Electrochemical studies of subsurface microorganisms

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ELECTROCHEMICAL STUDIES OF SUBSURFACE MICROORGANISMS
by
Yamini Jangir
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulllment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHYSICS)
August 2016
Copyright 2016 Yamini Jangir
To all my teachers.
ii
No matter what part of Nature one studies
- Microbes or Milky Way -
There is a point where one begins, but never ends.
by Ray, Hans Augusto (1898-1977)
iii
Acknowledgments
The thesis dissertation marks the end of a long and an eventful journey for which
I acknowledge all the sacrices that my family made to ensure that I received
an excellent education. Next, I confess that without my husbands' tremendous
understanding, this thesis would not have been possible. He brings peace and
tranquility to my life. Here, I also seek blessings from my family-in-law who have
encouraged all my scientic pursuits.
During my years at the University of Southern California, I have certainly
grown intellectually, thanks to Moh El-Naggar. He inspired me to explore the eld
of biophysics, ensured that I get the best possible training and ultimately got me
to work with world{renowned astrobiologists! Under his benevolent guidance I
have begun to appreciate the numbers in nature. I express my sincere gratitude
towards him for his constant patience, motivation, and immense knowledge.
Besides my adviser, I also thank Ken Nealson and Jan Amend for their insight-
ful comments and encouragement throughout these years. Their respective group
members have guided me through various microbiological techniques and allowed
me to borrow/steal/use their multitude apparatus. I am obligated to Ken Nealson
who, through his candid ways, constantly reminded me the purpose of pursuing
iv
Ph.D. { sheer joy of science { and Jan Amend for indulging my curiosity in ex-
ploratory science. The `Life Underground' team at USC has introduced me to the
plethora of microbes present in nature. I am truly indebted to them!
My stay at USC would have been no fun without the kinship of Edmond Leung,
Matt Xu, Annie Rowe, Ben Gross, Ian McFarlane, Sahand Pirbadian, Hyesuk
Byun, and Amruta Karbelkar! Your constant enthusiasm and hard work has been
extremely infectious. I could not have asked for a better blend of people and a
more cheerful work environment during these years.
I nally acknowledge the funding agencies for this thesis: NASA Astrobiology
Institute 41 under cooperative agreement NNA13AA92A. I am also grateful for
a 2015-2016 merit fellowship from the USC Women in Science and Engineering
(WiSE) program.
v
Contents
Dedication ii
Epigraph iii
Acknowledgments iv
List of Figures ix
List of Tables xv
Abstract xvii
1 Introduction 1
2 Background 8
2.1 Energetics of life . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 Anaerobic respiration and dissimilatory metal reduction . . . . . . . 10
2.3 Extracellular electron transfer mechanisms . . . . . . . . . . . . . . 12
2.4 Subsurface bacteria and electrochemical enrichment . . . . . . . . . 15
2.5 Microbial community analysis . . . . . . . . . . . . . . . . . . . . . 17
2.6 Isolation Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.7 Electrochemical characterisation . . . . . . . . . . . . . . . . . . . . 21
3 Biophysical Mechanisms of Interfacial Electron Transfer inS.onei-
densis MR-1 24
3.1 Shewanella oneidensis MR-1: nanowire as outer membrane extensions 25
3.1.1 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.1.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.1.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.2 Disentangling extracellular electron transfer in Shewanella oneiden-
sis MR-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.1 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
vi
3.2.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4 Isolation and characterization of electrochemically active subsur-
face Delftia and Azonexus species 44
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.2 Material and Methods . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.2.1 Sampling Site and Initial Enrichment . . . . . . . . . . . . . 46
4.2.2 Electrochemical Enrichment Bioreactor . . . . . . . . . . . . 47
4.2.3 Bacterial Community Analysis of Enrichments . . . . . . . . 50
4.2.4 Isolation and Electrochemical Measurement of Pure Cultures 52
4.2.5 Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.3 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.3.1 Sampling Site . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.3.2 Microbial Community Analysis in the Electrochemical Biore-
actor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.3.3 Isolation of Delftia and Azonexus strains . . . . . . . . . . . 59
4.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5 In Situ Electrochemical Enrichment of Subsurface Bacteria at the
Sanford Underground Research Facility 66
5.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.2 Material and Methods . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.2.1 Field Measurements . . . . . . . . . . . . . . . . . . . . . . 68
5.2.2 In Situ Electrochemical Reactor Deployment . . . . . . . . . 69
5.2.3 Secondary Electrochemical Enrichment . . . . . . . . . . . . 72
5.2.4 Active Microbial Community Analysis . . . . . . . . . . . . 73
5.2.5 Isolations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.2.6 DNA extraction from isolates . . . . . . . . . . . . . . . . . 75
5.2.7 Electrochemical Measurement of Isolates . . . . . . . . . . . 75
5.2.8 Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.3 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.3.1 Current production and microbial community structure in
the in situ reactor . . . . . . . . . . . . . . . . . . . . . . . 76
5.3.2 Current production and microbial community structure in
laboratory electrochemical enrichment . . . . . . . . . . . . 81
5.3.3 Isolation of microbial strains and initial electrochemistry . . 85
5.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6 Summary and Conclusion 90
Bibliography 93
vii
Appendices 117
A Appendix: Isolation and characterization of electrochemically ac-
tive subsurface Delftia and Azonexus species 118
A.1 Ferrozine Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
A.2 Microbial community . . . . . . . . . . . . . . . . . . . . . . . . . . 119
A.3 Delftia acidovorans . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
A.4 Control run - without poised eletrodes . . . . . . . . . . . . . . . . 126
B Appendix: In Situ Electrochemical Enrichment of Subsurface
Bacteria at Sanford Underground Research Facility 127
B.1 In Situ Reactor Deployment and Sampling Techniques . . . . . . . 127
B.1.1 Electrical Connections . . . . . . . . . . . . . . . . . . . . . 127
B.1.2 Aqueous Chemistry . . . . . . . . . . . . . . . . . . . . . . . 129
B.1.3 Microbiological Sampling . . . . . . . . . . . . . . . . . . . . 130
B.2 In situ Reactor Chronoamperometry . . . . . . . . . . . . . . . . . 132
B.2.1 Decommissioning the in situ reactor . . . . . . . . . . . . . 132
B.3 Laboratory Electrochemical Reactor: Chronoamperometry and De-
commissioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
B.4 RNA extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
B.5 Mineral Preperation . . . . . . . . . . . . . . . . . . . . . . . . . . 139
B.6 Microbial community . . . . . . . . . . . . . . . . . . . . . . . . . . 142
B.7 Isolations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
viii
List of Figures
2.1 Schematic of the electron transport chain (ETC) in respiratory or-
ganisms. Electron transfer through the components of the ETC
drives the pumping of protons across the inner (mitochondrial) mem-
brane. The resulting proton motive force then powers ATP synthase
for the production of ATP. Figure adapted from [1]. . . . . . . . . . 9
2.2 Vertical proles of nutrients in common (left) freshwater (Lake Michi-
gan) and (right) marine (Black Sea) environments.The upper oxic
regions are compatible with eukaryotic life, whereas lower anoxic re-
gions are dominated by bacteria. The depth of oxygen depletion is a
function of the amount of organic carbon that reaches the sediment.
Figure from [2]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Reduction potentials of physiologically relevant reactions in electron
tranport chain employed by various bacteria. Standard reduction
potential (E
0
[mV vs SHE, 25C, pH = 7]) are indicated by dashed
lines. If physiological or environmental conditions are known to
shift the potential from the E
0
, redox windows are indicated by
solid lines. Phenazine1 = Phenazine-1-carboxylic acid; Phenazine2
= Phenazine-1-carboxamide. c-type cytochromes can cover a broad
range of redox potentials as indicated. Figure adapted from [3, 4]. . 11
2.4 Extracellular electron transfer mechanisms in Shewanella oneiden-
sis MR-1. The direct EET occurs via cell-surface attachment or
bacterial nanowire, whereas indirected EET requires redox active
molecules such as
avins. Figure from [5]. . . . . . . . . . . . . . . 13
2.5 (a) Heme b and Heme c are the two dierent heme types found
in biological systems. The structure of heme c includes covalent
thioether linkages to the protein. (b) Heme c binds to peptide chain
via the C-X1-X2-CH-binding motif (providing the proximal axial
histidine ligand) and with histidine as the distal axial ligand. Figure
from [6] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
ix
2.6 Extracellular electron transfer pathway in Shewanella oneidensis
MR-1. From the quinol pool, the respiratory electrons are trans-
ferred to cytochrome cymA located in the inner membrane and
subsequently to exterior via mtrCAB complex. Figure from [7] . . . 15
2.7 Secondary structure of Escherichia coli 16S rRNA. At the bottom,
16S rRNA gene illustrating the conserved (green) and hypervariable
(grey) regions. Figure from [8, 9] . . . . . . . . . . . . . . . . . . . 18
3.1 Cells containing periplasmic and cytoplasmic Green Fluorescent Pro-
tein (GFP). S. oneidensis MR-1 expressing GFP fused with the
twin-arginine translocation (Tat) signal peptide from the E. coli
TorA. Cells were imaged on an agar pad following LB growth and
IPTG induction. Due to the presence of the signal peptide, the fu-
sion protein is exported to the periplasm, resulting in
uorescence
limited primarily to the periphery of the cell (Scale bar = 1 m.
Inset: Fluorescence image of S. oneidensis MR-1 constitutively ex-
pressing GFP with no signal sequence for export, resulting in a uni-
form cellular
uorescence pattern throughout the cytoplasm. Scale
bar = 2 m. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2 Bacterial nanowires from S. oneidensis MR-1 strains containing GFP
only in the cytoplasm (Upper) or in the periplasm as well (Lower).
The green and red channels monitor GFP (Left) and FM 4-64FX
(right)
uorescence, respectively. The nanowires display green
uo-
rescence only when GFP is present in the periplasm (C) (Scale bar
= 2 m). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3 Left: Electrochemical activity of fabricated microelectrodes using
the standard ferri/ferro cyanide redox couple. The CV of a single
working electrode exposed to a solution of 10 mM K
3
Fe(CN)
6
in 1 M
KNO
3
, depicts the classic expected ultramicroelectrode (UME i.e.
with a critical dimension smaller than the length scale of the sur-
rounding depletion later) behavior with a sigmoidal quasi-reversible
transition between the oxidized and reduced states [10]. Right: Elec-
tromicrograph of the fabricated microelectrode depicts the working
electrode visible through the nonpassivated windows and central
counter electrode (Scale bar = 100 m). . . . . . . . . . . . . . . . 38
x
3.4 Chronoamperometry of S. oneidensis MR-1 interacting with a ITO
(top left) and carbon cloth (top right) working electrode poised at
0.422 and 0.44 V vs. SHE. Turnover cyclic voltammetry (CV) re-
veals one catalytic wave with onset at 0.1 V vs. SHE corresponding
to
avin-dependent EET pathway in ITO biolm (bottom left) and
two catalytic waves corresponding to
avin-dependent and direct
EET via multiheme cytochrome pathways in Carbon cloth biolm
(bottom right) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.1 Schematic diagram of the electrochemical enrichment reactor. The
four carbon cloth working electrodes were poised at 272 mV, 372
mV, 472 mV, 572 mV vs. SHE, respectively, using a 4-channel
potentiostat. The reactor and medium reservoir were continuously
purged with lter-sterilized N
2
gas to maintain anaerobic conditions. 49
4.2 Electrochemical enrichment. The increasing anodic current vs. time
points to enrichment of microbial communities capable of mediating
extracellular electron transfer to the 4 working electrodes. Arrows
indicate times of medium change, which resulted in an abrupt de-
crease of anodic current before recovery of the increasing trend. . . 56
4.3 Scanning electron microscopy of the enrichment electrodes and as-
sociated biomass (1 m scale bar). . . . . . . . . . . . . . . . . . . 57
4.4 Family level microbial community analysis of the electrochemical
enrichment. Families were determined using 16S rRNA tagged py-
rosequencing analysis of DNA extracts. A batch culture (acetate
as electron donor and Fe(III)-NTA as electron acceptor) was the
inoculum for the electrochemical reactor. . . . . . . . . . . . . . . . 58
4.5 Chronoamperometry of the isolated strains (A) Delftia sp. WE1-13,
(B) Azonexus sp. WE2-4. In both cases, the working electrode was
poised at 522 mV vs SHE, acetate (10 mM) was used as electron
donor, and reactors were purged with N
2
to maintain anaerobic
conditions. Insets show the total protein content measured at time
points labeled 1 and 2. . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.6 Further electrochemical and microscopic characterization of Delftia
sp. WE1-13. (A) Medium change resulted in a sudden decrease
of anodic current, as observed in the enrichment cultures. Inset
shows total protein content measured at the time points marked 1
and 2. (B) Scanning electron microscopy (1 m scale bar) shows
attachment of the rod-shaped of Delftia sp. WE1-13 on the carbon
cloth bers. (C) In vivo
uorescent image (FM 4-64 FX membrane
stain) conrms the attachment of Delftia sp. WE1-13 cells to the
electrode during electrochemical analysis (10 m scale bar). . . . . 62
xi
5.1 A) Schematic diagram of the in situ electrochemical enrichment re-
actor. The four carbon cloth working electrodes were poised at -0.19,
+0.01, +0.26 and +0.53 V vs. SHE, respectively, using a 4-channel
potentiostat to enrich for electrode oxidising and reducing microbes
in a single reactor. DUSEL 3A borehole water was fed directly into
the reactor and the euent was pumped out to Yates shaft. B)
Geological map of the 4850 ft L at SURF showing drilled boreholes
intersecting Precambrian Yates member and tertiary rhyolite. the
reactor was deployed to enrich for potential electrode oxidising and
reducing bacteria from borehole DUSEL 3A water. C) Photograph
of the installed the reactor near borehole DUSEL 3A manifold. . . . 70
5.2 Schematic diagram of the lab electrochemical reactors employed to
enrich for electrode oxidising and reducing microbes. Carbon cloth
was used as working electrodes (1 cm x 1 cm) with platinum act-
ing as counter electrode and Ag/AgCl (saturated in 1 M KCL) as
reference electrode. . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.3 Microbial community of the borehole DUSEL 3A versus dierently
poised electrodes in the reactor depicts enrichment of unclassied/novel
microbiota at the reducing potentials (-0.19 V and 0.01 V vs SHE)
where microbes use the electrode as the electron donor. . . . . . . . 78
5.4 Bacterial community grouped with an increasing (a), decreasing (b),
and random pattern (c,d) at poised electrodes of in situ electrochem-
ical reactor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.5 A) Mean anodic and cathodic currents observed in laboratory en-
richments at WE1 (-0.19 V vs SHE) and WE4 (+0.53 V vs SHE)
were higher than their respective abiotic controls. Similarly, dier-
ence was observed in cyclic voltammetry of electrode oxidising (B)
and reducing (C) biolms compared to the abiotic controls. This
indicates that the electron transfer was mediated by the resident
microorganisms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.6 Scanning electron microscopy of the enrichment electrodes associ-
ated biomass at (A) reducing potential (-0.19 V vs SHE) and (B)
oxidising potential (0.53 V vs SHE). Enriched biolms at reducing
potentials comprised of long thin lamentous like bacteria whereas
biolms at oxidising potentials contain extracellular appendages as
observed in case of Shewanella and Geobacter. Scale bar A) 100m
and B) 1 m. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.7 Family level bacterial community analysis of the electrochemical
enrichment. Families were determined using 16S rRNA tagged py-
rosequencing analysis of RNA extracts. . . . . . . . . . . . . . . . . 84
xii
5.8 Chronoamperometry of the isolated strains (A) Comamonas sp.
(strain WE1-1D1) and (B) Bacillus sp. (strain WE4-1A1-BC).
While Comamonas sp. (strain WE1-1D1) was enriched on work-
ing electrode poised at -0.19 V vs. SHE with CO
2
as the carbon
source and oxygen as the electron donor, Bacillus sp. (strain WE4-
1A1-BC) was enriched on working electrode poised at 0.53 V vs.
SHE with acetate as the electron donor and carbon source. Arrow
indicate the time point when the bioreactor was inoculated with the
cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
A.1 Family-level bacterial lineages of NDW2 groundwater and following
enrichments. Down-
ow Hanging Sponge reactor (DHS), purging
with 80:20 v:v H
2
:CO
2
, targeted autotrophic Fe- and Mn-reducing
microbes. The rst electrochemical reactor run (Run1) mimicked
the conditions of the DHS reactor. Further enrichments was fo-
cused on isolating Fe-reducers on acetate as electron donor and
Fe(III)-NTA as electron acceptor. The two batch bacterial commu-
nity were obtained from DNA extracts performed on two dierent
dates from the same sample. This batch culture was the inoculum
for the second electrochemical reactor with dened medium. . . . . 123
A.2 Chronoamperometry (center) of Delftia sp. WE1-13, Azonexus sp.
WE2-4 and D. acidovorans (DSMZ DSM-39) shows production of
current in hundreds of nA range for all. Current production by
Delftia sp. WE2-4 and D. acidovorans began declining at approxi-
mately 20 hrs, while Azonexus sp. WE2-4 produced a steady current
for 90 hours before decreasing. In all the cases working electrode
poised at 522 mV vs. SHE, acetate (10 mM) as the electron donor
and reactors were purged with N
2
to maintain anaerobic conditions.
The current production was correlated with a constant planktonic
cell density (left) and slight increase in total protein content (right)
in case of D. acidovorans. The protein content values include pro-
tein measured from poised carbon cloth (CC) associated cells and
planktonic cells (Plank) . . . . . . . . . . . . . . . . . . . . . . . . 124
B.1 Electrical connections in the in situ reactor to avoid any electrical
shock (top). This circuit forced all the electrical instruments to turn
o during electrical outage or a water leakage (bottom right) and
had not aect during normal operations (bottom left). . . . . . . . 128
B.2 Chronoamperometry from the in situ reactor and continuous aque-
ous chemistry measurements. . . . . . . . . . . . . . . . . . . . . . 131
xiii
B.3 Chronoamperometry from the laboratory enrichment reactor indi-
cates growth of electrode oxidisers at the lowest potential (-0.19 V
vs. SHE) and electrode reducers at the higher potentials (+0.26 and
+0.53 V vs. SHE). . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
B.4 Raw data of phylum-level bacterial community abundances of the
dehumidier water, DUSEL 3A water and in situ reactor. . . . . . 145
B.5 Electron micrograph of A) Bacillus sp. (strain WE4-1A1-BC) and
B) Comamonas sp. (strain WE1-1D1) on poised electrode. . . . . . 146
xiv
List of Tables
3.1 Strains and plasmids used in this study. . . . . . . . . . . . . . . . . 29
4.1 Chemical composition of water collected from Nevares Deep Well 2
(Death Valley, CA, USA). . . . . . . . . . . . . . . . . . . . . . . . 47
5.1 Chemical composition of water from DUSEL 3A borehole, 4850 ft
level, Sanford Underground Research Facility (SD, USA). . . . . . . 69
5.2 Mean current (last 40 days) observed at various redox potentials
in the in situ reactor and aqueous water chemistry of the in situ
reactor euent. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
A.1 Normalised abundances of bacterial phyla in groundwater from Nevares
Deep Well 2 (Death Valley, CA, USA). . . . . . . . . . . . . . . . . 120
A.2 Sequences used for tagged sequencing of 16S rRNA genes from the
samples. Linker primer sequence was GTGCCAGCMGCCGCG-
GTAA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
A.3 Family-level bacterial lineages identied in NDW2 groundwater and
following enrichments. Down-
ow Hanging Sponge reactor (DHS),
purging with 80:20 v:v H
2
:CO
2
, targeted autotrophic Fe-and Mn-
reducing microbes. The rst electrochemical reactor run (Run1)
mimicked the conditions of the DHS reactor. Further enrichments
was focused on isolating Fe-reducers on acetate as electron donor
and Fe(III)-NTA as electron acceptor. The two batch bacterial com-
munity were obtained from DNA extracts performed on two dierent
dates from the same sample. This batch culture was the inoculum
for the second electrochemical reactor with dened medium. . . . . 122
A.4 Heme-binding sites Cys-X1-X2-Cys-His (CXXCH) detected in D.
acidovorans SPH-1 and further putative localisation as predicted
by PSORTB. These localisation are compared with some proteins
containing heme-binding sites in E. coli str. K-12 substr. MG1655
and S. oneidensis MR-1. . . . . . . . . . . . . . . . . . . . . . . . . 125
xv
B.1 Sequences used for tagged sequencing of 16S rRNA genes from the
samples from SURF (top) and laboratory electrochemical enrich-
ments (bottom). Linker primer sequence was GTGCCAGCMGC-
CGCGGTAA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
B.2 Phyla-level bacterial lineages of SURF DUSEL 3A water, dehumid-
ier water, and the in situ reactor. . . . . . . . . . . . . . . . . . . 149
B.3 Phyla-level bacterial lineages in laboratory electrochemical enrich-
ment reactor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
B.4 Strains isolated, the respective strategies involved and observed oxy-
gen tolerance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
xvi
Abstract
Microorganisms are the unknown majority life forms on the Earth with approx-
imately 10
30
cells. While the marine subsurface biomass has been estimated at
1.5 { 22 petagrams of carbon, the continental subsurface biomass (top 2 km) ac-
counts for 14 { 135 petagrams of carbon. Subsurface microorganisms can thrive in
extreme environmental conditions (e.g., pressure, temperature, pH, connement),
and derive energy via diverse metabolic processes. While these environments are
geologically diverse, potentially allowing energy harvesting by microorganisms that
catalyze redox reactions, many of the abundant electron donors and acceptors are
insoluble and therefore not directly bioavailable. Extracellular electron transfer
(EET) is a metabolic strategy that microorganisms can deploy to either donate
or receive electrons to and from insoluble substrates outside of the cell. Despite
its environmental importance and technological promise, EET is mechanistically
characterized in only two model dissimilatory metal reducing bacteria (DMRB)
Shewanella and Geobacter.
This thesis contributes towards an understanding of the EET mechanisms in
the model system S. oneidensis MR-1. Firstly, I focus on in vivo imaging of
nanowires formed by S. oneidensis MR-1 expressing periplasmic or cytoplasmic
green
uorescent protein (GFP). These images points towards the composition of
xvii
Shewanella nanowires as outer membrane extensions. Secondly, I describe dier-
ential EET behaviour of S. oneidensis MR-1 dependent on the electrode surface
(tin doped indium oxide vs. carbon cloth), and an experimental system which
allows simultaneous analysis of both the
avin-independent and
avin-dependent
routes of EET between S. oneidensis MR-1 and the poised electrodes.
Since EET can be mimicked on the poised electrodes which can serve as un-
limited source of energy, it opens the door to electrochemical techniques to enrich
for and quantify the activities of environmental microorganisms beyond the well
studied model systems. I report the electrochemical enrichment of microorganisms
from a deep fractured-rock aquifer in Death Valley, California, USA. In experiments
performed in mesocosms containing a synthetic medium based on aquifer chem-
istry, four working electrodes were poised at dierent redox potentials (272, 373,
472, 572 mV vs. SHE) to serve as electron acceptors, resulting in anodic currents
coupled to the oxidation of acetate during enrichment. The anodes were dominated
by Betaproteobacteria from the families Comamonadaceae and Rhodocyclaceae. A
representative of each dominant family was subsequently isolated from electrode-
associated biomass. The EET abilities of the isolated Delftia strain (designated
WE1-13) and Azonexus strain (designated WE2-4) were conrmed. Both genera
have been previously observed in mixed communities of microbial fuel cell enrich-
ments, but this is the rst direct measurement of their electrochemical activity.
While alternate metabolisms (e.g. nitrate reduction) by these organisms were pre-
viously known, our observations suggest that additional hidden interactions with
external electron acceptors are also possible.
Next, I describe the rst in situ potentiostatically controlled electrochemical en-
richment of subsurface microorganisms from borehole DUSEL 3A located at 4850
ft below the surface in Sanford Underground Research Facility, South Dakota,
xviii
USA. The mine is hosted in quartz-veined, sulde rich segments of an Early Pro-
terozoic, carbonate-facies iron-formation providing in situ study of subsurface mi-
croorganims from the early Earth. Here, we poised electrodes at electron-donating
(reducing/cathodic) and electron-accepting (oxidising/anodic) redox potentials, to
enrich for potential mineral oxidising and reducing microorganisms. We were able
to capture the bacterial community present in DUSEL 3A onto the poised elec-
trodes; all the other bacterial families were represented on at least one of the
poised electrodes. This demonstrates that in situ electrochemical techniques are
promising for studying subsurface bacterial communities. The results also show a
dominance of unclassied bacterial lineages at reducing potentials highlighting the
reducing power available in the subsurface. Finally, we successfully isolated several
microbes from genera including Bacillus, Anaerospora, Comamonas, Cupriviadus
and Azonexus from these poised electrodes. Initial electrochemical studies on a
representative electrode oxidising gram-negative Comamonas sp. (strain WE1-
1D1) and a representative electrode reducing gram-positive Bacillus sp. (strain
WE4-1A1-BC) are reported.
xix
Chapter 1
Introduction
Although it was only in the last 300 years that electron
ow was harnessed to
power much of the technology used in modern society, it has existed in all livings
organisms from animals, plants to bacteria for their sustenance since the origin
of life [11, 12]. While respiratory organisms transfer high energy electrons from
an electron donor, through various steps to the terminal electron acceptor, photo-
synthetic organisms derive energy from sunlight for electron transfer. Simply put
this
ow of electrons through various macromolecules drives the energetics of life
[13, 14].
In respiration, hydrogen atoms are extracted from complex organic molecules
to form hydrogen carriers through glycolysis and the citric acid cycle. These hy-
drogen carriers are later regenerated in the electron transport chain located in the
cellular inner-membrane of microbes (or in mitochondrial membrane in eukary-
otes). The electron transport chain consists of a series of membrane bound protein
complexes in which the hydrogen carriers are split into protons and electrons. The
electrons are passed down the chain and then reduce the terminal electron accep-
tor. A proton gradient is created across the inner membrane as a result, and is
used to generate adenosine triphosphate (ATP), the energy currency of life [15].
This process is known as oxidative phosphorylation. Chemical energy stored in
ATP is later used for biosynthesis of macromolecules and converted into various
forms of energy including mechanical work. ATP is also formed via substrate level
1
phosphorylation (SLP) where high-energy phosphate is transferred directly from a
substrate/organic molecule to adenosine diphosphate (ADP).
Within our ever inspiring nature, microorganisms have a ubiquitous presence
[16], from the Earth's atmosphere in clouds [17] to the continental and marine
subsurface [18]. In aerobic metabolism for energy/ATP production, oxidation of
complex organic molecules is coupled to oxygen reduction through the electron
transport chain, known as oxidative phosphorylation. That said, microorganisms
are renowned for their metabolic versatility. They can benet from numerous
alternate electron donors/acceptors determined by the environmental conditions
in which they thrive in. These include employing soluble electron donors such
as dissolved organic carbon, H
2
, methane and soluble electron acceptors such as
nitrate and sulfate [2, 19{23]. Recently, microorganisms that can utilize insoluble
electron donors and acceptors (iron, manganese and sulfur) are gaining attention
[24{30], as this form of metabolism may have existed in reducing environments of
early Earth [31].
The isolation of dissimilatory metal-reducing bacteria (DMRBs) in late 1980's,
such as the Gammaproteobacteria Geobacter and Deltaproteobacteria Shewanella,
pioneered research on underlying principles of electron transfer to/from the min-
erals or poised electrodes through the cellular membrane, termed as extracellular
electron transfer (EET) [32, 33]. Multiheme cytochromes residing in the periplas-
mic and outer-membrane of the bacteria play a crucial role in transferring electrons
from inner-membrane to external substrates [7, 34, 35]. These outer-membrane cy-
tochromes may either directly transfer electrons to the insoluble electron acceptor
(direct electron transfer) or interact with redox active molecules such as humic
2
substances and
avins to indirectly transfer electrons to the external surfaces [36{
40]. A third EET strategy requires production of extracellular appendages called
bacterial nanowires [41{43].
Bacterial nanowires were rst discovered in Geobacter sulfurreducens DL-1 and
Shewanella oneidensis MR-1 (2006), and were conductive across their width under
non-physiological conditions [41, 43]. While bacterial nanowires of Geobacter were
3 { 5 nm in diameter and up to 20m in length [43, 44], Shewanella nanowires were
reported as lament bundles approximately 50 { 100 nm in diameter and extend
by many microns from the cell surface [41, 42]. Geobacter nanowire were postu-
lated to be type IV pili composed of the monomer PilA. Type IV pili are widely
expressed among gram-negative bacteria facilitate biolm formation, transfer of
DNA, motility and bacterial attachment to various surfaces. Deletion of the gene
encoding for PilA in Geobacter resulted in fewer laments per cell and complete
loss of ability to reduce insoluble Fe(III) oxide minerals [43]. While the structure
of Geobacter nanowire is widely accepted as composed of pilin protein, the compo-
sition of Shewanella nanowires was unknown and were only referred to as pilus-like
[41]. In chapter 3, I report my contribution towards understanding the composi-
tion and structure of Shewanella oneidensis MR-1, by localizing green
uorescent
proteins inside the cellular periplasm and imaging the production of nanowire in
vivo in physiological conditions. This observation demonstrated that the nanowire
of S. oneidensis MR-1 are cellular outer-membrane extensions. This was a crucial
step for further experiments which resulted in proposing Shewanella nanowire as
outer-membrane extensions comprising of outer-membrane cytochromes [45].
Mediated electron transfer is an alternate mechanism of EET which requires
interaction of outer-membrane multiheme cytochrome with a mobile small redox
molecule. Such molecules get reduced at the cell surface by the outer-membrane
3
cytochromes, diuse in solution to reach the terminal electron acceptor and get oxi-
dized at the metal oxide surface (acting as an electron shuttles). Such a shuttle pro-
vides a mechanism for an indirect reduction process. Some examples of molecules
that act as redox shuttles are humic substances, quinones, phenazines, and
avins,
but anything that is reversibly redox active and has the right redox potential (e.g.,
poised between the reduction potential of the electron donor/reductant and the
electron acceptor/oxidant) could serve the purpose. This process is proposed to
occur in cycles, transferring electrons at a fast enough rate to support the respi-
ration rate of cells [46]. However, this mechanism requires a large concentration
of shuttle molecules and relatively fast diusion (i.e. large diusion coecient) in
solution.
Mediated electron transfer had been implicated as a primary mechanism of
extracellular electron transfer to insoluble electron acceptors in anaerobic cultures
of the S. oneidensis [37]. The underlying principles of mediated electron trans-
fer is just now coming into light. Recently, the x-ray crystal structures of the
Shewanella outer-membrane cytochromes, including decaheme MtrC, revealed a
conserved disulde bond which when reduced in presence of
avin mononucleotide
(FMN) results in the formation of a stable
avocytochrome complex [47]. This
indicated that
avin bound to the outer-membrane cytochromes may enhance the
electron transfer process. This structural evidence conformed to the hypothesis
proposed earlier by Okamato et al. [39, 48]. In chapter 3, I describe initial ob-
servation of dierential EET behavior of S. oneidensis MR-1 dependent on the
electrode surface (tin doped indium oxide vs. carbon cloth). While direct EET
via multiheme cytochromes was observed in tin doped indium oxide (ITO) based
reactors, carbon cloth based reactors were able to capture
avin dependent and
4
independent routes of EET. These results inclined us to employ carbon based re-
actor for further electrochemical analyses of Shewanella biolms to disentangle
the
avin-independent and
avin-dependent routes of EET between S. oneidensis
MR-1 and poised electrodes [49].
While rigorous studies are being conducted to understand the EET mecha-
nisms active in Shewanella and Geobacter, since their isolation many other bac-
terial genera have been shown to associate with both the anode and cathode of
an electrochemical cell [50{55]. Novel organisms have been isolated and shown
capable of using poised electrodes as electron donors or acceptors [56{62] in place
of naturally occurring minerals. The relevance of EET in the subsurface [24{
30], where microorganisms may harness insoluble electron donors/acceptors (iron,
manganese and sulfur) for energy or sustenance led us to build electrochemical en-
richment techniques which enrich for microbes at specic redox potentials. Since
electrodes act as unlimited electron sources or sinks they are powerful tools for
enriching subsurface organisms potentially capable of EET. This thesis is focused
on developing and harnessing electrochemical enrichment techniques to reveal the
diversity of subsurface microorganisms. Our approaches have been applied both ex
situ and in situ to mimic the complex interactions and energetic gradients present
in two NASA Astrobiology Institute (NAI) eld sites - Nevares Deep Well 2, Death
Valley, USA and Sanford Underground Research Facility, South Dakota, USA.
This thesis consists of 6 chapters. Chapter 2 introduces the bioenergetics of
life, known mechanisms of extracellular electron transfer (EET) and the relevance
of EET to subsurface environments. This chapter also includes strategies involv-
ing microbial community analysis, isolation of pure colonies, and electrochemical
characterizations.
5
Chapter 3 describes the two experimental platforms which were instrumental
in understanding EET mechanisms employed by S. oneidensis MR-1. Firstly, by
expressing green
uorescent protein in the periplasmic space of the gram negative
S. oneidensis MR-1 nanowires, we show that the nanowires produced are outer
membrane extensions via in vivo imaging. Secondly, micro-liter and milli-liter
electrochemical bioreactors were designed to tease apart the
avin dependent and
independent mechanisms using voltammetric techniques. In this study we found
that the
avins accelerate EET primarily as cytochrome-bound cofactors, rather
than as a free soluble mediator.
Chapter 4 reports the laboratory based electrochemical enrichment and fur-
ther isolation of Delftia and Azonexus from a deeply sourced artesian well located
in Death Valley, California, USA. While alternative anaerobic metabolisms (e.g.
nitrate reduction) by Delftia and Azonexus species were previously known, our
study suggests that these organisms can also gain some energy for survival and/or
persistence by electron transfer to external surfaces.
Chapter 5 describes the installation of the rst in situ electrode cultivation
(ISEC) reactor at Sanford Underground Research Facility (SURF) in the former
Homestake Gold Mine (South Dakota, USA), with electrodes poised at electron-
donating (reducing/cathodic) and electron-accepting (oxidising/anodic) redox po-
tentials, to enrich for bacteria capable of oxidising and/or reducing minerals.
We successfully isolated several microbes including Bacillus, Anaerospora, Coma-
monas,Cupriviadus and Azonexus from these poised electrodes and initial elec-
trochemical studies on electrode oxidiser Comamonas sp. (strain WE1-1D1) and
electrode reducer Bacillus sp. (strain WE4-1A1-BC) are elucidated. .
Finally, chapter 6 concludes this dissertation. We review the work presented in
this thesis and look at possible future improvements to the in situ electrochemical
6
enrichments techniques and intriguing questions which can be asked to advance
our knowledge of extracellular electron transfer mechanisms in the already isolated
subsurface microorganisms.
7
Chapter 2
Background
2.1 Energetics of life
Microorganisms are the dominant life forms on Earth comprising of approximately
10
30
cells [16]. They make up about 40-60% of the biomass of the terrestrial
subsurface [63]. Being highly ordered systems compared to the environment, they
continuously gain free energy from their surroundings to maintain this imbalance
of order [64]. They derive their ability to thrive and survive in a wide range
of environments from the great metabolic versatility that exists among them [65].
They store energy in the form of adenosine triphosphate (ATP). The ATP molecule
with high energy phosphate bonds releases energy when a single phosphate group
is hydrolyzed, converting ATP into adenosine diphosphate (ADP). This released
energy is used to power the energy demanding processes of life.
All respiratory organisms gain energy (ATP) from electron transfer which cou-
ples oxidation of a high energy electron donor to reduction of low energy electron
acceptor. They share the basic catabolic pathway such as glycolysis for sugar
metabolism and citric acid cycle oxidizing the acetyl carbons in acetyl-CoA to
carbon dioxide producing NADH and FADH
2
with low energy electrons. In oxida-
tive phophorylation (Figure 2.1), the low energy electrons travel through electron
transport chain comprising of carrier sites and proton pumps situated in the in-
ner membrane of the bacteria (or inner mitochondrial membrane) and ultimately
8
Figure 2.1: Schematic of the electron transport chain (ETC) in respiratory organ-
isms. Electron transfer through the components of the ETC drives the pumping of
protons across the inner (mitochondrial) membrane. The resulting proton motive
force then powers ATP synthase for the production of ATP. Figure adapted from
[1].
reduce the terminal electron acceptor. As electrons transport through the car-
rier sites, protons are pumped out to create an electrochemical gradient across
the inner membrane. Subsequently, these protons move to the interior through
ATP synthase [66]. The energy from the proton gradient dissipation is harnessed
by ATP synthase to transfer phosphoryl group to ADP leading to formation of
ATP [15]. An alternative way to create ATP is through substrate-level phophory-
lation, where a high-energy phosphoryl is transferred directly from a substrate
to ADP thus forming ATP. In case of S. oneidensis MR-1 substrate-level phos-
phorylation is the primary anaerobic energy conservation strategy [67]. Lastly,
During fermentation, pyruvate is metabolized to various compounds such as lactic
acid, ethanol and carbon dioxide or other acids. Fermentation without substrate
level phosphorylation uses an endogenous electron acceptor, which is usually an
9
Figure 2.2: Vertical proles of nutrients in common (left) freshwater (Lake Michi-
gan) and (right) marine (Black Sea) environments.The upper oxic regions are com-
patible with eukaryotic life, whereas lower anoxic regions are dominated by bacte-
ria. The depth of oxygen depletion is a function of the amount of organic carbon
that reaches the sediment. Figure from [2].
organic compound (lactic acide and ethanol fermentation). Each ATP molecule
stores about 40-60 kJ/mol free energy under physiological conditions which can
be released through ATP hydrolysis [68] during synthesis of macromolecules like
proteins, complex lipids, RNA and DNA.
2.2 Anaerobic respiration and dissimilatory
metal reduction
With oxygen being the most favorable electron acceptor (Figure 2.3), many mi-
crobes use it as the terminal electron acceptor via aerobic respiration. After organic
matter is degraded in sediments by aerobic respiration, sediments quickly become
anoxic with depth resulting in other electron acceptors getting consumed, known
10
Figure 2.3: Reduction potentials of physiologically relevant reactions in electron tranport chain employed by various
bacteria. Standard reduction potential (E
0
[mV vs SHE, 25C, pH = 7]) are indicated by dashed lines. If physiological
or environmental conditions are known to shift the potential from the E
0
, redox windows are indicated by solid lines.
Phenazine1 = Phenazine-1-carboxylic acid; Phenazine2 = Phenazine-1-carboxamide. c-type cytochromes can cover a
broad range of redox potentials as indicated. Figure adapted from [3, 4].
11
as anaerobic respiration (Fig 2.2) [2]. Some of these organisms, known as dissim-
ilatory metal reducing bacteria (DMRB), have evolved to use insoluble mineral
oxides to dispose of their low energy electrons extracellularly. Much of what is
known about extracellular electron tranfer (EET) comes from studies of model
DMRB including Shewanella and Geobacter, involving the movement of electrons
out of the cell to insoluble metal oxides [69, 70].
Shewanella oneidensis MR-1, isolated primarily as manganese oxide reducing
organism in 1988 [71], is a facultative anaerobe capable of using many electron
acceptors for growth. Although growth was conrmed in presence of manganese
oxide, extracellular electron transfer coupled with proton translocation was con-
rmed later by using protonophores and electron transport inhibitors [72]. Re-
cently, DMRB have also shown to extracellularly reduce solid phase electron ac-
ceptors, such as poised carbon electrodes or indium tin oxide lms, which has
powered research in microbial fuel cells (MFC )[51, 73] and single cell research
[74].
2.3 Extracellular electron transfer mechanisms
Extracellular electron transfer (EET) is the capability of microorganisms to trans-
port electrons to and from insoluble substrates outside of the cell. The mechanisms
of EET that have been characterized include interaction of the cellular electron
transport chain with an insoluble substrate through surface proteins (outer mem-
brane multiheme cytochromes) [7, 34, 35], small molecules (
avins, humic acids,
etc.) that are proposed to shuttle electrons between the cell surface and substrate
[36{40], and extracellular appendages (nanowires) [41{43].
12
Figure 2.4: Extracellular electron transfer mechanisms in Shewanella oneidensis
MR-1. The direct EET occurs via cell-surface attachment or bacterial nanowire,
whereas indirected EET requires redox active molecules such as
avins. Figure
from [5].
Outer-membrane cytochromes are multi-heme cytochromes that play a crucial
role in EET. Cytochromes are a large family of proteins that contain heme pros-
thetic groups to function as electron carriers either in photosynthesis or respira-
tion [75]. In c-type cytochromes, the heme cofactor covalently binds, via thioether
bonds ({ S {), with the cysteines of the heme binding motif Cys-X1-X2-Cys-His
(CXXCH) in the protein (Figure 2.5). Here, histidine provides an axial ligand for
the heme iron and X can be any amino acid [76]. Outer-membrane cytochromes
13
Figure 2.5: (a) Heme b and Heme c are the two dierent heme types found in
biological systems. The structure of heme c includes covalent thioether linkages to
the protein. (b) Heme c binds to peptide chain via the C-X1-X2-CH-binding motif
(providing the proximal axial histidine ligand) and with histidine as the distal axial
ligand. Figure from [6]
contain multiple c-heme cofactors [6, 77, 78], and hence can transfer electrons at
relatively long distances. S. oneidensis MR-1 contains 39 genes encoding c-type
cytochromes (14 of which contain 4 or more hemes) [79] compared to E. coli which
has only 6 genes encoding c-type cytochromes. On the other hand, G. sulfurre-
ducens contains 111 genes encoding c-type cytochromes (73 of which contain more
than 2 hemes) [80].
These multiheme cytochromes carry the electrons from the inner membrane
(the electron transport chain) to the periplasm, through a porin in the outer
membrane, and to the exterior of the cell [81, 82]. As part of this chain, `porin-
cytochrome' modules are found to transport electrons across the outer membrane
[82, 83]. Overall, this array of cytochromes help conduct electrons necessary for
electron transfer (Figure 2.6)).
14
Figure 2.6: Extracellular electron transfer pathway in Shewanella oneidensis MR-
1. From the quinol pool, the respiratory electrons are transferred to cytochrome
cymA located in the inner membrane and subsequently to exterior via mtrCAB
complex. Figure from [7]
2.4 Subsurface bacteria and electrochemical en-
richment
The subsurface geological environment oers insoluble electron donors and accep-
tors in the form of redox active elements in minerals (e.g. S, Fe, Mn) associated
with sediments and rocks, after other bioavailable soluble electron donors and ac-
ceptors are utilized [27, 29, 32]. Furthermore, many organisms beyond Shewanella
15
oneidensis [84] and Geobacter sulfurreducens [85], have been shown to utilise elec-
trodes as a surrogate for terminal electron acceptors including novel isolated organ-
isms such as Thermincola sp. strain JR, Ochrobactrum anthropi YZ-1, Arcobacter
butzleri, and Comamonas denitricans [57, 86{88]. These studies employed two
electrode set-up (microbial fuel cell) where the current-carrying electrodes are also
used for measuring voltage drop across the whole electrochemical cell. However
three electrode set-up is necessary to study the electrochemical behaviour of mi-
crobes. It provides precise control of the redox conditions of electrochemical inter-
face (cell-electrode interaction) by poising very specic potentials at the working
electrode with reference to a constant reference electrode.
Three-electrode electrochemical cell has been widely used to study underlying
principles of EET in case of Shewanella and Geobacter [89{91]. Selection of diverse
anode respiring bacteria from wastewater-activated sludge performed via electro-
chemical cell at various potentials (E
anode
= 0.15, 0.09, +0.02, and +0.37 V vs.
SHE) resulted in predominance of G. sulfurreducens at the lowest potentials and a
high diversity of bacterial community at the highest potential [92]. In our lab, we
have also investigated growth of S. oneidensis MR-1 on tin-doped indium oxide
(ITO) electrodes poised at 0.2 V vs Ag/AgCl- KCl sat. electrode (E
0
0
vs SHE =
0.222 V).
Microbes capable of electron uptake were also studied using electrodes as the
electron donors. G. metallireducens was the rst reported organism to take up elec-
trons from electrodes poised at {0.5 V vs. Ag/AgCl while simultaneously reducing
nitrate as the terminal electron acceptor [93]. Later, many microbes were shown to
potentially oxidize negatively poised electrodes and the review by Rosenbaum et al.
16
[55] elucidated putative electron uptake mechanism involving direct electron trans-
fer via cytochrome c and hydrogenases. Direct studies of natural iron-oxidising
microbes using electrode as the electron donor was demonstrated in Maripro-
fundus ferroxidans [59], Rhodopseudomonas palustris TIE-1 [60], Methanococcus
maripaludis [94] and Acidithiobacillus ferrooxidans [61]. Furthermore, the impor-
tance of diverse electron uptake mechanisms from electrode was demonstrated by
isolating phylogenetically dierent marine microbes within Gammaproteobacteria
and Alphaproteobacteria family from single electrochemical incubation [62]. Rising
interest in isolating and enriching electrode oxidising bacteria will result in a bet-
ter understanding of the underlying principles of electron uptake from electrodes,
which until now remains illusive.
2.5 Microbial community analysis
With more than 99% of the microorganisms dicult to culture using standard
culturing techniques known as the 'great plate count anomaly' [95], the primary
source of information for these uncultured but viable representatives is derived by
employing culture-independent nucleic acid approaches. These techniques include
analyses of whole genomes or selected genes such as 16S and 18S rRNA (ribosomal
RNA) for prokaryotes and eukaryotes, respectively. The 16S small ribosomal RNA
subunit gene as a taxonomic marker for microbes was rst suggested by Woese
and Fox [96], and since then has been widely accepted as a gold standard [97].
Several aspects make the 16S rRNA gene an optimal marker for microbial
diversity analysis (Figure 2.7). These include (i) its presence in almost all bacteria
and archaea; (ii) its preserved function over time, suggesting that any random
sequence change acts as a `molecular clock', and (iii) the 16S rRNA gene with
17
Figure 2.7: Secondary structure of Escherichia coli 16S rRNA. At the bottom,
16S rRNA gene illustrating the conserved (green) and hypervariable (grey) regions.
Figure from [8, 9]
about 1550 bp length contains 8 highly conserved regions (U1-U8), which can be
used for designing amplication primers across taxa, as well as nine hypervariable
regions (V1-V9), which can be used to distinguish between taxa [98, 99].
The development of DNA sanger sequencing [100{103] aided in obtaining se-
quence length of < 1000 bp of 16S rRNA gene for microbial diversity analysis.
18
However with the advent of next-generation sequencing technologies (`pyrosequenc-
ing'), the focus shifted from sequencing the full length 16S rRNA gene to short sub-
regions of the gene providing large datasets [104, 105]. Amongst next-generation
sequencing platforms, 454 Life Sciences had been a preferred choice, as it pro-
vided read length of 200-500 bp while reading the fragment from one end to the
other (single-end reads). Recently, the popularity of Illumina technologies has es-
calated due to increase in sequencing read lengths (up o paired-end 150-300 bp)
via paired-end reading where reading of the fragment occurs from both ends.
Eective analysis of microbial diversity via short length sequences calls for, (i)
judicial selection of sequencing strategy (Illumina vs. 454 platform); (ii) appropri-
ate hypervariable region to be sequenced; (iii) best clustering technique of the 16S
rRNA sequences into Operational Taxonomic Units (OTUs), and (iv) the choice
of the taxonomic classier. Each sequencing platform has its own limitations and
comes with its own sets of sequencing errors [106, 107]. Hence, employing rele-
vant denoising algorithm [108, 109] and removing chimeras [110] for minimizing
sequence reads errors becomes essential. There has been extensive studies on the
choice of primers and hypervariable region to target for sequencing [111]. Amongst
them, universal primer set F515 { R806 spanning hypervariable V4 region was
adopted by Earth Microbiome Project because it amplies a broad range of bacte-
rial and archaeal phyla [105, 112, 113]. However, a recent study suggests that this
primer underestimates (e.g. SAR11) and overestimates (e.g. Gammaproteobacte-
ria) some commonly found marine taxa [114]. Hence, for marine microbial studies
the usage of alternate primers was put forth. For studying microbial community
sequence data, there are primarily two softwares - Mothur [115] and QIIME [116].
19
In mothur OTU's are clustered based on distance matrix threshold method (re-
quires more computational resource), whereas QIIME employs high-scoring local
and global alignments clusters via UCLUST [117] (requires less computational re-
source). For classication of clustered OTU's, Na ve Bayesian classier has been a
popular choice [118] since it classies sequences to genus level at 95.7% accuracy
with 80% bootstrap cuto for hypervariable region V4 [119].
2.6 Isolation Strategies
While microbial community analysis based on 16S rRNA gene provides an insight
to the putative environmental microbiome, the ability to isolate subsurface mi-
crobes provides physiological understanding and the physical/chemical context in
which these organisms thrive. This allows us to predict which geochemical envi-
ronments can support subsurface life on Earth. To tackle the `great plate count
anomaly', described in previous section, mimicking the environmental geochem-
istry in base media design becomes crucial [120, 121]. With nitrogen, sulfur and
phosphorous being major nutrients required for growth [122], media design should
incorporate the elemental composition of N:P:S in bacteria. The elemental com-
position of N:P in many microbial biomass was measured to be approximately 9:1
[123] and further ratio of P:S was approximated to be 1:1 [124]. For isolations
purposes, the media used in this thesis took the geochemistry of the environment
and the elemental requirements of the bacteria (N:P:S' 10:1:1, [125]) into consid-
eration. Once the base medium is designed, it was amended with a carbon source,
electron donor/acceptor pairs, vitamins and minerals, pH buer, free and dissolved
gases.
20
2.7 Electrochemical characterisation
The study of electrochemically active bacteria (EAB), interacting with electrodes,
via electrochemical techniques plays a crucial role in revealing electron transfer
processes employed by the EABs. The interactions between microorganisms and
electrodes take place at physiologically important potentials unique to the microbe.
The potentials are aected by electrode surface, pH, temperature, light conditions
etc. Also, a mixed community of bacteria will give rise to complex electrode-biolm
interaction providing broader picture of electrode-biolm interaction. Hence, it
is best recommended to isolate pure strains and grow monospecies biolms for
electrochemical characterization [126].
The electrochemical behavior of a biolm is characterized by a three-electrode
electrochemical cell, where a potential is applied at the working electrode with
respect to a known reference electrode (in our case Ag/AgCl saturated with 1 M
KCl). Any current
ow is measured between the working electrode and a counter
electrode. This conguration allows the potential of the working electrode to be
measured without compromising the stability of that reference electrode by passing
current through it.
One of the simplest electrochemical technique, chronoamperometry, monitors
the current over a period of time with a constant potential applied. This tech-
nique has been widely used to perform electrochemical enrichments as described
in section 2.4. The choice of potential applied at the working electrode depends
on whether microbes enriched are electrode oxidizers or reducers. A wise choice of
potential to be applied depends on the environmental geochemistry - concentration
of available electron donor/acceptors, pH and temperature. This potential should
be calculated using the Nernst Equation (2.1).
21
E =E
0

RT
nF
ln
[Products]
[Reactants]
(2.1)
whereE is the reduction potential of the mineral,E
o
is the standard reduction
potential at temperature = 25
o
C,R is the gas constant (8.3145 J/molK), T is
the temperature, n is the number of electron transfer, F is the faraday constant
(95484.56 C/mol), [Products] is the concentration of products and [Reactant] is
the concentration of reactants.
The electrochemical behavior of microbes are widely characterized by cyclic
voltammetry (CV). It involves sweeping a potential (usually at a scan rate of
1-10 mV/s) back and forth between a negative, or reducing potential, and a pos-
itive or oxidizing potential and studying the behavior of the current. During the
forward (reducing to oxidising potential) and reverse (oxidising to reducing poten-
tial), anodic and cathodic current peaks appear around the reduction potential of
the redox couple. In a CV, the current is the sum of faradaic and non-faradaic
current. Faradaic current, is the current generated by the an electrochemical re-
action at an electrode. Non-faradaic current is a background current caused by
charging of double layer. Due to the complexity of the biolm-electrode interac-
tion, all the voltammetric studies on biolms require an abiotic control in order
to perceive any change due to biotic activity. Electrochemically active bacteria
can interact with the electrodes in two ways, as an electrode reducer (anodic) and
electrode oxidizer (cathodic). With a plethora of research on anodic respiration
by Shewanella and Geobacter, it is widely known that when bacteria respires on
electrodes while consuming electron donors it produces a sigmoidal voltammogram
(turn-over conditions). However, the sigmoidal shape is lost during starving con-
ditions with no electron donor [127]. If the catalytic currents are much lower than
22
the non-faradaic/background currents, observation of the sigmoidal shape will be
close to impossible. Geobacter was the rst organism to be studied as an electrode
oxidizer. However, this data was not supplemented with any cyclic voltammetry
studies [93]. Recently, cyclic voltammograms from cathodic biolms exhibit a wide
range of shapes ranging from sigmoidal [59, 128{130] to not-so-sigmoidal [60, 62]
implying that the mechanisms of electron uptake is still illusive and requires further
research.
Dierential pulse voltammetry has been recently gaining attention [39, 131].
Here, the potential is applied in pulses and is superimposed on a slowly changing
base potential. Current is measured at two points for each pulse, just before the
application of the pulse and at the end of the pulse. These sampling points are
selected to allow for the decay of the nonfaradaic (charging) current. The dierence
between current measurements at these points for each pulse is determined and
plotted against the base potential. DPV allows for decay in nonfaradaic current, it
is suitable to study biolms with large nonfaradaic current compared to catalytic
current.
23
Chapter 3
Biophysical Mechanisms of
Interfacial Electron Transfer in S.
oneidensis MR-1
Extensive research on S. oneidensis MR-1 has resulted in proposed EET mecha-
nisms which are broadly classied into direct and indirect, depending on whether
cell-mineral/electrode contact is required. For direct EET, S. oneidensis MR-
1 transports respiratory electrons from inner membrane to extracellular electron
acceptors through a network of periplasmic and outer-membrane multi-heme cy-
tochromes that extend along the cellular surface [7, 34], and through micro-meter
long conductive extracellular appendages called bacterial nanowires [41, 42]. For
indirect EET, naturally occurring or biogenic small molecules (
avins, humic acids,
etc.) are proposed to shuttle electrons between the cell surface and substrate acting
as soluble mediators [36{39].
This chapter will discuss the direct and indirect routes of EET in S. oneidensis
MR-1. In the rst section, I will describe how expression of GFP in the periplasmic
space of S. oneidensis MR-1 helped reveal the structure of the nanowire (direct
EET) as an outer-membrane and a periplasmic extension. This work was included
in a pathbreaking publication providing detailed study of rst in vivo measurement
of bacterial nanowires [45]. In the second section, I will demonstrate dierential
24
interaction of S. oneidensis MR-1 with poised electrodes based on their surface
structure (ITO vs. carbon cloth) and anity for
avins. The carbon cloth based
electrochemical reactor built in this thesis was employed to further investigate
detailed electrochemical interaction of S. oneidensis MR-1 biolms with poised
electrodes, ultimately providing us a insight into electron transfer through
avins
(earlier proposed as indirect EET) [49].
3.1 Shewanella oneidensis MR-1: nanowire as
outer membrane extensions
Prior to our work, bacterial nanowires were never directly observed or studied in
vivo, and therefore a number of fundamental issues were unresolved. Most re-
search focused on conductance measurement of bacterial nanowires made under
ex situ dry conditions using solid-state techniques [41{43, 132]. Little was known
about the molecular structure, identity of the charge carriers, and interfacial elec-
tron transport mechanisms responsible for the high electron mobility of bacterial
nanowires. Geobacter nanowires are thought to be type IV pili, and their conduc-
tance is proposed to stem from a metallic-like band transport mechanism resulting
from the stacking of aromatic amino acids along the subunit PilA [132]. This mech-
anism remains controversial [133, 134]. In contrast, the molecular composition of
bacterial nanowires from Shewanella, had not yet been reported and had always
Expression of GFP in periplasmic region of S. oneidensis MR-1 (this chapter) was instru-
mental in the following publication - Pirbadian, S., Barchinger, S.E., Leung, K.M., Byun, H.S.,
Jangir, Y., et al. 2014. Shewanella oneidensis MR-1 nanowires are outer membrane and
periplasmic extensions of the extracellular electron transport components. Proceedings of the
National Academy of Sciences, 111(35), pp.12883-12888.
25
been referred as pilus-like structures. Shewanella nanowire conductance corre-
lates with the ability to produce outer membrane redox proteins [42], suggesting a
multistep redox hopping mechanism for EET [135, 136].
In this collaboratory work, in vivo imaging of bacterial nanowires from S. onei-
densis MR-1 illustrated correlation of nanowire growth with increased cellular re-
ductase activity. These produced nanowire also stained with membrane-selective
stain FM R
4-64FX. This indicated that membranes were a substantial component
of these nanowires. Next question arises whether nanowires are just outer mem-
brane extensions comprising of just the periplasmic components, or whether they
also contain inner membrane with cytoplasmic components. To determine this, I
expressed either GFP fused to a signal sequence that enables GFP export to the
periplasm.
Proteins located outside the cytoplasm are normally synthesized with amino-
terminal signal peptides that target them to either the Sec (secretory) or the Tat
(twin-arginine translocation) protein export pathway. The Sec system is involved
in both the secretion of unfolded proteins across the cytoplasmic membrane and the
insertion of membrane proteins into the cytoplasmic membrane. The Tat system
has mostly been implicated in the secretion of folded and/or cofactor containing
proteins [137{140]. Exceptionally diverse set of protein export pathway (except
for the type IVa secretion system (T4aSS)) has been observed in genomic analysis
of all sequenced Shewanella [141]. In previous work by Luo et al. [142], the Tat
machinery of S. oneidensis MR-1 was shown to be responsible for the transport of
PetA across the membrane, allowing the cells to respire on oxygen by completing
electron transfer to the terminal oxidases. In that same study, they constructed the
IPTG-inducible plasmid (pHGE-PtacTorAGFP) expressing the Tat signal peptide
26
of TorA (Trimethylamine oxide reductase) fused to GFP (green
uorescent pro-
tein). This plasmid was able to transfer folded GFP to the periplasmic space of S.
oneidensis MR-1.
To understand the molecular make-up of the nanowire, this section reports real-
time monitoring of in vivo nanowire production by S. oneidensis MR-1 expressing
GFP in the periplasmic and cytoplasmic region. We expressed IPTG-inducible
tac promoter plasmid pHGE-PtacTorAGFP (obtained from Luo et al.) with the
ability to tranfer folded GFP to the periplasmic region [142] of S. oneidensis MR-1.
We also employed lac promoter plasmid p519nGFP [143, 144] with the ability to
express GFP in cytoplasmic region of S. oneidensis MR-1. Using live
uorescence
measurements we nd that the Shewanella nanowires are membrane- rather than
pilin-based, and are associated with outer membrane vesicles.
3.1.1 Experimental
Cytoplasmic and periplasmic green
uorescent protein (GFP) imaging
To express GFP in the cytoplasm, S. oneidensis MR-1 was transformed with plas-
mid p519ngfp [143, 144] via electroporation. In short, electrocompetent S. onei-
densis MR-1 were prepared by aerobically growing the culture to mid-log phase:
cells were centrifuged for 1 min at 12000 g, washed once in 1.5 mL of 1 M
D-sorbitol (pH: 7.59, room temperature), centrifuged again at 12000 g for 1
min and nal resuspension in 250 L of 1 M D-sorbitol and placed on ice for 15
min. Plasmid p519ngfp (100-500 ng) was introduced to 100 L freshly prepared
electrocompetent S. oneidensis MR-1 by electroporation using 0.2-cm cuvettes
(Gene Pulser R
/MicroPulser
TM
electroporation cuvettes, 0.2 cm, BIO-RAD) and
27
a ECM R
399 (1.8 kV for 5 sec). After electroporation, the cells were immedi-
ately suspended in 500 L LB broth, followed by incubation at 30
o
C for 90 min
and plated on LB augmented with 50 g/mL kanamycin [145]. The colonies were
picked after 24 hours of growth and checked for GFP expression by
uorescence
microscopy.
Localization of GFP in the periplasm of S. oneidensis MR-1 was achieved by
employing plasmid pHGE-PtacTorAGFP received from [142]. Initially, the plas-
mid was expressed in EC100 chemically competent E. coli. Brie
y, 100 L of the
competent cells were thawed on ice, 500 ng of the plasmid was added to the cells
and incubated on ice for 30 min, followed by a heat shock at 42
o
C for 30 sec and
back on ice for 2 min. Then, 250 L SOC media was added and the cells were re-
covered by incubation at 37
o
C for 60 min with horizontal shaking of 225 rpm. The
cells were plated on LB augmented with 50 g/mL kanamycin, the transformed
colonies were picked the next day and transferred to LB culture (with 50 g/mL
kanamycin). After conrmation of periplasmic GFP expression in the transformed
E. coli, plasmid was extracted using PureLink R
Quick Plasmid Miniprep Kit (In-
vitrogen, Carlsbad, CA). S. oneidensis was transformed via electroporation (as
mentioned earlier) with 2000 ng of plasmid in 100 L of freshly prepared elec-
trocompetent S. oneidensis. pHGE-PtacTorAGFP is an IPTG-inducible plasmid
expressing GFP fused to the twin-arginine translocation (Tat) signal peptide from
E. coli TorA. Fusion to the Tat signal peptide enables GFP to be exported to the
periplasm [142]. S. oneidensis MR-1 strain expressing periplasmic GFP will be
referred as S. oneidensis MR-1 periGFP.
Since the Tat system exports fully folded proteins from the cytoplasm, the cel-
lular
uorescence pattern of this S. oneidensis periGFP strain is not necessarily
28
limited to the periplasm only, and can change over time depending on the induc-
tion, export rate, or post-export cleavage of the signal peptide [142]. Therefore, to
assess the successful periplasmic localization of GFP prior to use in the perfusion
experiments, we imaged this strain on 3% agar on a slide (no nutrients or fur-
ther IPTG induction) which resulted in the characteristic peripheral
uorescence
pattern consistent with periplasmic localization (Figure 3.1).
Cell growth and in vivo imaging
S. oneidensis periGFP strain was grown aerobically in 50 mL of sterile LB broth
(Sigma L3022, 20 g in 1 L of deionized water) augmented with 50g/mL kanamycin
in 125-mL
asks from a frozen (-80
o
C) stock, at 30
o
C, shaking at 150 rpm, and
0.1 mM IPTG was added to induce the expression of TorA
sp
-GFP at OD
600
= 0.4
ahead of use in the perfusion imaging experiment as described [45]. The cells were
grown up to an OD
600
of 2.4 - 2.8. 15 mL of the preculture was centrifuged at 4226
g for 5 min, the pelleted cells were washed twice, and nally resuspended in 45
mL of a dened medium consisting of 30 mM PIPES, 60 mM sodium DL-lactate
as an electron donor, 28 mM NH
4
Cl, 1.34 mM KCl, 4:35 mM NaH
2
PO
4
, 7:5 mM
NaOH, 30 mM NaCl, 1 mM MgCl
2
, 1mM CaCl
2
, and 0.05 mM ferric nitrilotriacetic
acid. In addition, vitamins, amino acids, and trace mineral stock solutions were
Plasmid Description
Source
pHGE-PtacTorAGFP
Km
R
, Ptac, TorA
sp
-GFP
(periplasmic GFP)
[142]
p519ngfp
Km
R
, Plac/Pnpt-2, mob
+
, GFP
(cytoplasmic GFP)
[143, 144]
Table 3.1: Strains and plasmids used in this study.
29
Figure 3.1: Cells containing periplasmic and cytoplasmic Green Fluorescent Pro-
tein (GFP). S. oneidensis MR-1 expressing GFP fused with the twin-arginine
translocation (Tat) signal peptide from the E. coli TorA. Cells were imaged on
an agar pad following LB growth and IPTG induction. Due to the presence of
the signal peptide, the fusion protein is exported to the periplasm, resulting in

uorescence limited primarily to the periphery of the cell (Scale bar = 1 m. In-
set: Fluorescence image of S. oneidensis MR-1 constitutively expressing GFP with
no signal sequence for export, resulting in a uniform cellular
uorescence pattern
throughout the cytoplasm. Scale bar = 2 m.
used to supplement the medium as described in [41]. The medium was adjusted
to an initial pH of 7.2.
The membrane stain FM R
4-64FX (Life Technologies) stains cells, membrane
vesicles, and bacterial nanowires, and hence was used (25g in 100 mL of me-
dia) to visualize the nanowires/vesicles. TRITC (Nikon lter set G-2E/C) ex-
citation/emission channel were used for
uorescence imaging of the FM 4-64FX
stained cells, whereas GFP was visualised under FITC (Nikon lter set B-2E/C)
excitation/emission channel [45].
30
3.1.2 Results
Earlier experiments, through in vivo imaging of bacterial nanowires with lipid
(FM R
4-64FX) and protein (NanoOrange) stain, revealed S. oneidensis MR-1
nanowires were derived from cell membrane and may be related to bacterial vesi-
cles [45]. Most known bacterial vesicles are composed primarily of outer mem-
brane and periplasm. In this section, we expressed either GFP in the periplasm
and the cytoplasm to determine whether S. oneidensis MR-1 nanowires contain
periplasm. While
uorescence was observed along the bacterial nanowires in S.
oneidensis MR-1 periGFP strain, no
uorescence was detected along nanowires
from a strain expressing cytoplasmic-only GFP (Figure 3.2). This result indicates
that S. oneidensis MR-1 nanowires are outer membrane extensions containing sol-
uble periplasmic components.
3.1.3 Conclusion
Bacterial nanowires from Shewanella oneidensis MR-1 were shown to be conductive
under nonphysiological conditions. While the above section demonstrates bacterial
nanowires from Shewanella oneidensis MR-1 are outer membrane extensions con-
taining soluble periplasmic components, it does not rule out the presence of pilin
proteins on the nanowire (as earlier proposed for the case of G. sulfurreducens).
To test whether pili played a role in S. oneidensis MR-1 nanowire production
many studies were done, including in vivo imaging of nanowire production in mu-
tants lacking the type IV pilin major subunit (pilA) or both the type IV and
msh pilus biogenesis systems (pilMQ/mshHQ). Both these mutants produced
bacterial nanowires and displayed an increase in reductase activity. Furthermore,
localization of the decaheme cytochromes MtrC and OmcA along nanowires was
31
A B
C D
Figure 3.2: Bacterial nanowires from S. oneidensis MR-1 strains containing GFP
only in the cytoplasm (Upper) or in the periplasm as well (Lower). The green
and red channels monitor GFP (Left) and FM 4-64FX (right)
uorescence, respec-
tively. The nanowires display green
uorescence only when GFP is present in the
periplasm (C) (Scale bar = 2 m).
observed along the membrane-stained bacterial nanowires via immuno
uorescence
with MtrC- and OmcA-specic antibodies. Since S. oneidensis MR-1 nanowires
contain periplasm colocalised with electron transport proteins; therefore, it is also
possible that outer-membrane proteins, periplasmic proteins and soluble redox co-
factors may contribute to electron transport through these extensions. This study
further motivates detailed understanding of the biomolecular composition, electron
transport mechanisms, and physiological impact of bacterial nanowires.
32
3.2 Disentangling extracellular electron transfer
in Shewanella oneidensis MR-1
For S. oneidensis, outer-membrane multiheme cytochromes (such as MtrCAB-
OmcA [146]) tranfer electron outside the bacterial cell envelope by establishing
direct contact with mineral surfaces [147{149] and also indirectly using electron
shuttles such as
avins and FMN [37, 38, 91, 150]. The molecular pathways (i.e.
Mtr and
avins) of these EET mechanisms to the poised electrodes are not mutu-
ally exclusive. They may be operational at dierent conditions [151, 152], and can
interact.
There are two distinct hypotheses on how multiheme cytochrome interaction
with
avin occurs. While, rst hypothesis proposes biogenically produced ri-
bo
avin (RF) and
avin mononucleotide (FMN) function as indirect mediators
that shuttle between S. oneidensis and substrates [37, 150]. This interaction is
supported by studies on a mutant of S. oneidensis lacking bacterial
avin adenine
dinucleotide exporter bfe, which led to 75% decrease in the ability of S. oneiden-
sis to reduce iron oxides or poised electrodes [153]. Later, studies revealed that

eeting interactions existed between the redox mediators (including RF, FMN)
and outer-membrane cytochromes close to the hemes [154]. These results pointed
towards shuttling of redox mediators back and forth between the cell surface and
external substrate. The second hypothesis, supported by voltammetry studies in
anaerobic conditions, invoked a more intimate interaction and the formation of a
stable
avocytochrome where the
avins enhance EET as redox cofactors bound
Carbon-based bioelectrochemical reactor (built in this thesis) was employed in the following
publication - Xu, S., Jangir, Y., and El-Naggar, M. Y. (2016). Disentangling the roles of free
and cytochrome-bound
avins in extracellular electron transport from Shewanella oneidensis
MR-1. Electrochimica Acta, 198, 49-55.
33
to the multiheme cytochromes MtrC and OmcA [39] instead as free-form
avins.
These two seemingly disparate hypothesis were reconciled by structural studies
which revealed that the Shewanella outer-membrane multiheme cytochromes con-
tain
avin-binding sites [47, 155] and they can switch from cytochromes to
av-
ocytochromes. This switching mechanism is controlled by conserved redox-active
disuldes that respond to oxygen presence which may be operative under dierent
conditions.
Electrochemical studies of S. oneidensis MR-1 strain on graphite electrode,
have been earlier reported to have two catalytic centers with E
m
value 0.33 V
and 0.07 V vs. Ag/AgCl (sat. KCl, 0.195 V vs. SHE) in turnover conditions
[156]. Carmonas et al., attributed the rst center to mediated electron transfer
(MET), since its redox potential was within the range of soluble electron shuttles
found in Shewanella [37]. The second redox potential was attributed to direct elec-
tron transfer (DET) mechanism based on formal potentials measured of puried
outer-membrane cytochromes [157{160]. A year later, similar studies were per-
formed on Shewanella loihica biolm on graphite and ITO electrodes. Both cases,
resulted in three catalytic waves centered at approximately 0.44 V (RF), 0.35 V
(quinone derivatives) and 0.07 V (DET) vs. Ag/AgCl [161]. The report showed
that the dominant EET mechanism was dependent on the electrode material on
which biolms are grown. S. loihica PV-4 biolm show predominant DET at ITO
electrodes and mixed DET/MET when grown on graphite electrodes.
It is known that free
avin signature is typically detected using glassy car-
bon electrodes that have a high adsorption anity for
avins, but generally lower
surface area (i.e. compared to our carbon cloth) for cellular attachment, thereby
34
favoring
avin-mediated EET [37, 161] . At the same time, direct EET through cy-
tochromes and cytochrome-bound
avins has been detected using ITO electrodes,
which have a low adsorption anity for
avins [39]. Using voltammetric tech-
niques, this section will identify dierential behavior of S. oneidensis MR-1 based
on the poised substrates (ITO vs carbon cloth). We will also set out to identify
the experimental design which is capable of electrochemically distinguishing the
contributions of diusible
avins vs. cytochrome-bound
avin cofactors to the cat-
alytic EET current (i.e. turnover conditions in the presence of an electron donor)
between S. oneidensis MR-1 cells and anodes [49].
3.2.1 Experimental
Initial Cell Growth
For electrochemical characterization of Shewanella on tin doped indium oxide
(ITO), S. oneidensis MR-1 preculture in LB (OD
600
= 2.1) was spun down at
7000 rpm for 10 mins and resuspended in 50 mL dened medium (in 125 mL
ask)
containing 50 mM PIPES, 28 mM NH
4
Cl, 1.3 mM KCl, 4.3 mM NaH
2
PO
4
H
2
O,
20 mM sodium DL-lactate as the electron donor, with vitamins and amino acids
as described previously [41], and grown aerobically at 30
o
C and shaking at 150
rpm shaking. After 3 hours, 1 mL of the culture was injected in 10 mL N
2
purged
dened media. This mixture was introduced into the perfusion chamber set up.
For electrochemical characterization of Shewanella on carbon cloth, S. oneiden-
sis MR-1 preculture in LB (OD
600
= 2.4-2.9) was transferred to 350 mLs anaerobic
dened media with 50X dilution. The dened media contained 50 mM PIPES, 28
mM NH4Cl, 1.3 mM KCl, 4.3 mM NaH
2
PO
4
H
2
O, 40 mM sodium DL-lactate as
the electron donor, and 30 mM sodium fumarate as the electron acceptor, with
35
vitamins and amino acids as described previously [41]. After this culture grew to
an OD
600
of 0.3, it was introduced to the bioelectrochemical bioreactor with 350
mLs of N
2
purged dened media (without fumarate).
ITO microelectrodes fabrication and perfusion chamber set-up
Glass wafers (50 mm x 43 mm) were washed by three step cleaning protocol 5 min
sonication in acetone, isopropyl alcohol (IPA) and deionized water, followed by
nitrogen blow dry. The wafers were then kept for drying on hotplate at 150
o
C for
10 min. Subsequently, organics on the glass wafer was removed by Tegal Plasma
Asher for 2 min at 200 W. For eective bonding of the photoresist, glass wafers were
kept in HMDS (Hexamethyldisilazane) tank for 10 min. This process produces a
thick sticking layer on the surface of the glass wafer. AZ5214 a positive photoresist
of thickness about 1.6 m was deposited on the glass wafer using a spin coater
spun at 3000 rpm for 30 sec. A soft bake of 100
o
C for 2 min was required prior to
exposure. The exposure was performed using Karl Suss Aligner MA6 for 15 sec,
with hard contact having 30 m gap and lamp power 8 W. The lithography mask
was designed using DoubleCAD and printed (emulsion down) (Outputcity, Bandon,
OR). Following photolithography, the wafers were developed in a beaker with 5:1
water:AZ400K developer with constant gentle agitation. Immediate rinsing in

owing DI water for 2 min was performed followed by nitrogen blow dry. Post
bake was done at 150
o
C for 10 min proceeded by another plasma cleaning by
Tegal Plasma Asher for 2 min at 200 W. 100 nm thick indium tin oxide was
deposited on the wafer using Denton Discovery 550 reactive sputterer at 180 W
with 30 sscm Ar gas and no oxygen. The remaining unexposed photoresist was
removed by sonicating the glass chip in acetone for 10 min. Subsequently, cleaning
36
was performed by 5 min sonication in IPA and DI water. Lastly, all the chips were
kept on hot plate set at 150
o
C for 10 min for complete dehydration. Now, for
passivation approximately 1 m SU-8 2000.5 negative photoresist (MicroChem)
was spin coated on the chips. The chips were then baked at 90
o
C for 1 min in
order to remove the solvent. The second photomask, which denes the openings
for ITO features, is aligned with each chip using the Karl Suss Aligner set for 23 s
exposure with a 23 m gap on soft contact. The chips are post baked for 10 min at
90
o
C in order to cross link the SU-8 polymer and harden the exposed resist. After
post baking, the chips are submerged in SU-8 developer for 1 minute, rinsed with
isopropyl alcohol for 10 s, rinsed with milliQ water, and developed again for 10 s
in fresh developer. The second development step ensures the removal of any traces
which might have reattached from the solution during rst development step. The
result is a ready-to-use mostly passivated electrochemical chip with small exposed
windows at the tips of the ITO working electrodes, and a large exposed central
ITO counter electrode (Figure 3.3).
The perfusion chamber comprised of 1) Ag/AgCl reference electrode (Warner
Instruments, CT) housed on the top, and 2) custom designed transparent electro-
chemical chip containing separate patterned ITO microelectrodes (as the working
and counter electrode)on a standard coverslip. Constant
ow of media within the
perfusion chamber from the serum bottle was established by pressurizing the sup-
ply bottle's head space with N
2
using a 6-inch septum needle, while the serum
bottle was held slightly higher than the end of the outlet tube of the chamber.
The working electrodes were poised at 0.442 V vs. SHE, acted as the sole electron
acceptor. The resulting current production was monitored by a potentiostat (Ref-
erence 600, Gamry Instruments, PA). At conclusion, cyclic voltammetry (CV) was
37
Figure 3.3: Left: Electrochemical activity of fabricated microelectrodes using the
standard ferri/ferro cyanide redox couple. The CV of a single working electrode
exposed to a solution of 10 mM K
3
Fe(CN)
6
in 1 M KNO
3
, depicts the classic
expected ultramicroelectrode (UME i.e. with a critical dimension smaller than
the length scale of the surrounding depletion later) behavior with a sigmoidal
quasi-reversible transition between the oxidized and reduced states [10]. Right:
Electromicrograph of the fabricated microelectrode depicts the working electrode
visible through the nonpassivated windows and central counter electrode (Scale
bar = 100 m).
performed with a scan window of -0.658 to 0.942 V vs. SHE, using 5 mV/s scan
rate.
Bioelectrochemical Reactor
Four PW06 Carbon cloth (Zoltek, St. Louis, MO) working electrodes measuring
25 mm 45 mm, a platinum wire counter electrode (Sigma, St. Louis, MO), and a
Ag/AgCl reference electrode (1M KCl) (CH Instruments, Austin, TX) were tted
into a custom built bioreactor (600 mL volume). Brie
y, a 1 L polypropylene wide
mouth jar was modied to provide separate ports for 1) media inlet, 2) media
outlet, 3) gas purging, 4) vent. All electrical connections were made via sheathed
38
Ti wires (Keegotech, MudWatt, CA). The carbon cloth working electrode rests on
a PTFE structure which sits inside the poylpropylene jar.
The working electrodes were sonicated in ethanol and distilled de-ionized
(milliQ) water before placement in the bioreactor, for cleaning and to enhance
hydrophilicity. The working electrode potentials were maintained and monitored
using a four-channel potentiostat (eDAQ, Colorado Springs, CO). Cultures were
transferred to the reactor through a bench top pumping system (Watson-Marlow,
Wilmington, MA), and anoxic conditions were maintained with constant purging of
puried N
2
gas. The working electrodes were poised at 0.422 V vs. SHE using the
four-channel potentiostat, and acted as the sole electron acceptors (electrochemical
cultivation) for 20 hours before voltammetric measurements. Cyclic voltammetry
(CV) were performed on the bioelectrodes with a Gamry 600 potentiostat (Gamry,
Warminster, PA). CV was performed with a scan window of -0.578 to 0.422 V vs.
SHE, using 1 mV/s scan rate.
Microscopy
Electrodes were examined using scanning electron microscopy (SEM) using a JEOL
JSM 7001F eld-emission microscope. SEM samples were subjected to a serial de-
hydration protocol using increasing concentrations of ethanol, and the dehydrated
samples were then critical-point dried (Tousimis Autosamdri 815, Tousimis Inc.,
Rockville, MD) ahead of SEM.
3.2.2 Results
An anodic current was immediately observed following inoculation of S. oneiden-
sis MR-1 into, 1) the perfusion chamber set-up with the ITO electrochemical chip
39
and 2) the carbon cloth bioelectrochemical reactor, containing working electrodes
poised at 0.442 and 0.422 V vs. SHE respectively (Figure 3.4). This current,
which increased steadily to 30 nA (35 hours) and 400A (20 hours), re
ects
oxidation of lactate coupled to EET by cells to the electrode surface ITO and car-
bon cloth respectively. This phenomenon has been observed previously [37, 161].
Following the incubation, electrochemical interaction between S. oneidensis MR-1
and dierent electrodes was further probed using cyclic voltammetry (CV). Dis-
similar voltammograms (Figure 3.4) at the two distinct surfaces (ITO and carbon)
highlights the dierent anities for
avins leading to dierent EET mechanisms
active on the surface. While carbon cloth is able to able to capture both the
avin
dependent (free-form and cytochrome bound) and
avin independent EET mecha-
nisms (cytochromes), ITO in our case is only able to detect direct electron transfer
via cytochromes [37, 39].
Turnover CV of S. oneidensis MR-1 on carbon cloth has two signicant cat-
alytic waves with onset potentials of { 0.24 V and 0.20 V (vs. SHE). The obser-
vation of two redox pathways linking cells to the electrode is also consistent with
previous reports on S. oneidensis MR-1 EET [152, 156, 162, 163]; the wave with an
onset at { 0.24 V re
ects a
avin-dependent mechanism, while the higher-potential
wave with an onset at 0.20 V is facilitated by direct EET through multiheme cy-
tochromes. To test whether the
avin-dependent EET current stems from free se-
creted or cytochrome-bound
avins, we obtained a derivative of the turnover CV.
Compared to free-form soluble
avins, the redox reaction of cytochrome-bound

avins is expected to be shifted approximately 100 mV in the positive direction
[39]. This feature was observed in case of biolms on S. oneidensis MR-1 on car-
bon cloth (Figure 3.4E). On the other hand, turnover CV of S. oneidensis MR-1
40
Figure 3.4: Chronoamperometry of S. oneidensis MR-1 interacting with a ITO (top
left) and carbon cloth (top right) working electrode poised at 0.422 and 0.44 V
vs. SHE. Turnover cyclic voltammetry (CV) reveals one catalytic wave with onset
at 0.1 V vs. SHE corresponding to
avin-dependent EET pathway in ITO biolm
(bottom left) and two catalytic waves corresponding to
avin-dependent and direct
EET via multiheme cytochrome pathways in Carbon cloth biolm (bottom right)
41
on ITO did not us give a clear catalytic wave. However, after dierentiating the
CV, a peak was observed at 0.15 V vs SHE (Figure 3.4 E), corresponding to outer-
membrane cytochromes, indicating interaction with ITO requires direct electron
transfer. This does not show ITO requires DET. Just that DET is the only rele-
vant mechanism for ITO. It is also relevant for carbon cloth, but it is one of the
mechanisms. This observation highlights S. oneidensis MR-1 dierential interac-
tion to poised electrodes based on the substrate, as observed previously in case of
S. loihica [161].
The carbon cloth based bioelectrochemical reactor designed in this thesis
was able to disentangle multiple EET mechanisms by capturing both the
avin-
dependent (both soluble and cytochrome-bound) as well as
avin-independent (cy-
tochrome hemes) routes of cell-to-anode EET in one experimental system (Figure
3.4 E). Further experiments conrmed that
avins enhance the catalytic current
primarily as redox cofactors under our experimental conditions [49].
3.2.3 Conclusion
The described experimental system and electrochemical analyses simultaneously
captured both the
avin-independent and
avin-dependent routes of EET between
S. oneidensis MR-1 and carbon electrodes. By studying the electrochemical sig-
nature of secreted redox-active molecules, these measurements led to the ndings
that
avins accelerate EET as cytochrome-bound cofactors, rather than free sol-
uble molecular shuttles. Electrochemical analysis of S. oneidensis MR-1 on ITO
electrodes only pointed towards direct electron transfer mechanism. This study
stresses the impact of electrode materials and surface properties when dening
42
specic microbial redox signatures and further motivates structural and biophys-
ical studies to unravel the location and precise mechanistic pathway that allows
cytochrome-bound
avins to mediate electron transfer at the interface of cellular
proteins and electrodes.
43
Chapter 4
Isolation and characterization of
electrochemically active
subsurface Delftia and Azonexus
species
4.1 Introduction
Beyond the well characterized model systems the Deltaproteobacteria Geobacter
and the Gammaproteobacteria Shewanella, electrochemical enrichments coupled
with 16S-based rDNA surveys from a variety of environments, especially marine
sediments, suggest that more physiologically and phylogenetically diverse microor-
ganisms are capable of using electrodes as electron acceptors [50{53]. It is im-
portant to note that the survival or enrichment of a microbe on an electrode is
not itself evidence of EET ability since operational conditions, including poten-
tial oxygen leakage or the use of partially fermentable substrates, may support
subpopulations of heterotrophic aerobes or fermentative organisms, for instance.
This chapter is published as: Jangir, Y., French, S., Momper, L.M., Moser, D.P., Amend,
J.P., and El-Naggar., M.Y. (2016). Isolation and characterization of electrochemically active
subsurface Delftia and Azonexus species. Frontiers in Microbiology, 7, 756.
44
This makes isolation and electrochemical characterization fundamental to investi-
gating EET processes. As such, several enrichments have led to the isolation of
pure cultures that do interact with electrodes [57, 86{88], indicating that microbial
electrochemical activity at redox-active surfaces may be advantageous in a wide
variety of habitats. It now appears that microbial EET may be widespread in
nature, and that electrode-based techniques are critical for both shedding light on
the phylogenetic diversity and for identication and mechanistic measurements of
organisms whose electrochemical activity would have been missed using traditional
cultivation techniques.
Our study was motivated by both the apparent extent of EET in nature, and
the particular relevance to the continental subsurface that may present opportuni-
ties for microbial interaction with redox-active abiotic surfaces. Here, we utilized
electrodes poised at anodic (electron-acceptor) redox potentials for enrichment
of electrochemically active bacteria from a deeply sourced artesian well that is
supplied by a large regional
ow system in Death Valley, California, USA. The
enrichments at dierent anodic potentials, subsequent isolation of pure cultures,
and small currents observed by electrochemical testing of these pure cultures un-
der well-dened conditions, suggest that isolated Delftia and Azonexus strains may
gain an advantage by passing electrons to external surfaces. Our results contribute
to the emerging view that microbial electrochemical activity is a widespread phe-
nomenon that may impact survival and activity in energy-limited conditions.
45
4.2 Material and Methods
4.2.1 Sampling Site and Initial Enrichment
The sampling site (Nevares Deep Well 2 NDW2) was drilled without drilling
uid
in 2009, penetrating the Nevares Spring Mound, a prominent travertine assemblage
located at the western foot of the Funeral Mountains in Death Valley, California,
USA [164]. The hole was drilled to 103 meters below land surface, but cased only
to a depth of 18 m, below which it is open and intersects with fractures of the
Furnace Creek Fault Zone. Once in the fault, the hole opens into a rubble zone
from which artesian
ow ( 300 L/min) emanates (Michael King, Hydrodynamics
Group LLC, and Richard Friese, National Park Service personal communication).
After
ushing the hole with 200 well volumes, freshwater samples were collected
for aqueous chemistry (Table 4.1) and microbiological analyses.
Aqueous geochemical analyses (Table 4.1) were performed at the Analytical Lab
San Bernardino (ALSB). Brie
y, anions were measured according to Environmental
Protection Agency method 300.0 using ion chromatography. Cations and metals
were determined using Inductively Coupled Plasma Mass Spectrometry (ICP-MS),
and Atomic Absorption (AA), respectively. Water samples were used for an initial
enrichment in a Down-
ow Hanging Sponge (DHS) reactor [165], targeting Fe- and
Mn-reducing microbes. The minerals 2-line ferrihydrite (containing Fe-III) and -
MnO
2
(containing Mn-IV) were freshly made for inoculation of the DHS reactor
[71, 166]. Aqueous medium was designed to mimic the geochemical composition
in situ (Table 4.1). Specically, 1 L of medium contained: NH
4
Cl, 0.1 g; NH
4
SO
4
,
0.1 g; KH
2
PO
4
, 0.2 g; MgSO
4
, 0.1 g; CaCl
2
, 0.2 g. After autoclaving, pH was
set at 7.2 with HCO

3
buer and DSMZ medium 141 trace metals and vitamins
46
were lter sterilized into medium. The gas phase was composed of 80:20 v:v
H
2
:CO
2
. Aqueous and gaseous media were provided at 0.1 mL per minute via
a peristaltic pump (Cole-Parmer Master
ex L/S, model 28 number 7551-10) and
media was kept anaerobic with 100M Ti-NTA [freshly made from 20% TiCl
3
and
nitrilotriacetic acid (NTA) in saturated Na
2
CO
3
]. Resazurin (0.001%) was used as
a redox indicator.
4.2.2 Electrochemical Enrichment Bioreactor
The electrochemical enrichment bioreactor, schematized in Figure 4.1, was con-
structed from a polypropylene wide mouth jars (1 L) which contained a PTFE
assembly of four threaded rods, each of which supported a working electrode (WE)
NDW2
Depth (m) 103
pH 7
Temperature (
o
C) 31
Conductance (S/cm) 790
Dissolved Constituents
dO
2
(mg/L) 0.4
SO
2
4
(mg/L) 170
Na
+
(mg/L) 160
Ca
2+
(mg/L) 44
Mg
2+
(mg/L) 21
DOC (mg/L) 0.12
Fe
2+
(mM) 0.01
H
2
S (mg/L) BDL
Cl

(mg/L) 35.8
Br

(mg/L) 0.25
NO

3
(mg/L) 0.13
PO
3
4
(mg/L) 0.06
NO
2
2
(mg/L) BDL
Table 4.1: Chemical composition of water collected from Nevares Deep Well 2
(Death Valley, CA, USA).
47
composed of 3 cm 2 cm carbon cloth (PW06, Zoltek, St. Louis, MO, USA). The
bioreactor is a standard half-cell that contained, in addition to the four carbon
cloth working electrodes, a common platinum counter electrode (VWR, Radnor,
PA, USA) and a common Ag/AgCl reference electrode (1 M KCl, CH Instru-
ments, TX, USA). During enrichment, the four working electrodes (WE1, WE2,
WE3, and WE4) were poised at 272, 372, 472, 572 mV vs. SHE, respectively,
using a four-channel potentiostat (EA164 Quadstat, EDaq, USA). All electrical
connections were made using insulated titanium wires. The complete set-up was
autoclaved with working and counter electrodes, and the ethanol-sterilized refer-
ence electrode was inserted before use. The bioreactor was fed with media using a
media reservoir. Both the bioreactor and media reservoir were continuously purged
with sterile ltered inert gas to maintain anaerobic conditions.
During the rst phase of electrochemical enrichment, designed to match the
conditions of the DHS reactor, DHS euent was used as an inoculum and the en-
richment medium consisted of autoclaved NDW2 water supplemented with (L
1
):
NH
4
Cl, 0.025 g; (NH
4
)
2
SO
4
, 0.132 g; KH
2
PO
4
, 0.095 g; with DSMZ medium 141
trace metals and vitamins. The media reservoir was constantly bubbled with ster-
ile ltered N
2
to maintain anaerobic conditions, while the bioreactor was bubbled
with H
2
/CO
2
(80:20, v:v) to maintain similar conditions as the DHS bioreactor.
This enrichment was performed for 30 days, and WE4 (+572 mV vs. SHE) was
used as an inoculum for a batch enrichment of Fe(III)-reducers in the same medium
using 10 mM sodium acetate as electron donor and 5 mM Fe(III)-NTA (nitrilotri-
acetic acid) as electron acceptor. After growth (increase in cell counts conrmed by
DAPI staining) and reduction of Fe(III)-NTA [detected visually by color change of
Fe(III) to colorless Fe(II), and conrmed by ferrozine assay], the batch culture was
48
Figure 4.1: Schematic diagram of the electrochemical enrichment reactor. The
four carbon cloth working electrodes were poised at 272 mV, 372 mV, 472 mV,
572 mV vs. SHE, respectively, using a 4-channel potentiostat. The reactor and
medium reservoir were continuously purged with lter-sterilized N
2
gas to maintain
anaerobic conditions.
further enriched in the electrochemical bioreactor. During this second electrochem-
ical phase, NDW2 medium was designed where sodium acetate (10 mM) served
as electron donor, and basal salts and nutrients were (L
1
): MgSO
4
.2H
2
O, 0.19 g;
CaCl
2
.2H
2
O, 0.15 g; NH
4
Cl, 0.025 g; (NH
4
)
2
SO
4
, 0.132 g; KH
2
PO
4
, 0.095 g; vita-
mins and trace minerals [167]. Both the media reservoir and the bioreactor were
bubbled constantly with N
2
to maintain anaerobic conditions. This enrichment
was operated continuously for 6 months, with media reservoir change performed
every 10{30 days. The dilution rate in the bioreactor amounted to one reactor
volume per day. Acetate concentration was monitored by high performance liquid
chromatography (HPLC), as described previously [84].
49
4.2.3 Bacterial Community Analysis of Enrichments
Working electrode sections (1 cm 1 cm) and planktonic cells (500 L) from the
electrochemical bioreactor were suspended with 700 L lysozyme buer (25 mM
Tris HCl, pH 8.0, 2.5 mM EDTA with 2 mg/mL lysozyme) in a microcentrifuge
tube, followed by heating on a microtube shaking incubator at 65
o
C for 10 min.
The mixture was vortexed at maximum speed for 10 min. This was further incu-
bated at 65
o
C with vigorous shaking for 20{30 min after adding 700 L of TESC
buer (100 mM Tris HCl pH 8.0, 100 mM EDTA pH 8.0, 1.5 M NaCl, 1% w/v
cTAB, 1% w/v SDS, 30 adjusted nal pH of TESC buer to 10, added 100g/mL
Proteinase K just before use) for cell lysis. Further, vortexing for 10 s was per-
formed, followed by incubation on ice for 10{20 min. Cellular debris was removed
by spinning at 18000 g for 10 min. The supernatant was mixed with 25:24:1
phenol:chloroform:isoamyl alcohol. The mixture was spun down at 18000 g for
10 min and the supernatant was transferred to a fresh microcentrifuge tube. This
collected supernatant was mixed with 750 L of isopropanol (no vortexing) and
incubated at room temperature for 5 min. Final spin was performed at 18000 g
for 10 min and the pellet was washed with 95% ice-cold ethanol. The above step
of centrifuging the pellet followed by washing was repeated 2-3 times. The pellet
was dried overnight until no liquid was left. The extracted DNA was suspended
in 100 L of TE buer (pH 8.0, 10 mM Tris HCl, 1 mM EDTA). Finally, DNA
concentration and purity were assessed using a nanodrop spectrophotometer.
The near-complete bacterial 16S rRNA gene was amplied using the extracted
DNA with primers 8f: 5' - AGA GTT TGA TCC TGG CTC AG - 3 and 1492r: 5
- GGT TAC CTT GTT ACG ACT T 3 to perform microbial community analysis
[168]. PCR amplication was carried out in 20L reaction volumes using TaKaRa
50
ExTaq DNA Polymerase (TaKaRa Biosciences, Mountainview, CA), and the prod-
ucts were cloned using pCR
TM
2.1-TOPO (Invitrogen, Carlsbad, CA) following
manufacturer's instructions. Transformed Giga competent cells (E. coli DHS)
were grown overnight in Luria-Bertani (LB) agar plates containing Isopropyl-D-
1- thiogalactopyranoside (IPTG), 5-bromo-4-chloro-3-indolyl--galactopyranoside
(X-gal), and 50g/mL kanamycin. Approximately 40 recombinant colonies (white)
were picked from each sample (originating from the WE1, WE2, WE3, WE4, and
planktonic), and grown in LB broth with kanamycin resistance overnight. Plasmid
extraction was performed using PureLink quick plasmid DNA miniprep kit (In-
vitrogen, Carlsbad, CA, USA). This DNA was sent for unidirectional sequencing
to Genewiz (South Plaineld, NJ, USA). The sequences (600 bp) were analyzed
using Geneious software and compared to other published sequences using Na ve
Bayesian Classier provided by Ribosomal Database Project (RDP) classier tool.
High{throughput pyrosequencing was also applied to the same DNA extracts
used to construct clone libraries for bacterial community analysis across the sam-
ples. Variable region V4 of the bacterial 16S rRNA gene amplicon was amplied
using barcoded fusion primer 515f: 5 GTG CCA GCM GCC GCG GTA A - 3 and
806r: 5 GGA CTA CHV GGG TWT CTA AT- 3. All the samples were pooled
in equimolar concentrations and puried using Agencourt Ampure beads (Angen-
court Biosciences Corporation, MA, USA). Samples were sequenced at MR DNA
(Shallowater, TX, USA) with Roche 454 FLX titanium instrument. The resulting
data were processed using MOTHUR [115]. Sequences were depleted of barcodes
and the primer, followed by trimming to remove sequences with any ambiguous
base calls and homopolymer runs greater than 8 bp. Average sequence lengths
of 260 bp and a total of approximately 12,000 unique sequences were obtained
51
for all samples. The trimmed sequences were aligned with RDP infernal aligner
[169], chimeras were removed using UCHIME [110], and a distance matrix was
created. The sequences were clustered to identify unique operational taxonomical
units (OTUs) at the 97% level, and taxonomy was assigned using RDP classier.
The raw sequences have been uploaded to NCBI SRA database (accession number:
SRP071268).
4.2.4 Isolation and Electrochemical Measurement of Pure
Cultures
Electrode-attached biomass from each WE was streaked on R2A agar plates [170]
to obtain colonies at 30 C. Morphologically dierent colonies were restreaked on
fresh R2A agar plates, resulting in multiple isolates. For taxonomic classication,
isolates were grown to late exponential phase in R2A media at 30 C and DNA
extraction was performed using the UltraClean Microbial DNA Isolation kit (Mo
Bio laboratories, Carlsbad, CA, USA). The bacterial 16S rRNA gene was PCR-
amplied from the extracted DNA with primers 8f and 1492r, and the puried
PCR product (PureLink PCR Purication Kit, Life Technologies, CA, USA) was
sequenced (Genewiz, South Plaineld, NJ, USA) from the 1492r primer. The
600 bp length sequences of the isolated strains have been deposited to Genbank
(accessionnumbers: KU836931, KU836932).
Chronoamperometry measurements of each isolate were performed in standard
three-electrode glass electrochemical cells (50 mL volume). Carbon cloth (1 cm
1 cm) was used as the working electrode, with platinum wire as counter electrode,
and a 1 M KCl Ag/AgCl reference electrode. Each isolate was aerobically grown
from frozen stocks ({80
o
C in 20% glycerol) in R2A media and used to inoculate
52
500 mL of NDW2 medium at 1% (v:v). After reaching mid-exponential phase
(OD
600
0.3) aerobic growth 30
o
C, the culture was pelleted by centrifugation at
6000 g for 10 min, washed 2 and resuspended in 10 mL fresh NDW2 medium
without electron donor. 5 mL of this nal resuspension was stored in 200 mM
NaOH for measuring protein content and the remaining 5 mL was introduced to
the electrochemical cell, which already contained 50 mL NDW2 medium (acetate as
electron donor) with the WE poised at +522 mV vs. SHE, after the abiotic current
has stabilized to a constant baseline. Filter-sterilized N
2
gas was used throughout
to maintain anaerobic conditions, and the poised WE acted as the sole electron
acceptor. Cell densities were determined by plate counts (for Delftia) or DAPI
staining (for Azonexus). The cell measurements techniques diered because of slow
growth of Azonexus on synthetic NDW2 media plates. To measure protein content,
samples were digested in 200 mM NaOH at 100
o
C for 90 min, with vigorous
vortexing at 15 min intervals. Soluble protein in the extracts was measured with
Pierce BCA Protein Assay Kit (Thermo Scientic, CA, USA) using bovine serum
albumin (BSA) protein as standard.
4.2.5 Microscopy
For scanning electron microscopy (SEM), electrode samples were xed overnight
in 2.5% glutaraldehyde (GradeI,specially puried for use as an electron microscopy
xative, SigmaAldrich). Samples were then subjected to an ethanol dehydration
series (25%, 50%, 75%, 90%, and 100% v/v ethanol, for 15 min each treatment) and
critical point drying (Autosamdri 815 critical point drier, Tousimis Inc., Rockville,
MD, USA). The samples were then mounted on aluminum stubs, coated with Au
(Sputter coater 108, Cressington Scientic), and imaged at 5 keV using a JEOL
53
JSM 7001F low vacuum eld emission SEM. Samples were also imaged using
uo-
rescence microscopy on a Nikon Eclipse Ti-E inverted microscope. Glutaraldehyde
xed sample were stained with FM 4-64FX (Life Technologies) membrane stain (5
g/mL) and imaged using the TRITC channel (Nikon lter set G-2E/C).
4.3 Result and Discussion
4.3.1 Sampling Site
The Death Valley Flow System (DVFS) consists of highly fractured mostly
carbonate-rock aquifers that form a regional groundwater
ow system covering
hundreds of square km; extending from recharge zones associated with Central
Nevada Uplands to large discharge springs in the Amargosa Valley and Furnace
Creek area of Death Valley [171{173]. The source of water used for this study,
Nevares Deep Well 2, is located in the Death Valley portion of the discharge zone
of this system and intersects the Furnace Creek Fault Zone, which is inferred to
represent a major conduit for
uid
ow through the Funeral Mountains [173, 174].
Given the natural artesian
ow associated with this site, the mostly uncased na-
ture of the hole, and the fact that it was drilled without drilling
uid, NDW2 can
be regarded as a unique window into a deep continental microbial ecosystem.
4.3.2 Microbial Community Analysis in the Electrochemi-
cal Bioreactor
Anodic currents began developing within 20 days of the start of the electrochem-
ical enrichment (Figure 4.2), at all working electrode conditions (WE1: 272 mV,
WE2: 372 mV, WE3: 472 mV, and WE4: 572 mV vs. SHE). Separate testing
54
with sterile media under identical operating conditions (abiotic control) did not
result in any anodic current, indicating that the electron transfer was mediated by
the resident microorganisms. The observed anodic currents were correlated with
a modest decrease of electron donor (acetate) concentration, from 10 mM to 7.5
mM as measured by HPLC over a one month period, consistent with the activity
of a microbial community that couples the oxidation of acetate to anodic reduc-
tion. Acetate was chosen as the electron donor in recognition of its important
role as an intermediate in anaerobic degradation of organic matter, and in light
of previous observations that acetate-oxidizing microorganisms were prevalent on
anodes inserted into sediments [175]. Taken collectively, the overall increase in
potentiostatic current, reduction in acetate concentration, and presence of cells on
the WEs, point to the enrichment of a microbial community capable of electron
transfer to the reactors working electrodes across a wide range of oxidizing poten-
tials. This was also consistent with SEM images of the working electrodes, which
revealed the formation of biolms containing various cellular morphologies (Figure
4.3).
Along with the increasing trend of anodic current over time, we observed an
abrupt decrease in current with every change of medium (Figure 4.2). This may
indicate a role for soluble components that act as redox shuttles carrying electrons
from cells to the electrodes (indirect EET), as previously proposed for
avins se-
creted by Shewanella to mediate EET to electrodes [37, 38]. While the observed
decline with medium change is usually attributed to such redox shuttling mecha-
nisms, it has alternatively been suggested that such a decline may originate from
a decreased binding of the soluble redox components to cell-surface proteins that
perform direct EET to external surfaces [39]. While the precise mechanism(s) of
55
Figure 4.2: Electrochemical enrichment. The increasing anodic current vs. time
points to enrichment of microbial communities capable of mediating extracellular
electron transfer to the 4 working electrodes. Arrows indicate times of medium
change, which resulted in an abrupt decrease of anodic current before recovery of
the increasing trend.
cell-to-anode electron transfer in our enrichment culture is more dicult to pin-
point compared to the canonical metal-reducing model systems such as Shewanella,
it is interesting to also note that SEM imaging (Figure 4.3) revealed the presence
of extracellular laments morphologically similar to bacterial nanowires [41].
The microbial community analyses, performed using 16S rRNA gene pyrose-
quencing, revealed shifts between the initial samples, the DHS and batch enrich-
ments used as inocula for the electrochemical reactor, and the communities at-
tached to the electrodes (see Appendix A.2). The original groundwater inoculum
56
Figure 4.3: Scanning electron microscopy of the enrichment electrodes and associ-
ated biomass (1 m scale bar).
was dominated by members of the Proteobacteria (78%), with presence of Nitro-
spirae (10%), Firmicutes (2.5%), Chloro
exi (2.3%), Spirochaetes (1.24%), and
other unclassied organisms (2.7%) The sequences obtained from the DHS euent
belonged primarily to Alphaproteobacteria (60%) and Betaproteobacteria (16%),
in addition to Actinobacteria (7%) and Bacteroidetes (5%). The Fe(III)-reducing
batch culture used as inoculum for electrochemical reactor resulted in an increase
of Betaproteobacteria, and this trend continued in the electrochemical enrichment
with all electrodes dominated by the Comamonadaceae and Rhodocyclaceae fami-
lies in the Betaproteobacteria (Figure 4.4).Clone libraries (approximately 40 clones)
comprised of 21 clones from the genus Delftia (bootstrap 100%) and 1 clone from
57
Figure 4.4: Family level microbial community analysis of the electrochemical en-
richment. Families were determined using 16S rRNA tagged pyrosequencing anal-
ysis of DNA extracts. A batch culture (acetate as electron donor and Fe(III)-NTA
as electron acceptor) was the inoculum for the electrochemical reactor.
the genus Acidovorax (bootstrap 85%) within the Comamonadaceae family in ad-
dition to 8 clones from the genus Dechloromonas (bootstrap ranging from 68%
to 95%) and 5 clones from the genus Azonexus (bootstrap ranging from 46% to
69%) within the Rhodocyclaceae family. One clone each was obtained from the
genera Ignavibacterium (bootstrap 100%), Mesorhizobium (boostrap 100%) and
Aestuariimicrobium (bootstrap 100%).
With acetate as the electron donor and electrodes poised within the redox
potential range (300-750 mV vs. SHE) of nitrogen-based redox couples like ni-
trate in neutral waters [176], the electrochemical enrichment of Comamonadaceae
and Rhodocyclaceae, reported here, is consistent with their previous description
as short-chain fatty acid-utilising denitriers [177, 178]. Rhodocyclaceae are also
known to reduce chlorate and iron [179, 180]. Indeed, both families have been ob-
served on anodes in previous microbial fuel cell (MFC) enrichments [178]. Speci-
cally, Comamonadaceae were previously found to dominate the anodic community
58
of cellulose-fed MFCs inoculated with rumen microorganisms [181], and Rhodocy-
claceae dominated the anodic biolm of an acetate-fed MFC enrichment inoculated
with anaerobic digester slurry [182]. In both cases, it was suggested that some
members of these families could utilize electrodes as electron acceptors, in the ab-
sence of nitrate, while oxidizing short-chain fatty acids. However, no particular
members of these families from these previous enrichments have been isolated or
tested to conrm their EET ability. To date, the only conrmatory reports of EET
from relevant pure cultures are limited to Rhodoferax ferrireducens [183] and Co-
mamonas denitricans [88], from the Comamonadaceae, both of which have been
shown to produce current at MFC anodes in the absence of alternate soluble elec-
tron acceptors (e.g. oxygen or nitrate). With this in mind, we proceeded to isolate
representative members of the dominant groups in our electrochemical enrichment
to directly test these pure cultures for electrochemical activity in half-cell reactors.
4.3.3 Isolation of Delftia and Azonexus strains
Two strains, designated WE1-13 and WE2-4, were isolated as pure colonies orig-
inating from the biomass associated with the 272 mV vs. SHE (WE1) and 372
mV vs. SHE (WE2) electrodes. Phylogenetic analyses of their 16S rRNA gene
sequences demonstrated that these strains belonged to the genera Delftia and
Azonexus, respectively. Remarkably, previous studies have hinted at potential
EET activity from these genera, based on their minor appearance within MFC
anode communities fed with glucose and acetate, respectively [184, 185]. Since the
mere presence of microbes in anode biolms is not direct evidence of EET capa-
bility [53], these previous reports noted the need for additional studies targeting
their electrochemical activity [185]. Motivated by these reports, and the potential
59
Figure 4.5: Chronoamperometry of the isolated strains (A) Delftia sp. WE1-13,
(B) Azonexus sp. WE2-4. In both cases, the working electrode was poised at
522 mV vs SHE, acetate (10 mM) was used as electron donor, and reactors were
purged with N
2
to maintain anaerobic conditions. Insets show the total protein
content measured at time points labeled 1 and 2.
relevance of these microorganisms to subsurface environments by virtue of being
enriched from a deeply-sourced spring, we tested Delftia sp. strain WE1-13 and
Azonexus sp. strain WE2-4 as pure cultures in half-cell reactors. In addition to
the reference and counter electrodes, each anoxic reactor contained a carbon cloth
electrode (same material used in the electrochemical enrichment) poised at 572
mV vs. SHE, and acetate (10 mM) was used as electron donor.
Upon inoculation of Delftia sp. WE1-13 to reactors, the anodic current im-
mediately started rising above the typical 50 nA background established from the
sterile medium (Figure 4.5a), leveling o at 400 nA after within 11 hr. This rise
in current, indicative of microbial EET current to the anode, was accompanied
by near doubling in total protein content from 5.931 0.109 mg (inoculum) to
10.645 0.288 mg (sum of contributions from both planktonic cells and electrode-
attached biomass after 12 h). Similar chronoamperometry results were obtained
from Azonexus sp. WE2-4, whose inoculation resulted in a slower rise of anodic
60
current from a 70 nA sterile-medium background current to 325 nA over 60 hr,
concomitant with total protein increase from 5.487 0.095 mg to 9.069 0.301
mg (Figure 4.5b).
While dierent for the two strains, the time frame of the rise in anodic current
is in both cases consistent with previous observations of electrochemically active
microbes [37, 57, 183]. However, the magnitude of anodic currents detected from
the Delftia and Azonexus strains reported here is signicantly lower than these
typically reported for the metal-reducing microbes usually investigated as EET
model systems. Shewanella oneidensis MR-1, for example, produced three orders
of magnitude more current in reactors operated using identical electrode types
and under similar conditions while using lactate as an electron donor [49]. While
this may be partially attributed to growth and medium conditions that have not
been systematically optimized for EET activity, the low currents observed here may
point to the isolated strains lower ability to gain cellular energy from external redox
active surfaces. The latter notion is consistent with our observation of very low
cellular decay rates, but no growth, from the planktonic cell counts performed over
the course of chronoamperometry. Specically, the Delftia sp. WE1-13 cell density
changed from 7 2 10
7
CFUs/mL to 5 2 10
7
CFUs/mL, while Azonexus sp.
WE2-4 showed a slight decrease from from 6.9 1.4 10
8
CFUs/mL to 4.1 1.4
10
8
CFUs/mL (between the time points labeled 1 and 2 on Figure 4.5). Taken
collectively, the small decrease in cell density, modest increase in protein content,
and low EET currents observed, suggest low energy gain that may be directed for
cell maintenance and decreased cellular decay, rather than growth. Considering
the above, we heuristically compare EET rates in the isolated strains relative to
the well-characterized S. oneidensis MR-1 that is known to harness EET to gain
61
Figure 4.6: Further electrochemical and microscopic characterization of Delftia sp.
WE1-13. (A) Medium change resulted in a sudden decrease of anodic current, as
observed in the enrichment cultures. Inset shows total protein content measured
at the time points marked 1 and 2. (B) Scanning electron microscopy (1 m scale
bar) shows attachment of the rod-shaped of Delftia sp. WE1-13 on the carbon
cloth bers. (C) In vivo
uorescent image (FM 4-64 FX membrane stain) conrms
the attachment of Delftia sp. WE1-13 cells to the electrode during electrochemical
analysis (10 m scale bar).
energy for growth. Assuming an average bacterial protein content of 0.2 g/mL
[186] and average cell volume of Delftia and Azonexus as 0.4m
3
(estimated from
SEM images e.g. Figure 4.6b), we calculated the average respiration rate of Delftia
as 174.7 e

/sec/cell (carbon cloth protein content 1 mg; current 350 nA) and
of Azonexus as 326.4 e

/s/cell (protein content increase 0.39 mg; current 255
nA). These values are 3 orders of magnitude less than respiration rate (6.2 10
5
e

/sec/cell) observed in case of Shewanella oneidensis MR-1 [74].
We further investigated impact of medium change on Delftia sp. WE1-13, in
light of the observation that the original enrichments anodic current decreased
62
abruptly in response to this procedure (Figure 4.2). A separate reactor was inoc-
ulated with Delftia sp. WE1-13 after the establishment of a background sterile-
medium current, as described above (Figure 4.6). Following the expected rise in
current (up to 500 nA), during which the total planktonic cell density remained
nearly constant (from 5 3 10
7
to 6 1 10
7
CFUs/mL), the spent medium
and associated planktonic cells were replaced with fresh medium. Consistent with
the enrichments observations (Figure 4.2), this resulted in an abrupt decrease of
anodic current down to 60 nA followed by recovery to 200 nA over 6 hr (Figure
4.6a). At the conclusion of the experiment, the nal total protein content (both
planktonic and electrode biomass) was measured to be 12.764 0.466 mg, com-
pared to the inoculums 6.128 0.115 mg. SEM and
uorescence imaging (Figures
4.6b.c) conrmed cellular attachment and the formation of a Delftia biolm on the
working electrode, consistent with measured protein from the electrode biomass.
The drop in current upon medium change is consistent with loss of putative se-
creted components that may enhance EET by Delftia sp. WE1-13 to anodes.
Alternatively,the medium change may have temporarily disrupted biolm compo-
nents that directly mediate electron transfer to the electrodes. These observations
motivate further investigations into the identity of the charge carriers and precise
pathway(s) involved. In this context, it is also interesting to note that the phyloge-
netically closely related Delftia acidovorans, an organism dominant in subsurface
gold-associated mine communities [187], has been shown to produce secondary
metabolites responsible for extracellular gold reduction and precipitation, provid-
ing protection from toxic soluble gold [188]. To check whether D. acidovorans is
also electrochemically active, we obtained this strain from DSMZ DSM no. 39,
63
and tested it under identical conditions as Delftia sp. WE1-13. Our tests re-
vealed that D. acidovorans is indeed electrochemically active (data not shown)
with similar anodic currents, cell survival, and modest increase in protein content.
Clearly, additional studies are required to pinpoint both the precise mechanism of
EET, and its relevance to cellular protection and energy generation in subsurface
environments.
4.4 Conclusion
While alternative anaerobic metabolisms (e.g. nitrate reduction) by Delftia and
Azonexus species were previously known, this study provides direct evidence that
these organisms can also gain some energy for survival and/or persistence by elec-
tron transfer to external surfaces. The electrochemical enrichment and isolation
of these organisms from a deeply sourced spring raises interesting questions about
the relevance of EET in energy-limited continental subsurface environments, where
the diversity of host rocks may provide insoluble electron acceptors that greatly
expand the range of redox couples available for the resident microorganisms. Sig-
nicantly, these modes of energy acquisition from external surfaces, detectable
using electrochemical enrichment and analytical techniques, are easily missed us-
ing traditional cultivation strategies. In addition to helping identify new microbial
candidates for renewable energy and biocatalytic applications, ex situ and in situ
electrochemical approaches expand the range of cultivation conditions available for
subsurface microorganisms, and may shed light on potential mechanisms for cellu-
lar survival and detoxication activity in these environments. It should be noted
that while the focus here was on a continental subsurface site, similar relevance of
64
EET and electrochemically-active microorganisms may be of interest in the vast
marine subsurface or even in environments on extraterrestrial planetary bodies.
65
Chapter 5
In Situ Electrochemical
Enrichment of Subsurface
Bacteria at the Sanford
Underground Research Facility
5.1 Background
Continental subsurface environments can present signicant energetic challenges to
the resident microorganisms. Following molecular oxygen consumption by the ox-
idation of organic matter near the surface, alternate electron acceptors and donors
become prevalent [2]. Although strong emphasis has been given to studies of solu-
ble electron donors and acceptors in marine [19, 189, 190] and continental [191{193]
deep subsurface, alternate source of energy in form of insoluble minerals (elemental
sulfur, iron, manganese) have only recently gained attention [24{30]. Microbes ca-
pable of performing extracellular electron transfer (EET), allowing cellular electron

ow from or to inorganic minerals, may therefore have a selective advantage for
energy acquisition by uptake or disposal of electrons at the biotic{abiotic interfaces
prevalent in subsurface habitats.
Majority of this chapter will go into a paper in prep: Jangir et al. (2016)
66
Many physiologically and phylogenetically diverse microorganisms, beyond the
well-characterized dissimilatory metal reducing bacteria the Deltaproteobacterium
Geobacter and the Gammaproteobacterium Shewanella, are capable of using poised
electrodes as electron donors or acceptors [50{53, 55, 59{62] in substitution of nat-
urally occuring insoluble minerals. Since electrodes can act as unlimited electron
source or sinks, they are powerful tools for enriching subsurface organisms poten-
tially capable of EET.
The Sanford Underground Research Facility (SURF) in the former Homestake
Gold Mine (South Dakota, USA), active from 1876 to 2001, was reopened as a
state-run science facility in 2011. The mine is hosted in quartz-veined, sulde rich
segments of an Early Proterozoic, carbonate-facies iron-formation [194, 195]. Hy-
drological modeling indicates older (>10,000 yrs)
uids reaching the deeper levels,
especially on the northern ledges [196]. The mine provides an in situ study of
a signicant block of the Earth's Paleoproterozoic crust 2.7 x 3 x 5 kms and 6.5
kms of plunge length with >500 km of drift. Recently, by combining microbial
physiotype abundances (16S rRNA gene) with energy yields, the NASA Astrobiol-
ogy Institute (NAI) Life Underground team identied putative chemolithotrophic
metabolism present in the mine, including oxidation of iron, sulfur and methane,
and reduction of iron [197]. With evidence of potentially chemolithotrophic mi-
crobes, SURF provides a unique portal to the deep subsurface to enrich EET
capable microorganisms using electrochemical methods.
Here, we describe the rst in situ electrochemical reactor installed at SURF,
with electrode poised at electron-donating (reducing/cathodic) and electron-
accepting (oxidising/anodic) redox potentials, to enrich for potential mineral ox-
idising and reducing bacteria as predicted by Osburn et al. [197]. In addition
67
to demonstrating that the in situ electrodes can capture important bacterial lin-
eages active in the subsurface, we successfully isolated several bacteria including
Bacillus, Anaerospora, Comamonas,Cupriviadus and Azonexus from these poised
electrodes. Initial electrochemical studies on electrode oxidiser Comamonas sp.
(strain WE1-1D1) and electrode reducer Bacillus sp. (strain WE4-1A1-BC) are
elucidated.
5.2 Material and Methods
5.2.1 Field Measurements
Samples were acquired from the euent of an exploratory borehole (DUSEL 3A)
located at 4850 ft level close to Yates shaft of the Sanford Underground Research
Facility in Lead, South Dakota (USA). The borehole, drilled in 2009 with depth
200 m, intersects the Precambrian amphibolite metamorphic schists [194, 195]
and has been capped with manifolds to prevent
ooding. Initial aqueous chem-
istry of DUSEL 3A was obtained for thermodynamic modeling in February 2014
[197]. Further, detailed geochemical data were acquired during deployment (De-
cember 2014) and conclusion (May 2015) of the in situ electrochemical enrich-
ments. Aqueous chemistry (Table 5.1) was measured as described earlier [197].
Brie
y, oxidation-reduction potential (ORP), conductivity, pH, temperature, and
total dissolved solids (TDS) were measured in situ with an Ultrameter II 6PFCE
(Myron L Company), redox sensitive species were measured using Hach DR/2400
portable eld spectrophotometers and associated reaction kits (Hach Company,
Loveland, CO), and anions and cations were measured using ion chromatography
68
with Metrohm column Metrosep A SUPP 150 and Metrosep C6 - 250/4.0 respec-
tively. For detailed sampling techniques see Appendix B.1.
5.2.2 In Situ Electrochemical Reactor Deployment
For the in situ electrochemical incubation, schematized in Figure 5.1, borehole
DUSEL 3A water was pumped through the in situ electrochemical bioreactor
via peristaltic pumps (MasterFlex L/S Digital Drive, EW-07522-20, Cole-Parmer,
USA) at a
ow rate of 1 mL/min, providing a dilution of 1 per day. Aqueous chem-
istry (ORP, conductivity, pH and temperature) of the euent of the bioreactor was
Dec 2014 May 2015
pH 7.08 7.75
Temperature (
o
C) 18.7 19.31
Conductance (S/cm) 6514 7869
TDS (ppm) 3257 3929
ORP (mV) 59.2 -50
Dissolved Constituents
dO
2
(mg/L) 3.79 0
S
2
(g/L) 17 17
Fe
2+
(mg/L) 0.6 0.6
NO

3
(mg/L) 0.4 0.2
NH
3
(mg/L) 0.11 0.07
SiO
2
(mg/L) 24.7 9.7
Mn
2+
(mg/L) 0.4 0.9
PO
3+
4
(mg/L) 1.24 1.09
NO

2
(mg/L) BDL 0.001
Br
2
(mg/L) BDL 0.01
Li
+
(mg/L) 0.024 0.035
Na
+
(mg/L) 10.619 5.698
K
+
(mg/L) 0.229 0.751
Cl

(mg/L) 0.316 0.385
SO
2
4
(mg/L) 14.049 46.696
Table 5.1: Chemical composition of water from DUSEL 3A borehole, 4850 ft level,
Sanford Underground Research Facility (SD, USA).
69
A)
B) C)
Figure 5.1: A) Schematic diagram of the in situ electrochemical enrichment reac-
tor. The four carbon cloth working electrodes were poised at -0.19, +0.01, +0.26
and +0.53 V vs. SHE, respectively, using a 4-channel potentiostat to enrich for
electrode oxidising and reducing microbes in a single reactor. DUSEL 3A borehole
water was fed directly into the reactor and the euent was pumped out to Yates
shaft. B) Geological map of the 4850 ft L at SURF showing drilled boreholes
intersecting Precambrian Yates member and tertiary rhyolite. the reactor was de-
ployed to enrich for potential electrode oxidising and reducing bacteria from bore-
hole DUSEL 3A water. C) Photograph of the installed the reactor near borehole
DUSEL 3A manifold.
70
logged using multiparameter HI9829 meter (Hanna Instruments, USA). The biore-
actor is a standard half-cell, custom designed from a 1 L spinner
ask (CLS-1410,
Chemglass Life Sciences) to incorporate Ag/AgCl reference electrode (MF-2052,
BASi, USA) and platinum counter electrode (CHI115, CH Instruments, USA). The
bioreactor holds a PTFE assembly of ve threaded rods, each of which supported
a working electrode (WE) composed of 6 cm 7 cm carbon cloth (PW06, Zoltek,
St. Louis, MO, USA). During enrichment, four working electrodes (WE1, WE2,
WE3, and WE4) were poised at -0.19, +0.01, +0.26 and +0.53 V vs. SHE, re-
spectively, using a four-channel potentiostat (EA164 Quadstat, EDaq, USA); while
one working electrode (WE0) was kept at the open circuit potential. All electrical
connections were made using insulated titanium wires. The complete set-up was
autoclaved with the working and the counter electrodes. The ethanol-sterilized
reference electrode was inserted before deployment. Applied redox potentials and
current production were controlled and recorded via the eCorder eCHART software
(eDAQ Inc., Colorado Springs, CO). Before the deployment of the electrochemi-
cal reactor in December 2014, modications in the manifold design were made by
SURF ocials that may lead to back
ow from the dehumidier water. To avert
this contamination, a check valve was installed upstream of the solenoid valve
in December 2014. The reactor was incubated with continous
ow of borehole
DUSEL 3A water for 5 months beginning December 2014, monitored monthly for
proper operation, and nally retrieved in May 2015. During decommission, small
sections of carbon cloth from each potential were collected for microscopy, active
microbial community analysis and further laboratory electrochemical enrichments
as described below.
71
Figure 5.2: Schematic diagram of the lab electrochemical reactors employed to
enrich for electrode oxidising and reducing microbes. Carbon cloth was used as
working electrodes (1 cm x 1 cm) with platinum acting as counter electrode and
Ag/AgCl (saturated in 1 M KCL) as reference electrode.
5.2.3 Secondary Electrochemical Enrichment
At conclusion of the reactor deployment, the carbon cloth working electrodes (3 cm
6 cm) were placed into autoclaved 125 mL serum bottles containing ltered (0.22
m) DUSEL 3A euent water. To minimize oxygen exposure, the serum bottles
were stored in Mylar bags with oxygen absorbers during transportation. In the
laboratory, secondary electrochemical enrichments were performed using the poised
carbon cloth from the reactor in four separate three-electrode electrochemical cells
(50 mL) for further enrichment of electrode oxidising and reducing microbes. SURF
base media was designed to mimic the aqueous chemistry of borehole DUSEL 3A
water, with basal salts (L
1
): NaCl, 0.029 g; KH
2
PO
4
, 0.041 g; NH
4
Cl, 0.160 g;
FeSO
4
.7H
2
O, 0.042 g; Na
2
SO
4
, 0.014 g; yeast extract, 0.0005 g; vitamins and trace
minerals [167]. Media was maintained at pH 7.5 using phosphate buer (10
mM). Enrichment of electrode oxidisers was performed by augmenting SURF base
72
media with sodium bicarbonate (10 mM), electrodes poised at -0.19 and +0.01 V
vs SHE, and constant purging of gas mix (CO
2
:air v:v 20:80) in the electrochemical
cell. To enrich for electrode reducers, media was augmented with sodium acetate
(5 mM), electrodes poised at +0.26 and +0.53 V vs SHE and constant purging with
N
2
to maintain anaerobic conditions (Figure 5.2). The secondary electrochemical
enrichments performed in batch mode (media change every 7-10 days), ran for 2
months.
5.2.4 Active Microbial Community Analysis
Active microbial community was analysed by extracting RNA from: 1) dehumidi-
er water, 2) Euent of DUSEL 3A, 3) Planktonic media of the in situ reactor, 4)
Biolms of carbon cloth in the in situ reactor, and 5) Biolms of carbon cloth from
lab electrochemical enrichments. Microbiological samples were collected in 0.22m
sterivex lters (Millipore,USA) from the euent of the borehole DUSEL 3A, dehu-
midier water, along with planktonic media (For details, please see Appendix B.1).
RNA was puried from the lters and the biolms on the working electrodes (in
situ reactor and secondary lab electrochemical enrichments) via physical (freeze,
thaw, vortex) and chemical (lysozyme) disruption of the cell wall prior to phenol-
chloroform extraction described earlier [198]. Any residual DNA was digested using
TURBO DNA-free
TM
Kit (ThermoFisher Scientic, USA). Detailed RNA extrac-
tion protocol is provided in Appendix B.4. Puried RNA samples were sent to
Molecular Research DNA (Shallowater, TX, USA) for reverse transcription and
high throughput sequencing. Hypervariable region V4 of the bacterial 16S rRNA
gene amplicon was amplied using barcoded fusion primers 515f and 806r [105],
according to previously described methods [199]. Following amplication, all PCR
73
products from dierent samples were mixed in equal concentrations and puried
using Agencourt Ampure beads (Agencourt Bioscience Corporation, MA, USA).
Samples were sequenced on Roche 454 FLX titanium instruments using recom-
mended reagents and following manufacturer's guidelines.
The sequence data were processed using MOTHUR [115]. Sequences were de-
pleted of barcodes and the primer, followed by trimming to remove any ambiguous
base calls and homopolymer runs greater than 8 bp. Average sequence lengths of
260 bp and a total of 590000 unique sequences were obtained for all samples.
The trimmed sequences were aligned with the Ribosomal Data Project (RDP) in-
fernal aligner [169], chimeras were removed using UCHIME [110], and a distance
matrix was created. The sequences were clustered to identify unique operational
taxonomical units (OTUs) at the 97% level, and taxonomy was assigned using
RDP classier. The resultant OTUs were manually curated to remove microbial
community any contaminants from dehumidier water and extraction reagents for
downstream analysis.
5.2.5 Isolations
Poised carbon cloth with biolm was chosen for further isolation of bacterial
strains. For manganese oxidisers, soluble MnCl
2
(with sodium bicarbonate as the
carbon source) and manganese (II) acetate was used as the electron donor. Re-
duced iron substrates such as FeCO
3
[200], FeS [201] and (NH
4
)
2
Fe(SO
4
)
2
.6H
2
O
was used for iron oxidisers. Sulfur oxidisers were enriched using colloidal sulfur
[62]. To isolate mineral oxidisers, nitrate and oxygen served as terminal electron
acceptors and sodium bicarbonate as the carbon source. Iron reducers were en-
riched on sodium acetate and Fe(III)-NTA or poorly crystalline iron oxides (PCIO)
74
[202]. Manganese reducers were enriched with sodium acetate and -MnO
2
[71].
Gradient tubes containing 1% agar and 1.5% agar plates were used to isolate indi-
vidual colonies. Microbial strains were also isolated on rich media R2A [170]. The
detail for preparing each mienral is provided in Appendix B.5.
5.2.6 DNA extraction from isolates
DNA extraction of pure cultures was performed by UltraClean Microbial DNA Iso-
lation kit (MoBio laboratories, USA) following manufacturers instructions. Bac-
terial 16S rRNA gene was amplied from the extracted DNA with primers 8f: 5 -
AGA GTT TGA TCC TGG CTC AG - 3 and 1492r: 5 - GGT TAC CTT GTT
ACG ACT T 3, and the puried PCR product (PureLink PCR Purication Kit,
Life Technologies, CA, USA) was sent for sequencing to Genewiz (South Plaineld,
NJ, USA).
5.2.7 Electrochemical Measurement of Isolates
Chronoamperometry measurement of each isolate was performed in a standard
three-electrode glass electrochemical cell (50 mL volume). Carbon cloth (1 cm
1 cm) was used as the working electrode (WE), with platinum wire as counter
electrode, and a 1 M KCl Ag/AgCl reference electrode. Each isolate was aero-
bically grown from frozen stocks (-80
o
C) in R2A media to stationary phase and
this culture was used to inoculate 500 mL of SURF base medium at 1% (v:v).
After reaching mid-exponential phase, the culture was pelleted by centrifugation
at 7000 g for 10 min, washed 2 and resuspended in 10 mL fresh SURF base
medium without electron donor. 5 mL of this suspension was introduced to the
75
electrochemical cell, already containing 50 mL SURF base medium with the work-
ing electrode poised at relevant potential, after the abiotic current has stabilized
to a constant baseline. Cell densities were determined by plate counts.
5.2.8 Microscopy
For scanning electron microscopy (SEM), electrode samples were xed overnight
in 2.5% glutaraldehyde. Samples were then subjected to an ethanol dehydration
series (25%, 50%, 75%, 90%, and 100% v/v ethanol, for 15 min each treatment) and
critical point drying (Autosamdri 815 critical point drier, Tousimis Inc., Rockville,
MD, USA). The samples were then mounted on aluminum stubs, coated with Au
(Sputter coater 108, Cressington Scientic), and imaged at 5 keV using a JEOL
JSM 7001F low vacuum eld emission SEM.
5.3 Result and Discussion
5.3.1 Current production and microbial community struc-
ture in the in situ reactor
The redox potentials applied in situ mimicked elemental sulfur oxidation (-0.19
V vs SHE), iron oxidation (+0.01 V vs SHE), manganese reduction (+0.53 V vs
SHE), all predicted as putative energy yielding metabolism for microbial commu-
nity inhabiting the borehole DUSEL 3A water [197]. Since, reduction potential of
many cytochromes range from -0.5 to +0.3 V vs SHE [3], a fourth redox potential
was also applied at +0.26 V vs SHE. Therefore, the in situ reactor consisted of two
working electrode (WE1: -0.19 V vs SHE, WE2: +0.01 V vs SHE) operating at
potentially electron donating conditions, while rest two (WE3: +0.26 V vs SHE,
76
WE4: +0.53 V vs SHE) were operating at potentially electron accepting condi-
tions. Detailed chronoamerometry data is available in Appendix B.2. In May 2015
(during decommissioning of the in situ reactor), the ORP of DUSEL 3A water and
the euent of the in situ reactor was -50 mV (Table 5.1) and 200 mV (Table 5.2)
respectively; indicating oxygen leakage in the in situ reactor. Increase in cathodic
currents with decrease in the poised potential of electrode (Table 5.2) suggests
possible interplay of abiotic (O
2
reduction) and biotic electron exchange at the
electrode surfaces.
The microbial community analyses was performed using phylogenetic 16S rRNA
pyrosequencing. The obtained data was curated to remove any contamination from
the dehumidier water, the nucleic acid extraction reagents. These contaminants
mainly comprised of Actinobacteria, Verrumicrobia, Acidobacteria, Enterobacteri-
aceae, Moraxellaceae and Porphyromonadaceae (Appendix B.7). Borehole DUSEL
3A water was dominated by members of the Proteobacteria (84.78%), with presence
of Bacteriodetes (7.64%), Firmicutes (0.07%), Chloro
exi (0.03%), Planctomycetes
Applied Voltage
Average current (A)
standard deviation
-0.19 V vs SHE -276.68 43.80
+0.01 V vs SHE -168.76 26.82
+0.26 V vs SHE -0.15 0.09
+0.53 V vs SHE -1.47 0.11
the in situ reactor euent aqueous chemistry
pH 7.5
Temperature (
o
C) 19
Conductance (S/cm) 11000
ORP (mV) 200
Table 5.2: Mean current (last 40 days) observed at various redox potentials in the
in situ reactor and aqueous water chemistry of the in situ reactor euent.
77
Figure 5.3: Microbial community of the borehole DUSEL 3A versus dierently
poised electrodes in the reactor depicts enrichment of unclassied/novel microbiota
at the reducing potentials (-0.19 V and 0.01 V vs SHE) where microbes use the
electrode as the electron donor.
(0.01%), and other unclassied organisms (7.47%). Within Proteobacteria, major-
ity organisms were unclassieds (37.03 %) followed by Deltaproteobacteria (32.33
%), Gammaproteobacteria (7.04 %), Alphaproteobacteria (6.68%) and Betapro-
teobacteria (1.48 %) (Figure 5.3).
Majority of the bacteria falling under unclassieds populated the reducing po-
tentials highlighting the value of alternate sources of reducing power available in
the subsurface [26, 30] (Figure 5.3). Some bacterial families dominant in bore-
hole DUSEL 3A water, such as Desulfobacteraceae, Desulfobulbaceae, Desulfovib-
rionaceae and Myxococcales, were underrepresent in the in situ reactor. Oxygen
leakage in the in situ reactor may have hampered growth of bacteria belonging to
the families Desulfobacteraceae and Desulfovibrionaceae comprising of strict anaer-
obes species deriving energy by reduction of sulfate. The presence of sulde in the
borehole DUSEL 3A water should have enhanced the growth of sulde oxidising
bacterial species belonging to the family Desulfobulbaceae [203{205] in in situ re-
actor; however, its dearth may be a resultant of abiotic sulde oxidation from the
78
oxidised metal compounds such as Fe(III) and Mn(III, IV) phases that are e-
cient chemical oxidants for sulde [206]. All bacterial families present in DUSEL
3A water were represented on at least one of the poised electrodes in the in situ
reactor. Taken together, this rst unique opportunity of deploying an electrochem-
ical reactor close to the environment of our interest was valuable to recognise the
signicance of in situ electrochemical reactor in capturing the indigenous bacterial
community.
Based on the bacterial data collected with this experiment, sequences closely
matching each bacterial family appeared to respond dierentially to each working
electrode. Some preferential bacterial attachment to the poised electrodes may
be explained by their known physiotypes. The families Flavobacteraceae, Bacil-
laceae and Rhodobacteraceae containing some putative metal oxidising members
[207{213] grouped at the reducing potentials (-0.19 and 0.01 V vs SHE) (Figure:
5.4 A,B), and the canonical sulfate reducing family Desulfovibrionaceae and the
methylotrophic family Methylococcaceae were enriched more at oxidising potential
(+0.53 V vs SHE) [204, 214] (Figure: 5.4 D). The presence of methylotrophic bacte-
ria at higher potentials suggests potential bacterially mediated methane oxidation
coupled to electrode reduction. We also observed some unexpected assemblage
of the bacterial families on the poised electrodes. The family Bradyrhizobiaceae
and Hypomicrobiaceae containing prominent iron- oxidising members and the fam-
ily Desulfobulbaceae containing prominent sulfur- oxidising members, dominated
the electrodes poised at the higher potential (+0.26 and +0.53 V vs SHE, re-
spectively). Since electron uptake at an electrode potential of +0.4 V vs. SHE
has been previously observed for chemolithotrophic Acidithiobacillus ferrooxidans
[61]; in our case, members of the family Bradyrhizobiaceae, Hypomicrobiaceae and
79
Figure 5.4: Bacterial community grouped with an increasing (a), decreasing (b),
and random pattern (c,d) at poised electrodes of in situ electrochemical reactor.
80
Desulfobulbaceae may be capable of electron uptake form the more positively poised
electrodes for either survival or growth.
Many bacterial families preferred both lower and higher potentials. For ex-
ample, the family Caulobacteraceae comprising of prosthecate bacteria with some
having the capability of Mn(II) oxidation [211], the family Rhodospirillaceae con-
taining some methylotrophic bacteria [215], putative metal oxidiser families Co-
mamonadaceae and Rhodocyclaceae [212, 216], and sulfate reducing family Desul-
fobacteraceae [203] were dominated on electrodes poised at -0.19 V and 0.26 V
vs SHE (Figure 5.4 E). Whereas methylotrophic Mythylocystaceae [214], hydrogen
oxidiser Hydrogenophilaceae [217] and metal reducers like Desulfuromonadaceae
and Geobacteraceae [218, 219] dominated the electrodes poised at 0.01 V and 0.53
V vs SHE (Figure 5.4 F). Taken collectively, subsurface bacteria may be more
physiologically versatile to interact with the minerals (oxidising or reducing) in
the extreme environment compared to their analogues on the surface.
5.3.2 Current production and microbial community struc-
ture in laboratory electrochemical enrichment
After 5 months of in situ incubations, sections of carbon cloth working elec-
trodes from the reactor were transferred to laboratory reactors as described earlier.
Within 8 days, while cathodic currents were observed at the reducing potentials
(-0.19 and +0.01 V vs SHE), anodic currents were observed at the oxidising poten-
tial WE4 (+0.26 and +0.53 V vs SHE). For detailed chronoamerometry data, see
Appendix B.3. Separate incubation with sterile media under identical operating
conditions (abiotic control) resulted in an average current of -24.976 0.74A at
WE1 (background oxygen reduction/negative current production) whereas 0.925
81
A)
B)
C)
Figure 5.5: A) Mean anodic and cathodic currents observed in laboratory enrich-
ments at WE1 (-0.19 V vs SHE) and WE4 (+0.53 V vs SHE) were higher than their
respective abiotic controls. Similarly, dierence was observed in cyclic voltamme-
try of electrode oxidising (B) and reducing (C) biolms compared to the abiotic
controls. This indicates that the electron transfer was mediated by the resident
microorganisms.
0.23A at WE4. Since the WE3 (+0.26 V vs. SHE) was lost due to an electrical
short between the counter and the working electrode on 42nd day of incubation,
further analysis was performed only on WE1 and WE4 of the laboratory electro-
chemical enrichments(Figure 5.5). Cyclic voltammetry (CV) was performed with
a scan rate of at 10 mV/s on the biolms of colonized electrodes of WE1 and WE4
and compared to the abiotic control (Figure 5.5). The classic turnover CV corre-
sponding to active multi-step oxidation of acetate [49] observed on WE4 points to
enrichment of a putative metal-reducer. The single sigmoid wave was shifted to
82
A) B)
Figure 5.6: Scanning electron microscopy of the enrichment electrodes associated
biomass at (A) reducing potential (-0.19 V vs SHE) and (B) oxidising potential
(0.53 V vs SHE). Enriched biolms at reducing potentials comprised of long thin
lamentous like bacteria whereas biolms at oxidising potentials contain extracel-
lular appendages as observed in case of Shewanella and Geobacter. Scale bar A)
100 m and B) 1 m.
positive redox potentials indicating that the enrichment may harvest energy from
electron acceptors with higher redox potentials. This value lies between redox
potential triggering current
ow in G. sulfurreducens [220] and redox potential of
outer-membrane cytochromes of S. oneidensis MR-1 [39]. The overall increase in
potentiostatic current, dierence in the cyclic voltammogram and cell biomass on
the electrodes observed through SEM images (Figure 5.6), point to the enrichment
of a microbial community capable of electron transfer to/from the electrode.
The bacterial community analysis of the laboratory electrochemical enrich-
ments indicate growth of many unclassied bacteria at the reducing potential sim-
ilar to the in situ reactor (Figure 5.3). The dominance of the Desulfuromonadaceae
and the Geobacteraceae observed at the most oxidising potential of +0.53 V vs.
SHE (Figure 5.7) explains the high anodic current and the classic turn over CV.
The members of these family are capable of anaerobic respiration utilizing a va-
riety of compounds as electron acceptors, including sulfur, Mn(IV), Fe(III), and
83
Figure 5.7: Family level bacterial community analysis of the electrochemical enrich-
ment. Families were determined using 16S rRNA tagged pyrosequencing analysis
of RNA extracts.
84
poised electrodes [53, 221]. The Firmicutes were more dominant than the Desul-
furomonadaceae and the Geobacteraceae at WE3 (+0.26 V vs. SHE), suggesting
that the Firmicutes were able to survive the electrical short possibly due to their
ability of forming an endospore. Interestingly, Comamonadaceae and Rhodocy-
claceae were present in all the four poised electrodes with decreasing abundance
from most reducing potential (-0.19 V vs SHE) to most oxidising potential (+0.53
V vs SHE). Many bacterial families were enriched at the lower/reducing potentials
(-0.19 and +0.01 V vs. SHE) including the Chitinophagaceae, Plactomycetaceae,
Caulobacteraceae, Bradyrhizobiaceae, Hyphomicrobiaceae, Sphingomonadaceae and
Hydrogenophilaceae. The family Caulobacteraceae and Hyphomicrobiaceae are
prosthecate budding bacteria. Some members withing these family are capable
of oxidising Mn(II). Within Bradyrhizobiaceae, Rhodopseudomonas palustris oxi-
dises Fe(II) and perform electron uptake [60] for growth. The Hydrogenophilaceae
family contain two genura Hydrogenophilus and Thiobacillus capable of gaining
energy from oxidising hydrogen and sulfur. Principally, the species belonging
to Plactomycetaceae are chemoorganotrophic and obligate aerobes, Chitinopha-
gaceae are aerobes or facultative anaerobes, and most of the Sphingomonadaceae
are chemoorganotrophic. The diversity of organisms from the subsurface enriched
on lower potentials stress further experiments to investigate electron uptake via
microbial extracellular electron transfer.
5.3.3 Isolation of microbial strains and initial electrochem-
istry
Bacterial strains were isolated as pure colonies originating from the biomass asso-
ciated with the -0.19 V vs. SHE (WE1) and +0.53 V vs. SHE (WE4) electrodes.
85
Phylogenetic analyses of their 16S rRNA gene sequences demonstrated that these
strains belonged to the genera Bacillus, Anaerospora, Cupriavidus, Azonexus and
Comamonas. While Bacillus, Anaerospora, Cupriavidus, Azonexus were isolated
from WE4 (0.53 V vs SHE), two strains of Comamonas were isolated from both
WE1 (-0.19 V vs SHE) and WE4 (0.53 V vs SHE). Further electrochemical studies
were performed on isolated Comamonas sp. (strain WE1-1D1) from WE1 (-0.19
V vs. SHE) and Bacillus sp. (strain WE4-1A1-BC) from WE4 (+0.53 V vs SHE).
Both Gram-negative Comamonas and Gram-positive Bacillus have appeared
in microbial fuel cell communities. While, the genus Comamonas has been pre-
viously enriched on both cathodic and anodic community of microbial fuel cell
[88, 184, 222], the genus Bacillus has been dominant as the anodic community
[223{225]. Some species of these genera are capable of interacting with minerals.
The Comamonas sp. strain IST-3 oxidises Fe(II) for energy gain [216] and Bacil-
lus can reduce Fe(IV) but without conserving energy [202]. Isolation of both these
genera from subsurface and its relevance in the EET research, led us to further
perform electrochemistry on their pure cultures in half cell reactors.
Comamonas sp. (strain WE1-1D1) was studied in a half cell reactor for elec-
tron uptake at working potential of -0.19 V vs SHE. The media was augmented
with 10 mM sodium bicarbonate and constantly purged with gas mix CO
2
:Air v:v
20:80. Upon inoculation of Comamonas sp. (strain WE1-1D1) to the reactors,
the cathodic current (electron uptake from electrodes) initially decreased in the
rst 5 hours and then increased from -5 A to -8 A. Initial decrease in cathodic
current may be attributed to rapid utilization of dissolved oxygen by inoculated
microbes, subsequently decreasing the abiotic oxygen reduction at the electrode.
86
Figure 5.8: Chronoamperometry of the isolated strains (A) Comamonas sp. (strain
WE1-1D1) and (B) Bacillus sp. (strain WE4-1A1-BC). While Comamonas sp.
(strain WE1-1D1) was enriched on working electrode poised at -0.19 V vs. SHE
with CO
2
as the carbon source and oxygen as the electron donor, Bacillus sp.
(strain WE4-1A1-BC) was enriched on working electrode poised at 0.53 V vs.
SHE with acetate as the electron donor and carbon source. Arrow indicate the
time point when the bioreactor was inoculated with the cells.
Eventually, once the biolm is formed, the biotic cathodic current develops over-
shadows the background abiotic oxygen reduction. The increase in planktonic cell
density from 3 1 10
7
CFU/mL to 5 1 10
8
CFU/mL, along with larger
cathodic current observed in CV of Comamonas sp. compared to abiotic control
adheres to our hypothesis that Comamonas sp. is accepting electrons from the
electrodes for gaining energy (Figure 5.8 A). Killed control CV resulted in similar
voltammogram as the abiotic control conrming the decay at -0.1 V vs. SHE in
cyclic voltammogram was biotic. Taken together, the rise in the palntonic cell
density, increase in the cathodic current, dierence in the biotic cyclic voltamo-
gram, and presence of cells on the electrodes (Figure B.5), point to electron uptake
from the poised electrode by Comamonas sp. (strain WE1-1D1) for growth. The
few electrochemical studies performed on the mineral oxidising bacteria show a
87
diversity of of EET mechanisms [59, 61, 62, 94]. Further electrochemical analysis
are underway to investigate the precise EET mechanism active in Comamonas sp.
(strain WE1-1D1).
Bacillus sp. (strain WE4-1A1-BC) was studied in a half cell reactor with work-
ing electrode poised at 0.53 V vs SHE. Upon inoculation of Bacillus sp. (strain
WE4-1A1-BC) to the reactors, the anodic current increased from 0.1 A to 1.6
A. In this case, the decay by two order of magnitude in planktonic cell counts
were observed from 2 1 10
8
CFU/mL to 4 1 10
6
CFU/mL indicating
possible cell lysis as free planktonic cells. Cyclic voltammetry shows a reversible
peak centered at 0 V vs. SHE in biolm associated working electrode. To study
whether this peak arose from any soluble redox mediators released by the organ-
ism, cyclic voltammetry was performed on the spent media (planktonic cells were
spun down 15 at 7000 g). Cyclic voltamogram of the spent media also lacked
the reversible peak indicating that redox components bound to the Gram-positive
bacteria (Figure 5.8 B) may be responsible for extracellular electron transfer. Fur-
thermore, HPLC measurements conrmed a decrease in acetate concentration by
0.8 mM (12%) in a span of 13 hours. Based on these measurements, the coulombic
eciency was calculated as 1.81% using the ratio of the total coulombs produced
during the experiment to the theoretical amount of coulombs available from the
oxidation of acetate to CO
2
. While coulombic ecieny of a well known electrode
reducer S. oneidensis MR-1 for electricity generation from lactate (to acetate) is
20% [226]. Taken collectively, the overall increase in anodic current, reduction
in acetate concentration, the shift in biotic cyclic voltamogram, and presence of
cells on the WEs (Figure B.5), point to Bacillus sp. (strain WE4-1A1-BC) capable
of electron transfer to the poised electrode. Further electrochemical analysis are
88
underway to investigate the precise EET mechanism active in Bacillus sp. (strain
WE4-1A1-BC).
5.4 Conclusion
SURF provides a portal to deep subsurface microorganisms inhabiting sulde rich
segments of an Early Proterozoic (2 { 2.5 billion years ago), carbonate-facies iron-
formation. Our study provides the rst report of a potentiostatically controlled
in situ enrichment of subsurface bacteria. We were able to capture most mem-
bers of the bacterial community present in DUSEL 3A; all bacterial families were
represented in at least one of the poised electrodes in the reactor. This illus-
trates the promise of electrochemical techniques for enriching and studying life in
the subsurface. This work led to isolation of putative electrode reducers such as
Bacillus, Anaerospora, Cupriavidus, Azonexus, and a putative electrode oxidiser
Comamonas. Initial electrochemistry performed on a representative electrode ox-
idiser Comamonas sp. and Bacillus sp. indicated presence of potential electron
uptake and donate mechanisms, which needs to be further investigated. The elec-
trochemical enrichment of the subsurface microbes elucidates the relevance of EET
in the subsurface environment which expands the energy sources available to resi-
dent microbes.
89
Chapter 6
Summary and Conclusion
Every cell must solve the problem of energy generation in order to survive. While
many organisms use soluble electron donors and acceptors to gain energy, this form
of metabolism may be limiting in the subsurface. Interestingly, a diverse group
of microorganisms have solved this problem by gaining energy using extracellular
electron transfer (EET) to/from the insoluble substrates on the surface and the
subsurface. This mode of energy acquisition from external surfaces, detectable
using electrochemical enrichment which mimics the environmental conditions, are
easily missed using traditional cultivation strategies.
The analytical techniques developed in this thesis examined EET mechanisms
in a model dissimilatory metal reducing bacteria S. oneidensis MR-1. We demon-
strated that S. oneidensis MR-1 nanowires as extensions of outer-membrane and
periplasm, rather than pilin based structures. We also identied dierential EET
interaction between S. oneidensis MR-1 with the poised electrodes based on the
electrode material, and showed that the
avins accelerate EET as cytochrome-
bound cofactors, rather than free soluble molecular shuttles. This study moti-
vates structural and electrochemical approaches to unravel precise EET mechanis-
tic pathway in Shewanella and other candidate organisms.
Electrochemical approaches expand the range of enrichment conditions avail-
able for subsurface microorganisms. In this thesis, ex situ electrochemical en-
richment led to isolation of two subsurface microorganisms Delftia and Azonexus
90
from Nevares Deep Well 2, Death Valley. This thesis provides direct evidence that
these organisms may also gain energy for survival and/or persistence by electron
transfer to external surfaces. This thesis also reported rst ever potentiostati-
cally controlled in situ electrochemical reactor of the subsurface bacteria at 4850
ft depth located in the Sanford Underground Research Facility. This reactor was
able to capture the bacterial community in subsurface on the electrodes. The value
of alternate sources of reducing power available in the subsurface was also high-
lighted by dominance of unclassied bacteria at the reducing potentials. Many
putative electrode oxidisers and reducers were isolated from the electrochemical
enrichments. Studies focusing on understanding detailed mechanistic pathway of
EET operable in these isolated microbes should be pursued in future.
This study sheds light on potential mechanisms for cellular survival in
subsurface environments, and may nd applications in the eld of subsur-
face/extraterrestrial sensing of biosignatures. This novel in situ electrochemical
enrichment methods oers a powerful tool to understand the physiology and ecol-
ogy elusive in subsurface microbes. To our knowledge, this project is the rst
to integrate redox controlled cultivation with a comprehensive subsurface detec-
tion and sampling eort, while studying the role that redox-active surfaces play
in energy-exchange with microbes. The ability to cultivate subsurface microbes
impacts our understanding of their physiology and the physical/chemical context
in which these organisms thrive. To conclude, this thesis has helped resolve some
profound question regarding EET mechanisms of a model DMRB Shewanella, and
isolated subsurface organisms (capable of EET) using novel custom designed elec-
trochemical reactor. This opens a `window' to explore the subsurface microbial
ecology and physiology to understand the role of microbes in biogeochemical cycles
91
of elements, and numerous opportunities to identify novel biochemical pathways
and electron transfer mechanisms.
92
Bibliography
[1] Oxidative phosphorylation - Reference pathway. URL http://www.genome.
jp/kegg/pathway/map/map00190.html. Accessed: 2016-03-01. ix, 9
[2] Kenneth H Nealson. Sediment bacteria: Who's there, What are they doing,
and What's new? Annual Review of Earth and Planetary Sciences, 25(1):
403{434, 1997. ix, 2, 10, 12, 66
[3] Frauke Kracke, Igor Vassilev, and Jens O Kr omer. Microbial electron trans-
port and energy conservation{the foundation for optimizing bioelectrochem-
ical systems. Frontiers in microbiology, 6, 2015. ix, 11, 76
[4] What is the redox potential of a cell? URL http://book.bionumbers.org/
what-is-the-redox-potential-of-a-cell/. Accessed: 2016-03-01. ix, 11
[5] Mohamed Y. El-Naggar and Steven E. Finkel. Live wires: Discoveries of
microbial communities that transfer electrons between cells and across rela-
tively long distances are launching a new eld of microbiology. The Scien-
tist. URL http://www.the-scientist.com/?articles.view/articleNo/
35299/title/Live-Wires/. ix, 13
[6] Marian Breuer, Kevin M. Rosso, Jochen Blumberger, and Julea N. Butt.
Multi-haem cytochromes in Shewanella oneidensis mr-1: structures, func-
tions and opportunities. Journal of The Royal Society Interface, 12(102),
2014. ISSN 1742-5689. doi: 10.1098/rsif.2014.1117. URL http://rsif.
royalsocietypublishing.org/content/12/102/20141117. ix, 14
[7] Liang Shi, Kevin M Rosso, Tomas A Clarke, David J Richardson, John M
Zachara, and James K Fredrickson. Molecular underpinnings of Fe(III) ox-
ide reduction by Shewanella oneidensis MR-1. Frontiers in microbiology,
page 46, 2012. x, 2, 12, 15, 24
[8] Center for molecular biology of rna, . URL http://rna.ucsc.edu/
rnacenter/ribosome_images.html. Accessed: 2016-03-01. x, 18
[9] Dna sequence analysis, . URL http://www.alimetrics.net/en/index.
php/dna-sequence-analysis. Accessed: 2016-03-01. x, 18
93
[10] Allen J Bard, Larry R Faulkner, Johna Leddy, and Cynthia G Zoski. Elec-
trochemical methods: fundamentals and applications, volume 2. Wiley New
York, 1980. x, 38
[11] David Deamer and Arthur L Weber. Bioenergetics and life's origins. Cold
Spring Harbor perspectives in biology, 2(2):a004929, 2010. 1
[12] Chiaolong Hsiao, I-Chun Chou, C Denise Okafor, Jessica C Bowman, Eric B
O'Neill, Shreyas S Athavale, Anton S Petrov, Nicholas V Hud, Roger M
Wartell, Stephen C Harvey, et al. RNA with iron (ii) as a cofactor catalyses
electron transfer. Nature chemistry, 5(6):525{528, 2013. 1
[13] Benjamin E Ramirez, Bo G Malmstr om, Jay R Winkler, and Harry B Gray.
The currents of life: the terminal electron-transfer complex of respiration.
Proceedings of the National Academy of Sciences, 92(26):11949{11951, 1995.
1
[14] Harry B Gray and Jay R Winkler. Electron
ow through proteins. Chemical
physics letters, 483(1):1{9, 2009. 1
[15] Peter Mitchell. Coupling of phosphorylation to electron and hydrogen trans-
fer by a chemi-osmotic type of mechanism. Nature, 191(4784):144{148, 1961.
1, 9
[16] William B Whitman, David C Coleman, and William J Wiebe. Prokaryotes:
the unseen majority. Proceedings of the National Academy of Sciences, 95
(12):6578{6583, 1998. 2, 8
[17] Pierre Amato. Clouds provide atmospheric oases for microbes. Issues, 2010.
2
[18] Frank Reith. Life in the deep subsurface. Geology, 39(3):287{288, 2011. 2
[19] Francis H Chapelle, Kathleen O'Neill, Paul M Bradley, Barbara A Meth e,
Stacy A Ciufo, LeRoy L Knobel, and Derek R Lovley. A hydrogen-based
subsurface microbial community dominated by methanogens. Nature, 415
(6869):312{315, 2002. 2, 66
[20] Yun-Juan Chang, Aaron D Peacock, Philip E Long, John R Stephen,
James P McKinley, Sarah J Macnaughton, AKM Anwar Hussain, Arnold M
Saxton, and David C White. Diversity and characterization of sulfate-
reducing bacteria in groundwater at a uranium mill tailings site. Applied
and Environmental Microbiology, 67(7):3149{3160, 2001.
94
[21] Thomas M McCollom. Methanogenesis as a potential source of chemical en-
ergy for primary biomass production by autotrophic organisms in hydrother-
mal systems on europa. Journal of Geophysical Research: Planets, 104(E12):
30729{30742, 1999.
[22] Karsten Pedersen. The hydrogen driven intra-terrestrial biosphere and its
in
uence on the hydrochemical conditions in crystalline bedrock aquifers. In
Hydrogeology of Crystalline Rocks, pages 249{259. Springer, 2000.
[23] Norman K Fry, James K Fredrickson, Susan Fishbain, Michael Wagner, and
David A Stahl. Population structure of microbial communities associated
with two deep, anaerobic, alkaline aquifers. Applied and Environmental Mi-
crobiology, 63(4):1498{1504, 1997. 2
[24] Derek R Lovley and Francis H Chapelle. Deep subsurface microbial processes.
Reviews of Geophysics, 33(3):365{381, 1995. 2, 5, 66
[25] Wolfgang Bach and Katrina J Edwards. Iron and sulde oxidation within
the basaltic ocean crust: implications for chemolithoautotrophic microbial
biomass production. Geochimica et Cosmochimica Acta, 67(20):3871{3887,
2003.
[26] Katrina J Edwards, Wolfgang Bach, and Thomas M McCollom. Geomi-
crobiology in oceanography: microbe{mineral interactions at and below the
sea
oor. TRENDS in Microbiology, 13(9):449{456, 2005. 78
[27] Beth N Orcutt, Jason B Sylvan, Nina J Knab, and Katrina J Edwards.
Microbial ecology of the dark ocean above, at, and below the sea
oor. Mi-
crobiology and Molecular Biology Reviews, 75(2):361{422, 2011. 15
[28] James K Fredrickson and John M Zachara. Electron transfer at the microbe{
mineral interface: a grand challenge in biogeochemistry. Geobiology, 6(3):
245{253, 2008.
[29] Gordon Southam. Minerals as substrates for life: the prokaryotic view. Ele-
ments, 8(2):101{106, 2012. 15
[30] Jan P Amend and Andreas Teske. Expanding frontiers in deep subsurface mi-
crobiology. Palaeogeography, Palaeoclimatology, Palaeoecology, 219(1):131{
155, 2005. 2, 5, 66, 78
[31] Karrie A Weber, Laurie A Achenbach, and John D Coates. Microorganisms
pumping iron: anaerobic microbial iron oxidation and reduction. Nature
Reviews Microbiology, 4(10):752{764, 2006. 2
95
[32] Kenneth H. Nealson, Andrea Belz, and Brent McKee. Breathing metals
as a way of life: geobiology in action. Antonie van Leeuwenhoek, 81(1):
215{222. ISSN 1572-9699. doi: 10.1023/A:1020518818647. URL http:
//dx.doi.org/10.1023/A:1020518818647. 2, 15
[33] Derek R Lovley. Dissimilatory metal reduction. Annual Reviews in Micro-
biology, 47(1):263{290, 1993. 2
[34] Charles R Myers and Judith M Myers. Localization of cytochromes to the
outer membrane of anaerobically grown Shewanella putrefaciens MR-1. Jour-
nal of Bacteriology, 174(11):3429{3438, 1992. 2, 12, 24
[35] Byoung-Chan Kim, Xinlei Qian, Ching Leang, Maddalena V Coppi, and
Derek R Lovley. Two putative c-type multiheme cytochromes required for
the expression of OmcB, an outer membrane protein essential for optimal Fe
(III) reduction in Geobacter sulfurreducens. Journal of bacteriology, 188(8):
3138{3142, 2006. 2, 12
[36] Dianne K Newman and Roberto Kolter. A role for excreted quinones in
extracellular electron transfer. Nature, 405(6782):94{97, 2000. 3, 12, 24
[37] Enrico Marsili, Daniel B Baron, Indraneel D Shikhare, Dan Coursolle, Jef-
frey A Gralnick, and Daniel R Bond. Shewanella secretes
avins that medi-
ate extracellular electron transfer. Proceedings of the National Academy of
Sciences, 105(10):3968{3973, 2008. 4, 33, 34, 35, 40, 55, 61
[38] Harald Von Canstein, Jun Ogawa, Sakayu Shimizu, and Jonathan R Lloyd.
Secretion of
avins by Shewanella species and their role in extracellular elec-
tron transfer. Applied and Environmental Microbiology, 74(3):615{623, 2008.
33, 55
[39] Akihiro Okamoto, Kazuhito Hashimoto, Kenneth H Nealson, and Ryuhei
Nakamura. Rate enhancement of bacterial extracellular electron transport
involves bound
avin semiquinones. Proceedings of the National Academy of
Sciences, 110(19):7856{7861, 2013. 4, 23, 24, 34, 35, 40, 55, 83
[40] Akihiro Okamoto, Koichiro Saito, Kengo Inoue, Kenneth H Nealson,
Kazuhito Hashimoto, and Ryuhei Nakamura. Uptake of self-secreted
avins
as bound cofactors for extracellular electron transfer in Geobacter species.
Energy Environ. Sci., 7(4):1357{1361, 2014. 3, 12
[41] Yuri A Gorby, Svetlana Yanina, Jerey S McLean, Kevin M Rosso, Dianne
Moyles, Alice Dohnalkova, Terry J Beveridge, In Seop Chang, Byung Hong
Kim, Kyung Shik Kim, et al. Electrically conductive bacterial nanowires
produced by Shewanella oneidensis strain MR-1 and other microorganisms.
96
Proceedings of the National Academy of Sciences, 103(30):11358{11363, 2006.
3, 12, 24, 25, 30, 35, 36, 56
[42] Mohamed Y El-Naggar, Greg Wanger, Kar Man Leung, Thomas D Yuzvin-
sky, Gordon Southam, Jun Yang, Woon Ming Lau, Kenneth H Nealson,
and Yuri A Gorby. Electrical transport along bacterial nanowires from She-
wanella oneidensis MR-1. Proceedings of the National Academy of Sciences,
107(42):18127{18131, 2010. 3, 24, 26
[43] Gemma Reguera, Kevin D McCarthy, Teena Mehta, Julie S Nicoll, Mark T
Tuominen, and Derek R Lovley. Extracellular electron transfer via microbial
nanowires. Nature, 435(7045):1098{1101, 2005. 3, 12, 25
[44] Gemma Reguera, Kelly P Nevin, Julie S Nicoll, Sean F Covalla, Trevor L
Woodard, and Derek R Lovley. Biolm and nanowire production leads to
increased current in Geobacter sulfurreducens fuel cells. Applied and envi-
ronmental microbiology, 72(11):7345{7348, 2006. 3
[45] Sahand Pirbadian, Sarah E Barchinger, Kar Man Leung, Hye Suk Byun,
Yamini Jangir, Rachida A Bouhenni, Samantha B Reed, Margaret F Romine,
Daad A Saarini, Liang Shi, et al. Shewanella oneidensis mr-1 nanowires
are outer membrane and periplasmic extensions of the extracellular electron
transport components. Proceedings of the National Academy of Sciences, 111
(35):12883{12888, 2014. 3, 24, 29, 30, 31
[46] ME Hernandez and DK Newman. Extracellular electron transfer. Cellular
and Molecular Life Sciences CMLS, 58(11):1562{1571, 2001. 4
[47] Marcus J Edwards, Gaye F White, Michael Norman, Alice Tome-Fernandez,
Emma Ainsworth, Liang Shi, Jim K Fredrickson, John M Zachara, Julea N
Butt, David J Richardson, et al. Redox linked
avin sites in extracellular
decaheme proteins involved in microbe-mineral electron transfer. Scientic
reports, 5, 2015. 4, 34
[48] Akihiro Okamoto, Shafeer Kalathil, Xiao Deng, Kazuhito Hashimoto, Ryuhei
Nakamura, and Kenneth H Nealson. Cell-secreted
avins bound to mem-
brane cytochromes dictate electron transfer reactions to surfaces with diverse
charge and ph. Scientic reports, 4, 2014. 4
[49] Shuxai Xu, Yamini Jangir, and Mohamed Y El-Naggar. Disentangling the
extracellular electron transfer mechanisms of Shewanella oneidensis MR-1
on graphite electrodes. Electrochimica Acta, 2016. 5, 25, 35, 42, 61, 82
[50] DE Holmes, DR Bond, RA Oneil, CE Reimers, LR Tender, and DR Lovley.
Microbial communities associated with electrodes harvesting electricity from
97
a variety of aquatic sediments. Microbial ecology, 48(2):178{190, 2004. 5, 44,
67
[51] Korneel Rabaey and Willy Verstraete. Microbial fuel cells: novel biotechnol-
ogy for energy generation. TRENDS in Biotechnology, 23(6):291{298, 2005.
12
[52] CE Reimers, P Girguis, HA Stecher, LM Tender, N Ryckelynck, and P Whal-
ing. Microbial fuel cell energy from an ocean cold seep. Geobiology, 4(2):
123{136, 2006.
[53] Daniel R Bond. Electrodes as electron acceptors, and the bacteria who love
them. Springer, 2010. 44, 59, 67, 85
[54] Shun'ichi Ishii, Shino Suzuki, Trina M Norden-Krichmar, Angela Wu, Yuko
Yamanaka, Kenneth H Nealson, and Orianna Bretschger. Identifying the
microbial communities and operational conditions for optimized wastewater
treatment in microbial fuel cells. Water research, 47(19):7120{7130, 2013.
[55] Miriam Rosenbaum, Federico Aulenta, Marianna Villano, and Largus T. An-
genent. Cathodes as electron donors for microbial metabolism: Which extra-
cellular electron transfer mechanisms are involved? Bioresource Technology,
102(1):324 { 333, 2011. ISSN 0960-8524. doi: http://dx.doi.org/10.1016/
j.biortech.2010.07.008. URL http://www.sciencedirect.com/science/
article/pii/S096085241001196X. Special Issue: Biofuels - II: Algal Biofu-
els and Microbial Fuel Cells. 5, 17, 67
[56] KC Wrighton, JC Thrash, RA Melnyk, JP Bigi, KG Byrne-Bailey, JP Remis,
D Schichnes, M Auer, CJ Chang, and JD Coates. Evidence for direct electron
transfer by a gram-positive bacterium isolated from a microbial fuel cell.
Applied and environmental microbiology, 77(21):7633{7639, 2011. 5
[57] Viatcheslav Fedorovich, Matthew C Knighton, Eulyn Pagaling, F Bruce
Ward, Andrew Free, and Igor Goryanin. Novel electrochemically active bac-
terium phylogenetically related to Arcobacter butzleri, isolated from a mi-
crobial fuel cell. Applied and environmental microbiology, 75(23):7326{7334,
2009. 16, 45, 61
[58] Nathaniel R Glasser, Suzanne E Kern, and Dianne K Newman. Phenazine
redox cycling enhances anaerobic survival in Pseudomonas aeruginosa by
facilitating generation of ATP and a proton-motive force. Molecular micro-
biology, 92(2):399{412, 2014.
[59] Zarath M Summers, Jerey A Gralnick, and Daniel R Bond. Cultivation
of an obligate fe (II)-oxidizing lithoautotrophic bacterium using electrodes.
98
MBio, 4(1):e00420{12, 2013. URL http://mbio.asm.org/content/4/1/
e00420-12. 17, 23, 67, 88
[60] Arpita Bose, Emily Jeanette Gardel, Charles Vidoudez, EA Parra, and Pe-
ter R Girguis. Electron uptake by iron-oxidizing phototrophic bacteria. Na-
ture communications, 5, 2014. doi: 10.1038/ncomms4391. URLhttp://www.
nature.com/ncomms/2014/140226/ncomms4391/full/ncomms4391. 17, 23,
85
[61] Takumi Ishii, Satoshi Kawaichi, Hirotaka Nakagawa, Kazuhito Hashimoto,
and Ryuhei Nakamura. From chemolithoautotrophs to electrolithoau-
totrophs: Co2 Fixation by Fe(II)-Oxidizing bacteria coupled with direct
uptake of electrons from solid electron sources. Frontiers in Microbiology,
6(994), 2015. ISSN 1664-302X. doi: 10.3389/fmicb.2015.00994. URL
http://www.frontiersin.org/microbiotechnology,_ecotoxicology_
and_bioremediation/10.3389/fmicb.2015.00994/abstract. 17, 79, 88
[62] Annette Ruth Rowe, Prithiviraj Chellamuthu, Bonita Lam, Akihiro
Okamoto, and Kenneth Nealson. Marine sediments microbes capable
of electrode oxidation as a surrogate for lithotrophic insoluble substrate
metabolism. Frontiers in Microbiology, 5(784), 2015. doi: 10.3389/fmicb.
2014.00784. URL http://www.frontiersin.org/microbial_physiology_
and_metabolism/10.3389/fmicb.2014.00784/abstract. 5, 17, 23, 67, 74,
88, 140
[63] Sean McMahon and John Parnell. Weighing the deep continental biosphere.
FEMS microbiology ecology, 87(1):113{120, 2014. 8
[64] Erwin Schr odinger. What Is Life? the Physical Aspect of the Living Cell and
Mind. Dublin, 1943. 8
[65] Jr Peter Jurtshuk. Chapter 4: Bacterial Metabolism. Medical Microbiology,
4th edition, Galveston (TX): University of Texas Medical Branch at Galve-
ston, 1996. 8
[66] Rob Phillips, Jane Kondev, Julie Theriot, and Hernan Garcia. Physical
biology of the cell. Garland Science, 2012. 9
[67] Kristopher A Hunt, Jerey M Flynn, Bel en Naranjo, Indraneel D Shikhare,
and Jerey A Gralnick. Substrate-level phosphorylation is the primary source
of energy conservation during anaerobic respiration of Shewanella oneidensis
strain mr-1. Journal of bacteriology, 192(13):3345{3351, 2010. 9
[68] Ron Milo and Rob Phillips. Cell biology by the numbers. 2014. 10
99
[69] Derek R Lovley. Dissimilatory Fe (III) and Mn (IV) reduction. Microbiolog-
ical reviews, 55(2):259{287, 1991. 12
[70] Kenneth H Nealson and Daad Saarini. Iron and manganese in anaerobic
respiration: environmental signicance, physiology, and regulation. Annual
Reviews in Microbiology, 48(1):311{343, 1994. 12
[71] Charles R Myers and Kenneth H Nealson. Bacterial manganese reduction
and growth with manganese oxide as the sole electron acceptor. Science, 240
(4857):1319{1321, 1988. 12, 46, 75, 139
[72] Charles R Myers and Kenneth H Nealson. Respiration-linked proton translo-
cation coupled to anaerobic reduction of manganese (IV) and iron (III) in
Shewanella putrefaciens MR-1. Journal of Bacteriology, 172(11):6232{6238,
1990. 12
[73] Bruce E Logan. Exoelectrogenic bacteria that power microbial fuel cells.
Nature Reviews Microbiology, 7(5):375{381, 2009. 12
[74] Benjamin J Gross and Mohamed Y El-Naggar. A combined electrochemical
and optical trapping platform for measuring single cell respiration rates at
electrode interfaces. Review of Scientic Instruments, 86(6):064301, 2015.
12, 62
[75] L Th ony-Meyer. Biogenesis of respiratory cytochromes in bacteria. Micro-
biology and Molecular Biology Reviews, 61(3):337{376, 1997. 13
[76] Ivano Bertini, Gabriele Cavallaro, and Antonio Rosato. Cytochrome c:
Occurrence and Functions. Chemical Reviews, 106(1):90{115, 2006. doi:
10.1021/cr050241v. URL http://dx.doi.org/10.1021/cr050241v. PMID:
16402772. 13
[77] Christopher G. Mowat and Stephen K. Chapman. Multi-heme cytochromes-
new structures, new chemistry. Dalton Trans., pages 3381{3389, 2005. doi:
10.1039/B505184C. URL http://dx.doi.org/10.1039/B505184C. 14
[78] Shailesh Sharma, Gabriele Cavallaro, and Antonio Rosato. A systematic
investigation of multiheme c-type cytochromes in prokaryotes. JBIC Jour-
nal of Biological Inorganic Chemistry, 15(4):559{571, 2010. ISSN 1432-
1327. doi: 10.1007/s00775-010-0623-4. URL http://dx.doi.org/10.1007/
s00775-010-0623-4. 14
[79] John F Heidelberg, Ian T Paulsen, Karen E Nelson, Eric J Gaidos, William C
Nelson, Timothy D Read, Jonathan A Eisen, Rekha Seshadri, Naomi Ward,
100
Barbara Methe, et al. Genome sequence of the dissimilatory metal ion{
reducing bacterium Shewanella oneidensis. Nature biotechnology, 20(11):
1118{1123, 2002. 14
[80] B. A. Meth e, K. E. Nelson, J. A. Eisen, I. T. Paulsen, W. Nelson, J. F. Hei-
delberg, D. Wu, M. Wu, N. Ward, M. J. Beanan, R. J. Dodson, R. Madupu,
L. M. Brinkac, S. C. Daugherty, R. T. DeBoy, A. S. Durkin, M. Gwinn,
J. F. Kolonay, S. A. Sullivan, D. H. Haft, J. Selengut, T. M. Davidsen,
N. Zafar, O. White, B. Tran, C. Romero, H. A Forberger, J. Weidman,
H. Khouri, T. V. Feldblyum, T. R. Utterback, S. E. Van Aken, D. R. Lovley,
and C. M. Fraser. Genome of Geobacter sulfurreducens: Metal reduction in
subsurface environments. Science, 302(5652):1967{1969, 2003. ISSN 0036-
8075. doi: 10.1126/science.1088727. URL http://science.sciencemag.
org/content/302/5652/1967. 14
[81] Liang Shi, David J Richardson, Zheming Wang, Sebastien N Kerisit, Kevin M
Rosso, John M Zachara, and James K Fredrickson. The roles of outer mem-
brane cytochromes of Shewanella and Geobacter in extracellular electron
transfer. Environmental Microbiology Reports, 1(4):220{227, 2009. 14
[82] David J Richardson, Julea N Butt, Jim K Fredrickson, John M Zachara,
Liang Shi, Marcus J Edwards, Gaye White, Nanakow Baiden, Andrew J
Gates, Sophie J Marritt, et al. The `porin{cytochrome model for microbe-
to-mineral electron transfer. Molecular microbiology, 85(2):201{212, 2012.
14
[83] Gaye F White, Zhi Shi, Liang Shi, Zheming Wang, Alice C Dohnalkova,
Matthew J Marshall, James K Fredrickson, John M Zachara, Julea N Butt,
David J Richardson, et al. Rapid electron exchange between surface-exposed
bacterial cytochromes and Fe (III) minerals. Proceedings of the National
Academy of Sciences, 110(16):6346{6351, 2013. 14
[84] Orianna Bretschger, C.M. Cheung, Andrea, Florian Mansfeld, and Ken-
nethH. Nealson. Comparative microbial fuel cell evaluations of Shewanella
spp. Electroanalysis, 22(7-8):883{894, 2010. ISSN 1521-4109. doi: 10.1002/
elan.200800016. URL http://dx.doi.org/10.1002/elan.200800016. 16,
49
[85] Daniel R Bond and Derek R Lovley. Electricity production by Geobacter sul-
furreducens attached to electrodes. Applied and environmental microbiology,
69(3):1548{1555, 2003. 16
[86] Kelly C Wrighton, Peter Agbo, Falk Warnecke, Karrie A Weber, Eoin L
Brodie, Todd Z DeSantis, Philip Hugenholtz, Gary L Andersen, and John D
101
Coates. A novel ecological role of the Firmicutes identied in thermophilic
microbial fuel cells. The ISME journal, 2(11):1146{1156, 2008. 16, 45
[87] Yi Zuo, Defeng Xing, John M Regan, and Bruce E Logan. Isolation of the
exoelectrogenic bacterium Ochrobactrum anthropi yz-1 by using a u-tube
microbial fuel cell. Applied and environmental microbiology, 74(10):3130{
3137, 2008.
[88] Defeng Xing, Shaoan Cheng, Bruce E Logan, and John M Regan. Isolation of
the exoelectrogenic denitrifying bacterium Comamonas denitricans based
on dilution to extinction. Applied microbiology and biotechnology, 85(5):
1575{1587, 2010. 16, 45, 59, 86
[89] Juan Pablo Busalmen, Abraham Esteve-Nu~ nez, and Juan Miguel Feliu.
Whole cell electrochemistry of electricity-producing microorganisms evidence
an adaptation for optimal exocellular electron transport. Environmental sci-
ence & technology, 42(7):2445{2450, 2008. doi: 10.1021/es702569y. URL
http://pubs.acs.org/doi/full/10.1021/es702569y. 16
[90] Enrico Marsili, Janet B Rollefson, Daniel B Baron, Raymond M Hozalski,
and Daniel R Bond. Microbial biolm voltammetry: direct electrochemical
characterization of catalytic electrode-attached biolms. Applied and envi-
ronmental microbiology, 74(23):7329{7337, 2008.
[91] Dan Coursolle, Daniel B Baron, Daniel R Bond, and Jerey A Gralnick.
The Mtr respiratory pathway is essential for reducing
avins and electrodes
in Shewanella oneidensis. Journal of bacteriology, 192(2):467{474, 2010. 16,
33
[92] C esar I Torres, Rosa Krajmalnik-Brown, Prathap Parameswaran, An-
drew Kato Marcus, Greg Wanger, Yuri A Gorby, and Bruce E Rittmann.
Selecting anode-respiring bacteria based on anode potential: phylogenetic,
electrochemical, and microscopic characterization. Environmental science
& technology, 43(24):9519{9524, 2009. doi: 10.1021/es902165y. URL
http://pubs.acs.org/doi/full/10.1021/es902165y. 16
[93] Kelvin B. Gregory, Daniel R. Bond, and Derek R. Lovley. Graphite electrodes
as electron donors for anaerobic respiration. Environmental Microbiology, 6
(6):596{604, 2004. ISSN 1462-2920. doi: 10.1111/j.1462-2920.2004.00593.x.
URL http://dx.doi.org/10.1111/j.1462-2920.2004.00593.x. 16, 23
[94] J org S Deutzmann, Merve Sahin, and Alfred M Spormann. Extracellular
enzymes facilitate electron uptake in biocorrosion and bioelectrosynthesis.
MBio, 6(2):e00496{15, 2015. doi: 10.1128/mBio.00496-15. URL http://
mbio.asm.org/content/6/2/e00496-15. 17, 88
102
[95] Philip Hugenholtz et al. Exploring prokaryotic diversity in the genomic era.
Genome Biol, 3(2):1{0003, 2002. 17
[96] Carl R. Woese and George E. Fox. Phylogenetic structure of the prokaryotic
domain: The primary kingdoms. Proceedings of the National Academy of
Sciences, 74(11):5088{5090, 1977. doi: 10.1073/pnas.74.11.5088. URL http:
//www.pnas.org/content/74/11/5088.abstract. 17
[97] Norman R. Pace. A molecular view of microbial diversity and the bio-
sphere. Science, 276(5313):734{740, 1997. ISSN 0036-8075. doi: 10.1126/
science.276.5313.734. URL http://science.sciencemag.org/content/
276/5313/734. 17
[98] Carl R Woese. Bacterial evolution. Microbiological reviews, 51(2):221, 1987.
18
[99] Jon Jonasson, Margaretha Olofsson, and Hans-Jrg Monstein. Classi-
cation, identication and subtyping of bacteria based on pyrosequencing
and signature matching of 16s rdna fragments. APMIS, 110(3):263{272,
2002. ISSN 1600-0463. doi: 10.1034/j.1600-0463.2002.100309.x. URL
http://dx.doi.org/10.1034/j.1600-0463.2002.100309.x. 18
[100] F Sanger, GG Brownlee, and BG Barrell. A two-dimensional fractionation
procedure for radioactive nucleotides. Journal of molecular biology, 13(2):
373{IN4, 1965. 18
[101] Frederick Sanger, Steven Nicklen, and Alan R Coulson. Dna sequencing
with chain-terminating inhibitors. Proceedings of the National Academy of
Sciences, 74(12):5463{5467, 1977.
[102] Frederick Sanger. Sequences, sequences, and sequences. Annual review of
biochemistry, 57(1):1{29, 1988.
[103] Lloyd M Smith, Jane Z Sanders, Robert J Kaiser, Peter Hughes, Chris Dodd,
Charles R Connell, Cheryl Heiner, SB Kent, and Leroy E Hood. Fluorescence
detection in automated dna sequence analysis. Nature, 321(6071):674{679,
1985. 18
[104] Susannah G Tringe and Philip Hugenholtz. A renaissance for the pioneering
16S rRNA gene. Current opinion in microbiology, 11(5):442{446, 2008. 19
[105] J Gregory Caporaso, Christian L Lauber, William A Walters, Donna Berg-
Lyons, Catherine A Lozupone, Peter J Turnbaugh, Noah Fierer, and Rob
Knight. Global patterns of 16S rRNA diversity at a depth of millions of
sequences per sample. Proceedings of the National Academy of Sciences, 108
(Supplement 1):4516{4522, 2011. 19, 73
103
[106] Chengwei Luo, Despina Tsementzi, Nikos Kyrpides, Timothy Read, and
Konstantinos T Konstantinidis. Direct comparisons of Illumina vs. Roche
454 sequencing technologies on the same microbial community DNA sample.
PloS one, 7(2):e30087, 2012. 19
[107] Patrick D Schloss, Sarah L Westcott, Matthew L Jenior, and Sarah K High-
lander. Sequencing 16S rRNA gene fragments using the PacBio SMRT DNA
sequencing system. Technical report, PeerJ PrePrints, 2015. 19
[108] Patrick D Schloss, Dirk Gevers, and Sarah L Westcott. Reducing the eects
of PCR amplication and sequencing artifacts on 16S rRNA-based studies.
PloS one, 6(12):e27310, 2011. 19, 119, 144
[109] James J Kozich, Sarah L Westcott, Nielson T Baxter, Sarah K Highlander,
and Patrick D Schloss. 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
(17):5112{5120, 2013. 19
[110] Robert C Edgar, Brian J Haas, Jose C Clemente, Christopher Quince, and
Rob Knight. UCHIME improves sensitivity and speed of chimera detection.
Bioinformatics, 27(16):2194{2200, 2011. 19, 52, 74
[111] David AW Soergel, Neelendu Dey, Rob Knight, and Steven E Brenner. Se-
lection of primers for optimal taxonomic classication of environmental 16S
rRNA gene sequences. The ISME journal, 6(7):1440{1444, 2012. 19
[112] William A Walters, J Gregory Caporaso, Christian L Lauber, Donna Berg-
Lyons, Noah Fierer, and Rob Knight. PrimerProspector: de novo design and
taxonomic analysis of barcoded polymerase chain reaction primers. Bioin-
formatics, 27(8):1159{1161, 2011. 19
[113] Justin Kuczynski, Christian L Lauber, William A Walters, Laura Wegener
Parfrey, Jos e C Clemente, Dirk Gevers, and Rob Knight. Experimental
and analytical tools for studying the human microbiome. Nature Reviews
Genetics, 13(1):47{58, 2012. 19
[114] Alma E Parada, David M Needham, and Jed A Fuhrman. Every base mat-
ters: assessing small subunit rrna primers for marine microbiomes with mock
communities, time series and global eld samples. Environmental microbiol-
ogy, 2015. 19
[115] Patrick D Schloss, Sarah L Westcott, Thomas Ryabin, Justine R Hall, Mar-
tin Hartmann, Emily B Hollister, Ryan A Lesniewski, Brian B Oakley,
104
Donovan H Parks, Courtney J Robinson, et al. Introducing mothur: open-
source, platform-independent, community-supported software for describing
and comparing microbial communities. Applied and environmental microbi-
ology, 75(23):7537{7541, 2009. 19, 51, 74
[116] J Gregory Caporaso, Justin Kuczynski, Jesse Stombaugh, Kyle Bittinger,
Frederic D Bushman, Elizabeth K Costello, Noah Fierer, Antonio Gonzalez
Pena, Julia K Goodrich, Jerey I Gordon, et al. QIIME allows analysis of
high-throughput community sequencing data. Nature methods, 7(5):335{336,
2010. 19
[117] Robert C Edgar. Search and clustering orders of magnitude faster than
BLAST. Bioinformatics, 26(19):2460{2461, 2010. 20
[118] Qiong Wang, George M Garrity, James M Tiedje, and James R Cole. Na ve
bayesian classier for rapid assignment of rRNA sequences into the new bac-
terial taxonomy. Applied and environmental microbiology, 73(16):5261{5267,
2007. 20
[119] Marcus J Claesson, Orla O'Sullivan, Qiong Wang, Janne Nikkil a, Julian R
Marchesi, Hauke Smidt, Willem M de Vos, R Paul Ross, and Paul W O'Toole.
Comparative analysis of pyrosequencing and a phylogenetic microarray for
exploring microbial community structures in the human distal intestine. PloS
one, 4(8):e6669, 2009. 20
[120] Stephanie A Connon and Stephen J Giovannoni. High-throughput methods
for culturing microorganisms in very-low-nutrient media yield diverse new
marine isolates. Applied and Environmental Microbiology, 68(8):3878{3885,
2002. 20
[121] Beate K opke, Reinhard Wilms, Bert Engelen, Heribert Cypionka, and Hen-
rik Sass. Microbial diversity in coastal subsurface sediments: a cultivation
approach using various electron acceptors and substrate gradients. Applied
and environmental microbiology, 71(12):7819{7830, 2005. 20
[122] David P Clark, P Dunlap, M Madigan, and J Martinko. Brock Biology of
Microorganisms. Scientic Publisher, Beijing, 2009. 20
[123] Joel C Goldman, David A Caron, and Mark R Dennett. Regulation of gross
growth eciency and ammonium regeneration in bacteria by substrate C:N
ratio. Limnol. Oceanogr, 32(6):1239{1252, 1987. 20
[124] Kjell Magne Fagerbakke, Mikal Heldal, and Svein Norland. Content of car-
bon, nitrogen, oxygen, sulfur and phosphorus in native aquatic and cultured
bacteria. Aquatic Microbial Ecology, 10(1):15{27, 1996. 20
105
[125] Dianne Newman and Jared Leadbetter. Microbial diversity course, 2014. 20
[126] Haluk Beyenal and Jerome T Babauta. Biolms in Bioelectrochemical Sys-
tems: From Laboratory Practice to Data Interpretation. John Wiley and
Sons, 2015. 21
[127] E LaBelle and DR Bond. Cyclic voltammetry for the study of microbial elec-
tron transfer at electrodes. Bioelectrochemical systems: From extracellular
electron transfer to biotechnological application. London. IWA Publishing. p,
pages 137{152, 2009. 22
[128] Daniel E Ross, Jerey M Flynn, Daniel B Baron, Jerey A Gralnick, and
Daniel R Bond. Towards electrosynthesis in Shewanella: energetics of re-
versing the Mtr pathway for reductive metabolism. PloS one, 6(2):e16649,
2011. 23
[129] Kyla P Gregoire, Sarah M Glaven, Judson Hervey, Baochuan Lin, and
Leonard M Tender. Enrichment of a high-current density denitrifying mi-
crobial biocathode. Journal of The Electrochemical Society, 161(13):H3049{
H3057, 2014.
[130] Jerome T Babauta, Lewis Hsu, Erhan Atci, Je Kagan, Bart Chadwick, and
Haluk Beyenal. Multiple cathodic reaction mechanisms in seawater cathodic
biolms operating in sediment microbial fuel cells. ChemSusChem, 7(10):
2898{2906, 2014. 23
[131] Akihiro Okamoto, Kazuhito Hashimoto, and Kenneth H Nealson. Flavin
redox bifurcation as a mechanism for controlling the direction of electron

ow during extracellular electron transfer. Angewandte Chemie International
Edition, 53(41):10988{10991, 2014. 23
[132] Nikhil S Malvankar, Madeline Vargas, Kelly P Nevin, Ashley E Franks, Ching
Leang, Byoung-Chan Kim, Kengo Inoue, T unde Mester, Sean F Covalla,
Jessica P Johnson, et al. Tunable metallic-like conductivity in microbial
nanowire networks. nature nanotechnology, 6(9):573{579, 2011. 25
[133] Rachel M Snider, Sarah M Strycharz-Glaven, Stanislav D Tsoi, Jerey S
Erickson, and Leonard M Tender. Long-range electron transport in Geobacter
sulfurreducens biolms is redox gradient-driven. Proceedings of the National
Academy of Sciences, 109(38):15467{15472, 2012. 25
[134] Sarah M Strycharz-Glaven and Leonard M Tender. Reply to the com-
ment on \on electrical conductivity of microbial nanowires and biolms"
'by ns malvankar, mt tuominen and dr lovley, energy environ. sci., 2012, 5,
106
doi: 10.1039/c2ee02613a. Energy & Environmental Science, 5(3):6250{6255,
2012. 25
[135] Sahand Pirbadian and Mohamed Y El-Naggar. Multistep hopping and ex-
tracellular charge transfer in microbial redox chains. Physical Chemistry
Chemical Physics, 14(40):13802{13808, 2012. 26
[136] Nicholas F Polizzi, Spiros S Skourtis, and David N Beratan. Physical con-
straints on charge transport through bacterial nanowires. Faraday discus-
sions, 155:43{61, 2012. 26
[137] William Wickner and Randy Schekman. Protein translocation across biolog-
ical membranes. science, 310(5753):1452{1456, 2005. 26
[138] Philip A Lee, Danielle Tullman-Ercek, and George Georgiou. The bacterial
twin-arginine translocation pathway. Annual review of microbiology, 60:373,
2006.
[139] Paolo Natale, Thomas Br user, and Arnold JM Driessen. Sec-and Tat-
mediated protein secretion across the bacterial cytoplasmic membranedis-
tinct translocases and mechanisms. Biochimica et Biophysica Acta (BBA)-
Biomembranes, 1778(9):1735{1756, 2008.
[140] Tracy Palmer and Ben C Berks. The twin-arginine translocation (Tat) pro-
tein export pathway. Nature Reviews Microbiology, 10(7):483{496, 2012. 26
[141] Margaret F Romine. Genome-wide protein localization prediction strategies
for gram negative bacteria. BMC genomics, 12(1):1, 2011. 26
[142] Qixia Luo, Yangyang Dong, Haijiang Chen, and Haichun Gao. Mislocaliza-
tion of Rieske protein PetA predominantly accounts for the aerobic growth
defect of tat mutants in Shewanella oneidensis. PLoS One, 8(4):e62064, 2013.
26, 27, 28, 29
[143] Ann G Matthysse, Serina Stretton, Catherine Dandie, Nicholas C McClure,
and Amanda E Goodman. Construction of gfp vectors for use in gram-
negative bacteria other than Escherichia coli. FEMS microbiology letters,
145(1):87{94, 1996. 27, 29
[144] Jerey S McLean, Greg Wanger, Yuri A Gorby, Martin Wainstein, Je Mc-
Quaid, Shunichi Ishii, Orianna Bretschger, Haluk Beyenal, and Kenneth H
Nealson. Quantication of electron transfer rates to a solid phase electron
acceptor through the stages of biolm formation from single cells to multi-
cellular communities. Environmental science & technology, 44(7):2721{2727,
2010. 27, 29
107
[145] CR Myers and JM Myers. Replication of plasmids with the p15A origin in
Shewanella putrefaciens MR-1. Letters in applied microbiology, 24(3):221{
225, 1997. 28
[146] Carolina Reyes, Julie N Murphy, and Chad W Saltikov. Mutational and
gene expression analysis of mtrDEF, omcA and mtrCAB during arsenate
and iron reduction in Shewanella sp. ANA-3. Environmental microbiology,
12(7):1878{1888, 2010. 33
[147] Daniel E Ross, Shane S Ruebush, Susan L Brantley, Robert S Hartshorne,
Thomas A Clarke, David J Richardson, and Ming Tien. Characterization of
protein-protein interactions involved in iron reduction by Shewanella onei-
densis MR-1. Applied and environmental microbiology, 73(18):5797{5808,
2007. 33
[148] Daniel E Ross, Susan L Brantley, and Ming Tien. Kinetic characterization of
OmcA and MtrC, terminal reductases involved in respiratory electron trans-
fer for dissimilatory iron reduction in Shewanella oneidensis MR-1. Applied
and environmental microbiology, 75(16):5218{5226, 2009.
[149] Robert S Hartshorne, Catherine L Reardon, Daniel Ross, Jochen Nuester,
Thomas A Clarke, Andrew J Gates, Paul C Mills, Jim K Fredrickson, John M
Zachara, Liang Shi, et al. Characterization of an electron conduit between
bacteria and the extracellular environment. Proceedings of the National
Academy of Sciences, 106(52):22169{22174, 2009. 33
[150] Evan D Brutinel and Jerey A Gralnick. Shuttling happens: soluble
avin
mediators of extracellular electron transfer in Shewanella. Applied microbi-
ology and biotechnology, 93(1):41{48, 2012. 33
[151] Huan Liu, Greg J Newton, Ryuhei Nakamura, Kazuhito Hashimoto, and
Shuji Nakanishi. Electrochemical characterization of a single electricity-
producing bacterial cell of Shewanella by using optical tweezers. Angewandte
Chemie International Edition, 49(37):6596{6599, 2010. 33
[152] Jared N Roy, Soa Babanova, Kristen E Garcia, Jose Cornejo, Linnea K Ista,
and Plamen Atanassov. Catalytic biolm formation by Shewanella oneidensis
MR-1 and anode characterization by expanded uncertainty. Electrochimica
Acta, 126:3{10, 2014. 33, 40
[153] Nicholas J Kotloski and Jerey A Gralnick. Flavin electron shuttles dominate
extracellular electron transfer by Shewanella oneidensis. MBio, 4(1):e00553{
12, 2013. 33
108
[154] Catarina M Paquete, Bruno M Fonseca, Davide R Cruz, Tiago M Pereira,
Isabel Pacheco, Cl audio M Soares, and Ricardo O Louro. Exploring the
molecular mechanisms of electron shuttling across the microbe/metal space.
Frontiers in microbiology, 5, 2014. 33
[155] Thomas A Clarke, Marcus J Edwards, Andrew J Gates, Andrea Hall, Gaye F
White, Justin Bradley, Catherine L Reardon, Liang Shi, Alexander S Beliaev,
Matthew J Marshall, et al. Structure of a bacterial cell surface decaheme
electron conduit. Proceedings of the National Academy of Sciences, 108(23):
9384{9389, 2011. 34
[156] Alessandro A Carmona-Martinez, Falk Harnisch, Lisa A Fitzgerald, Justin C
Binger, Bradley R Ringeisen, and Uwe Schr oder. Cyclic voltammetric anal-
ysis of the electron transfer of Shewanella oneidensis MR-1 and nanolament
and cytochrome knock-out mutants. Bioelectrochemistry, 81(2):74{80, 2011.
34, 40
[157] Mackenzie Firer-Sherwood, G ok ce Su Pulcu, and Sean J Elliott. Electro-
chemical interrogations of the Mtr cytochromes from Shewanella: opening
a potential window. JBIC Journal of Biological Inorganic Chemistry, 13(6):
849{854, 2008. 34
[158] Robert S Hartshorne, Brian N Jepson, Tom A Clarke, Sarah J Field, Jim
Fredrickson, John Zachara, Liang Shi, Julea N Butt, and David J Richard-
son. Characterization of Shewanella oneidensis MtrC: a cell-surface deca-
heme cytochrome involved in respiratory electron transport to extracellular
electron acceptors. JBIC Journal of Biological Inorganic Chemistry, 12(7):
1083{1094, 2007.
[159] Carrick M Eggleston, Janos V or os, Liang Shi, Brian H Lower, Timothy C
Droubay, and Patricia JS Colberg. Binding and direct electrochemistry of
OmcA, an outer-membrane cytochrome from an iron reducing bacterium,
with oxide electrodes: a candidate biofuel cell system. Inorganica Chimica
Acta, 361(3):769{777, 2008.
[160] Leisa A Meitl, Carrick M Eggleston, Patricia JS Colberg, Nidhi Khare,
Catherine L Reardon, and Liang Shi. Electrochemical interaction of She-
wanella oneidensis MR-1 and its outer membrane cytochromes OmcA and
MtrC with hematite electrodes. Geochimica et Cosmochimica Acta, 73(18):
5292{5307, 2009. 34
[161] Anand Jain, Jack O Connolly, Richard Woolley, Satheesh Krishnamurthy,
and Enrico Marsili. Extracellular electron transfer mechanism in Shewanella
109
loihica PV- 4 biolms formed at indium tin oxide and graphite electrodes. In-
ternational Journal of Electrochemical Science, 8:1778{1793, February 2013.
URL http://oro.open.ac.uk/36671/. 34, 35, 40, 42
[162] Anand Jain, Xiaoming Zhang, Gabriele Pastorella, Jack O Connolly, Ni-
amh Barry, Robert Woolley, Satheesh Krishnamurthy, and Enrico Marsili.
Electron transfer mechanism in Shewanella loihica PV-4 biolms formed at
graphite electrode. Bioelectrochemistry, 87:28{32, 2012. 40
[163] Victor Bochuan Wang, Nathan D Kirchhofer, Xiaofen Chen, Melissa Yuan Li
Tan, Krishnakumar Sivakumar, Bin Cao, Qichun Zhang, Staan Kjelleberg,
Guillermo C Bazan, Say Chye Joachim Loo, et al. Comparison of
avins and
a conjugated oligoelectrolyte in stimulating extracellular electron transport
from Shewanella oneidensis MR-1. Electrochemistry Communications, 41:
55{58, 2014. 40
[164] DP Moser. Some new windows into terrestrial deep subsurface microbial
ecosystems. In AGU Fall Meeting Abstracts, volume 1, page 01, 2011. 46
[165] Hiroyuki Imachi, Ken Aoi, Eiji Tasumi, Yumi Saito, Yuko Yamanaka, Yayoi
Saito, Takashi Yamaguchi, Hitoshi Tomaru, Rika Takeuchi, Yuki Morono,
et al. Cultivation of methanogenic community from subsea
oor sediments
using a continuous-
ow bioreactor. The ISME journal, 5(12):1913{1925,
2011. 46
[166] Udo Schwertmann and Rochelle M Cornell. Iron oxides in the laboratory:
preparation and characterization. John Wiley & Sons, 2008. 46, 141
[167] TL Kieft, JK Fredrickson, TC Onstott, YA Gorby, HM Kostandarithes,
TJ Bailey, DW Kennedy, SW Li, AE Plymale, CM Spadoni, et al. Dis-
similatory reduction of Fe (III) and other electron acceptors by a Thermus
isolate. Applied and Environmental Microbiology, 65(3):1214{1221, 1999. 49,
72
[168] Sean Turner, Kathleen M Pryer, Vivian PW Miao, and Jerey D Palmer. In-
vestigating deep phylogenetic relationships among cyanobacteria and plastids
by small subunit rrna sequence analysis. Journal of Eukaryotic Microbiology,
46(4):327{338, 1999. 50
[169] Eric P Nawrocki, Diana L Kolbe, and Sean R Eddy. Infernal 1.0: inference
of RNA alignments. Bioinformatics, 25(10):1335{1337, 2009. 52, 74
[170] DJ Reasoner and EE Geldreich. A new medium for the enumeration and
subculture of bacteria from potable water. Applied and environmental mi-
crobiology, 49(1):1{7, 1985. 52, 75
110
[171] Isaac J Winograd and FJ Pearson. Major carbon 14 anomaly in a regional
carbonate aquifer: Possible evidence for megascale channeling, south central
great basin. Water Resources Research, 12(6):1125{1143, 1976. 54
[172] Wayne R Belcher, MS Bedinger, Jennifer T Back, and Donald S Sweetkind.
Interbasin
ow in the great basin with special reference to the southern
funeral mountains and the source of furnace creek springs, death valley, cal-
ifornia, us. Journal of hydrology, 369(1):30{43, 2009.
[173] Wayne R Belcher and Donald S Sweetkind. Death valley regional groundwa-
ter
ow system, nevada and california-hydrogeologic framework and transient
groundwater
ow model. Technical report, US Geological Survey Professional
Paper 1711, 2010. 54
[174] James M Thomas, Duane P Moser, Jenny C Fisher, Jessica Reihle, Alexandra
Wheatley, Ronald L Hershey, Cristi Baldino, and Darrick Weissen
uh. Using
water chemistry, isotopes and microbiology to evaluate groundwater sources,

ow paths and geochemical reactions in the death valley
ow system, usa.
Procedia Earth and Planetary Science, 7:842{845, 2013. 54
[175] Colin Wardman, Kelly P Nevin, and Derek R Lovley. Real-time monitoring
of subsurface microbial metabolism with graphite electrodes. Frontiers in
microbiology, 5, 2014. 55
[176] Rudolf K Thauer, Kurt Jungermann, and Karl Decker. Energy conservation
in chemotrophic anaerobic bacteria. Bacteriological reviews, 41(1):100, 1977.
58
[177] Maneesha P Ginige, J urg Keller, and Linda L Blackall. Investiga-
tion of an acetate-fed denitrifying microbial community by stable iso-
tope probing, full-cycle rrna analysis, and
uorescent in situ hybridization-
microautoradiography. Applied and environmental microbiology, 71(12):
8683{8691, 2005. 58
[178] Ruud A Timmers, Michael Rothballer, David PBTB Strik, Marion Engel,
Stephan Schulz, Michael Schloter, Anton Hartmann, Bert Hamelers, and
Cees Buisman. Microbial community structure elucidates performance of
Glyceria maxima plant microbial fuel cell. Applied microbiology and biotech-
nology, 94(2):537{548, 2012. 58
[179] Wenjie Sun, Reyes Sierra-Alvarez, Lily Milner, and Jim A Field. Anaerobic
oxidation of arsenite linked to chlorate reduction. Applied and environmental
microbiology, 76(20):6804{6811, 2010. 58
111
[180] Tomoyuki Hori, Alexandra M uller, Yasuo Igarashi, Ralf Conrad, and
Michael W Friedrich. Identication of iron-reducing microorganisms in
anoxic rice paddy soil by 13C-acetate probing. The ISME journal, 4(2):
267{278, 2010. 58
[181] Hamid Rismani-Yazdi, Ann D Christy, Burk A Dehority, Mark Morrison,
Zhongtang Yu, and Olli H Tuovinen. Electricity generation from cellulose
by rumen microorganisms in microbial fuel cells. Biotechnology and bioengi-
neering, 97(6):1398{1407, 2007. 59
[182] Abhijeet P Borole, Choo Y Hamilton, Tatiana Vishnivetskaya, David Leak,
and Calin Andras. Improving power production in acetate-fed microbial fuel
cells via enrichment of exoelectrogenic organisms in
ow-through systems.
Biochemical Engineering Journal, 48(1):71{80, 2009. 59
[183] Swades K Chaudhuri and Derek R Lovley. Electricity generation by direct
oxidation of glucose in mediatorless microbial fuel cells. Nature biotechnology,
21(10):1229{1232, 2003. 59, 61
[184] Shunichi Ishii, Bruce E Logan, and Yuji Sekiguchi. Enhanced electrode-
reducing rate during the enrichment process in an air-cathode microbial fuel
cell. Applied microbiology and biotechnology, 94(4):1087{1094, 2012. 59, 86
[185] Chih-Yu Chen, Tzu-Yu Chen, and Ying-Chien Chung. A comparison of
bioelectricity in microbial fuel cells with aerobic and anaerobic anodes. En-
vironmental technology, 35(3):286{293, 2014. 59
[186] Ron Milo and Rob Phillips. Cell Biology by the Numbers. Garland Science,
2015. 62
[187] Frank Reith, Lintern Fairbrother, Gert Nolze, Oliver Wilhelmi, Peta L Clode,
Adrienne Gregg, John E Parsons, Steven A Wakelin, Allan Pring, Robert
Hough, et al. Nanoparticle factories: Biolms hold the key to gold dispersion
and nugget formation. Geology, 38(9):843{846, 2010. 63
[188] Chad W Johnston, Morgan A Wyatt, Xiang Li, Ashraf Ibrahim, Jeremiah
Shuster, Gordon Southam, and Nathan A Magarvey. Gold biomineralization
by a metallophore from a gold-associated microbe. Nature chemical biology,
9(4):241{243, 2013. 63
[189] Todd O Stevens and James P McKinley. Lithoautotrophic microbial ecosys-
tems in deep basalt aquifers. Science, 270(5235):450, 1995. 66
[190] Robert T Anderson, Francis H Chapelle, and Derek R Lovley. Evidence
against hydrogen-based microbial ecosystems in basalt aquifers. Science,
281(5379):976{977, 1998. 66
112
[191] RY Morita. Is h2 the universal energy source for long-term survival? Micro-
bial ecology, 38(4):307{320, 1999. 66
[192] Karsten Pedersen. Microbial life in deep granitic rock. FEMS Microbiology
Reviews, 20(3-4):399{414, 1997.
[193] Karsten Pedersen. Exploration of deep intraterrestrial microbial life: current
perspectives. FEMS microbiology letters, 185(1):9{16, 2000. 66
[194] Stanton W Caddey and Richard L Bachman. The hornestake gold mine, an
early proterozoic iron-formation-hosted gold deposit, lawrence county, south
dakota. 1991. 67, 68
[195] K Hart, TC Trancynger, W Roggenthen, and J Heise. Topographic, geologic,
and density distribution modeling in support of physics experiments at the
sanford underground research facility (SURF). In Proceedings of the South
Dakota Academy of Science, volume 93, page 33, 2014. 67, 68
[196] Larry C Murdoch, Leonid N Germanovich, Herb Wang, TC Onstott, Derek
Elsworth, Larry Stetler, and David Boutt. Hydrogeology of the vicinity of
homestake mine, South Dakota, USA. Hydrogeology Journal, 20(1):27{43,
2012. 67
[197] Magdalena R Osburn, Douglas E LaRowe, Lily M Momper, and Jan P
Amend. Chemolithotrophy in the continental deep subsurface: Sanford un-
derground research facility (SURF), usa. Front. Microbiol., 5(610), 2014. 67,
68, 76
[198] Heath J Mills, Brandi Kiel Reese, and Cruz St Peter. Characterization of
microbial population shifts during sample storage. Frontiers in Microbiology,
3. 73, 135
[199] Scot E Dowd, Todd R Callaway, Randall D Wolcott, Yan Sun, Trevor McKee-
han, Robert G Hagevoort, and Thomas S Edrington. Evaluation of the bacte-
rial diversity in the feces of cattle using 16s rDNA bacterial tag-encoded FLX
amplicon pyrosequencing (bTEFAP). BMC microbiology, 8(1):125, 2008. 73
[200] Lqtta Hallbeck, Fredrik St ahl, and Karsten Pedersen. phytogeny and phe-
notypic characterization of the stalk-forming and iron-oxidizing bacterium
Gallionella ferruginea. Microbiology, 139(7):1531{1535, 1993. 74, 141
[201] Hans H Hanert. The genus Gallionella. In The prokaryotes, pages 4082{4088.
Springer, 1992. 74, 141
[202] Derek Lovley. Dissimilatory Fe (III) and Mn (IV) reducing prokaryotes. In
The prokaryotes, pages 635{658. Springer, 2006. 75, 86, 142
113
[203] Jan Kuever. The family Desulfobacteraceae. In The Prokaryotes, pages 45{
73. Springer, 2014. 78, 81
[204] Jan Kuever. The family Desulfovibrionaceae. In The Prokaryotes, pages
107{133. Springer, 2014. 79
[205] Jan Kuever. The family Desulfobulbaceae. In The Prokaryotes, pages 75{86.
Springer, 2014. 78
[206] George W Luther, Alyssa J Findlay, Daniel J MacDonald, Shannon M Ow-
ings, Thomas E Hanson, Roxanne A Beinart, and Peter R Girguis. Thermo-
dynamics and kinetics of sulde oxidation by oxygen: a look at inorganically
controlled reactions and biologically mediated processes in the environment.
2011. 79
[207] Mary J Carmichael, Sarah K Carmichael, Cara M Santelli, Amanda Strom,
and Suzanna L Br auer. Mn (II)-oxidizing bacteria are abundant and environ-
mentally relevant members of ferromanganese deposits in caves of the Upper
Tennessee River Basin. Geomicrobiology Journal, 30(9):779{800, 2013. 79
[208] Bradley M Tebo, Hope A Johnson, James K McCarthy, and Alexis S Tem-
pleton. Geomicrobiology of manganese (II) oxidation. TRENDS in Microbi-
ology, 13(9):421{428, 2005.
[209] Nicky C Caiazza, Douglas P Lies, and Dianne K Newman. Phototrophic fe
(II) oxidation promotes organic carbon acquisition by Rhodobacter capsulatus
SB1003. Applied and environmental microbiology, 73(19):6150{6158, 2007.
[210] Kenneth H Nealson. The manganese-oxidizing bacteria. In The prokaryotes,
pages 222{231. Springer, 2006.
[211] Eileen Gregory and James T Staley. Widespread distribution of ability to
oxidize manganese among freshwater bacteria. Applied and environmental
microbiology, 44(2):509{511, 1982. 81
[212] Sabrina Hedrich, Michael Schl omann, and D Barrie Johnson. The iron-
oxidizing Proteobacteria. Microbiology, 157(6):1551{1564, 2011. 81
[213] David Emerson, Emily J Fleming, and Joyce M McBeth. Iron-oxidizing
bacteria: an environmental and genomic perspective. Annual review of mi-
crobiology, 64:561{583, 2010. 79
[214] John Bowman. The methanotrophsthe families Methylococcaceae and Methy-
locystaceae. In The prokaryotes, pages 266{289. Springer, 2006. 79, 81
114
[215] Christopher Anthony. Biochemistry of methylotrophs. Academic Press, 1982.
81
[216] Marco Bl othe and Eric E Roden. Microbial iron redox cycling in a
circumneutral-ph groundwater seep. Applied and environmental microbiol-
ogy, 75(2):468{473, 2009. 81, 86
[217] Johann Orlygsson and Jakob K Kristjansson. The family Hydrogenophi-
laceae. In The Prokaryotes, pages 859{868. Springer, 2014. 81
[218] Anthony C Greene. The family Desulfuromonadaceae. In The Prokaryotes,
pages 143{155. Springer, 2014. 81
[219] Wilfred FM R oling. The family Geobacteraceae. In The Prokaryotes, pages
157{172. Springer, 2014. 81
[220] Hanno Richter, Kelly P Nevin, Hongfei Jia, Daniel A Lowy, Derek R Lovley,
and Leonard M Tender. Cyclic voltammetry of biolms of wild type and
mutant Geobacter sulfurreducens on fuel cell anodes indicates possible roles
of OmcB, OmcZ, type IV pili, and protons in extracellular electron transfer.
Energy & Environmental Science, 2(5):506{516, 2009. 83
[221] Jonathan P Badalamenti, Zarath M Summers, Chi Ho Chan, Jerey A Gral-
nick, and Daniel R Bond. Isolation and genomic characterization of desul-
furomonas soudanensis wtl, a metal-and electrode-respiring bacterium from
anoxic deep subsurface brine. Frontiers in Microbiology, 7:913, 2016. 85
[222] Yanmei Sun, Jincheng Wei, Peng Liang, and Xia Huang. Microbial commu-
nity analysis in biocathode microbial fuel cells packed with dierent materi-
als. AMB express, 2(1):21, 2012. 86
[223] Vanita Roshan Nimje, Chien-Yen Chen, Chien-Cheng Chen, Jiin-Shuh Jean,
A Satyanarayana Reddy, Cheng-Wei Fan, Kuan-Yeu Pan, Hung-Tsan Liu,
and Jia-Lia Chen. Stable and high energy generation by a strain of Bacillus
subtilis in a microbial fuel cell. Journal of Power Sources, 190(2):258{263,
2009. 86
[224] Zainab Z Ismail and Ali Jwied Jaeel. Sustainable power generation in con-
tinuous
ow microbial fuel cell treating actual wastewater: In
uence of bio-
catalyst type on electricity production. The Scientic World Journal, 2013,
2013.
[225] Jia Tang, Guiqin Yang, Junlin Wen, Zhen Yu, Shungui Zhou, and Zhi Liu.
Bacillus thermophilum sp. nov., isolated from a microbial fuel cell. Archives
of microbiology, 196(9):629{634, 2014. 86
115
[226] Valerie J Watson and Bruce E Logan. Power production in mfcs inoculated
with shewanella oneidensis mr-1 or mixed cultures. Biotechnology and bio-
engineering, 105(3):489{498, 2010. 88
[227] PD Schloss, D Gevers, and SL Westcott. 454 sop, 2015. URL http://www.
mothur.org/wiki/454_SOP. 119, 144
[228] Charles E Grant, Timothy L Bailey, and William Staord Noble. Fimo:
scanning for occurrences of a given motif. Bioinformatics, 27(7):1017{1018,
2011. 126
[229] Y Yu Nancy, James R Wagner, Matthew R Laird, Gabor Melli, S ebastien
Rey, Raymond Lo, Phuong Dao, S Cenk Sahinalp, Martin Ester, Leonard J
Foster, et al. Psortb 3.0: improved protein subcellular localization predic-
tion with rened localization subcategories and predictive capabilities for all
prokaryotes. Bioinformatics, 26(13):1608{1615, 2010. 126
[230] Duane P Moser and Kenneth H Nealson. Growth of the facultative anaerobe
shewanella putrefaciens by elemental sulfur reduction. Applied and environ-
mental microbiology, 62(6):2100{2105, 1996. 140
[231] Marie-Agn es Dillies, Andrea Rau, Julie Aubert, Christelle Hennequet-Antier,
Marine Jeanmougin, Nicolas Servant, C eline Keime, Guillemette Marot,
David Castel, Jordi Estelle, et al. A comprehensive evaluation of normal-
ization methods for illumina high-throughput rna sequencing data analysis.
Briengs in bioinformatics, 14(6):671{683, 2013. 145
116
Appendices
117
Appendix A
Appendix: Isolation and
characterization of
electrochemically active
subsurface Delftia and Azonexus
species
A.1 Ferrozine Assay
The ferrozine assay (originaaly by Stookey (1970), amended by AK in 2003 in
Woods Hole) was used to colorimetrically determine ferrous [Fe(II)] and total iron
[Fe(II) + Fe(III)] content in the enrichments. In this assay, dissolved ferrous
iron, [Fe(II)], reacts with three ferrozine molecules [Na
2
-3-(2-pyridyl)-5,6-bis(4-
bephenylsulfonate)-1,2,3-triazine] to form an intensively purple-colored complex
which can be quantied spectrophotometrically at 562 nm. Dissolved Fe(II) was
assessed by adding 0.1 mL of ltered sample (under anoxic conditions) to 0.9 mL
1 M HCl, in an microcentrifuge tube and mixed. Afer 2 minutes, 0.1 mL of the
mixture was transferred to 0.9 mL 1 M HCl in disposable cuvettes and 1 mL fer-
rozine solution (0.1% ferrozine, w/v, in 50% w/v ammonium acetate) was added.
118
Solutions were mixed, left standing for 10 min and then measured for absorbance
at 562 nm. FeCl
2
standards were prepared (in 1 M HCl) at 1, 2, 3, 4, 5 and 10 mM
concentrations and treated with the same procedure. Total dissolved iron [Fe(II)
+ Fe(III)] was assessed by adding to 0.1 mL of ltered sample (under anoxic con-
ditions) to 0.9 mL HAHC (10% w/v hydroxylamine hydrochloride in 1 M HCl)
and mixing. After 2 minutes, 0.1 mL of the mixture was transferred to 0.9 mL 1
M HCl in disposable cuvettes and 1 mL ferrozine solution was added. Solutions
were mixed, left standing for 10 min and then measured for absorbance at 562
nm. Where necessary, samples were diluted with sterile media to fall within the
absorbance range covered by the standard curve. 1 M HCl was used to blank the
iron readings on the spectrophotometer.
A.2 Microbial community
Four sequence les (.s) were obtained for 4 dierent batches of DNA extracts.
They included: 1) WE1, WE2, planktonic cells from rst electrochemical reactor
(run1), 2) WE3, WE4, DHS from rst electrochemical reactor (run1), 3) batch
culture , 4) WE1, WE2, WE3, WE4, planktonic cells from second electrochemical
reactor (run2). Sequences used for tagged sequencing of 16S rRNA genes from the
samples are provided in Table A.2. The standard operating procedure (SOP) was
followed as outlined by Schloss et al., [108, 227] geared towards current data, by
merging all the les after initial trimming using quality scores with parameters
as maxambig=0, maxhomop=8, qaverage=20, minlength=200, maxlength=280.
Next, unique sequences were obtained to simplify the dataset. Sequences were
aligned with RDP infernal aligner. The aligned sequences were screened to ensure
that all of the sequences overlap in the same alignment space by setting a hard
119
start (471) and end (1470) positions. Next, the screened sequences were ltered
to only overlap in the same region and to remove any columns in the alignment
that do not contain data. Unique sequences from the resulting dataset were ac-
quired. To reduce sequencing error preclustering was performed to merge sequence
counts that are within 2 bp of a more abundant sequence (1 bp per 100 bp of se-
quence length). Chimeric sequences were detected using UCHIME and removed.
Sequences were classied using mothur-formatted version of the RDP training set.
NDW2
Proteobacteria 77.51
Nitrospirae 10.54
unclassied 2.68
Firmicutes 2.45
Chloro
exi 2.33
Spirochaetes 1.24
OP1 0.77
Acidobacteria 0.49
OP3 0.46
Planctomycetes 0.39
NC10 0.31
SC4 0.19
Elusimicrobia 0.13
Thermi 0.11
Bacteroidetes 0.11
Synergistetes 0.08
AC1 0.08
OD1 0.04
GAL15 0.04
Verrucomicrobia 0.02
Gemmatimonadetes 0.02
Fusobacteria 0.01
Table A.1: Normalised abundances of bacterial phyla in groundwater from Nevares
Deep Well 2 (Death Valley, CA, USA).
120
Lineages classied as `Chloroplast', `Mitochondria', `Archaea', `Eukaryota' or `un-
known' were removed. OTU's were clustered after creating a distance matrix with
cuto as 0.15. Sub-sample (least number of sequences amongst all) from all the
samples was taken for further analysis. Finally, each OTU was assigned a taxon-
omy with 0.03 distance levels. This output le was used for further analysis. The
dierent OTU's obtained each samples were normalised across each sample.
The microbial community shift was observed - from initial NDW2 groundwater,
the DHS euent, rst electrochemical reactor run mimicking DHS conditions - to
batch enrichments for iron reduction, the nal electrochemical reactor run. The
percentage of sequences observed at the phyla level in NDW2 groundwater are
provided in Table A.1. Detailed bacterial community in each enrichments is avail-
able in Table A.3 and further visualised in Figure A.1. With successive enrichment
Sample ID
Barcode
Sequence
DHS GATGCATC
WE1 run1 GAGTCAGA
WE2 run1 GAGTCTCA
WE3 run1 GATGAGGT
WE4 run1 GATGCAAG
Plank run1 GAGTCTGT
Fe NTA 10 08 2013 GAGAGTGT
Fe NTA 09 09 2014 GAGATCAG
Plank run2 GATCCTTC
WE1 run2 GATCAGGA
WE2 run2 GATCCAAC
WE3 run2 GATCCATG
WE4 run2 GATCCTAG
Table A.2: Sequences used for tagged sequencing of 16S rRNA genes from the
samples. Linker primer sequence was GTGCCAGCMGCCGCGGTAA
121
Label NDW2 site DHS bioreactor Run1-Planktonic cells Run1-WE1: 272 mV Run1-WE2: 372 mV Run1-WE3: 472 mV Run1-WE4: 572 mV
Acidobacteria Acidobacteria Gp3 family incertae sedis - - - - - - -
Actinobacteria Micrococcaeae - 0.056287 - 0.081915 0.075704 0.075269 0.068939
Actinobacteria Nocardiaceae - 0.010379 - 0.158290 0.160920 0.247312 0.049354
Actinobacteria Nocardioidaceae - - - 0.030471 0.037654 0.017921 0.009009
Armatimonadetes Armatimonadetes gp5 family incetae sedis - - - - - 0.005974 -
Bacteroidetes Flavobacteriaceae - - 0.007053 - - - 0.016060
Bacteroidetes Chitinophagaceae - 0.007984 - - - - -
Bacteroidetes Cytophagaceae - 0.009980 - - - - -
Bacteroidetes Sphingobacteriaceae - 0.031537 - - - - -
Chlorobi Ignavibacteriaceae - - - - - - -
Alphaproteobacteria Caulobacteraceae - 0.030339 0.020376 0.029284 0.027745 0.032258 0.067372
Alphaproteobacteria Bradyrhizobiaceae - 0.027545 0.026646 0.080728 0.095918 0.072879 0.092440
Alphaproteobacteria Hyphomicrobiaceae - - - - - - -
Alphaproteobacteria Methylobacteriaceae - - - - - - -
Alphaproteobacteria Phyllobacteriaceae - - 0.010972 0.017016 0.011494 0.022700 0.052879
Alphaproteobacteria Rhizobiaceae - 0.023154 - - - - -
Alphaproteobacteria Sphingomonadaceae - 0.512974 0.051724 0.137317 0.165279 0.166866 0.087740
Alphaproteobacteria Unclassied - 0.018363 - - - - -
Alphaproteobacteria 0.003155 - - - - - -
Betaproteobacteria Comamonadaceae - 0.014770 0.861677 0.379106 0.331748 0.271605 0.471994
Betaproteobacteria Oxalobacteraceae - 0.139321 - 0.007123 0.010702 - 0.012143
Betaproteobacteria Burkholderiales Unclassied - - - - - - -
Betaproteobacteria Rhodocyclaceae - 0.017565 - 0.010289 0.013476 - 0.007442
Betaproteobacteria Unclassied - - - - - - -
Betaproteobacteria 0.001841 - - - - - -
Gammaproteobacteria Xanthomonadaceae - - - - - 0.027479 0.009792
Deltaproteobacteria 0.760360 - - - - - -
Gammaproteobacteria 0.009518 - - - - - -
Unclassied - - - - - - -
Others 0.225126 0.099800 0.021552 0.068461 0.069362 0.059737 0.054837
Label Batch Culture 10/08/2013 Batch Culture 09/09/2014 Run2-Planktonic cells Run2-WE1: 272 mV Run2-WE2: 372 mV Run2-WE3: 472 mV Run2-WE4: 572 mV
Acidobacteria Acidobacteria Gp3 family incertae sedis 0.019600 - - - - - -
Actinobacteria Micrococcaeae - - - - - - -
Actinobacteria Nocardiaceae 0.132889 0.528458 - - - - -
Actinobacteria Nocardioidaceae - 0.016996 - - - - -
Armatimonadetes Armatimonadetes gp5 family incetae sedis - - - - - - -
Bacteroidetes Flavobacteriaceae 0.036064 0.005929 - - - - -
Bacteroidetes Chitinophagaceae - - - - - - -
Bacteroidetes Cytophagaceae - - - - - - -
Bacteroidetes Sphingobacteriaceae - - - - - - -
Chlorobi Ignavibacteriaceae - 0.010277 - - - - -
Alphaproteobacteria Caulobacteraceae - - - - - - -
Alphaproteobacteria Bradyrhizobiaceae 0.010192 0.094862 - 0.005850 - - 0.008970
Alphaproteobacteria Hyphomicrobiaceae 0.024696 - - - - - -
Alphaproteobacteria Methylobacteriaceae - 0.011462 - - - - -
Alphaproteobacteria Phyllobacteriaceae - 0.024111 - - - - -
Alphaproteobacteria Rhizobiaceae - - - - - - -
Alphaproteobacteria Sphingomonadaceae 0.037632 - 0.028963 0.017161 0.018316 0.008974 0.021061
Alphaproteobacteria Unclassied - - - - - - -
Alphaproteobacteria - - - - - - -
Betaproteobacteria Comamonadaceae 0.605253 0.094071 0.641096 0.597894 0.525721 0.499805 0.480109
Betaproteobacteria Oxalobacteraceae - - - - - - -
Betaproteobacteria Burkholderiales Unclassied - - 0.037965 0.021061 0.019486 0.014436 0.019111
Betaproteobacteria Rhodocyclaceae 0.086241 0.113043 0.244618 0.284711 0.360873 0.422552 0.396256
Betaproteobacteria Unclassied - - 0.015656 0.021451 0.016758 0.015217 0.020671
Betaproteobacteria - - - - - - -
Gammaproteobacteria Xanthomonadaceae - - - - - - -
Deltaproteobacteria - - - - - - -
Gammaproteobacteria - - - - - - -
Unclassied - 0.040316 0.007828 0.030031 0.030008 0.013656 0.026131
Others 0.047432 0.060474 0.023875 0.021841 0.028839 0.025361 0.027691
Table A.3: Family-level bacterial lineages identied in NDW2 groundwater and
following enrichments. Down-
ow Hanging Sponge reactor (DHS), purging with
80:20 v:v H
2
:CO
2
, targeted autotrophic Fe-and Mn-reducing microbes. The rst
electrochemical reactor run (Run1) mimicked the conditions of the DHS reactor.
Further enrichments was focused on isolating Fe-reducers on acetate as electron
donor and Fe(III)-NTA as electron acceptor. The two batch bacterial community
were obtained from DNA extracts performed on two dierent dates from the same
sample. This batch culture was the inoculum for the second electrochemical reactor
with dened medium.
122
Figure A.1: Family-level bacterial lineages of NDW2 groundwater and following
enrichments. Down-
ow Hanging Sponge reactor (DHS), purging with 80:20 v:v
H
2
:CO
2
, targeted autotrophic Fe- and Mn-reducing microbes. The rst electro-
chemical reactor run (Run1) mimicked the conditions of the DHS reactor. Further
enrichments was focused on isolating Fe-reducers on acetate as electron donor and
Fe(III)-NTA as electron acceptor. The two batch bacterial community were ob-
tained from DNA extracts performed on two dierent dates from the same sample.
This batch culture was the inoculum for the second electrochemical reactor with
dened medium.
123
Figure A.2: Chronoamperometry (center) of Delftia sp. WE1-13, Azonexus sp.
WE2-4 and D. acidovorans (DSMZ DSM-39) shows production of current in hun-
dreds of nA range for all. Current production by Delftia sp. WE2-4 and D.
acidovorans began declining at approximately 20 hrs, while Azonexus sp. WE2-4
produced a steady current for 90 hours before decreasing. In all the cases work-
ing electrode poised at 522 mV vs. SHE, acetate (10 mM) as the electron donor
and reactors were purged with N
2
to maintain anaerobic conditions. The current
production was correlated with a constant planktonic cell density (left) and slight
increase in total protein content (right) in case of D. acidovorans. The protein
content values include protein measured from poised carbon cloth (CC) associated
cells and planktonic cells (Plank)
.
for electrochemically active microbes, sequences matching Rhodocyclaceae and Co-
mamonadaceae present in NDW2 groundwater with relative abundance <0.002%
became dominant on the electrodes.
A.3 Delftia acidovorans
Delftia acidovorans strain was obtained from DSMZ (DSM no. 39) and was tested
for electrochemical activity in a three electrode set-up with working electrode
poised at +522 mV vs. SHE (similar to each isolate as described in 4.2.4). Once
abiotic current is stabilised to 40 nA, aerobically grown cells resuspended in
124
electron donor free media were introduced in the N
2
purged reactor resulting in
the current development. The current stabilised to a nal value of 590 nA in
15 hours. The current production corresponded to modest increase in the total
protein content from 4.642 0.084 mg to 6.654 0.084 mg, with a constant plank-
tonic cell density A.2, indicating D. acidovorans ability to gain limited energy from
external redox sources.
SeqID Final Localization Final Localization Details
gij501157459jrefjWP 012201994.1jsulte dehydrogenase (cytochrome) subunit SorB [Delftia acidovorans] Unknown
gij501157905jrefjWP 012202414.1jMULTISPECIES: iron permease [Proteobacteria] CytoplasmicMembrane
gij501157923jrefjWP 012202430.1jMULTISPECIES: cytochrome C [Delftia] Unknown
gij501158575jrefjWP 012203041.1jnitrate reductase A subunit beta [Delftia acidovorans] CytoplasmicMembrane
gij501158726jrefjWP 012203184.1jthioredoxin [Delftia acidovorans] Cytoplasmic
gij501158983jrefjWP 012203427.1jhypothetical protein [Delftia acidovorans] Periplasmic
gij501159015jrefjWP 012203456.1jthiol oxidoreductase [Delftia acidovorans] Periplasmic
gij501159057jrefjWP 012203495.1jcytochrome C [Delftia acidovorans] Periplasmic
gij501159120jrefjWP 012203556.1jHNH endonuclease [Delftia acidovorans] Cytoplasmic
gij501159157jrefjWP 012203590.1jMULTISPECIES: cytochrome C [Delftia] Periplasmic
gij501159358jrefjWP 012203780.1jcytochrome CBB3 [Delftia acidovorans] Unknown (This protein may have multiple localization sites.)
gij501159631jrefjWP 012204051.1jmembrane protein [Delftia acidovorans] CytoplasmicMembrane
gij501159855jrefjWP 012204272.1j(2Fe-2S) ferredoxin [Delftia acidovorans] Unknown
gij501160005jrefjWP 012204418.1jcytochrome C [Delftia acidovorans] Periplasmic
gij501160143jrefjWP 012204556.1jhypothetical protein [Delftia acidovorans] Unknown
gij501160211jrefjWP 012204624.1jMULTISPECIES: peptidase S41 [Delftia] CytoplasmicMembrane
gij501160213jrefjWP 012204626.1jMULTISPECIES: cytochrome CBB3 [Delftia] Unknown
gij501160297jrefjWP 012204710.1jdihydropyrimidine dehydrogenase subunit B [Delftia acidovorans] Cytoplasmic
gij501160303jrefjWP 012204716.1jMULTISPECIES: alcohol dehydrogenase [Delftia] Cytoplasmic
gij501160327jrefjWP 012204740.1jcytochrome C peroxidase [Delftia acidovorans] Periplasmic
gij501160416jrefjWP 012204829.1jcytochrome C553 [Delftia acidovorans] Periplasmic
gij501160417jrefjWP 012204830.1jalcohol dehydrogenase [Delftia acidovorans] CytoplasmicMembrane
gij501160594jrefjWP 012205007.1jgluconate 2-dehydrogenase [Delftia acidovorans] CytoplasmicMembrane
gij501160777jrefjWP 012205190.1jalcohol dehydrogenase [Delftia acidovorans] Periplasmic
gij501160814jrefjWP 012205227.1jDNA-directed DNA polymerase [Delftia acidovorans] Unknown (This protein may have multiple localization sites.)
gij501161016jrefjWP 012205429.1jADP-ribose pyrophosphatase [Delftia acidovorans] Cytoplasmic
gij501161349jrefjWP 012205754.1jcytochrome C [Delftia acidovorans] CytoplasmicMembrane
gij501161414jrefjWP 012205819.1j2Fe-2S type ferredoxin [Delftia acidovorans] Unknown
gij501161646jrefjWP 012206039.1jcytochrome C [Delftia acidovorans] Periplasmic
gij501161877jrefjWP 012206260.1jMULTISPECIES: Fur family transcriptional regulator [Delftia] Cytoplasmic
gij501161964jrefjWP 012206341.1jthioredoxin reductase [Delftia acidovorans] Unknown
gij501162343jrefjWP 012206700.1jMULTISPECIES: recombinase RecR [Delftia] Cytoplasmic
gij501162603jrefjWP 012206949.1jmembrane protein [Delftia acidovorans] CytoplasmicMembrane
gij501162689jrefjWP 012207031.1jmolecular chaperone DnaJ [Delftia acidovorans] Cytoplasmic
gij501162941jrefjWP 012207262.1jFAD-linked oxidase [Delftia acidovorans] Cytoplasmic
gij501162978jrefjWP 012207299.1jMULTISPECIES: cytochrome C [Delftia] CytoplasmicMembrane
gij501163098jrefjWP 012207418.1jcytochrome C [Delftia acidovorans] CytoplasmicMembrane
gij501163332jrefjWP 012207637.1jzinc-binding dehydrogenase [Delftia acidovorans] Cytoplasmic
gij501163362jrefjWP 012207664.1jhypothetical protein [Delftia acidovorans] CytoplasmicMembrane
gij501163454jrefjWP 012207749.1jexcinuclease ABC subunit A [Delftia acidovorans] Unknown (This protein may have multiple localization sites.)
gij512584830jrefjWP 016450967.1jhypothetical protein [Delftia acidovorans] Periplasmic
gij760098114jrefjWP 043780359.1jcytochrome C [Delftia acidovorans] Periplasmic
gij760098469jrefjWP 043780713.1jMULTISPECIES: hypothetical protein [Delftia] Periplasmic
gij760098718jrefjWP 043780962.1jcytochrome C oxidase subunit II [Delftia acidovorans] CytoplasmicMembrane
gij760098799jrefjWP 043781042.1jmethionyl-tRNA synthetase [Delftia acidovorans] Cytoplasmic
gij760099198jrefjWP 043781436.1jcytochrome C' [Delftia acidovorans] Periplasmic
gij760099280jrefjWP 043781517.1jhypothetical protein [Delftia acidovorans] Unknown
gij760099814jrefjWP 043782045.1jFur family transcriptional regulator [Delftia acidovorans] Unknown
gij760100005jrefjWP 043782235.1jcytochrome C [Delftia acidovorans] Periplasmic
gij760100041jrefjWP 043782271.1jcytochrome C [Delftia acidovorans] Unknown
gij760100751jrefjWP 043782977.1jprotein-disulde isomerase [Delftia acidovorans] Periplasmic
gij915335191jrefjWP 050771846.1jhypothetical protein [Delftia acidovorans] Unknown
gij16128962jrefjNP 415516.1jtrimethylamine N-oxide (TMAO) reductase I cytochrome c-type subunit [Escherichia coli str. K-12 substr. MG1655] CytoplasmicMembrane
gij16129826jrefjNP 416387.1jTMAO reductase III (TorYZ) cytochrome c-type subunit [Escherichia coli str. K-12 substr. MG1655] Unknown (This protein may have multiple localization sites.)
gij16131390jrefjNP 417975.1jputative cytochrome C peroxidase [Escherichia coli str. K-12 substr. MG1655] Periplasmic
gij16131896jrefjNP 418494.1jnitrite reductase formate-dependent cytochrome [Escherichia coli str. K-12 substr. MG1655] Periplasmic
gij90111403jrefjNP 416707.4jnitrate reductase small cytochrome C550 subunit periplasmic [Escherichia coli str. K-12 substr. MG1655] Periplasmic
gij90111682jrefjNP 418495.2jnitrite reductase formate-dependent penta-heme cytochrome c [Escherichia coli str. K-12 substr. MG1655]" Periplasmic
NP 717386.1jextracelllular iron oxide respiratory system periplasmic decaheme cytochrome c component MtrA [Shewanella oneidensis MR-1] Unknown (This protein may have multiple localization sites.)
NP 717385.1jextracellular iron oxide respiratory system outer membrane component MtrB [Shewanella oneidensis MR-1] OuterMembrane
NP 717388.1jextracelllular iron oxide respiratory system surface decaheme cytochrome c component OmcA [Shewanella oneidensis MR-1] Unknown
NP 720107.1jmembrane anchored tetraheme cytochrome c CymA [Shewanella oneidensis MR-1] CytoplasmicMembrane
Table A.4: Heme-binding sites Cys-X1-X2-Cys-His (CXXCH) detected in D. aci-
dovorans SPH-1 and further putative localisation as predicted by PSORTB. These
localisation are compared with some proteins containing heme-binding sites in E.
coli str. K-12 substr. MG1655 and S. oneidensis MR-1.
125
Considering above, I performed pattern matching (through FIMO - Find Indi-
vidual Motif Occurences [228]) of the cytochrome c heme-binding motif Cys-X1-
X2-Cys-His (CXXCH) against the genome sequence of Delftia acidovorans SPH-1
(from NCBI GenBank), we identied 17 possible cytochrome c genes. Out of these,
9 sequences contained more than one cytochrome c heme-binding sites as anno-
tated by NCBI database. Further putative localisation of these protein sequences
were analysed through subcellular localisation algorithm (PSORTB [229]) and com-
pared with heme-binding protein sequences in E. coli str. K-12 substr. MG1655
and S. oneidensis MR-1 shown in Table A.4. While there are none decaheme cy-
tochromes in D. acidovorans SPH-1, there are many unknown cytochromes with
unknown localization. This clearly suggests further studies are required to identify
precise EET mechanisms active in D. acidovorans and the isolated organisms in
this study - Delftia sp. WE1-13 and Azonexus sp. WE2-4.
A.4 Control run - without poised eletrodes
A nal control experiment was performed to study the behaviour (the change in
the planktonic cell density and the total protein content) of Delftia sp. WE1-13
in an electron acceptor free environment. In this case, the aerobically grown cells
were introduced in the reactor without any poised electrodes (no electron acceptor)
with constant N
2
purging. Here, the planktonic cell density decreased from 4 ( 2)
10
7
CFU/mL to 2 ( 1) 10
7
CFU/mL with minor increase in protein content
5.107 ( 0.003) mg to 7.350 ( 0.233) mg (with carbon cloth associated biomass
containing 0.462 0.012 mg protein). This is attributed to slight oxygen leakage
(measured dO
2
being 0.11 ppm) in the bioreactor likely being used by Delftia for
survival in this set-up.
126
Appendix B
Appendix: In Situ
Electrochemical Enrichment of
Subsurface Bacteria at Sanford
Underground Research Facility
B.1 In Situ Reactor Deployment and Sampling
Techniques
B.1.1 Electrical Connections
Many precautionary measures were taken before installation of the in situ electro-
chemical reactor at the SURF. Electric shock in case of water leakage was a major
hazard to be avoided. Water leak detector circuit (Figure B.1) was employed to
simultaneously stop the
ow into the in situ reactor (controlled by a
oat switch
- Normally Open) and turn o all the electrical instruments including the tem-
perature controller, the peristaltic pumps, the potentiostat and the laptop. The
experiment ran on SURF-supplied 110 VAC power. Electrically operated uninter-
ruptible power supply (UPS) protected the connected equipments from damaging
surges and spikes that travel over utility. In case of a power outage the unit
127
Figure B.1: Electrical connections in the in situ reactor to avoid any electrical
shock (top). This circuit forced all the electrical instruments to turn o during
electrical outage or a water leakage (bottom right) and had not aect during normal
operations (bottom left).
128
switched to internal battery and started an audible alarm for reminding users to
shut down the system. In case of complete loss of power, all the electrical instru-
ments turned o and
ow stopped into the in situ reactor. Any loose electrical
connections were housed in secondary container to avert electrical shock.
B.1.2 Aqueous Chemistry
All samples were frozen immediately after collection, stored and shipped on dry
ice in a styrofoam box/cooler. Sucient head space was allowed when freezing for
expansion of the sample to avoid breaking the sample container.
Dissolved Organic Carbon and Total Dissolved Nitrogen Sampling
Samples were collected through a combusted GF/F lter directly into pre-
combusted 40mL glass EPA vials with PTFE lined septa, or acid cleaned plastic
(HDPE) bottles. Glass vials were well padded to prevent breakage in shipment.
The samples were sent to DOM Analytical Lab Services (UCSB, CA, USA) for
analysis. A minimum sample volume of 15 mL is required for each analysis.
Anion and Cation Sampling
Water sample for cation (Lithium, Sodium, Ammonium, Potassium) and anion
(Fluoride, Chloride, Nitrite, Nitrate, Sulfate) measurement through Ion Chro-
matography (IC) were taken in 25 mL HDPE bottles. While the anion sample
were unltered, the cation samples were ltered and acidied to nal concentra-
tion of 10 mM HNO
3
.
129
Organics Sampling
Samples can be collected in 25 mL HDPR bottles for organics measurement
through High Performace Liquid Chromatography (HPLC) by ltering the water
and acidifying to a nal concentration of 12.5 mM H
2
SO
4
.
Continuous Monitoring of In Situ Reactor Euent
The temperature, ORP, pH and electrical conductivity of the euent of the in situ
reactor was measured using a waterproof portable logging multiparameter meter
(HI9829, Hanna Instruments, USA). The multiparameter was inserted in a
ow cell
(HI 7698297, Hanna Instruments, USA) and tted downstream the in situ reactor,
after 38 days of incubation. Some data points (days 60-115) were not recorded
due to low battery (Figure B.2). The temperature
uctuated around 19
o
C and
ORP slightly decreased from about 220 mV to 180 mV indicative of reduction in
oxidising power of the euent. The pH and electrical conductivity increased from
7.3 to 7.7 and 9000 to 14000 S/cm. The value of the electrical conductivity lies
between the fresh water (150 to 500 S/cm) and marine water (50000 S/cm).
Microbial activity in the in situ reactor may have caused the increase in the pH
and electrical conductivity of the water.
B.1.3 Microbiological Sampling
Water samples (atleast 2 L) were ltered through 0.22 m sterivex lters (Mil-
lipore, USA) using a peristaltic pump, stored in labeled sterile falcon tubes and
immediately kept on dry ice. A piece ( 5 cm x 4 cm) of carbon cloth from
each working electrode were collected in labeled falcon tubes and immediately
kept on dry ice. These samples were shipped on dry ice and moved to -80
o
C
130
Figure B.2: Chronoamperometry from the in situ reactor and continuous aqueous
chemistry measurements.
.
131
upon arrival at the laboratory. Microbiological samples were collected on follow-
ing dates (DD-MM-YYYY): 09-12-2014, 27-01-2015, 28-01-2015, 16-04-2015 and
20-05-2015. RNA were extracted from samples collected on 27-01-2015 and be-
yond, since the back
ow from the dehumidier water may have contaminated the
rst microbiological sample.
B.2 In situ Reactor Chronoamperometry
The current vs. time plot (Figure B.2) shows the variation in current across the
4 working electrodes. The spikes in current every 30-60 days is caused due to
instrumentation (rebooting the potentiostat). The working electrodes at lower
potentials (-0.19 and +0.01 V vs. SHE) constantly produced cathodic currents.
The current on working electrode at +0.26 V vs. SHE was always close to zero,
while working electrode poised the most oxidising potential of +0.53 V vs. SHE
produced a cathodic current which decreased over time. The currents observed in
the in situ reactor are a combination of abiotic and biotic redox reaction occurring
on the surface of the electrode, and dicult to analyze the individual reactions.
One can infer the predominance of cathodic reaction on all the working electrodes
due to the oxygen leakage.
B.2.1 Decommissioning the in situ reactor
During decommissioning, samples were collected for: 1) laboratory enrichment and
isolation, 2) imaging under electron microscope, and 3) RNA extraction.
132
Sampling for Laboratory Enrichment
Working electrodes of the in situ reactor were cut into a pieces of 6 cm x 3 cm,
and stored in ltered DUSEL 3A water (using 0.2m syringe lter) in autoclaved
125 mL serum bottle. The planktonic media were also stored in autoclaved 125 mL
serum bottle. Serum bottles were lled up to the brim and stoppered with sterile
butyl stoppers for anaerobic conditions. The serum bottles were stored in Mylar
bags with oxygen scrubber, nally sealed with hot jaw sealer and transported on
ice packs. The serum bottles were transferred to the laboratory refrigerator (4
o
C)
until further enrichments. Laboratory enrichments were begun within 2 weeks of
arrival of the samples.
Sampling for Electron Microscopy
Working electrodes of the in situ reactor were cut into a pieces of 3 cm x 1 cm,
stored in 2.5% glutaraldehyde in a labeled sterile falcon tube. The planktonic
media (10 mL) was collected in a labeled sterile falcon tube to a nal concentra-
tion of 2.5% glutaraldehyde. The samples were collected in a zip-lock bag and
transported on ice packs. The serum bottles were transferred to the laboratory
refrigerator (
o
C) until imaging.
Microbiological Sampling
Working electrodes of the in situ reactor were cut into a pieces of 5 cm x 4 cm and
the planktonic media ( 1 L) of the in situ reactor was ltered through 0.22 m
sterivex lters (Millipore, USA) using a peristaltic pump. Both the samples were
stored in respectively labeled sterile falcon tubes, and immediately kept on dry
133
Figure B.3: Chronoamperometry from the laboratory enrichment reactor indicates
growth of electrode oxidisers at the lowest potential (-0.19 V vs. SHE) and elec-
trode reducers at the higher potentials (+0.26 and +0.53 V vs. SHE).
ice. These samples were shipped on dry ice and moved to {80
o
C upon arrival at
the laboratory.
134
B.3 Laboratory Electrochemical Reactor:
Chronoamperometry and Decommission-
ing
The current vs. time plot (Figure B.3) shows the variation in current across the
4 working electrodes in the laboratory electrochemical enrichment. Media change
every 7-10 days resulted in decrease in the current followed a steady increase in-
dicative of possible disruption of biolms on the electrodes or the loss of putative
redox active molecules responsible for extracellular electron transfer. Working elec-
trode poised at +0.26 V vs. SHE was lost due to electrical short, but it did regain
some activity close to decommissioning.
During decommissioning, samples were collected in sterile 20% glycerol in au-
toclaved cryovials for further isolations, and stored at {80
o
C. samples for electron
microscopy were kept in dark at 4
o
C in 2.5% glutaraldehyde. Microbiological
samples for RNA extraction were quickly moved to {80
o
C in a cryovials.
B.4 RNA extraction
The RNA extraction was done via physical (free-that-vortex) and chemical
(lysozyme) disruption of cell wall prior to phenol:chloroform extraction as pre-
viously described by Mills et al. [198].
Here, I will describe the procedure in detail. Firstly, prepare/buy (all molecular
biology grade) the following items:
1. Autoclaved Reagents
135
(a) 18.3 m
milliQ water (bring upto pH 8 using 0.5 M NaOH)
(b) 1 M Tris-HCl pH 8 (15568-025, Thermo Fisher Scientic, USA)
(c) 500 mM EDTA (pH 8) (15575-020, Thermo Fisher Scientic, USA)
(d) 3 M Sodium Acetate (AM9740, Thermo Fisher Scientic, USA)
(e) 40% Glucose (lter sterilized 5 times using 0.2 m syringe lter; store
at 4
o
C)
(f) 10% Sodium dodecyl sulfate (AM9823, Thermo Fisher Scientic, USA)
(g) Phenol:Chloroform:IAA solution v:v:v 25:24:1 (P2069-100ML, Sigma,
USA)
(h) Lysozyme store at 4
o
C (89833, Thermo Fisher Scientic, USA)
(i) Molecular biology grade ethanol (E7023-500ML, Sigma, USA)
(j) Turbo DNAase kit store at {20
o
C (AM1907M, Thermo Fisher Scientic,
USA)
(k) 0.01 M Sodium citrate pH 6.4 or RNA Storage Solution (AM7000,
Thermo Fisher Scientic, USA)
(l) Liquid N
2
2. Equipment
(a) 2 mL screw cap tubes free of RNase, DNase and phenol resistant (14-
222-613, Axygen, Fisher Scientic, USA)
(b) 2 mL microcentrifuge tubes free of RNase, DNase and phenol resistant
(14-222-181, Axygen, Fisher Scientic, USA)
(c) Dewar
136
(d) Waterbath (at 55
o
C)
(e) Vortex shaker
(f) Ceramic Beads baked in an sterile glass container at 450
o
C for 2
hours (13113325, MoBio Laboratories, USA)
3. Extraction Solutions
(a) Solution 1 (prepare fresh): Add 230 L of 40% Glucose, 200 L of 500
mM EDTA [chelating agent; binds Fe(III) and Ca(II)] and 250 L of 1
M Tris-HCl (buer) in a 10 mL tube and bring the volume to 10 mL
using sterile milliQ water.
(b) Lysozyme solution (prepare fresh): Add 4 mg of lysozyme powder per
1 mL of solution 1 (Add Proteinase K to better target Archaea).
(c) Phenol:chloroform:IAA solution: Use non buered phenol for RNA ex-
traction. However, for DNA extraction use Tris-buered with pH 8.0.
Extraction Procedure is dened below:
1. Add equal amounts of samples to each bead beating tube (0.5 g of ceramic
beads in 2 mL screw cap tubes).
2. Add 100 L of solution 1 to each tube and vortex to re-suspend. Adjust
water content for proper vortexing (not more than 250 L).
3. Freeze samples in liquid nitrogen (30{40 s), thaw in water-bath at 55
o
C for
30 s, vortex for 1 min. Repeat this step 5 times to physically disrupt and
lyse the cells.
137
4. Add solution 1 to a nal volume of 250L, 100L of Lysozyme solution (to
rupture the peptidoglycan layer), and 50L of 500 mM EDTA to each tube.
5. Vortex quickly and place tubes on a shaker at 30
o
C at 200 rpm for 10 min.
6. Add 800L of phenol:chloroform:IAA solution (to remove proteins) followed
quickly by 50 L 10% SDS (to break phospholipid membrane and disrupt
non-covalent bonds in proteins). Vortex for 1 min to form an emulsion.
7. Spin the tube at maximum speed (15000 rpm) for 3 min.
8. Transfer supernatent safely with a pipette, avoiding lower phenol phase and
solids, to a fresh screw-cap tube with 800 L of phenol:chloroform:IAA so-
lution already added.
9. Vortex for 1 min, and spin the tube at maximum speed (15000 rpm) for 3
min.
10. Transfer the supernatent to a fresh tube avoiding the phenol phase.
11. Add 50 L of 3 M Sodium acetate and 1000 L of 100% ethanol to each
tube.
12. Mix tubes gently by
ipping (do not vortex) and spin them at maximum
speed for 15 min at 4
o
C.
13. Discard the supernatent gently by pipetting (in order to not dislodge the
pellet). Leave the tube on its side and let it dry for 10 min. Longer wait
may lead to RNA decay.
14. Resuspend the extracted RNA in 50 L RNA Storage Solution.
138
15. Quantitate the nucleic acid concentration on Nanodrop.
16. Remove any DNA contamination by Turbo DNAase kit according to manu-
facturer's instructions.
17. Reverse transcription reaction can be performed by MRDNA Labs after sub-
mitting RNA samples for sequencing.
B.5 Mineral Preperation
I will describe the preparation of minerals used in this thesis.
MnO
2
This preparation is adapted from Myers et al. [71]. It is prepared by the oxidation
of Mn
2+
by potassium permanganate in the presence of sodium ion according to
the following reaction:
3Mn
2+
+ 2MnO
4
{
+ 2H
2
O! 5MnO
2
+ 4H
+
In 200 mL of milliQ water, 8 g of KMnO
4
were dissolved. The solution was
continuously mixed and heated to just below boiling temperature. 10 mL of 5 M
NaOH was added to neutralize the acid produced by the above reaction, thereby
keeping the reaction mixture alkaline and kinetically favoring the oxidation of
Mn
2+
. In a separate
ask, 15 g of MnCl
2
.4H
2
O were dissolved in 75 mL of milliQ
water. This solution was slowly added to the basic permanganate solution resulting
suspension was heated and mixed for about one hour to ensure complete reaction.
After cooling. the manganate was several times (5-10) by centrifugation (maximum
speed for 10 min) and resuspension in milliQ water. The nal pellet was freeze
139
dried (or speed vacuumed at 60
o
C at maximum speed overnight) and stored as
ne powder in -80
o
C.
Colloidal Elemental Sulfur
This preparation is taken from Rowe et al. [62]. Elemental sulfur powder was
heated in milliQ water to 90
o
C for an hour. The suspension was the centrifuged
(maximum speed for 10 min), resuspended in fresh milliQ water and heated again
to 90
o
C for an hour. The above step was repeated 3 times. The nal pellet was
freeze dried (or speed vacuumed at 60
o
C at maximum speed overnight) and stored
as ne powder in -80
o
C.
Polysuldes
This preparation is adapted from Moser et al. [230]. Mix 7.2 g of sulfur
owers and
24 g of Na
2
S.9H
2
O into boiling water for a nal volume of 100 mL. Boil the mixture
for 15 min with stirring. The resulting transparent, reddish solution was 2.25 M
alkaline polysulde stock solution. Elemental sulfur was precipitated directly into
the growth media by adding the required amounts of the polysulde stock.
Fe(III)-NTA
Add 8.2 g of NAHCO
3
, 12.8 g of sodium nitriloacetic acid (NTA), and 13.5 g of
FeCl
3
.6H
2
O to 70 mL of milliQ water. Adjust the pH 6.5 using 10 M NaOH.
Bring the solution to 100 mL and stir for 15 min. Transfer the solution in 125 mL
serum bottle and bubble with N
2
to make an anaerobic stock. Filter sterilize to
an anaerobic autoclaved serum bottle and store at 4
o
C.
140
2-line Ferrihydrite
This preparation is adapted from Schwertmann et al. [166]. A 500 mL solution
of 40 g Fe(NO
3
)
3
.9H
2
O was stirred continuously, followed by addition of 330 mL
of 1 M KOH to bring the pH to 7-8. Add the last 20 mL of KOH dropwise and
with constant pH monitoring. Rinse pH meter regularly. The brown suspension
formed is washed 5 times by centrifugation and resuspension in fresh milliQ water.
The nal pellet is freeze dried (or speed vacuumed at 60
o
C at maximum speed
overnight) and stored as ne powder in -80
o
C.
Fe(II)CO
3
This method is adapted from Hallbeck et al. [200]. Make 100 mL solution of
3.9 g Fe(NH
4
)
2
(S0
4
)
2
, and 100 mL solution of 1.O g anhydrous Na
2
CO
3
. The
carbonate solution was slowly poured into the ferrous solution at 100
o
C under
a N
2
atmosphere. The precipitated Fe(II)CO
3
was washed 5 times with boiling
double-distilled water and sterilized at 121
o
C for 20 min.
Fe(II)S
This method is adapted from The Prokaryotes [201]. Add 78 g of
Fe(NH
4
)
2
(SO
4
)
2
.6H
2
O with 44 g of Na
2
S.9H
2
O in milliQ water at 50
o
C for 15 min
while constant stirring. Wash Fe(II)S precipitate extensively with fresh milliQ wa-
ter to remove Na
+
, NH
+
4
, and S
2
until the precipitate is neutral. The separation
of adsorbed sulde ion from Fe(II)S precipitate is a slow process and can take 4-5
days (5-9 washings at interval of at least 4 h). After cautious washing, store the
Fe(II)S precipitate in glass stoppered bottles that are completely lled
to avoid oxidation.
141
Poorly Crystalline Iron Oxide (PCIO)
This method is adapted from Lovley et al. [202]. Add 12 g of FeCl
3
.6H
2
O in 200
mL of milliQ water (nal concentration of 0.4 M). Keep stirring constantly and
adjust pH to 7.0 by adding 10 M NaOH drop-wise. After pH 3, rinse the pH probe
repeatedly to remove and accumulation. To remove dissolved chloride, centrifuge
the suspension at 5,000 rpm for 15 min (repeat 6 times). The nal pellet is freeze
dried (or speed vacuumed at 60
o
C at maximum speed overnight) and stored as
ne powder in -80
o
C.
B.6 Microbial community
Two sequence les (.s) were obtained for 2 dierent batches of RNA extracts.
They included:
1. A total from 15 samples of RNA extracted from:
(a) The SURF dehumidier water (on 27-01-2015, 28-01-2015, 16-04-
2015, 20-05-2015) labeled 127SURFDE, 128SURFDE, 416SURFDE,
520SURFDE.
(b) The DUSEL Hole 3A water (on 27-01-2015, 28-01-2015, 16-04-
2015, 20-05-2015) labeled 127SURFH3A, 128SURFH3A, 416SURFH3A,
520SURFH3A.
(c) The biomass collected from the in situ reactor on working electrode:
WE1 (-0.19 V vs. SHE), WE2 (+0.01 V vs. SHE), WE3 (+0.26 V vs.
SHE), WE4 (+0.53 V vs. SHE), WE0 (open circuit), and planktonic
142
cells. These are labeled 520ISECWE1, 520ISECWE2, 520ISECWE3,
520ISECWE4, 520ISECWE0, 520ISECPL
(d) A reagent control labeled Blank
2. A total of 14 samples of RNA extracted from:
(a) A reagent control labeled BlankR
(b) A carbon cloth control labeled BlankC
(c) Electrode-attached biomass from laboratory electrochemical cell poised
-0.19 V vs SHE (from WE1) and its open circuit. These are labeled
WE1 and WE1OC respectively.
(d) Electrode-attached biomass from the blank run of laboratory electro-
chemical cell poised -0.19 V vs SHE labeled WE1Blank.
(e) Electrode-attached biomass from laboratory electrochemical cell poised
+0.01 V vs SHE (from WE2) and its open circuit. These are labeled
WE2 and WE2OC respectively.
(f) Electrode-attached biomass from the blank run of laboratory electro-
chemical cell poised +0.01 V vs SHE labeled WE2Blank.
(g) Electrode-attached biomass from laboratory electrochemical cell poised
+0.26 V vs SHE (from WE3) and its open circuit. These are labeled
WE3 and WE3OC respectively.
(h) Electrode-attached biomass from the blank run of laboratory electro-
chemical cell poised -0.26 V vs SHE labeled WE3Blank.
(i) Electrode-attached biomass from laboratory electrochemical cell poised
-0.53 V vs SHE (from WE4) and its open circuit. These are labeled
WE4 and WE4OC respectively.
143
(j) Electrode-attached biomass from the blank run of laboratory electro-
chemical cell poised +0.53 V vs SHE labeled WE4Blank.
Sequences used for tagged sequencing of 16S rRNA genes from the samples are
provided in Table B.1. The standard operating procedure (SOP) was followed as
outlined by Schloss et al., [108, 227] (also in earlier appendix A.2) geared towards
current data, by merging all the les after initial trimming using quality scores
with parameters as maxambig=0, maxhomop=8, qaverage=20,minlength=230,
maxlength=290. Each OTU was assigned a taxonomy with 0.03 distance levels.
This output le was used for further analysis. The phyla-level bacterial lineages
(Table B.2) observed in SURF indicates Actinobacteria and Verrumicrobia as a
possible contaminant from the dehumudier, and Acidobacteria were present only
in the in situ reactor. These communities were removed from further downstream
analysis. Families observed in blank RNA extraction such as Enterobacteriaceae
and Moraxellaceae were also discarded for further analysis. Further laboratory
electrochemical enrichments contaminants were removed such as Porphyromon-
adaceae and Enterobacteriaceae (Table B.3).
The manually curated list of OTU's from various samples was further analysis.
The OTU's were normalised within each sample (example: across the row in Ta-
ble B.2) to obtain the total abundances of each bacterial community (Figure 5.3,
B.4). Normalization was performed by dividing the number of sequences for each
family by the total number of sequences obtained in the sample. The shift in the
bacterial community in DUSEL hole 3A (from 27-01-2015 to 20-05-2015) indicated
an increase in unclassieds and a decrease in Bacteroidetes. The family Plancto-
mycetes were highly abundant in the in situ reactor compared to the DUSEL hole
3A water.
144
Figure B.4: Raw data of phylum-level bacterial community abundances of the
dehumidier water, DUSEL 3A water and in situ reactor.
Furthermore, the OTU's were normalised between samples for each family (ex-
ample: across the column in Table B.2) to study the the variation of bacterial
families across each potential (in the in situ [Figure 5.4) and laboratory (Figure
5.7) reactor]. In this scenario, normalisation was performed by dividing the num-
ber of sequences for each family by the maximum number of sequences obtained
across the samples.
It is important to note that normalisation is an essential step in the analysis of
expression of sequences. Normalization enables accurate comparisons of expression
between and within sample. A comprehensive review on normalisation techniques
employed for mRNA-Seq data is given by Dillies et al. [231]. Total counts normal-
isation strategy, where counts are divided by the total number of mapped reads
(or library size) associated with their lane and multiplied by the mean total count
145
Figure B.5: Electron micrograph of A) Bacillus sp. (strain WE4-1A1-BC) and B)
Comamonas sp. (strain WE1-1D1) on poised electrode.
.
across all the samples of the dataset, may be apt for our data. I will perform more
normalisation strategy to understand the inherent variation present in data.
B.7 Isolations
Phylogenetic analyses of the isolated strains revealed that majority belonged to
the genera Bacillus, Anaerospora, Cupriavidus, Azonexus and Comamonas. While
strains from the genera Bacillus, Anaerospora, Cupriavidus, Azonexus were isolated
from the electrode reducing potentials, strains from the genus Comamonas were
isolated from both electrode oxidising (-0.19 V vs. SHE) and electrode reducing
potentials (+0.53 V vs. SHE). Details are provided in Table B.4. Strains from
the genera Bacillus, Cupriavidus, Azonexus and Comamonas can be grown in rich
media (R2A). Strains from the genus Anaerospora has to be grown anaerobically on
sodium acetate and Fe(III)-NTA. Many enrichments were frozen in 20% glycerol
for further isolations. Preliminary electrochemical analyses were performed on
Bacillus sp. (strain WE4-1A1-BC) for electrode reduction and Comamonas sp.
146
(strain WE1-1D1) for electrode oxidation (Figure B.5). Further characterization
are ongoing for these individual strains.
147
Sample ID
Barcode
Sequence
127SURFDE GAGTACAG
127SURFH3A GAGTACTC
128SURFDE GAGTAGTG
128SURFH3A GAGTAGAC
416SURFDE GAGATGTG
416SURFH3A GAGATGAC
520SURFDE GAGATCTC
520SURFH3A GAGATCAG
520ISECPL GAGAGTGT
520ISECWE0 GAGTGACA
520ISECWE1 GAGTCACT
520ISECWE2 GAGTCAGA
520ISECWE3 GAGTCTCA
520ISECWE4 GAGTCTGT
Blank GAGTGAGT
BlankC GATCCATG
BlankR GATCCAAC
WE1 GAGTGACA
WE1Blank GAGTGTCT
WE1OC GAGTGAGT
WE2 GAGTGTGA
WE2Blank GAGTTCTG
WE2OC GAGTTCAC
WE3 GAGTTGAG
WE3Blank GATCACCA
WE3OC GAGTTGTC
WE4 GATCACGT
WE4Blank GATCAGGA
WE4OC GATCAGCT
Table B.1: Sequences used for tagged sequencing of 16S rRNA genes from the
samples from SURF (top) and laboratory electrochemical enrichments (bottom).
Linker primer sequence was GTGCCAGCMGCCGCGGTAA
148
Acidobacteria Actinobacteria Bacteroidetes Chloro
exi Firmicutes Planctomycetes Proteobacteria unclassied Verrucomicrobia
127SURFDE 0 66 290 0 2 2 27145 54 0
128SURFDE 0 92 2078 2 1 0 52286 106 0
416SURFDE 0 30 40 0 13 1 40874 82 0
520SURFDE 0 706 2352 0 0 3 21214 1447 3
127SURFH3A 0 0 5107 52 31 0 29350 681 0
128SURFH3A 0 0 5983 81 200 1 29832 968 0
416SURFH3A 0 2 1397 4 155 0 25766 3532 0
520SURFH3A 0 0 2468 9 24 2 27392 2413 0
520ISECPL 5 258 1021 2 3 1393 32430 1976 141
520ISECWE0 68 9 909 0 0 1518 33351 1074 3
520ISECWE1 38 119 4691 0 169 1802 17072 2180 5
520ISECWE2 80 63 6101 1 8 1586 34464 2717 14
520ISECWE3 74 42 646 0 0 2902 40249 1621 5
520ISECWE4 47 3 732 1 0 1208 40163 1212 7
Blank 0 14 8 0 2 1 1376 10 0
Table B.2: Phyla-level bacterial lineages of SURF DUSEL 3A water, dehumidier
water, and the in situ reactor.
Acidobacteria Actinobacteria Armatimonadetes Bacteroidetes Firmicutes Planctomycetes Proteobacteria unclassied Verrucomicrobia
BlankC 0 1 0 6 0 0 106 0 0
BlankR 0 0 0 33 1 0 258 0 0
WE1Blank 0 7173 0 0 393 0 1281 0 0
WE1 30 64 53 4754 0 488 16227 157 0
WE1OC 24 106 86 4401 1 759 15154 324 0
WE2Blank 0 5490 0 2 1 0 3773 0 0
WE2 1 655 0 2831 23 462 2970 842 2
WE2OC 2 833 0 4239 117 566 5043 706 8
WE3Blank 0 758 0 0 4925 0 2022 0 0
WE3 0 922 0 3199 2151 0 2895 2 0
WE3OC 0 1921 0 2016 2435 0 2902 0 0
WE4Blank 0 5 0 1 2562 0 2381 0 0
WE4 0 52 0 2817 191 0 8104 0 0
WE4OC 0 992 0 4130 1450 0 2556 0 0
Table B.3: Phyla-level bacterial lineages in laboratory electrochemical enrichment
reactor.
Inoculum: Potential (V vs. SHE) Primary Enrichment Isolation Comments
Bacillus 0.53 R2A R2A facultative anaerobe
Anaerospora 0.53 Fe(III)NTA and Sodium Acetate Fe(III)NTA and Sodium Acetate obligate anaerobe
Cupriviadus 0.53 Fe(III)NTA and Sodium Acetate R2A facultative anaerobe
Azonexus 0.53 Fe(III)NTA and Sodium Acetate R2A facultative anaerobe
Comamonas +0.53 and -0.19 R2A R2A facultative anaerobe
Table B.4: Strains isolated, the respective strategies involved and observed oxygen
tolerance.
149

Asset Metadata

Creator Jangir, Yamini (author)

Core Title Electrochemical studies of subsurface microorganisms

Contributor Electronically uploaded by the author (provenance)

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program physics

Publication Date 07/22/2016

Defense Date 07/22/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Azonexus,Delftia,electrichemical enrichment,electromicrobiology,extracellular electron transfer,OAI-PMH Harvest,Shewanella,subsurface microbiology

Format application/pdf (imt)

Language English

Advisor El-Naggar, Mohamed Y. (committee chair), Amend, Jan P. (committee member), Boedicker, James Q. (committee member), Haas, Stephen W. (committee member), Kresin, Vitaly V. (committee member)

Creator Email jangir@usc.edu,yamini.iitb@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-272771

Unique identifier UC11281122

Identifier etd-JangirYami-4583.pdf (filename),usctheses-c40-272771 (legacy record id)

Legacy Identifier etd-JangirYami-4583.pdf

Dmrecord 272771

Document Type Dissertation

Format application/pdf (imt)

Rights Jangir, Yamini

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Abstract (if available)

Abstract Microorganisms are the unknown majority life forms on the Earth. While the marine subsurface biomass has been estimated at 1.5-22 petagrams of carbon, the continental subsurface biomass (top 2 km) accounts for 14-135 petagrams of carbon. Subsurface microorganisms can thrive in extreme environmental conditions (e.g., pressure, temperature, pH), and derive energy via diverse metabolic processes. While these environments are geologically diverse, potentially allowing energy harvesting by microorganisms that catalyze redox reactions, many of the abundant electron donors and acceptors are insoluble and therefore not directly bioavailable. Extracellular electron transfer (EET) is a metabolic strategy that microorganisms can deploy to either donate or receive electrons to and from insoluble substrates outside of the cell. Despite its environmental importance and technological promise, EET is mechanistically characterized in only two model dissimilatory metal reducing bacteria (DMRB) Shewanella and Geobacter. ❧ This thesis contributes towards an understanding of the EET mechanisms in the model system S. oneidensis MR-1. Firstly, I focus on in vivo imaging of nanowires formed by S. oneidensis MR-1 expressing periplasmic or cytoplasmic green fluorescent protein (GFP). These images point towards the composition of Shewanella nanowires as outer membrane extensions. Secondly, I describe differential EET behaviour of S. oneidensis MR-1 dependent on the electrode surface (tin doped indium oxide vs. carbon cloth), and an experimental system which allows simultaneous analysis of both the flavin-independent and flavin-dependent routes of EET between S. oneidensis MR-1 and the poised electrodes. ❧ Since EET can be mimicked on the poised electrodes which can serve as unlimited source of energy, it opens the door to electrochemical techniques to enrich for and quantify the activities of environmental microorganisms beyond the well studied model systems. I report the electrochemical enrichment of microorganisms from a deep fractured-rock aquifer in Death Valley, California, USA. In experiments performed in mesocosms containing a synthetic medium based on aquifer chemistry, four working electrodes were poised at different redox potentials (272, 373, 472, 572 mV vs. SHE) to serve as electron acceptors, resulting in anodic currents coupled to the oxidation of acetate during enrichment. The anodes were dominated by Betaproteobacteria from the families Comamonadaceae and Rhodocyclaceae. A representative of each dominant family was subsequently isolated from electrode-associated biomass. The EET abilities of the isolated Delftia strain (designated WE1-13) and Azonexus strain (designated WE2-4) were confirmed. Both genera have been previously observed in mixed communities of microbial fuel cell enrichments, but this is the first direct measurement of their electrochemical activity. While alternate metabolisms (e.g. nitrate reduction) by these organisms were previously known, our observations suggest that additional hidden interactions with external electron acceptors are also possible. ❧ Next, I describe the first in situ potentiostatically controlled electrochemical enrichment of subsurface microorganisms from borehole DUSEL 3A located at 4850 ft below the surface in Sanford Underground Research Facility, South Dakota, USA. The mine is hosted in quartz-veined, sulfide rich segments of an Early Proterozoic, carbonate-facies iron-formation providing in situ study of subsurface microorganims from the early Earth. Here, we poised electrodes at electron-donating (reducing/cathodic) and electron-accepting (oxidising/anodic) redox potentials, to enrich for potential mineral oxidising and reducing microorganisms. We were able to capture the bacterial community present in DUSEL 3A onto the poised electrodes