APPLICATION OF STABLE ISOTOPE PROBING AND HIGH-THROUGHPUT
SEQUENCING TO IDENTIFY MICROORGANISMS IN MICROBIAL FUEL CELLS
By
Yang Song

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirement
for the degree of
Environmental Engineering – Master of Science
2014

ABSTRACT
APPLICATION OF STABLE ISOTOPE PROBING AND HIGH-THROUGHPUT
SEQUENCING TO IDENTIFY MICROORGANISMS IN MICROBIAL FUEL CELLS
By
Yang Song
Microbial fuel cells (MFCs) have the potential for use in both waste degradation as well as
energy generation. An understanding of anode chamber microbial community could
contribute to optimizing these applications. The combination of stable isotope probing (SIP)
and high-throughput sequencing can provide information on the activity and abundance of
microorganisms while existing in mixed cultures. For this study, eight sets of MFCs anode
chamber microcosms were analyzed to profile the microbial community and identify the
microorganisms involved in carbon uptake from the amended substrates. The MFCs were
amended with labeled (13C) or unlabeled sodium acetate and glucose and the external
resistance was manipulated to two levels (10 ohms and 1000 ohms). For the sodium acetate
amended MFCs, the dominant phyla were the Proteobacteria, Bacteroidetes, Firmicutes, and
Actinobacteria. For the glucose amended MFCs, similar dominant phyla were detected,
except Actinobacteria. Through comparing enrichment factors between the labeled and
unlabeled fractions, 14 phylotypes were found to be responsible for label uptake in the
sodium acetate amended MFCs and 13 phylotypes were responsible in the glucose amended
MFCs. Among these, unclassified Parachlamydiaceae (Chlamydia), Azospirillum
(Proteobacteria), and unclassified Rhodocyclaceae (Proteobacteria) were the primary
phylotypes uptaking the labeled carbon. To my knowledge, this is the first study to combine
SIP with high-throughput sequencing to identify the active microorganisms in MFCs.

ACKNOWLEDGEMENTS

I would like to acknowledge those people who helped make this thesis possible. Firstly, I
would like to thank my parents for all their love and support during this research. Without
their love, the research would be more stressful. Secondly, I owe a tremendous thank to my
fellow graduate students, Indumathy Jayamani, Fernanda Paes as well as Yogendra Kanitkar
who provide me with not only technical support but also spiritual encouragement to complete
this research. I also appreciate the cooperation and help from Wisconsin
University-Milwaukee environmental biotechnology & bioenergy laboratory. The MFCs were
set up and operated by researchers under the direction of Dr. Zhen He (Associate Professor,
University of Wisconsin-Milwaukee).
Finally, I would like to give my sincere thanks to my advisor Dr. Alison M. Cupples, who
directed me through whole research and gave me valuable courage to finish this thesis. I
would also like to thank other two committee members Dr. Susan J. Masten and Dr. Irene
Xagoraraki.

iii

TABLE OF CONTENTS

LIST OF TABLES……………………………………………………………………………..v
LIST OF FIGURES……………………………………………………………………...……vi
1. INTRODUCTION…………………………………………………………………………1
1.1 Introduction to MFCs……………………………………………………………...….1
1.2 High-throughput sequencing……………………………………………………….....5
1.2.1 Concept and principles………………………………………………...…..5
1.2.2 Categories……………………………………………………………….....6
1.2.2.1 454 Pyrosequencing technology…………………………...6
1.2.2.2 Illumina’s Solexa sequencing technology…………………7
1.2.2.3 SOLiD technology…………………………………………8
1.2.3 Advantages………………………………………………………………...9
1.2.4 Potential drawbacks………………………………………………………..9
1.3 Stable isotope probing…………………………………………………………….…10
2. MATERIALS AND METHOD…………………………………………………..………13
2.1 Chemicals……………………………………………………………………………13
2.2 Operation of MFCs……………………………………………….……………….....13
2.3 DNA extracts…………………………………………………………….…………..15
2.4 Isopycnic centrifugation……………………….…………………………………….15
2.5 High-throughput amplicon sequencing (Illumina MiSeq)…………………...……...16
3. RESULTS……………………………………………………………….………………18
3.1 Fraction generation and sequencing summary……………………………………...18
3.2 Illumina sequencing results for total DNA……………………………………….….20
3.2.1 Phyla from total DNA extracts…………………………………………...20
3.2.2 Families from total DNA extracts………………………………………..21
3.3 Identification of phylotypes responsible for label uptake...………………………...27
4. DISCUSSION…………………………………………………………………………...41
5. CONCLUSION…………………………………………………………………………45
REFERENCES………………………………………………………………………..……...46

iv

LIST OF TABLES

Table 1: Microbes used in MFCs……………………………………………………………..4
Table 2: MFCs running data summary……………………………………………………….14
Table 3: Summary of MiSeq Illumina data generated from MFCs total DNA samples as well
as fractions in labeled and unlabeled samples……………………………………………….19
Table 4: Buoyant density (BD) of fractions chosen for sequencing from MFCs sample (DUP
is abbreviation of Duplicate)…………………………………………………………………20
Table 5: Summary of genera enriched in both duplicates for sodium acetate (10 ohms/ 1000
ohms) and glucose (10 ohms/ 1000 ohms) fed MFCs samples………………………………39

v

LIST OF FIGURES

Figure 1: Working principle for MFCs……………………………………………………….2
Figure 2: Summary of nucleic acid based isotope probing method…………………………11
Figure 3: Comparison of relative abundance of sequences in total genomic DNA extracted
from eight MFCs samples (A: Sodium Acetate; G: Glucose; 10: 10ohms; 1000: 1000ohms; L:
Labeled; UL: Unlabeled; T: Total)……………………...……………………………………23
Figure 4: Relative abundance of Proteobacteria (A), Bacteroidetes (B), Firmicutes (C), and
Actinobacteria (D) from sodium acetate fed MFCs total DNA extracts classified at the family
level (unless unclassified) (A: Sodium acetate; 10:10 ohms; 1000: 1000 ohms; L: Labeled;
UL: Unlabeled; T: Total)……………………………………………………………..………24
Figure 5: Relative abundance of Proteobacteria (A), Bacteroidetes (B), and Firmicutes (C)
from glucose fed MFCs total DNA extracts classified at the family level (unless unclassified)
(G: Glucose; 10:10 ohms; 1000: 1000ohms; L: Labeled; UL: Unlabeled; T: Total)……...…26
Figure 6: Enrichment factor of select OTUs (at genus level unless unclassified) in the heavy
fractions of the labeled sodium acetate 10 ohms samples to the unlabeled sodium acetate 10
ohms samples for two duplicates (A and B)…………………………………………..……...31
Figure 7: Enrichment factor of select OTUs (at genus level unless unclassified) in the heavy
fractions of the labeled sodium acetate 1000 ohms samples to the unlabeled sodium acetate
1000 ohms samples for two duplicates (A and B)……………………………………..……..33
Figure 8: Enrichment factor of select OTUs (at genus level unless unclassified) in the heavy
fractions of the labeled glucose 10 ohms samples to the unlabeled glucose 10 ohms samples
for two duplicates (A and B)……………………………………………………………..…..35
Figure 9: Enrichment factor of select OTUs (at genus level unless unclassified) in the heavy
fractions of the labeled glucose 1000 ohms samples to the unlabeled glucose 1000 ohms
samples for two duplicates (A and B)……………………………………………………..…37

vi

1. INTRODUCTION
1.1 Introduction to MFCs
Microbial fuel cells are devices that use microorganisms to transfer chemical energy to
electrical energy. Typically, MFCs consist of anode and cathode chambers which are divided
by one cation specific membrane. In the anode chamber, anaerobic conditions maximize
reducing equivalent yield through promoting acidogenic fermentative metabolism. Organisms
in anode chambers are capable of reproduction as well as transfer reducing equivalents to an
exterior electron acceptor. Extracellular electron transfer can be achieved through
fermentative pathways, acquisition by soluble electron shuttle compounds with reduction and
through bacterial pili, i.e. nanowires (1). Electrons are transferred to cathode chambers with
an external electric loop circuit. In the cathode chamber, electrons are consumed in oxidative
conditions by terminal electron acceptors such as oxygen, nitrate and ferric ions (2). Figure 1
(3) illustrates a basic working principle for one MFCs with glucose as the electron donor.

1

Figure 1: Working principle for MFCs (3)

Recently, there has been increasing attention towards MFCs due to their dual functionality for
organic waste degradation as well as clean energy production. One ‘Scopus’ search survey on
the keywords “microbial fuel cell” illustrates there has been an increase of ~60 fold in the
papers published during 1998 to 2008 (4). During the past several years, multiple research
areas have focused on MFCs as a new bioenergy resource. For example, researchers have
focused on the comparison of anode bacterial communities based on different choices of
electron donor (5); the use of new permeable membrane in microbial fuel cells (6); the
influence of external resistance on electrogenesis (7); and the cathodic limitations in MFCs
(8).

2

Under different substrate addition conditions, various MFCs power generation efficiencies
have been reported. For instance, Logan reported that the power density produced in a
graphite fiber brush anodes cube MFC with acetate substrate can be as much as 2400 mW/m2
or 73mW/m3 (9). Also, Rabaey indicated that a glucose fed-batch MFC using 100 mM ferric
cyanide as cathode oxidant with power density up to 216 W/m3 (10). Other substrates, such
as lignocellulosic biomass, wastewater, landfill leachates, have also been explored (4). In
most cases, acetate is preferred as the anode electron donor due to its high coulombic
efficiency (CE). Glucose usually has high power density (PD) when used as an electron donor,
while it generates much less CE than acetate due to its fermentable characteristic (11) .
Researchers have also tested various inocula to MFCs for improving energy generation
efficiency. Clostridium cellulolyticum, G. sulfurreducens (12), Enterobacter cloacae (13),
Schewanella (14), domestic wastewater (15), or anaerobic sludge (16) have been added to
MFCs to evaluate the corresponding current density generation.

Although researchers have already established the platform for MFCs improvement and
industrial application, more knowledge is still needed to understand the microbial community
structure as well as the significance of different microorganisms for improving energy
production (17). Previous research has focused on the anode microbial community. For
example, strain ISO2-3, affiliated with the Aeromonas sp. within the Gammaproteobacteria
was reported to be important for current generation through oxidation of glucose or hydrogen
(18). Moreover, Jung et al. reported that the characterization of MFCs under different
substrate additions (acetate, lactate and glucose) with anaerobic sludge inoculum had shown

3

functional communities affiliated with Geobacter sulfurreducens based on 16S rDNA
targeted denaturing gradient gel electrophoresis (DGGE) (5). In a recent study (19), with
inoculum of primary clarifier effluent from a municipal wastewater plant, it was suggested
that the families Geobacteraceae and Desulfobulbaceae correlate to electricity generation in
the biofilm. Results of phylum level studies also showed Proteobacteria, Bacteroidetes, and
Firmicutes were relatively abundant. A few papers also mentioned Actinobacteria could
either act as a dominant phylum or at least be present in the tested microbial community (20,
21). A summary of microorganisms in MFCs is shown (Table 1) below (22).

Microbes

Table 1: Microorganisms used in MFCs
Substrate
Applications

Actinobacillus
succinogenes
Aeromonas hydrophila
Alcaligenes faecalis,
Enterococcus
gallinarum,
Pseudomonas
aeruginosa

Clostridium beijerinckii

Clostridium butyricum

Desulfovibrio
desulfuricans
Erwinia dissolven
Escherichia coli
Geobacter
metallireducens

Glucose

Neutral red or thionin as electron mediator

Acetate

Mediator-less MFC

Glucose

Self-mediate consortia isolated from MFC
with a maximal level of 4.31 W m− 2

Starch,
glucose,
lactate,
molasses
Starch,
glucose,
lactate,
molasses

Fermentative bacterium

Fermentative bacterium

Sucrose

Sulphate/sulphide as mediator

Glucose
Glucose
sucrose

Ferric chelate complex as mediators

Acetate

Mediator-less MFC

Mediators such as methylene blue needed

4

Table 1 (Cont’d)
Microbes

Substrate

Applications

Geobacter
sulfurreducens

Acetate

Mediator-less MFC

Gluconobacter oxydans

Glucose

Klebsiella pneumoniae

Glucose

Lactobacillus
plantarum
Proteus mirabilis

Mediator (HNQ, resazurin or thionine)
needed
HNQ as mediator biomineralized
manganese as electron acceptor

Glucose

Ferric chelate complex as mediators

Glucose

Thionin as mediator

Pseudomonas
aeruginosa

Glucose

Pyocyanin and phenazine-1-carboxamide
as mediator

Rhodoferax
ferrireducens

Glucose,
xylose
sucrose,
maltose

Mediator-less MFC

Shewanella oneidensis

Lactate

Shewanella
putrefaciens
Streptococcus lactis

Lactate,
pyruvate,
acetate,
glucose
Glucose

Anthraquinone-2,6-disulfonate (AQDS)
as mediator
Mediator-less MFC; but incorporating an
electron mediator like Mn (IV) or NR into
the anode enhanced the electricity
production
Ferric chelate complex as mediators

Many factors appear to affect the microbial community present, including substrate type,
inoculum, cathode chamber types, and the experimental conditions. More information on the
dominant microorganisms in the anode chamber has the potential to contribute to an
understanding of the electron transfer process. Also, such information could potentially be
used to optimize waste treatment with MFCs.

1.2 High-throughput Sequencing
1.2.1 Concept and principles

5

High throughput sequencing is an efficient approach for obtaining large amounts of genetic
information. The technique can follow DNA extraction and amplification of 16S rRNA genes
(23). More specifically, high throughput sequencing involves two processes: i) the generation
of DNA libraries through PCR clonal amplification; ii) DNA sequencing by synthesis which
is determined by sequential addition of nucleotides to the complementary strand without a
physical separation process.

1.2.2 Categories
Among the next generation sequencing platforms available, the Roche/454 FLX, the
Illumina/Solexa Genome Analyzer, and the Applied Biosystems (ABI) SOLiD Analyzer are
the predominant platforms that are broadly used (24). Other available platforms and
developing ones are likely to become more popular in next few years due to their contribution
to faster sequencing and lower prices (25). Newly emerging third-generation sequencing
techniques could run without an initial DNA amplification process (26). An overview of the
three platforms listed above is provided below.

1.2.2.1 454 Pyrosequencing technology
Pyrosequencing is a method of DNA sequencing based on synthesis. It depends on the
detection of photons produced during nucleotide incorporation (27). The Roche/454 FLX
genome sequencer, based on pyrosequencing technology (28), was available for purchase in
2004. Major procedures in this application include sample library preparation,
emulsion-based clonal amplification, bead recovery and enrichment, Pico TiterPlatTM

6

preparation, sequencing and detection. Sample preparation in 454 pyrosequencing technology
is more simplified than Sanger sequencing (traditional sequencing). To test the DNA
sequence, the four nucleotides are sequentially provided through the plates with a polymerase
enzyme and primer. Synthesis of the complementary strand can start and emit photons as a
result of pyrophosphate release, which is captured by a CCD camera. By comparing the light
intensity corresponding to different nucleotides addition sequence, strands sequences can be
identified (24). According to one comparison with the advanced Sanger-based capillary
electrophoresis platform, the 454 system can produce around 100 times higher throughput on
sequence data (27). Currently, the limitation for the 454 system generators is that this
technology can only contain relatively short read lengths and low ratings for some genomic
regions’ base reading accuracies.

1.2.2.2 Illumina’s Solexa sequencing technology
Since it became available on the commercial market in 2006, the Solexa sequencing platform
has been broadly accepted as the most adaptable and easiest to use genomic analyzer. The
gene library preparation process for Solexa sequencing is similar to 454 technology. However,
the following substrate template capture and amplification procedure is different. It depends
on solid-phase bridge PCR technique for amplifying targeted DNA that would randomly
connect to adaptors and form clusters on the surface of a flow cell (27). After replicating cell
clusters of approximately 1000 copies of one-stranded DNA fragments, the reaction mixture
for sequencing and DNA synthesis is added onto the surface. The mixture pool for the
following reactions contains primers, four reversible terminator nucleotides labeled with

7

fluorescent dye, and DNA polymerase. After integration with the DNA strands, the CCD
camera can capture the images of fluorescent dye on the terminator nucleotides and the
position where they incorporate with the DNA strands. Then the terminator group as well as
the fluorescent dye are removed followed by the next round of synthesis reactions. There are
reports showing that, at a minimum, 40 million pairs of strands can be synchronously
determined in parallel resulting in high sequence throughput (24). Illumina’s sequencing by
synthesis (SBS) technology works both for single read and paired-end libraries. The SBS
technology combines short inserts and longer reads, which increase the capability to fully
characterize genomes. Its wide sample preparation process enables various sequencing
applications, containing whole-genome sequencing, de novo sequencing, candidate region
targeted resequencing, DNA sequencing, RNA sequencing, methylation analysis, and
protein-nucleic acid interaction analysis
(http://www.illumina.com/technology/sequencing_technology.ilmn). In 2008, the updated
Genome Analyzer Illumina HiSeq 2000 was able to produce single reads of 2×100 base pairs,
and around 200 giga base pair (Gbp) of short sequences each run. The Illumina MiSeq
platform produces 250 bp paired reads.

1.2.2.3 SOLiD technology
The Applied Biosystem SOLiD sequencing system based on ligation was launched in 2007.
This technology applies an emulsion PCR approach with beads to amplify the DNA
fragments for parallel sequencing. For the sequencing process, after a primer is attached to
the adapter, a mixture of oligonucleotide octamers is hybridized to the DNA fragments

8

followed by adding the ligation mixture. The octamers are fluorescent labelled di-base probes
that compete for ligation through interrogating the first and second bases in each ligation
reaction. The first two bases are recognized by characterizing their corresponding fluorescent
labels. Then, the fluorescent labels are removed enzymatically with the departure of the last
three bases on the octamer. The hybridization and ligation cycles are then repeated, in which
bases 6 and 7 as well as bases 11 and 12 can be determined. Moreover, the sequencing
process can be continued through adding another primer, shorter by one base, to test the
remaining bases, for example, bases 0 and 1, bases 5 and 6 etc. The primer offsetting scheme
allows a universal primer to hybridize to DNA templates along the entire fragment within five
cycles. Also, each base can be sequenced twice during the entire cycle. Because each base is
determined by different fluorescent labels, the misreading rate is largely reduced and the
accuracy rate can be as much as 99.94%. The SOLiD platform can provide accurate data,
however the longer time needed for DNA library preparation can be a shortcoming (24, 25).

1.2.3 Advantages
High throughput sequencing can be fast and accurate. It can help to speed up the genotype
characterization and also broaden the pools of determination targets. Researchers have used
these approaches to investigate the expression and patterns of functional genes in microbial
communities (29, 30). Moreover, transcriptomics (30-32) as well as plasmids (33) can also be
targeted for high throughput sequencing.

1.2.4 Potential drawbacks

9

High throughput sequencing technologies are relatively expensive. The Roche 454
sequencing technology generates a smaller amount of data, which is usually between 0.25
and 1 Gbp sequence information per plate (34). Meanwhile, difficulties exist for 454
sequencing technology for dealing with homopolymeric DNA sequences. Both the Illumina
Solexa sequencing system as well as the ABI SOLiD technology generate a larger amount of
data. However, the Illumina Solexa sequencing system is limited by read lengths and the ABI
SOLiD technology usually requires a longer time for sequencing. For this research, the
Illumina platform was used because of local availability and support for the analysis of the
data produced (Mothur, see methods section).

1.3 Stable Isotope Probing
Stable isotope probing is a molecular technique that allows identification of metabolically
active microorganisms from diverse microbial communities through tracking the flow of
isotopically labeled atoms incorporated into biomass (35, 36). According to a review on SIP,
this technique broadens the scope for linking function with identification due to its
independence from cultivation (37).

SIP involves the exposure of the microbial community to a labeled substrate. Microorganisms
assimilate the stable isotope into biomass including their nucleic acids (37). Researchers have
used three major biomarkers for detection during SIP: polar lipid derived fatty acids (PLFAs),
DNA, and rRNA. Other biomarkers such as mRNA and proteins have recently been
introduced for their strong sensitivity to isotopic enrichment. While both DNA and RNA are

10

taxonomically informative, DNA is more often employed (38).

DNA based SIP (DNA-SIP) dates back to 2000, when Radajewski et al. (39) detected two
groups of bacteria, α-Proteobacteria and Acidobacterium, which were responsible for the
degradation of methanol. 13C is the most common isotope chosen for SIP based on the fact
that carbon is the most abundant element in DNA. Even though buoyant density of DNA is
affected by its guanine-cytosine (G+C) content, the heavy stable isotope components enhance
the buoyant density of labeled DNA (40). Labeled DNA and unlabeled DNA are separated
through isopycnic centrifugation based on different buoyant densities. The centrifugation is
typically conducted in a CsCl solution and the heavier DNA can be found at the bottom of
centrifugation tubes. It is also possible to visualize the DNA bands in tubes through adding
ethidium bromide (EtBr) before centrifugation, which can show the location of bands under
UV light. A figure summary of nucleic acid based stable isotope probing method (41) is
shown here (Figure 2).

Fractioning – separates heavy from light
DNA

Ultracentrifugation
Nucleic
acid
extraction
Samples are amended

Molecular analysis –
to identify organisms that
incorporated heavy label

Lighter fractions

labeled (13C) or unlabeled

Heavier fractions

Figure 2: Summary of nucleic acid based isotope probing method

Despite its advantages, SIP also has some limitations, for instance, the limited availability

11

and high cost of labeled substrates. When designing SIP-based experiments, researchers also
need to consider the proper substrate concentrations as well as the incubation times to
minimize potential cross-feeding and over-enrichment (38).

In this study, SIP and high-throughput sequencing (Illumina) were used to profile the
microbial community and identify the microorganisms involved in label uptake (13C) from
eight sets of MFCs anode DNA samples. The MFCs were amended with labeled (13C) or
unlabeled sodium acetate and glucose and the external resistance was manipulated to two
levels (10 ohms and 1000 ohms). Initially TRFLP was coupled to SIP for this research,
however, this method did not provide the resolution needed to identify the microorganisms
that uptaking the label. Therefore, the molecular analysis approach was switched to high
throughput sequencing.

This research is a collaboration with another research group. The MFCs were set up and
operated by researchers under the direction of Dr. Zhen He (Associate Professor, University
of Wisconsin-Milwaukee). These researchers removed the anode and sent them to MSU. All
other activities (DNA extraction, SIP, Illumina sequencing, data analysis) were completed at
MSU.

12

2. MATERIALS AND METHODS
2.1 Chemicals
Reagents were purchased from one or more of the following vendors: Fisher Bioreagent
(Thermo Fisher Scientific, NJ, USA), Integrated DNA Technologies (Coralville, IA, USA)
and Sigma-Aldrich (St.Louis, MO, USA).

2.2 Operation of MFCs
The MFCs operation and sample collection occurred at the University of
Wisconsin-Milwaukee. The inoculum source of the two-chamber MFCs system was
anaerobic sludge from a local municipal wastewater treatment plant. The anode material was
carbon cloth with surface area of 12 cm2 and the cathode material was carbon brush. Both
electrodes were soaked in a 100 mL solution in the anode and cathode chambers. The solution
in anode chamber contained 0.3 g/L NH4Cl, 1 g/L NaCl, 0.03 g/L MgSO4, 0.04 g/L CaCl2,
0.2 g/L NaHCO3, 5.3 g/L KH2PO4, 10.7 g/L K2HPO4 and 1 mL/L trace solution. Trace
solution contained 10000 mg/L FeCl2-4H2O; 2000 mg/L CoCl2-6H2O; 1000 mg/L EDTA;
500 mg/L MnCl2-4H2O; 142 mg/L NiCl2-6H2O; 123 mg/L Na2SeO3; 90 mg/L AlCl3-6H2O;
69 mg/L Na2MoO4-2H2O; 50 mg/L ZnCl2; 50 mg/L H3BO3; 38 mg/L CuCl2-2H2O; 1 mL/L
HCl (37.7% solution). The solution in the cathode chamber contained potassium ferricyanide
at a concentration of 500 mM/L. All water was deionized water. The MFCs system in this
study was the H-type system as described elsewhere (42). The anode and cathode chambers
were separated by a cation exchange membrane (Ultex CMI 7000, Membranes International,
lnc., Glen Rock, NJ, USA ).

13

The substrates for MFCs startup were unlabeled sodium acetate or glucose with an initial
concentration of 1 g/L. After ~30 days operation, labeled (13C) and unlabeled substrates were
amended in MFCs (1 g/L). Eight sets MFCs were amended with labeled (13C) or unlabeled
sodium acetate and glucose and the external resistance was manipulated to two levels (10
ohms and 1000 ohms). All MFCs were operated at room temperature. After ~14 days, anode
electrodes were collected from the MFCs and stored at -20 °C. A summary of the MFCs
running characteristics is shown below (Table 2).

Table 2: MFCs running data summary
Samples
12

C sodium acetate

Power density/

External
resistance/ohms

Current/ mA

Coulombic
efficiency/%

10

4.17

37.8

1.74

10

5.06

52.8

2.56

1000

0.62

11.8

3.84

1000

0.59

8.0

3.48

10

1.80

24

0.32

10

1.63

29.8

0.27

1000

0.60

13.5

3.60

1000

0.54

14.5

2.92

W/m3

10 ohms
13

C sodium acetate
10 ohms

12

C sodium acetate
1000 ohms

13

C sodium acetate
1000 ohms

12

C glucose 10
ohms

13

C glucose 10
ohms

12

C glucose 1000
ohms

13

C glucose 1000
ohms

14

2.3 DNA extraction
Anode chamber samples were collected after two weeks of MFCs operation and were then
sent to MSU. These samples were stored at -20 °C until DNA extraction occurred. Total
genomic DNA was extracted using the Power Soil DNA extraction kit, following the
manufacturer’s instruction (MO BIO Laboratories, Inc. Carlsbad, CA). Eight samples were
investigated, including materials obtained from a MFC amended with i) unlabeled sodium
acetate operated at 10 ohms or 1000 ohms, ii) labeled sodium acetate operated at 10 ohms or
1000 ohms, iii) unlabeled glucose operated at 10 ohms or 1000 ohms, iv) labeled glucose
operated at 10 ohms or 1000 ohms. Extracted DNA were quantified with the Nanodrop-1000
(Thermo Fisher Scientific lnc.).

2.4 Isopycnic centrifugation
The extracted DNA was ultracentrifuged in cesium chloride gradients separately to obtain
density-resolved gradients and fractions. For each MFC treatment, replicate DNA samples
were subject to ultracentrifugation (16 DNA samples were ultracentrifuged). For each sample,
approximately 10 µg of total genomic DNA (except for duplicate one of sodium acetate fed
1000 ohms which involved 20 µg) was mixed with a Tris-EDTA (pH 8.0) buffer and CsCl
solution. This mixture was added to a 5.1 mL Quick-Seal polyallomer tubes (1.3 x 5.1 cm,
Beckman Coulter) the buoyant density (BD) of this mixture was adjusted to around 1.72
g/mL using a model AR200 digital refractometer (Leica Microsystems Inc.) and then sealed
using a tube topper (Cordless quick-seal tube topper, Beckman). The tubes were then
centrifuged at 178,000 x g for 46 hours at 20 °C in a Wx Sorvall Ultra 80 ultracentrifuge

15

fitted with a Stepsaver 70 V6 Vertical Titanium Rotor (Thermo Fisher Scientific lnc.).

Each of the 16 ultracentrifuged samples were separated into 20 fractions (250 µL) by
displacing the samples with molecular grade water. A syringe pump attached to a needle (BD,
23G and 1 inch) was used to displace samples from the top of the tube. This resulted in
fractions being collected from heavy to light BD values. The heavier BD fractions contained
the labeled DNA. The BD of each fraction was calculated from the refractive index obtained
using a refractometer. DNA from each of the fraction was recovered using a glycogen and
ethanol precipitation. Precipitated DNA was then re-suspended in 30 µL PCR grade water
and stored at -20 °C for further analysis.

2.5 High-throughput amplicon sequencing (Illumina MiSeq)
For each isotope pair, and each duplicate, four heavy fractions from labeled sample and three
heavy fractions from unlabeled sample were chosen for sequencing. In total, 56 fractions as
well as 8 total DNA samples were submitted for Illumina sequencing. MFCs were amended
with unlabeled substrate to provide a control against high GC content microorganisms that
would naturally be found in all heavy fractions. By comparing the microorganism in the
heavy fractions of the labeled amended samples to those in the heavy fractions of the
unlabeled amended samples, this ensures only those involved in label uptake are identified.
To quantitatively determine which phylotypes were more abundant in the heavy fractions of
the samples (label amended) compared to the controls (unlabeled amended), an enrichment
factor was calculated. The enrichment factor was obtained by dividing the % relative

16

abundance of the phylotype in the labeled fraction by the % relative abundance of phylotype
in corresponding unlabeled fraction.

When the enrichment factor was larger than one, the phylotype was considered to be involved
in carbon assimilation. Total DNA samples were also analyzed to profile the microbial
community structure in microcosm.

PCR and Illumina sequencing were performed at RTSF (Research Technology Supply
Facility) at Michigan State University using previously developed protocols (43). In short,
this involved the amplification of V4 region from 16S rRNA gene, quantification of
individual reactions (Picogreen assay), purifications with Ampure XP beads as well as gel
purification, and finally use of the Illumina MiSeq platform using 2×250 bp paired end flow
cell and reagent cartridge.

The data generated from Illumina sequencing was analyzed by Mothur (44) using the MiSeq
standard operating procedure developed by the same laboratory (45). In brief, the analysis
process involves the formation of contigs, removal of error sequences, chimera removal,
sequences alignment for operational taxonomic units (OTUs), and taxonomical levels from
group of OTUs. The data generated provided abundance data for the microbial community as
well as data for the calculation of enrichment factors.

17

3. RESULTS
3.1 Fraction generation and sequencing summary
Total DNA samples and the heavy fractions generated following ultracentrifugation were
submitted for Illumina sequencing. For each pair (e.g. 12C 10 ohms glucose and 13C 10 ohms
glucose) and duplicate, seven fractions were submitted for further analysis. In total, 56
fractions as well as 8 total DNA samples were submitted for Illumina sequencing and the data
were analyzed through Mothur. One Mothur run included 8 total DNA samples and 8 other
Mothur runs were for fractions from various treatments, which contained four labeled
fractions and three unlabeled fractions. The sequencing results have been summarized (Table
3). A total of 10,218,342 sequences were obtained from the Illumina high-throughput
sequencing for all samples. Approximately 55.3 ± 0.9% of these sequences were excluded
during sequencing analysis. Sequences were excluded because they were greater than 275 bp,
they contained ambiguous bases, they contained homopolymer lengths of >8 or they had a
start position after 1968 or an end position before 11550. Further, additional edits were
performed to remove chimeric sequences or to remove those belonging to mitochondria or
chloroplast lineage. Following this, 4,119,441 sequences remained and 4.7% of these were
unique sequences. These sequences were classified into OTUs and taxons through splitting
into different bins and then clustering at the level of order with a 97% similarity cutoff level.
Afterwards, three fractions pairs from the label amended and unlabeled amended anodes were
chosen with similar BD values for the determination of enrichment factors (Table 4).

18

Table 3: Summary of MiSeq Illumina data generated from MFCs total DNA samples
as well as fractions in labeled and unlabeled samples

Total DNA (8 MFCs samples from
10/1000 ohms labeled/unlabeled
sodium acetate/glucose)
Sodium acetate 10 ohms duplicate
1 (7 fractions)
Sodium acetate 10 ohms duplicate
2 (7 fractions)
Sodium acetate 1000 ohms
duplicate 1 (7 fractions)
Sodium acetate 1000 ohms
duplicate 2 (7 fractions)
Glucose 10 ohms duplicate 1
(7 fractions)
Glucose 10 ohms duplicate 2
(7 fractions)
Glucose 1000 ohms duplicate 1
(7 fractions)
Glucose 1000 ohms duplicate 2
(7 fractions)

# of Sequences
following make
Contigs command

Final # of unique
sequences

Final # of
sequences

% Chimeric

OTUs per fraction or
sample (average ±std
dev)

1,290,981

24,101

529,495

9.976

1439±291

953,513

18,103

396,956

6.855

668±250

1,092,795

20,727

427,028

13.873

1365±167

1,174,370

33,619

463,030

8.401

1261±639

1,017,669

18,092

404,495

12.209

753±227

1,092,399

19,492

432,749

9.100

744±121

977,908

18,080

400,811

9.427

654±107

995,435

17,624

409,450

6.973

622±81

1,623,272

27,100

655,427

9.545

956±538

19

Table 4: Buoyant density (BD) of fractions chosen for sequencing from MFCs sample (DUP
is abbreviation of Duplicate)
Samples
Fraction
BD (g/mL)
13

12

C

C

Difference

Sodium
acetate 10
ohms DUP1

F1
F2
F3

1.784
1.778
1.774

1.782
1.774
1.765

0.002
0.004
0.009

Sodium
acetate 1000
ohms DUP1

F1
F2
F3

1.778
1.772
1.765

1.783
1.775
1.768

-0.004
-0.003
-0.002

Sodium
acetate 10
ohms DUP2

F1
F2
F3

1.776
1.766
1.755

1.763
1.757
1.755

0.013
0.010
0.000

Sodium
acetate 1000
ohms DUP2

F1
F2
F3

1.790
1.782
1.772

1.793
1.783
1.773

-0.002
-0.001
-0.001

Glucose 10
ohms DUP1

F1
F2
F3

1.787
1.781
1.773

1.790
1.782
1.773

-0.003
-0.001
0.000

Glucose
1000 ohms
DUP1

F1
F2
F3

1.783
1.778
1.772

1.780
1.772
1.764

0.003
0.007
0.008

Glucose 10
ohms DUP2

F1
F2
F3

1.784
1.773
1.764

1.784
1.771
1.762

0.000
0.002
0.002

Glucose
1000 ohms
DUP2

F1
F2
F3

1.784
1.778
1.773

1.782
1.774
1.766

0.002
0.004
0.007

3.2 Illumina sequencing results for total DNA
3.2.1 Phyla from total DNA extracts
Illumina Sequencing data from 8 total DNA (13C sodium acetate 10 ohms/ 1000 ohms; 12C

20

sodium acetate 10 ohms/ 1000 ohms; 13C glucose 10 ohms/ 1000 ohms; 12C glucose 10
ohms/ 1000ohms) showed a diverse phyla distribution (Figure 3). There were 17-21 phyla in
each total DNA extract with a portion of unclassified sequences. The most abundant phylum
was Proteobacteria with an average percentage of 63.9 ±6.3%. Other dominant phyla were
Bacteroidetes and Firmicutes. Actinobacteria was over 1.5% in three out of four sodium
acetate fed MFCs but only dominant in one glucose fed MFCs.

3.2.2 Families from total DNA extracts
The Proteobacteria, Bacteroidetes, Firmicutes, and Actinobacteria were the most abundant
phyla in sodium acetate fed MFCs microcosms. A family level classification within these four
phyla has been generated (Figure 4). Within the Proteobacteria (Figure 4a), Geobacteraceae
is the most abundant family, followed by Rhodocyclaceae, Moraxellaceae, and
Comamonadaceae. Within Bacteroidetes (Figure 4b), the most abundant family for the
sodium acetate fed 10 ohms unlabeled sample was Flavobacteriaceae. However, the
dominance switched to Porphyromonadaceae for the other three samples of DNA extracts.
Other abundant families included Cryomorphaceae for the10 ohms samples and unclassified
Flavobacteriales for the 1000 ohms samples. Within phylum of Firmicutes, families of
Clostridiales Incertae Sedis XI and Peptostreptococcaceae together contributed to >50%
abundance for the10 ohms samples. While in the 1000 ohms samples, Clostridiales Incertae
Sedis XI, Gracilibacteraceae, Ruminococcaceae, and unclassified Clostridiales were the most
dominant families (Figure 4c). Family Nocardiaceae was the most dominant within the
Actinobacteria in all four sodium acetate fed MFCs anode samples (Figure 4d).

21

For the glucose amended MFCs, three phyla exhibited the highest level of abundance,
including Proteobacteria, Bacteroidetes, and Firmicutes (Figure 5). Within the
Proteobacteria phylum (Figure 5a), all of the glucose amended MFCs contained
Enterobacteriaceae as the most abundant family except 1000 ohms labeled sample, which
shifted to Rhodocyclaceae. Other families including Geobacteraceae, Comamonadaceae, and
Desulfovibrionaceae were also present in these samples. In the Bacteroidetes phylum (Figure
5b), Prophyromonadaceae, Bacteroidaceae, unclassified Bacteroidales, Flavobacteriaceae,
unclassified Bacteroidetes, unclassified Flavobacteriales were among the list of the most
abundant families. Bacteroidaceae exhibited a higher abundance in the 10 ohms MFCs
samples than those in the 1000 ohms samples. Also, unclassified Flavobacteriales was more
dominant in the 1000 ohms MFCs samples than those in the 10 ohms MFCs samples. Within
the phylum of Firmicutes, the dominant families were Ruminococcaceae, Streptococcaceae,
Clostridiales Incertae Sedis XI, and unclassified Clostridiales (Figure 5c).

22

100
90
80

Relative Aboundance %

70
60
50
40
30
20
10
0
A10LT

A10ULT

A1000LT A1000ULT

G10LT

G10ULT

G1000LT G1000ULT

unclassified
TM7
SR1
Firmicutes
Verrucomicrobia
Thermotogae
Synergistetes
Spirochaetes
Proteobacteria
Planctomycetes
Lentisphaerae
Gemmatimonadetes
Fusobacteria
Elusimicrobia
Deinococcus-Thermus
Chloroflexi
Chlorobi
Chlamydiae
Caldiserica
Bacteroidetes
Armatimonadetes
Actinobacteria
Acidobacteria

Figure 3: Comparison of relative abundance of sequences in total genomic DNA extracted from eight MFCs samples
(A: Sodium Acetate; G: Glucose; 10: 10ohms; 1000: 1000ohms; L: Labeled; UL: Unlabeled; T: Total)

23

Proteobacteria

A

100
90

Others

80
Moraxellaceae

70
60

Geobacteraceae

50
Rhodocyclaceae

40
30

Comamonadaceae

Relative Abundance (%)

20

10

Burkholderiaceae

0
A10UL-T

A10L-T

A1000UL-T A1000L-T
B

Bacteroidetes
100

Others

90
80

unclassified Bacteroidetes

70
60

unclassified
Flavobacteriales

50

Chitinophagaceae

40
Cryomorphaceae

30
20

Porphyromonadaceae

10
Flavobacteriaceae

0
A10UL-T

A10L-T A1000UL-T A1000L-T

Figure 4: Relative abundance of Proteobacteria (A), Bacteoidetes (B), Firmicutes (C), and
Actinobacteria (D) from sodium acetate fed MFCs total DNA extracts classified at the family
level (unless unclassified) (A: Sodium acetate; 10:10 ohms; 1000: 1000 ohms; L: Labeled;
UL: Unlabeled; T: Total)

24

Figure 4 (cont’d)
C

Firmicutes
Others

Relative Abundance (%)

100
90
80
70
60
50
40
30
20
10
0

unclassified
Clostridiales
Ruminococcaceae
Peptostreptococcaceae
Gracilibacteraceae
Clostridiales Incertae
Sedis XI
A10UL-T

A10L-T

A1000UL-T A1000L-T

Actinobacteria

Clostridiaceae 1

D

100
90
Others

80
70

Microbacteriaceae

60
unclassified
Actinomycetales

50
40

Nocardiaceae

30
20

Corynebacteriaceae

10
0
A10UL-T

A10L-T A1000UL-T A1000L-T

25

Proteobacteria

A

100
90
80
70

Others

60

Enterobacteriaceae

50

Geobacteraceae

40

Desulfovibrionaceae

30

Rhodocyclaceae

20

Comamonadaceae

Relative Abundance (%)

10
0
G10UL-T

G10L-T

G1000UL-T G1000L-T

Bacteroidetes

B

100

Others

90
80

unclassified
Flavobacteriales

70

unclassified
Bacteroidetes

60
50

Flavobacteriaceae

40
30

unclassified
Bacteroidales

20

Bacteroidaceae

10
0

Porphyromonadaceae
G10UL-T

G10L-T

G1000UL-T G1000L-T

Figure 5: Relative abundance of Proteobacteira (A), Bacteoidetes (B), and Firmicutes (C)
from glucose fed MFCs total DNA extracts classified at the family level (unless unclassified)
(G: Glucose; 10:10 ohms; 1000: 1000ohms; L: Labeled; UL: Unlabeled; T: Total)

26

Figure 5 (cont’d)

Relative Abundance (%)

Firmicutes
100
90
80
70
60
50
40
30
20
10
0

C
Others
unclassified Clostridiales
Ruminococcaceae
Gracilibacteraceae
Clostridiales Incertae Sedis
XI
Streptococcaceae
Enterococcaceae

G10UL-T

G10L-T

G1000UL-T G1000L-T

3.3 Identification of phylotypes responsible for label uptake
To identify the microorganisms responsible for 13C uptake from the amended substrates
(sodium acetate and glucose), enrichment factors for all phylotypes were calculated. For this,
the relative abundance of each phylotype was compared between the heavy fractions of the
samples (label amended) to heavy fractions of the controls (unlabeled amended). Specifically,
the enrichment factor was obtained by dividing the % relative abundance of each phylotype
in the fraction from the label amended substrate by the % relative abundance each phylotype
in corresponding fraction from the unlabeled amended substrate. As discussed above, the
comparison to heavy fractions from unlabeled amended samples is necessary to control for
high GC content microorganisms (these would be found in all heavy fractions, regardless of
label uptake).

27

The analysis involved comparing three heavy fractions of similar buoyant density between
the samples (label amended) and controls (unlabeled amended). The comparison was
conducted for each duplicate for each treatment, resulting in eight comparisons. These
comparisons involved 13C sodium acetate 10 ohms vs 12 C sodium acetate 10 ohms, 13C
sodium acetate 1000 ohms vs 12 C sodium acetate 1000 ohms, 13C glucose 10 ohms vs 12 C
glucose 10 ohms, 13C glucose 1000 ohms vs 12 C glucose 1000 ohms. Each comparison also
involved a complete duplicate for the entire approach (starting from DNA extraction). The
phylotypes enriched in the labeled fractions over the controls were responsible for 13C uptake
from the added substrate. The phylotypes considered responsible for label uptake have been
summarized for the sodium acetate 10 ohms treatment duplicates (Figure 6), sodium acetate
1000 ohms treatment duplicates (Figure 7), glucose 10 ohms treatment duplicates (Figure 8)
and glucose 1000 ohms treatment duplicates (Figure 9).

For the sodium acetate amended 10 ohms MFCs anode comparisons for both replicates, 40
phylotypes exhibited an enrichment factor above 1 (Figure 6). Of these, 27 phylotypes
belonged to Proteobacteira, 7 belonged to Bacteroidetes, 3 classified as Firmicutes, 2
classified as Chlamydiae, and 1 belonged to the phylum TM7. For these comparisons, the
average enrichment factors varied between 1.1 and 99.2. In duplicate 1, the phylotypes with
the highest enrichment factors included unclassified Parachlamydiaceae (Chlamydiae),
Brevundimonas (Proteobacteria), Azospirillum (Proteobacteria), Azoarcus (Proteobacteria),
and Telmatospirillum (Proteobacteria) (Figure 6a). In replicate 2, the most enriched
phylotypes were unclassified Parachlamydiaceae (Chlamydiae), Thauera (Proteobacteria),

28

Gracilibacter (Firmicutes), Castellaniella (Proteobacteria), and unclassified Rhodocyclales
(Proteobacteria) (Figure 6b).

For the sodium acetate amended 1000 ohms MFCs anode comparisons for both duplicates, 36
phylotypes exhibited an enrichment factor above 1 (Figure 7). Of these, 26 belonged to
Proteobacteria, 4 to Firmicutes, 2 to Actinobacteria, 2 to Chlamydiae, 1 to Lentisphaerae,
and 1 to Planctomycetes. The average enrichment factors from these comparisons varied
between 1.2 and 115.7. In replicate 1, the five most enriched phylotypes were unclassified
Parachlamydiaceae (Chlamydiae), Azospirillum (Proteobacteria), Gordonia (Actinobacteria),
Kaistia (Proteobacteria), and Shinella (Proteobacteria) (Figure 7a). In replicate 2, the five
most enriched phylotypes were unclassified Parachlamydiaceae (Chlamydiae), Victivallis
(Lentisphaerae), Rhizobium (Proteobacteria), Lactococcus (Firmicutes), and unclassified
Gammaproteobacteria (Proteobacteria) (Figure 7b).

For glucose amended 10 ohms MFCs anode comparisons for both replicates, 54 phylotypes
exhibited an enrichment factor above 1 (Figure 8). Of these, 26 belonged to Proteobacteria,
12 classified as Firmicutes, 7 belonged to Bacteroidetes and 6 classified as Actinobacteria.
The average enrichment factors from three fractions varied between 1.2 and 599.6. In
replicate 1, the five most enriched phylotypes were Corynebacterium (Actinobacteria),
Sulfuricurvum (Proteobacteria), unclassified Verrucomicrobia (Verrucomicrobia),
unclassified Betaproteobacteria (Proteobacteria) and Alistipes (Bacteroidetes) (Figure 8a). In
replicate 2, the five most enriched phylotypes were unclassified Cellulomonadaceae

29

(Actinobacteria), unclassified Rhodocyclaceae (Proteobacteria), Aquabacterium
(Proteobacteria), Ralstonia (Proteobacteria), and unclassified Hyphomicrobiaceae
(Proteobacteria) (Figure 8b).

For glucose fed 1000 ohms MFCs anode comparisons for both replicates, 62 phylotypes
exhibited an enrichment factor above 1 (Figure 9). Of these, 38 belonged to Proteobacteria,
11 belonged to Bacteroidetes and 7 classified as Firmicutes. The average enrichment factors
from these comparisons varied between 1.3 and 23.8. In replicate 1 the five most enriched
phylotypes classified as Bacteroides (Bacteroidetes), Aeromonas (Proteobacteria),
unclassified Polyangiaceae (Proteobacteria) Chitinophaga (Bacteroides), and Lactococcus
(Firmicutes) (Figure 9a). In replicate 2, the five most enriched phylotypes classified as
Byssovorax (Proteobacteria), unclassified Parachlamydiaceae (Chlamydiae), Devosia
(Proteobacteria), unclassified Polyangiaceae (Proteobacteria), and Sphingobacterium
(Bacteroidetes) (Figure 9b).

Notably, some phyotypes were highly enriched (enrichment factors of 100-1000) over the
others. For example, unclassified Parachlamydiaceae were highly enriched in three of the
four comparisons amended with acetate. A summary table has been provided that ranks the
most enriched phylotypes as an average across both replicates (Table 5).

30

Bacteroidetes (3) Chlamydiae (1)

31

unclassified Gammaproteobacteria

Stenotrophomonas

Sinobacter

Pseudomonas

Acinetobacter

Geobacter

Desulfobulbus

A_10_DUP1_F2

Azoarcus

Thiobacillus

unclassified Burkholderiales

A_10_DUP1_F1

unclassified Comamonadaceae

unclassified Burkholderiales

Telmatospirillum

Azospirillum

250

Brevundimonas

unclassified Parachlamydiaceae

unclassified Flavobacteriales

Fluviicola

Terrimonas

Enrichment factor (Relative abundance in labeled fraction %/
Relative abundance in unlabeled fraction %)
300

A
A_10_DUP1_F3

200

150

100

50

0

Proteobacteria (15)

Figure 6: Enrichment factor of select OTUs (at genus level unless unclassified) in the heavy fractions of the labeled sodium acetate 10 ohms
samples to the unlabeled sodium acetate 10 ohms samples for two duplicates (A and B)

Bacteroidetes (4) Chlamydiae (1)

32

Proteobacteria (12)

TM7 genus incertae sedis

Clostridium III

Gracilibacter

Finegoldia

Pseudomonas

unclassified Moraxellaceae

Acinetobacter

unclassified Geobacteraceae

A_10_DUP2_F2

Thauera

Azoarcus

unclassified Comamonadaceae

Castellaniella

A_10_DUP2_F1

Azospirillum

Xanthobacter

250

unclassified Bradyrhizobiaceae

300

Brevundimonas

unclassified Parachlamydiaceae

unclassified Flavobacteriales

Fluviicola

Terrimonas

Paludibacter

Enrichment factors (Relative abundance in labeled
fraction% / Relative abundance in unlabeled fraction %)

Figure 6 (Cont’d)
B
A_10_DUP2_F3

200

150

100

50

0

Firmicutes (3) TM7 (1)

33

Proteobacteria (19)

samples to the unlabeled sodium acetate 1000 ohms samples for two duplicates (A and B)

unclassified Lachnospiraceae

unclassified Planococcaceae

unclassified Bacillaceae 1

unclassified Gammaproteobacteria

Stenotrophomonas

Dokdonella

Aeromonas

Salmonella

A_1000_DUP1_F2

unclassified Enterobacteriaceae

unclassified Polyangiaceae

unclassified Betaproteobacteria

unclassified Rhodocyclaceae

A_1000_DUP1_F1

Shinella

unclassified Oxalobacteraceae

unclassified Comamonadaceae

Comamonas

Sphingopyxis

Azospirillum

Kaistia

Devosia

Bosea

Brevundimonas

unclassified Planctomycetaceae

unclassified Parachlamydiaceae

Gordonia

Enrichment factors (Relative abundance in labeled fraction
%/ Relative abundance in unlabeled fraction %)
120
100
80
60
40
20
0

A
A_1000_DUP1_F3

Firmicutes (3)

Figure 7: Enrichment factor of select OTUs (at genus level unless unclassified) in the heavy fractions of the labeled sodium acetate 1000 ohms

Actinobacteria (1) Chlamydiae (1) Lentisphaerae (1)

34

Proteobacteria (7)

Lactococcus

unclassified Gammaproteobacteria

Pseudomonas

A_1000_DUP2_F2

unclassified Cystobacteraceae

Sphingopyxis

Azospirillum

A_1000_DUP2_F1

Rhizobium

Kaistia

300
250
200
150
100
50
0

Victivallis

unclassified Parachlamydiaceae

unclassified Actinomycetales

Enrichment factors (Relative abundance in labeled fraction
%/ Relative abundance in unlabeled fraction %)

Figure 7 (Cont’d)
B
A_1000_DUP2_F3

Firmicutes (1)

Actinobacteria (3) Bacteroidetes (4) Elusimicrobia (1)

35

Proteobacteria (14)

the unlabeled glucose 10 ohms samples for two duplicates (A and B)

unclassified Firmicutes

unclassified Clostridiales

Oscillibacter

Clostridium III

Clostridium XlVa

unclassified Clostridiales

unclassified Bacillales

Lactococcus

unclassified Verrucomicrobia

unclassified Synergistaceae

Cloacibacillus

unclassified Proteobacteria

unclassified Enterobacteriaceae

G_10_DUP1_F2

Sulfuricurvum

Arcobacter

Geobacter

unclassified Desulfovibrionaceae

Desulfovibrio

G_10_DUP1_F1

Desulfobulbus

unclassified Betaproteobacteria

unclassified Rhodocyclaceae

Thiomonas

Azospirillum

Pleomorphomonas

1000

Elusimicrobium

unclassified Bacteroidetes

Alistipes

Petrimonas

Parabacteroides

unclassified Actinomycetales

Gordonia

Corynebacterium

Enrichment factors (Relative abundance in labeled fraction
%/ Relative abundance in unlabeled fraction %)
1200

A
G_10_DUP1_F3

800

600

400

200

0

Synergistetes (1) Verrucomicrobia (1) Firmicutes (8)

Figure 8: Enrichment factor of select OTUs (at genus level unless unclassified) in the heavy fractions of the labeled glucose 10 ohms samples to

Acidobacteria (1) Actinobacteria (3) Bacteroidetes (3)

36

Proteobacteria (12)
Firmicutes (4)

TM7 genus incertae sedis

unclassified Erysipelotrichaceae

Anaerofilum

Gracilibacter

Sedimentibacter

Stenotrophomonas

Cellvibrio

G_10_DUP2_F2

unclassified Pasteurellaceae

unclassified Deltaproteobacteria

unclassified Desulfobacterales

unclassified Rhodocyclaceae

G_10_DUP2_F1

Shinella

Microvirgula

Aquabacterium

Ralstonia

unclassified Alphaproteobacteria

1000
900
800
700
600
500
400
300
200
100
0

unclassified Hyphomicrobiaceae

unclassified Flavobacteriales

Parabacteroides

Paludibacter

Streptomyces

unclassified Cellulomonadaceae

unclassified Acidimicrobiales

Gp6

Enrichment factors (Relative abundance in labeled fraction
%/ Relative abundance in unlabeled fraction %)

Figure 8 (Cont’d)
B
G_10_DUP2_F3

TM7 (1)

Victivallis

Lentisphaerae (1)

37

Proteobacteria (20)

to the unlabeled glucose 1000 ohms samples for two duplicates (A and B)

Acetoanaerobium

Lactococcus

Bacillus

unclassified Verrucomicrobia

Dokdonella

Acinetobacter

Aeromonas

unclassified Enterobacteriaceae

Sulfuricurvum

Arcobacter

unclassified Polyangiaceae

Byssovorax

Desulfovibrio

Desulfobulbus

G_1000_DUP1_F2

unclassified Rhodocyclaceae

Thiobacillus

unclassified Burkholderiales

Thiomonas

unclassified Alphaproteobacteria

Sphingopyxis

G_1000_DUP1_F1

Telmatospirillum

Azospirillum

unclassified Rhizobiales

Brucella

Elusimicrobium

Actinobacteria (1) Bacteroidetes (6)

Elusimicrobia (1)

unclassified Bacteroidetes

unclassified Chitinophagaceae

Terrimonas

Chitinophaga

Bacteroides

unclassified Porphyromonadaceae

unclassified Microbacteriaceae

Enrichment factors (Relative abundance in labeled fraction %/
Relative abundance in unlabeled fraction %)

1000
900
800
700
600
500
400
300
200
100
0

A
G_1000_DUP1_F3

Verrucomicrobia (1) Firmicutes (3)

Figure 9: Enrichment factor of select OTUs (at genus level unless unclassified) in the heavy fractions of the labeled glucose 1000 ohms samples

Bacteroidetes (5) Chlamydiae (1) Gemmatimonadetes (1)

38

Proteobacteria (18)

Oscillibacter

Clostridium XlVa

unclassified Bacillales

Bacillus

Rhodanobacter

unclassified Moraxellaceae

Acinetobacter

unclassified Deltaproteobacteria

unclassified Polyangiaceae

Byssovorax

unclassified Betaproteobacteria

unclassified Rhodocyclaceae

Shinella

unclassified Comamonadaceae

B
G_1000_DUP2_F2

Thiomonas

Sphingopyxis

Telmatospirillum

unclassified Rhizobiales

unclassified Phyllobacteriaceae

G_1000_DUP2_F1

Devosia

Brucella

Bosea

Gemmatimonas

1000
900
800
700
600
500
400
300
200
100
0

unclassified Parachlamydiaceae

Sphingobacterium

unclassified Chitinophagaceae

Terrimonas

Niabella

Chitinophaga

Enrichment factors (Relative abundance in labeled fraction %/
Relative abundance in unlabeled fraction %)

Figure 9 (Cont’d)

G_1000_DUP2_F3

Firmicutes (4)

Table 5: Summary of genera enriched in both duplicates for sodium acetate (10 ohms/ 1000
ohms) and glucose (10 ohms/ 1000 ohms) fed MFCs samples
Average
Sample
Phylum
Genus
enrichment
factor
Chlamydiae
unclassified Parachlamydiaceae
95.2
Proteobacteria
Brevundimonas
16.0
Proteobacteria
Azoarcus
7.4
Proteobacteria
Azospirillum
6.8
Sodium
acetate 10
Bacteroidetes
Terrimonas
3.7
ohms
Bacteroidetes
Fluviicola
3.4
Proteobacteria
Acinetobacter
2.7
Bacteroidetes
unclassified Flavobacteriales
2.6
Proteobacteria
unclassified Comamonadaceae
1.4

Sodium
acetate 1000
ohms

Chlamydiae
Proteobacteria
Proteobacteria
Proteobacteria
Proteobacteria

unclassified Parachlamydiaceae
unclassified Gammaproteobacteria
Azospirillum
Kaistia
Sphingopyxis

93.7
12.0
7.9
6.8
2.9

Glucose 10
ohms

Proteobacteria
Bacteroidetes

unclassified Rhodocyclaceae
Parabacteroides

54.4
5.1

Proteobacteria
Proteobacteria
Proteobacteria
Bacteroidetes
Proteobacteria
Bacteroidetes
Bacteroidetes
Proteobacteria
Proteobacteria
Proteobacteria
Firmicutes

Byssovorax
unclassified Polyangiaceae
Brucella
Terrimonas
Sphingopyxis
Chitinophaga
unclassified Chitinophagaceae
Thiomonas
Telmatospirillum
unclassified Rhodocyclaceae
Bacillus

13.1
6.4
4.7
4.6
4.0
3.3
2.3
2.1
1.9
1.8
1.3

Glucose
1000 ohms

The phylotypes in Table 5 were dominant in label uptake and were therefore important
microorganisms for MFC function. Within the MFCs amended with sodium acetate,
phylotypes unclassified Parachlamydiaceae (Chlamydia) and Azospirillum (Proteobacteria)

39

were enriched in both 10 ohms and 1000 ohms replicates. And within MFCs amended with
glucose, unclassified Rhodocyclaceae were enriched in both 10 ohms and 1000 ohms
replicates.

40

4. DISCUSSION
To the author’s knowledge, this is the first study that combines SIP with high-throughput
sequencing to identify active community members in MFCs. SIP, a cultivation independent
technique, enables researchers to identify microorganisms involved in label uptake.
High-throughput sequencing provides a greater amount of data compared to traditional
sequencing methods. The combination of these two methods enables the investigation of
potentially low abundance, but functionally important microorganisms. Although MiSeq
sequencing provides shorter reads than 454 pyrosequencing, this study as well as other
research demonstrated the availability to apply paired-end MiSeq sequencing to identify
diverse microbial communities (24, 46).

The sequencing results for the overall microbial community of the MFCs anodes illustrates a
diverse collection of microorgansims. The major phyla were Proteobacteria, Bacteroidetes,
Firmicutes, and Actinobacteria for sodium acetate amended MFCs. Proteobacteria,
Bacteroidetes, and Firmicutes were also the dominant phyla within glucose amended MFCs.
The dominant abundance of Proteobacteria is consistent with prior research results (Table 1).
For example, Aeromonas hydrophila (Gammaproteobacteria), Geobacter metallireducens
(Deltaproteobacteria), and Geobacter sulfurreducens (Deltaproteobacteria) were found in
acetate amended MFCs. Further, Alcaligenes faecalis (Betaproteobacteria), Pseudomonas
aeruginosa (Gammaproteobacteria), Gluconobacter oxydans (Alphaproteobacteria),
Klebsiella pneumoniae (Gammaproteobacteria), Proteus mirabilis (Gamaproteobacteria),

41

and Pseudomonas aeruginosa (Gammaproteobacteria) were identified in glucose fed MFCs
(22).

The occurrence of microorganisms in the phyla Bacteroidetes, Firmicutes, and Actinobacteria
in MFCs anode chambers was previously reported. One project reported Bacteroidetes could
contribute to the power generation in granular semicoke anodic MFC with sodium acetate as
the substrate (47). Also, Bacteroidetes and Firmicutes were detected within air-cathode MFCs
fed with acetate (48). Actinobacteria and Firmicutes were reported on the anode biofilm in
MFC fed with acetate and propionate (49). In addition, Bacteroidetes and Firmicutes were
detected in a membrane-less MFC with glucose as the carbon source (50). Jung indicated that
Firmicutes were only found within glucose fed MFCs (5). Another former study showed most
Firmicutes phylum were related to glucose (22). In the current study, both glucose and
sodium acetate fed MFCs microcosms contained Firmicutes. Phyla of Bacteroidetes,
Firmicutes, and Actinobacteria in MFCs anode chamber were also identified with other
substrates additions (20, 51).

Comparing relative abundance between the eight microcosms, the Proteobacteria in the
glucose amended MFCs had a relatively higher abundance than those in sodium acetate
amended MFCs. In contrast, microorganisms within the phylum Actinobacteria were more
abundant in the sodium acetate amended MFCs. Glucose is a fermentative substrate and
sodium acetate is a non-fermentative substrate. Fermentation acts as important role in the
utilization of complex substances in MFCs. The differences of fermentative and

42

non-fermentative substrates would have an effect on the structure of anode microbial
community (52). Some organisms in Proteobacteria would participate in the process of
degrading complex substrate into simple substances. Comparing the results from the two
types of external resistance, the phyla percentage change was complex. The only obvious
trend was that Firmicutes exhibited a higher abundance at the lower external resistance (10
ohms).

A different trend was described in a study using azo dye as substrate in MFCs. They found no
species belonging to Firmicutes phylum with the external resistance of 10 ohms. Whereas,
some microorganisms were identified as phylum of Firmicutes with 510 ohms external
resistance (53).

At the genus level, unclassified Parachlamydiaceae (Chlamydia), Azospirillum
(Proteobacteria) were enriched in all sodium acetate amended MFCs anodes, and
unclassified Rhodocyclaceae were enriched in all glucose fed MFCs anodes. To my
knowledge, the family Parachlamydiaceae has not been identified as an important phylotype
in MFCs in previous studies. Previously, researchers identified the Parachlamydiaceae in
drinking water, which may have been attributed to an infectious bovine abortion (54).
Parachlamydiaceae, often existing as endosymbionts of amoebae, have also been found in
waste water treatment plant samples (55). Consistent with this, the MFCs in the current study
were inoculated with activated sludge samples. The current study suggests microorganisms in
this family are important for MFCs electricity generation when acetate is the added substrate.

43

Interestingly, Azospirillum has previously been linked to electrochemical activity. This
phylotype was enriched in microbial electrolysis cell (MEC) bio-cathodes, which were
transferred from sediment MFC bio-anodes (56). Microorganisms from the family
Rhodocyclaceae were also noted in MFCs anode chambers from several previous studies (48,
57).

Notably, the enrichment factors from the glucose amended high external resistance (1000
ohms) MFCs were lower than other three treatments. One possible reason could be that
glucose contributes to other microbial processes, such as fermentation, and methanogenesis
under higher external resistance values and using terminal electron acceptors that do not lead
to electricity generation (58). This explanation would be also evidenced by the relative low
coulombic efficiency of higher resistance (glucose 1000 ohms) compared with lower
resistance (glucose 10 ohms) in Table 2. This change might transfer carbon to other forms
(e.g. CO2) rather than retain the carbon within the cells in the MFC.

44

5. CONCLUSION
Stable isotope probing (SIP) and high-throughput sequencing were used to i) profile overall
microbial community present and ii) identify the microorganisms responsible for carbon
uptake at the MFC anode over four experimental treatments (with replication). This involved
the analysis of eight samples of DNA extracted from the MFCs anodes. The MFCs samples
differed from each other by substrate type (labelled and unlabeled acetate or glucose) and
external resistance (10 ohms and 1000 ohms). The results illustrated the anodes consisted of a
diverse microbial community. For the sodium acetate amended MFCs, the dominant phyla
were Proteobacteria, Bacteroidetes, and Firmicutes. For the glucose amended MFCs, in
addition to the above three phyla, Actinobacteria was also detected as important phylum.
Through comparing enrichment factors between the labeled and unlabeled fractions, 14
phylotypes were enriched in both replicates of the acetate amended MFC anodes (including
both resistance levels). Also, 13 phylotypes were enriched in both replicates of the glucose
amended MFC anodes (including both resistance levels). Among these, unclassified
Parachlamydiaceae (Chlamydia), Azospirillum (Proteobacteria), and unclassified
Rhodocyclaceae (Proteobacteria) were the dominant label uptaking phylotypes at the MFCs
anode. This study demonstrated that combination of SIP and high-throughput sequencing is a
useful tool to characterize microbial community and identify functional organisms in MFCs.

45

REFERENCES

46

REFERENCES

1.

Ahmad F, Atiyeh MN, Pereira B, Stephanopoulos GN. 2013. A review of cellulosic
microbial fuel cells: Performance and challenges. Biomass & Bioenergy 56:179-188.

2.

Jang JK, Ryou YS, Kim JG, Kang YK, Lee EY. 2012. Use of Nitrate and Ferric Ion
as Electron Acceptors in Cathodes to Improve Current Generation in Single-cathode
and Dual-cathode Microbial Fuel Cells. Korean Journal of Microbiology and
Biotechnology 40:414-418.

3.

Rabaey K, Verstraete W. 2005. Microbial fuel cells: novel biotechnology for energy
generation. Trends in Biotechnology 23:291-298.

4.

Pant D, Van Bogaert G, Diels L, Vanbroekhoven K. 2010. A review of the
substrates used in microbial fuel cells (MFCs) for sustainable energy production.
Bioresource Technology 101:1533-1543.

5.

Jung S, Regan JM. 2007. Comparison of anode bacterial communities and
performance in microbial fuel cells with different electron donors. Applied
Microbiology and Biotechnology 77:393-402.

6.

Pant D, Van Bogaert G, De Smet M, Diels L, Vanbroekhoven K. 2010. Use of
novel permeable membrane and air cathodes in acetate microbial fuel cells.
Electrochimica Acta 55:7710-7716.

7.

Jung S, Regan JM. 2011. Influence of External Resistance on Electrogenesis,
Methanogenesis, and Anode Prokaryotic Communities in Microbial Fuel Cells.
Applied and Environmental Microbiology 77:564-571.

8.

Rismani-Yazdi H, Carver SM, Christy AD, Tuovinen IH. 2008. Cathodic
limitations in microbial fuel cells: An overview. Journal of Power Sources
180:683-694.

9.

Logan B, Cheng S, Watson V, Estadt G. 2007. Graphite fiber brush anodes for
increased power production in air-cathode microbial fuel cells. Environmental
Science & Technology 41:3341-3346.
47

10.

Rabaey K, Lissens G, Siciliano SD, Verstraete W. 2003. A microbial fuel cell
capable of converting glucose to electricity at high rate and efficiency. Biotechnology
Letters 25:1531-1535.

11.

Chae K-J, Choi M-J, Lee J-W, Kim K-Y, Kim IS. 2009. Effect of different
substrates on the performance, bacterial diversity, and bacterial viability in microbial
fuel cells. Bioresource Technology 100:3518-3525.

12.

Ren Z, Steinberg LM, Regan JM. 2008. Electricity production and microbial
biofilm characterization in cellulose-fed microbial fuel cells. Water Science and
Technology 58:617-622.

13.

Rezaei F, Xing DF, Wagner R, Regan JM, Richard TL, Logan BE. 2009.
Simultaneous Cellulose Degradation and Electricity Production by Enterobacter
cloacae in a Microbial Fuel Cell. Applied and Environmental Microbiology
75:3673-3678.

14.

Debabov VG. 2008. Electricity from microorganisms. Microbiology 77:123-131.

15.

Ahn Y, Logan BE. 2010. Effectiveness of domestic wastewater treatment using
microbial fuel cells at ambient and mesophilic temperatures. Bioresource Technology
101:469-475.

16.

Li XM, Cheng KY, Selvam A, Wong JWC. 2013. Bioelectricity production from
acidic food waste leachate using microbial fuel cells: Effect of microbial inocula.
Process Biochemistry 48:283-288.

17.

Xing DF, Cheng SA, Regan JM, Logan BE. 2009. Change in microbial
communities in acetate- and glucose-fed microbial fuel cells in the presence of light.
Biosensors & Bioelectronics 25:105-111.

18.

Chung KM, Okahe S. 2009. Characterization of Electrochemical Activity of a Strain
IS02-3 Phylogenetically Related to Aeromonas sp Isolated From a Glucose-Fed
Microbial Fuel Cell. Biotechnology and Bioengineering 104:901-910.

19.

Ishii Si, Suzuki S, Norden-Krichmar TM, Wu A, Yamanaka Y, Nealson KH,
Bretschger O. Identifying the microbial communities and operational conditions for
optimized wastewater treatment in microbial fuel cells. Water Research.

48

20.

Zhang GD, Zhao QL, Jiao Y, Wang K, Lee DJ, Ren NQ. 2012. Biocathode
microbial fuel cell for efficient electricity recovery from dairy manure. Biosensors &
Bioelectronics 31:537-543.

21.

Phung NT, Lee J, Kang KH, Chang IS, Gadd GM, Kim BH. 2004. Analysis of
microbial diversity in oligotrophic microbial fuel cells using 16S rDNA sequences.
FEMS Microbiology Letters 233:77-82.

22.

Du Z, Li H, Gu T. 2007. A state of the art review on microbial fuel cells: A promising
technology for wastewater treatment and bioenergy. Biotechnology Advances
25:464-482.

23.

Fremont M, Coomans D, Massart S, De Meirleir K. 2013. High-throughput 16S
rRNA gene sequencing reveals alterations of intestinal microbiota in myalgic
encephalomyelitis/chronic fatigue syndrome patients. Anaerobe 22:50-56.

24.

Ansorge WJ. 2009. Next-generation
Biotechnology 25:195-203.

25.

Zhang J, Chiodini R, Badr A, Zhang G. 2011. The impact of next-generation
sequencing on genomics. Journal of Genetics and Genomics 38:95-109.

26.

Di Bella JM, Bao Y, Gloor GB, Burton JP, Reid G. 2013. High throughput
sequencing methods and analysis for microbiome research. Journal of Microbiological
Methods 95:401-414.

27.

Strausberg RL, Levy S, Rogers Y-H. 2008. Emerging DNA sequencing technologies
for human genomic medicine. Drug Discovery Today 13:569-577.

28.

Margulies M, Egholm M, Altman WE, Attiya S, Bader JS, Bemben LA, Berka J,
Braverman MS, Chen YJ, Chen ZT, Dewell SB, Du L, Fierro JM, Gomes XV,
Godwin BC, He W, Helgesen S, Ho CH, Irzyk GP, Jando SC, Alenquer MLI,
Jarvie TP, Jirage KB, Kim JB, Knight JR, Lanza JR, Leamon JH, Lefkowitz SM,
Lei M, Li J, Lohman KL, Lu H, Makhijani VB, McDade KE, McKenna MP,
Myers EW, Nickerson E, Nobile JR, Plant R, Puc BP, Ronan MT, Roth GT,
Sarkis GJ, Simons JF, Simpson JW, Srinivasan M, Tartaro KR, Tomasz A, Vogt
KA, Volkmer GA, Wang SH, Wang Y, Weiner MP, Yu PG, Begley RF, Rothberg
JM. 2005. Genome sequencing in microfabricated high-density picolitre reactors.
Nature 437:376-380.

49

DNA

sequencing

techniques.

New

29.

Li B, Zhang XX, Guo F, Wu WM, Zhang T. 2013. Characterization of tetracycline
resistant bacterial community in saline activated sludge using batch stress incubation
with high-throughput sequencing analysis. Water Research 47:4207-4216.

30.

Yoder-Himes DR, Chain PSG, Zhu Y, Wurtzel O, Rubin EM, Tiedje JM, Sorek R.
2009. Mapping the Burkholderia cenocepacia niche response via high-throughput
sequencing. Proceedings of the National Academy of Sciences of the United States of
America 106:3976-3981.

31.

Marguerat S, Wilhelm BT, Bahler J. 2008. Next-generation sequencing:
applications beyond genomes. Biochemical Society Transactions 36:1091-1096.

32.

Passalacqua KD, Varadarajan A, Ondov BD, Okou DT, Zwick ME, Bergman NH.
2009. Structure and Complexity of a Bacterial Transcriptome. Journal of Bacteriology
191:3203-3211.

33.

Li LL, Norman A, Hansen LH, Sorensen SJ. 2012. Metamobilomics--expanding
our knowledge on the pool of plasmid encoded traits in natural environments using
high-throughput sequencing. Clinical microbiology and infection : the official
publication of the European Society of Clinical Microbiology and Infectious Diseases
18 Suppl 4:5-7.

34.

van Vliet AHM. 2010. Next generation sequencing of microbial transcriptomes:
challenges and opportunities. Fems Microbiology Letters 302:1-7.

35.

Friedrich MW. 2006. Stable-isotope probing of DNA: insights into the function of
uncultivated microorganisms from isotopically labeled metagenomes. Current
Opinion in Biotechnology 17:59-66.

36.

Madsen EL. 2006. The use of stable isotope probing techniques in bioreactor and
field studies on bioremediation. Current Opinion in Biotechnology 17:92-97.

37.

Uhlik O, Jecna K, Leigh MB, Mackova M, Macek T. 2009. DNA-based stable
isotope probing: a link between community structure and function. Science of the
Total Environment 407:3611-3619.

38.

Uhlik O, Leewis MC, Strejcek M, Musilova L, Mackova M, Leigh MB, Macek T.
2013. Stable isotope probing in the metagenomics era: A bridge towards improved
bioremediation. Biotechnology Advances 31:154-165.

50

39.

Radajewski S, Ineson P, Parekh NR, Murrell JC. 2000. Stable-isotope probing as a
tool in microbial ecology. Nature 403:646-649.

40.

Radajewski S, McDonald IR, Murrell JC. 2003. Stable-isotope probing of nucleic
acids: a window to the function of uncultured microorganisms. Current Opinion in
Biotechnology 14:296-302.

41.

Cupples AM. 2011. The use of nucleic acid based stable isotope probing to identify
the microorganisms responsible for anaerobic benzene and toluene biodegradation.
Journal of Microbiological Methods 85:83-91.

42.

Logan BE, Hamelers B, Rozendal RA, Schrorder U, Keller J, Freguia S,
Aelterman P, Verstraete W, Rabaey K. 2006. Microbial fuel cells: Methodology and
technology. Environmental Science & Technology 40:5181-5192.

43.

Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N,
Owens SM, Betley J, Fraser L, Bauer M. 2012. Ultra-high-throughput microbial
community analysis on the Illumina HiSeq and MiSeq platforms. The ISME journal
6:1621-1624.

44.

Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB,
Lesniewski RA, Oakley BB, Parks DH, Robinson CJ, Sahl JW, Stres B,
Thallinger GG, Van Horn DJ, Weber CF. 2009. Introducing mothur: Open-Source,
Platform-Independent, Community-Supported Software for Describing and
Comparing Microbial Communities. Applied and Environmental Microbiology
75:7537-7541.

45.

Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD. 2013.
Development of a Dual-Index Sequencing Strategy and Curation Pipeline for
Analyzing Amplicon Sequence Data on the MiSeq Illumina Sequencing Platform.
Applied and Environmental Microbiology 79:5112-5120.

46.

Liang B, Cheng H, Van Nostrand JD, Ma J, Yu H, Kong D, Liu W, Ren N, Wu L,
Wang A, Lee D-J, Zhou J. 2014. Microbial community structure and function of
Nitrobenzene reduction biocathode in response to carbon source switchover. Water
research 54:137-148.

47.

Sun Y, Wei J, Liang P, Huang X. 2011. Electricity generation and microbial
community changes in microbial fuel cells packed with different anodic materials.
Bioresource Technology 102:10886-10891.
51

48.

Borole AP, Hamilton CY, Vishnivetskaya T, Leak D, Andras C. 2009. Improving
power production in acetate-fed microbial fuel cells via enrichment of exoelectrogenic
organisms in flow-through systems. Biochemical Engineering Journal 48:71-80.

49.

Zhang GD, Zhao QL, Jiao Y, Wang K, Lee DJ, Ren NQ. 2012. Efficient electricity
generation from sewage sludge using biocathode microbial fuel cell. Water Research
46:43-52.

50.

Aldrovandi A, Marsili E, Stante L, Paganin P, Tabacchioni S, Giordano A. 2009.
Sustainable power production in a membrane-less and mediator-less synthetic
wastewater microbial fuel cell. Bioresource Technology 100:3252-3260.

51.

Zhang Y, Min B, Huang L, Angelidaki I. 2011. Electricity generation and microbial
community response to substrate changes in microbial fuel cell. Bioresource
Technology 102:1166-1173.

52.

Quan XC, Quan YP, Tao K, Jiang XM. 2013. Comparative investigation on
microbial community and electricity generation in aerobic and anaerobic enriched
MFCs. Bioresource Technology 128:259-265.

53.

Fernando E, Keshavarz T, Kyazze G. 2014. External resistance as a potential tool
for influencing azo dye reductive decolourisation kinetics in microbial fuel cells.
International Biodeterioration & Biodegradation 89:7-14.

54.

Wheelhouse N, Sait M, Gidlow J, Deuchande R, Borel N, Baily J, Caldow G,
Longbottom D. 2011. Molecular detection of Chlamydia-like organisms in cattle
drinking water. Veterinary Microbiology 152:196-199.

55.

Horn M, Wagner M. 2001. Evidence for additional genus-level diversity of
Chlamydiales in the environment. Fems Microbiology Letters 204:71-74.

56.

Pisciotta JM, Zaybak Z, Call DF, Nam JY, Logan BE. 2012. Enrichment of
Microbial Electrolysis Cell Biocathodes from Sediment Microbial Fuel Cell
Bioanodes. Applied and Environmental Microbiology 78:5212-5219.

57.

Quan X-c, Quan Y-p, Tao K. 2012. Effect of anode aeration on the performance and
microbial community of an air–cathode microbial fuel cell. Chemical Engineering
Journal 210:150-156.

52

58.

Logan BE, Regan JM. 2006. Electricity-producing bacterial communities in
microbial fuel cells. Trends in Microbiology 14:512-518.

53