to ensure that the ARB were not limited by a low anode
143
potential, we poised the anode at a fixed potential of – 0.3 V vs. Ag/AgCl (or ~ – 0.04 V vs.
144
standard hydrogen electrode (SHE) in our media) using a VMP3 digital potentiostat (Bio-Logic
145
USA, Knoxville, TN) by placing an Ag/AgCl reference electrode (BASI Electrochemistry, west
146
Lafayette, IN) about 0.5 cm away from the anode. We separated the anode chamber from the
147
cathode chamber with an anion exchange membrane (AMI 7001, Membranes International, Glen
148
Rock, NJ). The pH of the cathode chamber was adjusted to 12 by addition of 10 N NaOH to
149
facilitate the transport of hydroxide ions (OH–) from the cathode compartment to the anode
150
compartment through the anion-exchange membrane.
151
We inoculated the MEC’s anode chamber with 2 mL of biofilm inoculum from a
152
previously operated MEC fed with 5 mM glucose as the sole electron donor. After sparging the
153
MEC with ultra-high purity N2 gas (≥99.9%) for ~ 45 min, we fed the MEC with autoclaved (for
154
90 min at 121°C) glucose medium (initial pH of 8.1) containing: 5 mM glucose (or 38.4 me–eq),
155
100 mM phosphate buffer (KH2PO4/Na2HPO4), 2.1 g NaHCO3, and 10 mL of trace minerals as
8 ACS Paragon Plus Environment
Page 8 of 43
Page 9 of 43
Environmental Science & Technology
156
outlined in Parameswaran et al.8 We added different amounts of ammonium chloride as the sole
157
N-source, giving FAN (and TAN) concentrations of 0.02 g FAN/L (0.2 g TAN/L), 0.18 g FAN/L
158
(2 g TAN/L), and 0.37 g FAN/L (4 g TAN/L). We eliminated the possibility of ARB biofilm
159
needing to fix atmospheric N2 by using an ammonium concentration more than sufficient for
160
growth in the control experiments (0.2 g N/L or 14 mM-N). Prior to the MEC’s electron balance
161
experiments, we acclimated the inoculum by continuously feeding the developed biofilm with 5
162
mM glucose medium as the sole electron donor at relatively short hydraulic retention time (i.e.,
163
24 h), followed by performing 2 consecutive batch cycles to select for a microbial community
164
that was efficient at consuming glucose and producing electrical current. We controlled the
165
temperature at 30°C in a temperature-controlled room, and the liquid in both chambers was
166
mixed using a magnetic stirrer at 220 rpm. We performed all batch MEC experiments in
167
duplicate.
168
In order to confirm and expand the results from the batch MEC experiments, we
169
evaluated MEC performance with semi-continuous operation. We operated two independent
170
MECs in parallel and fed in a semi-continuous mode once every 2 days. The MECs were
171
operated at a fixed temperature of 30°C and fed with a 5-mM glucose medium having the same
172
composition as mentioned in the previous paragraph. The liquid in both chambers was mixed
173
using a magnetic stirrer at 220 rpm. We tested the MECs with different FAN concentrations
174
achieved with different combinations of TAN concentration and pH: (1) 0.003 g FAN/L (pH = 7
175
and TAN = 0.2 g/L), (2) 0.02 g FAN/L (pH = 8.1 and TAN = 0.2 g/L), 0.18 g FAN/L (pH = 8.1
176
and 2 g/L), and (4) 0.37 g FAN/L (pH = 8.1 and TAN = 4 g/L).
177
9 ACS Paragon Plus Environment
Environmental Science & Technology
178 179
Chemical analyses We measured TAN, in duplicate, using HACH kits (HACH, Ames, IA) after filtration
180
through a 0.22-µm membrane filter (PVDF GD/X, Whatman, GE Healthcare, Ann Arbor, MI).
181
We quantified liquid samples for organic fermentation products and glucose using high-
182
performance liquid chromatography (HPLC; Model LC-20AT, Shimadzu, Columbia, MD) with
183
an Aminex HPX-87H column (Bio-Rad Laboratories, Hercules, CA) after filtration through a
184
0.22-µm membrane filter according the method described in Mahmoud et al.21 Briefly, we used
185
2.5 mM sulfuric acid and 18-MΩ reverse-osmosis water as eluents for determining organic
186
fermentation products and glucose, respectively, at a constant flow rate of 0.6 mL/min. We
187
developed a five-point calibration curve for every set of analyses, performed duplicate assays,
188
and report the average concentrations.
189
We measured the volume of gas produced with a friction-free glass syringe of 10- or 25-
190
mL volume (Popper & Sons, Inc., New Hyde Park, NY, USA). We estimated gas percentages of
191
H2, CH4, and CO2 in samples taken with a gas-tight syringe (SGE 500 µL, Switzerland) using a
192
gas chromatograph (GC 2010, Shimadzu Corporation, Columbia, MD) equipped with a thermal
193
conductivity detector and a packed column (CarboxenTM 1010 PLOT Capillary Column,
194
Supleco, Inc.). Helium was the carrier gas at a constant flow rate of 10 mL/min and pressure of
195
42.3 kPa. Temperature conditions for column, injection, and detector were 80, 150, and 220 °C,
196
respectively. We employed analytical grade H2, CH4, and CO2 gases for standard curves, carried
197
out gas analyses in duplicate, and averaged the two values.
198
10 ACS Paragon Plus Environment
Page 10 of 43
Page 11 of 43
199 200
Environmental Science & Technology
Microbial community analyses DNA extraction. At the end of each batch experiment, we harvested the entire biofilm
201
biomass from each MEC anode by scraping it off with a sterilized pipette tip and suspending the
202
biomass sample in a sterile centrifuge tube containing DNA-free water. We then centrifuged the
203
contents at 10,000g (Eppendorf Centrifuge 5414 D, USA) for 10 min to concentrate the biomass,
204
which we stored at −20 °C prior to DNA extraction. We also centrifuged the entire liquid of the
205
MEC chamber to concentrate the suspended phase for extraction. We extracted the total
206
genomic DNA using the MOBIO Powersoil DNA extraction kit according to manufacturer’s
207
instructions, and determined the quality and quantity of the extracted DNA using a nanodrop
208
spectrophotometer (ND 1000, Thermo Scientific) by measuring absorbance at 260 and 280 nm.
209
Quantitative real-time PCR (qPCR). We evaluated the presence and abundance of
210
methanogenic Archaea and homoacetogens by targeting the mcrA and FTHFS genes,
211
respectively, using qPCR. We carried out all PCR reactions in optically clear tubes with caps in
212
an Eppendorf Realplex 4S realcycler with a 20 µL total reaction volume. We performed all
213
qPCR reactions in triplicate along with a six-point standard curve by following modified assays
214
for FTHFS gene9 and mcrA gene.26 We performed negative-control assays by using DNA-free
215
water instead of DNA templates. We reported the results as the number of gene copies per
216
reactor, after calculating the number of 16S rRNA genes per the entire biofilm or suspended
217
phase of each MEC.
218
Illumina sequencing. We amplified the extracted DNA using 16S rRNA gene-targeting
219
forward and reverse fusion primers at MR DNA laboratory (www.mrdnalab.com, Shallowater,
220
TX, USA). We performed bacterial and archaeal sequence using the bar-coded primer set 515F
221
(5′–GTGCCAGCMGCCGCGGTAA–3′)/806R (5′–GGACTACHVGGGTWTCTAAT–3′)27 and
11 ACS Paragon Plus Environment
Environmental Science & Technology
222
349F (5′–GYGCASCAGKCGMGAAW–3′)/806R (5′–GGACTACVSGGGTATCTAAT–3′),28
223
respectively, in a single-step 28 cycle PCR using the HotStarTaq Plus Master Mix Kit (Qiagen,
224
USA) with the following conditions: 94°C for 3 min, followed by 28 cycles of 94°C for 30 s,
225
53°C for 40 s, and 72°C for 1 min, after which a final elongation step at 72°C was for 5 min. In
226
order to determine the amplification success and the relative intensity of bands, PCR products
227
were checked in 2% agarose gel.29 Then, the amplified products were pooled in equal
228
proportions based on their molecular weight and DNA concentrations, purified using calibrated
229
Ampure XP beads, and used to prepare DNA libraries following the Illumina TruSeq DNA
230
library preparation protocol. Sequencing was performed on a MiSeq Illumina sequencer
231
(Illumina Inc., USA) following the manufacturer’s guidelines.
232
Bioinformatics analysis. We analyzed the sequences data using QIIME software package
233
version 1.9.130 after trimming off low-quality bases and discarding sequences shorter than 25 bp,
234
longer than 450 bp, or labeled as chimeric sequences. We performed taxonomic classification at
235
3% sequence divergence (97% sequence similarity) and assigned taxonomy to operational
236
taxonomic units (OTUs) by using the Ribosomal Database Project (RDP) classifier with a 50%
237
confidence threshold.31 We performed rarefaction on the OTU table at a depth of 100 sequences
238
in 10 replicates, and we analyzed the rarefaction measures with the same sequence numbers per
239
sample (i.e., 223,000 sequences per sample for bacteria population) with a python script in
240
QIIME software.31 We performed alpha- and beta-diversity calculations, including richness of
241
each samples with Chao1 index,32 the diversity with Shannon and Phylogenetic Distance Whole
242
Tree metrics,33 the evenness with the equitability coefficient on a normalized scale from zero, in
243
which community is perfectly even, to 1, in which community has one dominant OTU and many
244
singlets,29 and principal-coordinates analysis (PCoA) using python script in QIIME software.
12 ACS Paragon Plus Environment
Page 12 of 43
Page 13 of 43
Environmental Science & Technology
245
Sequence data sets are available at NCBI/Sequence Read Archive (SRA) under study with
246
BioProject accession number PRJNA343831.
247 248 249
Electron balance and electron flow into various sinks We established electron balances by estimating the electron equivalents of experimentally
250
measured glucose and fermentation products with the following equivalences:12 24 me– eq per
251
mmole glucose, 8 me– eq per mmole acetate, 14 me– eq per mmole propionate, 20 me– eq per
252
mmole butyrate, 12 me– eq per mmole lactate, 25 me– eq per mmole valerate, 0.32 me– eq per mL
253
CH4 (@30°C), and 0.08 me– eq per mL H2 (@30°C).
254
First, we established electron flow from glucose when hydrogenotrophic methanogens
255
are not suppressed or are completely suppressed. We assumed stoichiometric fermentation of
256
glucose (24 me– eq) into 2 moles of acetate (16 me– eq) and 4 moles of H2 (8 me– eq). Acetate is
257
efficiently consumed by acetate-consuming ARB to generate electric current, while H2 is
258
consumed by hydrogenotrophic methanogens that have thermodynamic, kinetic, and metabolic
259
advantages over H2-consuimng ARB and homoacetogens.8 In parallel, a fraction of electrons is
260
utilized by different microbial groups in the MEC’s anode to synthesize new microbial cells. We
261
quantify the fraction of electrons used for biomass synthesis based on fs° (me– eq biomass per
262
me– eq substrate) values of 0.18 for fermenters,12 0.08 for hydrogen-consuming methanogens,12
263
~0.02 for homoacetogens, 34,35 and ~ 0.09 for ARB.19 Supplementary Figure S1 shows the
264
different paths for electron flow from glucose when hydrogenotrophic methanogens are not
265
suppressed (Figure S1A) or are completely suppressed (Figure S1B). The resulting CE for the
266
electron flows in Figure S1A (when methanogenesis is not suppressed) is 50%, with 25% of the
13 ACS Paragon Plus Environment
Environmental Science & Technology
267
electrons routed to CH4 gas and 25% to biomass. CE increases to 76% when methanogenesis is
268
completely inhibited, with comparable electrons routed to biomass (24%) (Figure S1B).
14 ACS Paragon Plus Environment
Page 14 of 43
Page 15 of 43
269
Environmental Science & Technology
Results and discussion
270 271
Framework for explaining electrons distribution for batch MECs
272
Figure 1 presents the electron distribution from glucose into the possible electron sinks
273
over the course of the experiments with different FAN concentrations. The inoculum had been
274
pre-acclimated to glucose, and then we performed an electron balance on batch experiments with
275
glucose as the sole electron donor. With the lowest initial FAN (0.02 g FAN/L), current
276
production increased significantly during the first 2 days (Figure 1A). Simultaneously, acetate
277
production increased to ~ 32% of the total electrons, and glucose decreased below the detection
278
limit. Methane slowly accumulated within 2 days of operation and reached a plateau after day 6,
279
when methane generation accounted for ~ 15% of the total electrons supplied from glucose.
280
15% methane is less than the estimate in Figure 1A (25%), probably due to either H2
281
consumption by either ARB or homo-acetogenesis. The total electrons contained in other
282
fermentation products were small: ~ 3.3 % for propionate and ~ 3.1% for butyrate within the
283
first 4 days of operation. Propionate and butyrate were completely consumed at day 17 and day
284
8, respectively.
285
We detected no H2 in the headspace gas of the MEC’s anode. Thus, H2 produced during
286
glucose fermentation was quickly channeled to methane by H2-consuming methanogens, to
287
current by ARB, or to acetate and then current by homoacetogens and ARB.13,24 The Coulombic
288
recovery at the end of batch cycle (~ 57%) was slightly greater than the estimated value in Figure
289
S1, which suggests that a modest flow of electron where routed through H2 to the anode.
290 291
Figure 1B shows the distribution of electrons from glucose during 20 days of batch operation with 0.18 g FAN/L, where methane generation was partly inhibited. Compared to the
15 ACS Paragon Plus Environment
Environmental Science & Technology
292
MEC fed with 0.02 g FAN/L (Figure 1A), current slowly increased during the first 2 days and
293
was associated with complete glucose fermentation. However, lactate started to increase,
294
becoming the largest electron sink (~ 30%) at day 2, followed by acetate (~ 13%), propionate (~
295
5%), and iso-butyrate (~ 2%). Since lactate fermentation is more likely than direct anode
296
respiration, it has to be fermented first before it can be utilized by ARB. Lactate fermentation
297
proceeded rapidly, and the current increased to become the largest electron sink after day 6, at
298
which time methane accounted for ~ 9% of the total electrons supplied from glucose. Although
299
we did not detect any H2 in the headspace gas, formate was detected in low concentration (0.67
300
mM), accounting for nearly 2% of glucose’s total electrons. These results support that that 0.18
301
g FAN/L delayed lactate fermentation, rather than glucose fermentation, and significantly
302
inhibited methanogenesis. A key finding is that CE was ~ 76%, which illustrates the benefit to
303
inhibiting methanogenesis.
304
Further increasing the initial FAN concentration to 0.37 g FAN/L (Figure 1C) had a more
305
pronounced effect on fermentation, although the impact on methanogenesis was not much
306
greater than 0.18 g FAN/L. Similar to the lower FAN concentration, the rate of glucose
307
consumption did not change; however, 0.37 g FAN-N/L altered the fermentation kinetics and
308
pathway. During the first 4 days, lactate, iso-butyrate, propionate, butyrate, and acetate
309
accumulated, collectively accounting for ~ 25% of glucose’s electrons. After 4 days,
310
fermentation proceeded rapidly with concurrent increases in electric current, which became the
311
largest electron sink by the end of the batch cycle (~ 62%), suggesting that this high FAN
312
concentration might partially inhibit ARB. However, methanogenesis was not completely
313
inhibited, as it accounted for ~ 6% of glucose’s electrons.
16 ACS Paragon Plus Environment
Page 16 of 43
Page 17 of 43
314
Environmental Science & Technology
As a result of organic-acids accumulation, the pH at the end of all batch cycles decreased
315
from their initial pH value (8.1) to a narrow range of 6.85−7.10. Since the pH values were not
316
much different from each other, variations in pH should have had a minimal effect on
317
fermentation pathway and kinetics.36
318 319
Figure 1.
320 321 322
Electron balances for batch MECs Figure S2 presents the electron balances at the end of batch-cycle operation: 57±2.9%,
323
76±3.5%, and 62±2.7% of the total electrons from glucose ended up as electric current for MECs
324
fed with 0.02, 0.18, and 0.37 g FAN/L, respectively. Theoretically, ~ 25% (or 9.7 me– eq) of
325
glucose’s electrons (i.e., 38.4 me– eq) should have been ended up as methane, if all the generated
326
H2 were utilized by H2-consuming methanogens (Figure S1A). However, we detected as
327
methane only about 15%, 9%, and 6%, respectively, of the original electrons.
328
Electrons ending up as organic acids at the end of the experiment were by far the highest
329
for the MEC fed with 0.37 g FAN/L: ~ 16% of the influent electrons, versus ~ 0.2% and ~ 2%
330
for MECs fed 0.02 and 0.18 g FAN-N/L, respectively. Approximately 28%, 13%, and 16% of
331
the electron from glucose were unaccounted by directly measured components. Unidentified
332
components include biomass, soluble microbial products, and fermentation products not
333
measured by HPLC. Figure S1 shows an estimate of 25 and 24% of electrons ending up in
334
biomass when methanogenesis is active versus completely inhibited.
17 ACS Paragon Plus Environment
Environmental Science & Technology
335 336
Semi-continuous glucose fermentation in MECs To examine whether high FAN concentration affected fermentation, methanogenesis, or
337
both, we performed MEC experiments using semi-continuous operation by replacing the feed
338
medium (i.e., 5 mM glucose) once every two days. Figure 2 summarizes the results for both
339
MECs, which showed similar trends. The current density was stable over the influent FAN
340
range of 0.003 to 0.18 g FAN-N/L (Figure 2A), but a further increase of influent FAN to 0.37 g
341
N/L led to a gradual decrease in the current density after 2 semi-continuous cycles. This pattern
342
suggests that higher FAN slowed acetate oxidation by ARB, since acetate accumulated at the end
343
of semi-continuous cycle (Figure 2B).
344
Similar to the batch MECs, glucose was completely fermented for all FAN conditions,
345
confirming that high FAN concentration did not affect the first fermentation step. As the FAN
346
concentration increased, more electrons ended up in organic acids, and this was associated with
347
17 – 29% decreases in methane yield. Perhaps high FAN altered the fermenters’ intracellular pH
348
homeostasis, leading them to divert electrons away from H2 production and toward formation of
349
more-reduced organic products.37 A shift from H2 to organic acids is consistent with high FAN
350
maintaining a low H2 concentration in the MEC, a change that should be beneficial for
351
syntrophic interactions not involving hydrogenotrophic methanogens.
352 353
Figure 2.
354
355
Distribution of bacterial population in batch MECs
356
Figure 3 and Figure S3 report the sequence analyses of the V4 region in the bacterial 16S
357
rRNA gene in the batch-MEC samples. The majority of bacterial 16S rRNA genes in all samples 18 ACS Paragon Plus Environment
Page 18 of 43
Page 19 of 43
Environmental Science & Technology
358
belonged to 3 phyla: Proteobacteria, Firmicutes, and Bacteroidetes (Figure 3A), which is
359
consistent with previous studies.9,38,39 Several members of phyla Bacteroidetes and Firmicutes
360
are known to ferment sugar, whereas many members of Proteobacteria are known to perform
361
anode respiration.40,41 The lower abundance of Proteobacteria in the suspended-phase (SP)
362
samples agrees with the fact that most of ARB can only use the anode surface as terminal
363
electron acceptor.
364
Among the Proteobacteria, Deltaproteobacteria were the largest sub-group. The
365
predominant genus among Deltaproteobacteria was Geobacter, and their relative abundance
366
increased with higher FAN. Whereas biofilm (Bf) samples from 0.02 g FAN/L had ~ 2.6% of
367
the total genus sequences in Geobacter, samples from 0.18 and 0.37 g FAN/L had 34.5% and
368
62% of the total genus sequences, respectively (Figure 3B). The high abundance of
369
Geobacteraceae in 0.18 and 0.37 g FAN/L Bf samples agrees with their higher CEs compared to
370
the control MEC, and it confirms that the relatively high FAN concentration gave the ARB an
371
ecological advantage that enhance electron flow towards anode respiration versus
372
methanogenesis. Although we observed high relative abundance of Geobacteraceae for the MEC
373
fed with 0.37 g FAN/L, its CE was lower when compared to MEC fed with 0.18 g FAN/L (p-
374
value < 0.05; pairwise t-test). This probably was due to partial inhibition to the biofilm’s ARB at
375
higher FAN. Partial inhibition of acetate uptake by ARB in the biofilm anode made the
376
microbial community less functionally effective for anode respiration, even though the relative
377
abundance of Geobacteraceae in anode’s biofilm remained high.
378 379
Figure 3
19 ACS Paragon Plus Environment
Environmental Science & Technology
380
The relative abundance of Firmicutes was similar in all samples except for 0.37 g FAN/L
381
Bf, but their compositions varied significantly. Figure S4 shows community breakdown at the
382
family level within the phylum Firmicutes. An increase in FAN concentration led to the
383
emergence of Clostridiaceae and Ruminococcaceae families, which belong to the order
384
Clostridiales. Compensating losses occurred for the family Tissierellaceae. Recently, Müller et
385
al. 20 revealed that several members of the Clostridia class, mainly belonging to the orders
386
Clostridiales, can perform syntrophic acetate oxidation (SAO) (See Table 1), which was the
387
predominant pathway for methane production with elevated ammonia levels.20,42
388
We also detected an increase in the relative abundance of Lactobacillales, which include
389
genera Lactobacillus and Enterococcus, at the highest FAN conditions (0.18 and 0.37 g FAN/L).
390
Lactobacillales are reported to play a major role in lactic-acid production in fermentation,43 and
391
this is consistent with the observation of lactate accumulation during glucose fermentation in
392
MECs fed with 0.18 and 0.37 g FAN/L (Figure 1).
393
Supplementary Table S1 and Figure S5A show all the values for the bacterial community
394
diversity metrics for Bf and SP samples. The Chao1, Shannon, and PD values reflect that the Bf
395
samples developed more diverse bacterial community than the corresponding SP samples. Also,
396
the bacterial diversity of Bf and SP samples from 0.18 g FAN/L-fed MEC, which achieved the
397
highest CE (Figure 1), was greater than from the other two MECs. Consistent with the diversity
398
results and based on the equitability coefficient, the evenness increased as FAN concentration
399
increased from 0.02 to 0.18 g FAN/L, while the 0.37 g FAN/L-fed MEC was least even (Figure
400
S5B). Previous studies revealed that microbial communities with greater evenness are often
401
associated with more robustness and functional stability compared to less even microbial
402
communities.29,44
20 ACS Paragon Plus Environment
Page 20 of 43
Page 21 of 43
Environmental Science & Technology
The results of weighted PCoA analysis, Figure 4, show that principal components (PC) 1
403 404
and 2 explained 33.7% and 29.8% of the total bacterial community variations, respectively.
405
Increasing FAN concentration correlated with the PC2 vector, whereas the type of sample (Bf
406
versus SP) correlated with the PC1 vector. The trend along PC2 was associated with the emergence of the orders
407 408
Desulfuromonadales, Lactobacillales, and Synergistales at elevated FAN concentration; to
409
compensate, the relative abundances of other orders, such as Clostridiales and Bacteroidales,
410
decreased in response in change in FAN concentration (Figure 4). Several members of
411
Desulfuromonadales, a sub-group of Deltaproteobacteria, and Lactobacillales are well-known
412
ARB and lactic acid producers, respectively. Although Synergistales, a sub-group of phylum
413
Synergistetes, have been detected in anaerobic digesters treating different wastewater, their
414
functions are still mysterious.45,46 One study suggested that members of this order are potential
415
propionate consumers,47 but another study suggested that they might be SAOB.48 The
416
emergence of those orders, coupled with the decreases of Clostridiales and Bacteroidales, well-
417
known acetic acid-producing fermenters, supports that higher FAN altered glucose fermentation
418
towards the production of higher organic acids (e.g., lactate) and away from H2, rather than the
419
formation of 2 moles of acetate and 4 moles of H2 per mole of glucose, as illustrated in Figure
420
S1.
421
The trend along PC1 was associated with the emergence of the orders Bacteroidales and
422
Lactobacillales, which play an important role in fermentation of complex substrates.
423
Compensating for these increases, the relative abundance of Desulfuromonadales decreased
424
along PC1 vector. These results confirm that most of fermentation occurred in the suspension
425
phase, since we operated MECs in batch mode; ARB dominated the MECs’ biofilm.
21 ACS Paragon Plus Environment
Environmental Science & Technology
426
Figure 4.
427 428
429
Distribution of Archaea population in batch MECs Figure 5 summarizes the sequencing results for Archaea. Hydrogenotrophic
430
methanogens dominated all Bf and SP Archaea communities, regardless the FAN concentration,
431
which is consistent with previous work.9,40 The percentage of Archaea relative to bacteria, based
432
on the prokaryotic library results, decreased as FAN increased, supporting that the methanogens
433
were partially inhibited by higher FAN (Figure S6). For example, the percentage of Archaea
434
relative to Bacteria in Bf samples was about 10.8%, 3.8%, and 4.0% for 0.02 g FAN/L, 0.18 g
435
FAN/L, and 0.37 g FAN/L, respectively. The percentage of Archaea to bacteria was higher for
436
all SP samples than for Bf samples, confirming that the methanogens were predominantly in the
437
suspended phase of MECs operated in batch mode. This trend is reinforced by the qPCR results
438
targeting the mcrA gene (Figure 5A) and by the decrease in CH4 production in the MECs’
439
headspace fed with 0.18 g FAN/L and 0.37 g FAN/L (i.e., 9% and 6% of the original electrons,
440
respectively) compared to control MEC (i.e., 15% of the original electrons) (Figure 1). The
441
mcrA gene copies in the MECs’ anodes fed with high FAN concentrations (0.18 g FAN/L, and
442
0.37 g FAN/L) were statically significant lower (p-value < 0.05; pairwise t-test) compared to
443
control experiment (0.02 g FAN/L). The most likely reason for increasing the mcrA gene copies
444
per reactor volume in the SP sample is either biomass detachment from the biofilm into the
445
supernatant or DNA residual from the inoculum.
446
For all experimental runs, we did not detect any acetoclastic methanogens in all Bf and
447
SP samples, which were dominated by hydrogenotrophic methanogens (i.e., Methanobacterium
448
and Methanobrevibacter) (Figure 5B). For the 0.02 g FAN/L bf, Methanobacterium was
22 ACS Paragon Plus Environment
Page 22 of 43
Page 23 of 43
Environmental Science & Technology
449
dominant (~83% of sequence), while Methanobrevibacter (74% for 0.18 g FAN/L biofilm and
450
91% for 0.37 g FAN/L biofilm) and Methanobacterium (16% for 0.18 g FAN/L biofilm and 11%
451
for 0.37 g FAN/L biofilm) shared the archaeal community for higher FAN concentrations.
452
Several members of both genera are reported to be resistant to ammonia inhibition up to 400 mM
453
(i.e., 5.6 g TAN/L or ~ 0.04 g FAN/L at pH 6.8).49
454
The higher relative abundance of Methanobrevibacter with higher FAN suggests that
455
they contained the strains more resistant to FAN inhibition. Following the FAN increase,
456
Methanobacterium started to outcompete Methanobrevibacter (Figure 5B). A new uncultured
457
Archaea, classified as a candidate genus vadinCA11 (Thermoplasmata sp.), emerged as FAN
458
decreased. This genus was detected in an anaerobic fluidized bed fed with wine-distillation
459
waste,51 and it is potentially halophilic.52 The results suggest that the community shifted toward
460
methanogens tolerant to FAN.
461 462
Figure 5.
463 464 465
High FAN did not affect homo-acetogens Previous work revealed that suppressing methanogenesis was the main reason that high
466
CEs were achieved in MECs.9 Suppressing H2-oxidizing methanogens provides ideal conditions
467
for homoacetogens to oxidize H2 and grow, leading to the generation of acetate, the ideal
468
substrate for ARB. 8,9 If this scenario were true, we should have observed a larger population of
469
homoacetogens with higher FAN concentration. To test this hypothesis, we tracked the relative
470
abundance of homoacetogens by targeting the FTHFS gene using qPCR; the results are in Figure
471
S7. FTHFS gene copies for all Bf and SP samples were comparable, regardless of the FAN
23 ACS Paragon Plus Environment
Environmental Science & Technology
472
concentration. The qPCR data also are consistent with bacterial 16S rRNA gene sequence data,
473
where we detected genus Treponema, a well-known homoacetogenic Spirochaetes (Figure 3B),
474
but at a very low relative abundance (< 1% of total sequences) for all samples. Thus,
475
homoacetogens were consistently present, and changes in their numbers likely were not
476
responsible for enhancing CE.
477 478 479
New insights into metabolic flexibility in MECs Our results provide insight for possible way to decrease CH4 production and increase CE
480
by controlling the fermentation step using FAN, instead of inhibiting methanogens using
481
expensive or toxic chemical inhibitors. Altering fermentation led to less production of H2, which
482
is the “fuel” for methanogens, resulting in less CH4 production and higher CE, even though
483
methanogenesis was not completely suppressed. The lowering of H2 generation was
484
accompanied by more production of short-chain organic acids. However, relatively high FAN
485
(0.37 g FAN/L in my study) might promote SAOB, which are competitors for acetoclastic ARB,
486
introducing the possibility of a new pathway for acetate consumption in MECs fed with
487
fermentable substrates; it is a pathway to be avoided.
488
SAO might be the reason for the lower CE we observed with the MEC fed with 0.37 g
489
FAN/L, compared to MEC fed with 0.18 g FAN/L. Although acetate-based anode respiration is
490
thermodynamically favorable compared to SAO (See Table 1), the half-saturation constant (Ks)
491
for SAOB ranges from 4.7 to 13 mM,48,53 values ~2.5- to 7-fold higher than the maximum Ks
492
reported for mixed-culture ARB (1.86 mM).54 Despite the higher Ks value for acetate, the SAOB
493
may be more tolerant to high FAN, which would give them a kinetic advantage compared to
494
ARB, since ARB often are susceptible to a high FAN concentration. This observation coincides
24 ACS Paragon Plus Environment
Page 24 of 43
Page 25 of 43
Environmental Science & Technology
495
with emergence of Clostridiaceae and Ruminococcaceae families, which are potential SAOB.
496
Further investigation of SAOB in MECs is warranted.
497
An important implication of this study is that methanogenesis does not need to be
498
completely suppressed to achieve high current production and CE from MECs fed with
499
fermentable substrates. Using FAN to suppress methanogenesis is a realistic option for scaling-
500
up MECs during the biodegradation of fermentable substrates, particularly when the feed stream
501
has a high nitrogen content, such as animal manures and landfill leachate.
502 503
Acknowledgements
504 505
We thank Dr. Juan Maldonado, Dr. Zehra Esra Ilhan, and Dr. Prathap Parameswaran for helpful
506
discussion and guidance. This study was partially supported by the endowment fund from Mr.
507
Brian Swette to the Swette Center for Environmental Biotechnology and Graduate and
508
Professional Student Association at Arizona State University. Mohamed Mahmoud was
509
supported by an Egyptian government fellowship (GM 0931).
510
511
Conflicts of interest
512 513
The authors declare that there is no conflict of interest regarding the publication of this paper.
514 515 516 517
25 ACS Paragon Plus Environment
Environmental Science & Technology
518
Supporting Information
519 520
Available in the Supporting Information are: Figure S1-S8: Electron flows from glucose into
521
different electron sinks in an MEC’s anode, Electron balance of MECs at the end of batch-cycle
522
operation, bacterial community distribution at the class level, composition of phylum Firmicutes
523
at the family level, bacterial community diversity and richness results, archaea-to-bacteria ratio,
524
FTHFS gene copies per reactor determined by qPCR at the end of batch cycles; and Table S1:
525
Chao1, Shannon, and PD indices for bacterial sequences.
526
26 ACS Paragon Plus Environment
Page 26 of 43
Page 27 of 43
527
Environmental Science & Technology
References
528 529
1. Torres, C. I. On the importance of identifying, characterizing, and predicting fundamental
530
phenomena towards microbial electrochemistry applications. Curr. Opin. Biotechnol. 2014, 27,
531
107–114.
532 533
2. Lovley, D. R. The microbe electric: conversion of organic matter to electricity. Curr. Opin.
534
Biotechnol. 2008, 19, 564–571.
535 536
3. Angenent, L. T.; Karim, K.; Al-Dahhan, M. H.; Wrenn, B. A.; Domíguez-Espinosa, R.
537
Production of bioenergy and biochemicals from industrial and agricultural wastewater. Trends
538
Biotechnol. 2004, 22, 477–485.
539 540
4. Kiely, P. D.; Regan, J. M.; Logan, B. E. The electric picnic: synergistic requirements for
541
exoelectrogenic microbial communities. Curr. Opin. Biotechnol. 2011, 22, 378–385.
542 543
5. Logan, B. E.; Regan, J. M. Electricity-producing bacterial communities in microbial fuel cells.
544
Trends Microbiol. 2006, 14, 512–518.
545 546
6. Lovley, D. R.; Coates, J. D. Novel forms of anaerobic respiration of environmental relevance.
547
Opin. Microbiol. 2000, 3, 252–256.
548
27 ACS Paragon Plus Environment
Environmental Science & Technology
549
7. Thauer, R. K.; Jungermann, K.; Decker, K. Energy conservation in chemotrophic anaerobic
550
bacteria. Bacteriol. Rev. 1977, 41, 100–180.
551 552
8. Parameswaran, P.; Torres, C. I.; Lee, H. S.; Krajmalnik-Brown, R.; Rittmann, B. E.
553
Syntrophic interactions among anode respiring bacteria (ARB) and Non-ARB in a biofilm anode:
554
electron balances. Biotechnol. Bioeng. 2009, 103, 513–523.
555 556
9. Parameswaran, P.; Zhang, H.; Torres, C. I.; Rittmann, B. E.; Krajmalnik-Brown, R. Microbial
557
community structure in a biofilm anode fed with a fermentable substrate: the significance of
558
hydrogen scavengers. Biotechnol. Bioeng. 2010, 105, 69–78.
559 560
10. Freguia, S.; Rabaey, K.; Yuan, Z.; Keller, J. Syntrophic processes drive the conversion of
561
glucose in microbial fuel cell anodes. Environ. Sci. Technol. 2008, 42, 7937–7943.
562 563
11. Stams, A. J.; Plugge, C. M. Electron transfer in syntrophic communities of anaerobic bacteria
564
and archaea. Nat. Rev. Microbiol. 2009, 7, 568–577.
565 566
12. Rittmann, B. E.; McCarty, P. L. Environmental biotechnology: Principles and applications;
567
McGraw-Hill: New York, 2001.
568 569
13. Zinder, S. H. Physiological ecology of methanogens. In: Ferry JG, editor. Methanogenesis—
570
Ecology, physiology, biochemistry & genetics. New York, NY: Chapman & Hall. 1993, pp 128–
571
206.
28 ACS Paragon Plus Environment
Page 28 of 43
Page 29 of 43
Environmental Science & Technology
572
14. Schuchmann, K.; Müller, V. Autotrophy at the thermodynamic limit of life: a model for
573
energy conservation in acetogenic bacteria. Nat. Rev. Microbiol. 2014, 12, 809–821.
574 575
15. McInerney, M. J.; Struchtemeyer, C. G.; Sieber, J.; Mouttaki, H.; Stams, A. J.; Schink, B.;
576
Rohlin, L.; Gunsalus, R. P. Physiology, ecology, phylogeny, and genomics of microorganisms
577
capable of syntrophic metabolism. Ann. New York Acad. Sci. 2008, 1125, 58–72.
578 579
16. Ishii, S.; Shimoyama, T.; Hotta, Y.; Watanabe, K. Characterization of a filamentous biofilm
580
community established in a cellulose-fed microbial fuel cell. BMC Microbiol. 2008, 8, 1–12.
581 582
17. Zhu, X.; Siegert, M.; Yates, M. D.; Logan, B. E. Alamethicin suppresses methanogenesis and
583
promotes acetogenesis in bioelectrochemical systems. Appl. Environ. Microbiol. 2015, 81, 3863–
584
3868.
585 586
18. Chae, K. J.; Choi, M. J.; Kim, K. Y.; Ajayi, F. F.; Park, W.; Kim, C. W.; Kim, I. S.
587
Methanogenesis control by employing various environmental stress conditions in two-chambered
588
microbial fuel cells. Bioresour. Technol. 2010, 101, 5350–5357.
589 590
19. Mahmoud, M.; Parameswaran, P.; Torres, C. I.; Rittmann, B. E. Electrochemical techniques
591
reveal that total ammonium stress increases electron flow to anode respiration in mixed-species
592
bacterial anode biofilms. Biotechnol. Bioeng. 2017 (In press; DOI: 10.1002/bit.26246).
593
29 ACS Paragon Plus Environment
Environmental Science & Technology
594
20. Müller, B.; Sun, L.; Westerholm, M.; Schnürer, A. Bacterial community composition and fhs
595
profiles of low- and high-ammonia biogas digesters reveal novel syntrophic acetate-oxidising
596
bacteria. Biotechnol. Biofuels. 2016, 9, 48.
597 598
21. Mahmoud, M.; Parameswaran, P.; Torres, C. I.; Rittmann, B. E. Fermentation pre-treatment
599
of landfill leachate for enhanced electron recovery in a microbial electrolysis cell (MEC).
600
Bioresour. Technol. 2014, 151, 151–158.
601 602
22. Yenigün, O.; Demirel, B. Ammonia inhibition in anaerobic digestion: A review. Process
603
Biochem. 2013, 48, 901–911.
604 605
23. Zinder, S. H.; Koch, M. Non-aceticlastic methanogenesis from acetate: acetate oxidation by a
606
thermophilic syntrophic coculture. Arch. Microbiol. 1984, 138, 263–272.
607 608
24. Shimada T.; Morgenroth, E.; Tandukar, M.; Pavlostathis, S. G.; Smith, A.; Raskin, L.; Kilian,
609
R. E. Syntrophic acetate oxidation in two-phase (acid-methane) anaerobic digesters. Water Sci.
610
Technol. 2011, 64, 1812–1820.
611 612
25. Lee, H. S.; Torres, C. I.; Parameswaran, P.; Rittmann, B. E. Fate of H2 in an upflow single-
613
chamber microbial electrolysis cell using a metal-catalyst-free cathode. Environ. Sci. Technol.
614
2009, 43, 7971–7976.
615
30 ACS Paragon Plus Environment
Page 30 of 43
Page 31 of 43
Environmental Science & Technology
616
26. Steinberg, L. M.; Regan, J. M. mcrA-targeted real-time quantitative PCR method to examine
617
methanogen communities. Appl. Environ. Microbiol. 2009, 75, 4435–4442.
618 619
27. Caporaso, J. G.; Lauber, C. L.; Walters, W. A.; Berg-Lyons, D.; Huntley, J.; Fierer, N.;
620
Owens, S. M.; Betley, J.; Fraser, L.; Bauer, M.; Gormley, N.; Gilbert, J. A.; Smith, G.; Knight,
621
R. Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq
622
platforms. ISME J. 2012, 6, 1621–1624.
623 624
28. Takai, K.; Horikoshi, K. Rapid detection and quantification of members of the archaeal
625
community by quantitative PCR using fluorogenic probes. Appl. Environ. Microbiol. 2000, 66,
626
5066–5072.
627 628
29. Werner, J. J.; Knights, D.; Garcia, M. L.; Scalfone, N. B.; Smith, S.; Yarasheski, K.;
629
Cummings, T. A.; Beers, A. R.; Knight, R.; Angenent, L. T. Bacterial community structures are
630
unique and resilient in full-scale bioenergy systems. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,
631
4158–4163.
632 633
30. Caporaso, J. G.; Kuczynski, J.; Stombaugh, J.; Bittinger, K.; Bushman, F. D.; Costello, E. K.;
634
Fierer, N.; Pena, A. G.; Goodrich, J. K.; Gordon, J. I.; Huttley, G. A.; Kelley, S. T.; Knight, D.;
635
Koenig, J. E.; Ley, R. E.; Lozupone, C. A.; McDonald, D.; Muegge, B. D.; Pirrung, M.; Reeder,
636
J.; Sevinsky, J. R.; Tumbaugh, P. J.; Walters, W. A.; Widmann, J.; Yatsunenko, T.; Zaneveld, J.;
637
Knight, R. QIIME allows analysis of high-throughput community sequencing data. Nature
638
Methods 2010, 7, 335–336.
31 ACS Paragon Plus Environment
Environmental Science & Technology
639
31. Cole, J. R.; Wang, Q.; Cardenas, E.; Fish, J.; Chai, B.; Farris, R. J.; Kulam-Syed-Mohideen,
640
A. S.; McGarrell, D. M.; Marsh, T.; Garrity, G. M.; Tiedje, J. M. The Ribosomal Database
641
Project: Improved alignments and new tools for rRNA analysis. Nucl. Acids Res. 2009, 37,
642
D141–D145.
643 644
32. Chao, A. Estimating the population-size for capture recapture data with unequal catchability.
645
Biometrics 1987, 43, 783–791.
646 647
33. Faith, D. P. Conservation evaluation and phylogenetic diversity. Biol. Cons. 1992, 61, 1–10.
648 649
34. Ni, B. J.; Liu, H.; Nie, Y. Q.; Zeng, R. J.; Du, G. C.; Chen, J.; Yu, H. Q. Coupling glucose
650
fermentation and homoacetogenesis for elevated acetate production: Experimental and
651
mathematical approaches. Biotechnol. Bioeng. 2011, 108, 345–353.
652 653
35. Graber, J. R.; Breznak, J. A. Physiology and nutrition of Treponema primitia, an H2/CO2-
654
acetogenic spirochete from termite hindguts. Appl. Environ. Microbiol. 2004, 70, 1307–1314.
655 656
36. Lee, H. S.; Salerno, M. B.; Rittmann, B. E. Thermodynamic evaluation on H2 production in
657
glucose fermentation. Environ. Sci. Technol. 2008, 42, 2401–2407.
658 659
37. González-Cabaleiro, R.; Lema, J. M.; Rodríguez, J. Metabolic energy-based modelling
660
explains product yielding in anaerobic mixed culture fermentations. PLoS ONE 2015, 10,
661
e0126739.
32 ACS Paragon Plus Environment
Page 32 of 43
Page 33 of 43
Environmental Science & Technology
662
38. Mahmoud, M.; Parameswaran, P.; Torres, C. I.; Rittmann, B. E. Relieving the fermentation
663
inhibition enables high electron recovery from landfill leachate in a microbial electrolysis cell.
664
RSC Advances 2016, 6, 6658 – 6664.
665 666
39. Sanchez-Herrera, D.; Pacheco-Catalan, D.; Valdez-Ojeda, R.; Canto-Canche, B.;
667
Dominguez-Benetton, X.; Domínguez-Maldonado, J.; Alzate-Gaviria, L. Characterization of
668
anode and anolyte community growth and the impact of impedance in a microbial fuel cell. BMC
669
Biotechnol. 2014, 14, 102.
670 671
40. Siegert, M.; Li, X. F.; Yates, M. D.; Logan, B. E. The presence of hydrogenotrophic
672
methanogens in the inoculum improves methane gas production in microbial electrolysis cells.
673
Front. Microbiol. 2015, 5, 778.
674 675
41. Rismani-Yazdi, H.; Carver, S. M.; Christy, A. D.; Yu, Z.; Bibby, K.; Peccia, J.; Tuovinen, O.
676
H. Suppression of methanogenesis in cellulose-fed microbial fuel cells in relation to
677
performance, metabolite formation, and microbial population. Bioresour. Technol. 2013, 129,
678
281–288.
679 680
42. Westerholm, M.; Levén, L.; Schnürer, A. Bioaugmentation of syntrophic acetate-oxidizing
681
culture in biogas reactors exposed to increasing levels of ammonia. Appl. Environ. Microbiol.
682
2012, 78, 7619–7625.
683
33 ACS Paragon Plus Environment
Environmental Science & Technology
684
43. De Vrieze, J.; Saunders, A. M.; He, Y.; Fang, J.; Nielsen, P. H.; Verstraete, W.; Boon, N.
685
Ammonia and temperature determine potential clustering in the anaerobic digestion microbiome.
686
Water Res. 2015, 75, 312–323.
687 688
44. Wittebolle, L.; Marzorati, M.; Clement, L.; Balloi, A.; Daffonchio, D.; Heylen, K.; De Vos,
689
P.; Verstraete, W.; Boon, N. Initial community evenness favours functionality under selective
690
stress. Nature 2009, 458, 623–626.
691 692
45. Zamanzadeh, M.; Hagen, L. H.; Svensson, K.; Linjordet, R.; Horn, S. J. Anaerobic digestion
693
of food waste – Effect of recirculation and temperature on performance and microbiology. Water
694
Res. 2016, 96, 246–254.
695 696
46. Delbès, C.; Moletta, R.; Godon, J. J. Bacterial and archaeal 16S rDNA and 16S rRNA
697
dynamics during an acetate crisis in an anaerobic digestor ecosystem. FEMS Microbiol. Ecol.
698
2001, 35, 19–26.
699 700
47. Hagen, L. H.; Vivekanand, V.; Linjordet, R.; Pope, P. B.; Eijsink, V. G. H.; Horn, S. J.
701
Microbial community structure and dynamics during co-digestion of whey permeate and cow
702
manure in continuous stirred tank reactor systems. Bioresour. Technol. 2014, 171, 350–359.
703 704
48. Ito, T.; Yoshiguchi, K.; Ariesyady, H. D.; Okabe, S. Identification of a novel acetate-utilizing
705
bacterium belonging to Synergistes group 4 in anaerobic digester sludge. ISME J. 2011, 5, 1844–
706
1856.
34 ACS Paragon Plus Environment
Page 34 of 43
Page 35 of 43
Environmental Science & Technology
707
49. Sprott, G. D.; Patel, G. B. Ammonia toxicity in pure cultures of methanogenic bacteria. Syst.
708
Appl. Microbiol. 1986, 7, 358–363.
709 710
51. Godon, J. J.; Zumstein, E.; Dabert, P.; Habouzit, F.; Moletta, R. Molecular microbial
711
diversity of an anaerobic digestor as determined by small-subunit rDNA sequence analysis. Appl.
712
Environ. Microbiol. 1997, 63, 2802–2813.
713 714
52. Durbin, A. M.; Teske, A. Archaea in organic-lean and organic-rich marine subsurface
715
sediments: an environmental gradient reflected in distinct phylogenetic lineages. Front.
716
Microbiol. 2012, 3, 168.
717 718
53. Rivera-Salvador, V.; López-Cruz, I. L.; Espinosa-Solares, T.; Aranda-Barradas, J. S.; Huber,
719
D. H.; Sharma, D.; Toledo, J. U. Application of Anaerobic Digestion Model No. 1 to describe the
720
syntrophic acetate oxidation of poultry litter in thermophilic anaerobic digestion. Bioresour.
721
Technol. 2014, 167, 495–502.
722
723
54. Lee, H. S.; Torres, C. I.; Rittmann, B.E. Effects of substrate diffusion and anode potential on
724
kinetic parameters for anode-respiring bacteria. Environ. Sci. Technol. 2009, 43, 7571–7577.
35 ACS Paragon Plus Environment
Environmental Science & Technology
725
List of Tables
726 727
Table 1. Overview of reactions involving acetate and H2
36 ACS Paragon Plus Environment
Page 36 of 43
Page 37 of 43
728
Environmental Science & Technology
Table 1
729
Process (1) Glucose fermentation: C6H12O6 + 2H2O → 2 CH3COO− + 4H2 + 2CO2 + 2H+
∆G°rxn (kJ/e– eq) a – 8.59
(2) Acetoclastic methanogenesis: CH3COO− + H2O → CH4 + HCO3−
– 3.88
(3) Hydrogenotrophic methanogenesis: HCO3− + 4H2 + H+ → CH4 + 3H2O
−16.38
(4) Homoacetogenesis: 2 CO2 + 4H2 → CH3COO− + H+ + 2H2O
− 11.88
(5) Acetoclastic anode respiration b: CH3COO− + 2H2O → 2CO2 + 7H+ + 8e−
− 19.30
(6) Hydrogenotrophic anode respiration b: 2H2 → 2H+ + 2e−
− 9.65
(7) Acetate oxidation to H2: CH3COO− + 4H2O → 2HCO3− + 4H2 + H+
+13.10
730
a
We calculated ∆G°rxn at standard conditions (i.e., 298 K, 1 atm for gases, pH 7, and 1 M for
731
soluble reactants) based on the thermodynamic data provided in Thauer et al.11; b Near-optimum
732
anode potential = +0.2 V versus SHE.15
733
37 ACS Paragon Plus Environment
Environmental Science & Technology
734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756
Figures caption Figure 1. Electron distributions in MECs fed with 5-mM glucose at an initial pH of 8.1: (A) 0.02 g FAN/L or 0.2 g TAN/L (final pH = 6.85), (B) 0.18 g FAN/L or 2 g TAN/L (final pH = 7.10), and (C) 0.37 g FAN-N/L or 4 g TAN/L (final pH = 6.89). Figure 2. Performance of semi-continuous MECs fed with different FAN concentrations: (A) Current generation profile of duplicate MECs versus time. (B) Electron balance of MECs at the end of semi-continuous cycle operation in panel A. Figure 3. Bacterial community sequencing results. (A) bacterial community distribution at the phylum level. Phyla with less than 1% of total sequences are grouped as “others”. (B) bacterial community distribution at the genus level. Genera with less than 1% of total sequences are grouped as “others”. Figure 4. Weighted Unifrac analysis shows that the relative abundance of order-level phylotypes on the Principal Coordinates. FAN concentration determined the main phylotypes that drove the community structures in MECs. Figure 5. Archaeal community analyses. (A) mcrA gene copies per reactor determined by qPCR at the end of batch cycles. All error bars show standard deviations of triplicate measurements. (B) Archaeal community distribution at the genus level. Genera with less than 1% of total sequences are grouped as “others”.
38 ACS Paragon Plus Environment
Page 38 of 43
Page 39 of 43
757 758 759
Environmental Science & Technology
Figure 1
39 ACS Paragon Plus Environment
Environmental Science & Technology
760 761
762 763 764 765
Figure 2
40 ACS Paragon Plus Environment
Page 40 of 43
Page 41 of 43
Environmental Science & Technology
766 767
768 769 770
Figure 3
41 ACS Paragon Plus Environment
Environmental Science & Technology
771 772 773 774
Figure 4
42 ACS Paragon Plus Environment
Page 42 of 43
Page 43 of 43
Environmental Science & Technology
775
776 777 778 779 780
Figure 5
781
43 ACS Paragon Plus Environment