Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Exploring bacteria-mineral interactions
(USC Thesis Other)
Exploring bacteria-mineral interactions
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Exploring Bacteria-Mineral Interactions
by
Martin Van Den Berghe
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfilment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GEOLOGICAL SCIENCES)
May 2021
Copyright 2021 Martin Van Den Berghe
ii
Acknowledgments
Out of the frying pan, into the fire
If an advisor’s uncontrolled temper can force a student to take a big career gamble and seek
relief in another laboratory, the underlying imperatives of an abusive industry ultimately remain,
and will dictate the course and meaning of any research endeavor, no matter how intellectually
rewarding these might otherwise be. Quantified measures of academic success inform a student,
facing at best dismal career prospects, that raising money, receiving awards and gaining a
fabricated sense of status and prestige associated with a journal’s brand, are the only
accomplishments of any worth if one were hoping to earn more than minimum wage following a
decade-long investment into a University education. These realities ultimately explain how a
pyramid-schemed industry flourishes, feeding on two very powerful human emotions: anxiety and
pride. And in the end, scientific or intellectual reward only ever come second to the superlative
demands of a proverbial rodent competition. Unfortunately for young prospects interested in a
career in research, such insights are only learned through direct experience, at a stage already far
too deep into a commitment that offers no meaningful contingencies. Such an environment is
certainly not a healthy or rewarding one to be in, a reality that is all too evident in so many of the
industry’s members across all seniority levels.
Within this context, I still owe a very large debt of gratitude to Josh West and Ken Nealson
for welcoming me in their laboratories at a time of deep uncertainty, and giving me the chance and
freedom to pursue scientific interests which were, at least at the time, dear to me. Josh deserves
special recognition for patiently listening to, and putting up with, my sporadic proselytizing.
Lastly, endless help, insights and experience was gracefully shared by peers and colleagues who
deserve all my extended thanks: Babak Hassanzadeh, Casey Barr, Bonita Lam, Roman Barco, Moh
iii
El-Naggar, Frank Corsetti, Arkadiy Garber, Nancy Merino, Abby Lunstrum, Gerid Ollison, Yubin
Raut, Emily Burt, Pratixaben Savalia, Xiaopeng Bian, as well as Chris Suffridge and Daniele
Monteverde.
Maximum dynamic pressure
A lingering interest in aeronautics had fueled a desire to join one of USC’s student rocket
groups. Originally meant to be a weekend hobby, working at the Liquid Propulsion Laboratory
proved to be a very demanding, time consuming, challenging, and stressful commitment, more
often akin to a job rather than a pastime. Considering the ambitious nature of the propulsion
development program at play, as well as the high expectation level in members’ skillsets and
engineering background, it is truly remarkable that a geoscientist would be so welcomed and
mentored through such a very steep learning curve. Perhaps it was the unspoken acknowledgement
that in reality, every single person in the group was venturing into a project well over their heads,
that this kind of work was not just incredibly difficult for students, but even for all the brightest,
most seasoned and hardened aerospace engineers the world over. This was rocket science: the
force that let humans reach out to the Heavens, and made them take their place among the Gods
of old. This was the stuff of legends.
Success, by definition, demands a standard for measure, depending upon which, different
conclusions can be drawn. Not all engine hot fires went well. Footage records reveal hard starts,
undue clouds of smoke and soot, as well as fireballs; but also thundering roars and blazing Mach
diamonds. Years spent dabbling in design, manufacturing, computational fluid dynamics, finite
element analysis, integration and testing; countless hours spent clueless, simply trying to solve
problems for which no one in the lab had any answers, were in the end, always met with ever more
iv
reward and satisfaction. The learning experience was immense, the group bonding deep. For this,
I am forever grateful to the Liquid Propulsion Laboratory for offering this incredible experience,
ripe with long-lasting memories and friendships.
Escape velocity
A range of funding sources have supported the various activities I participated in during
the four years spent at USC, namely the National Science Foundation and the Dubai Future
Foundation, as well as various private and corporate benefactors such as Ansys, Nimbix, NX and
the Aerospace Corporation. As invaluable as these contributions were, one cannot forget that they
were only but the means to an end. That end might too often be the focus of such metrics as number
of publications, citations, impact factors, or the basis for further funding proposals: the fuel for a
positive feedback loop, forever seeking metrics-for-grants and grants-for-metrics. Or, on the other
hand, that end could simply be the profound gratification and fulfilment that comes with a new
and deeper understanding of the world we live in, and the role that we humans play in it. Through
exploration and discovery, science’s principal cargo is awe and bewilderment. If such an end is
achieved, inspiration and innovation will inevitably follow in its wake.
Scientifically, the results and rewards of this Ph.D. are here: enough to catch the attention
of any curious soul. But if success does indeed consist, as the old British Bulldog once said, in
going from failure to failure without loss of enthusiasm, it is abundantly clear that this experience
has thoroughly underdelivered: the most damaging loss being that of joy, excitement, and even
interest in the geosciences. Nonetheless, boundless beauty still does and always will permeate this
world, just demanding to be witnessed, and it is our own responsibility to capture and celebrate it.
Any endeavor that keeps this as its ultimate goal will be a success.
v
Table of Contents
Acknowledgments........................................................................................................................... ii
List of Figures .............................................................................................................................. viii
List of Abbreviations ..................................................................................................................... ix
Abstract ........................................................................................................................................... x
Chapter 1: Gaia’s navel ............................................................................................................... 1
1.1. Of life and minerals ................................................................................................... 1
1.2. Bioavailability ............................................................................................................ 7
1.3. Weathering ............................................................................................................... 12
1.4. Engineering .............................................................................................................. 14
Chapter 2: Let them eat rocks!................................................................................................... 19
2.1. Materials and methods ............................................................................................. 24
2.1.1. Olivine ........................................................................................................... 24
2.1.1.1. Preparation ............................................................................................ 24
2.1.1.2. Imaging and characterization ................................................................ 24
2.1.2. S. oneidensis experimental setup................................................................... 25
2.1.2.1. Filtrate amendments .............................................................................. 25
2.1.2.2. Cell counts ............................................................................................. 26
2.1.3. P. aeruginosa experimental setup ................................................................. 26
2.1.3.1. Growth experiments .............................................................................. 26
2.1.3.2. Cell counts ............................................................................................. 27
2.2. Results ...................................................................................................................... 27
2.2.1. Mineral composition ..................................................................................... 27
2.2.2. S. oneidensis experiments ............................................................................. 28
2.2.2.1. Growth patterns ..................................................................................... 28
2.2.2.2. Olivine surface characterization ............................................................ 29
2.2.3. P. aeruginosa experiments ............................................................................ 31
2.2.3.1. Growth patterns ..................................................................................... 31
2.2.3.2. Olivine surface characterization ............................................................ 32
2.3. Discussion ................................................................................................................ 34
2.4. Conclusions .............................................................................................................. 36
Chapter 3: Regards to self interests ........................................................................................... 38
3.1. Materials and methods ............................................................................................. 41
3.2. Results ...................................................................................................................... 41
3.3. Discussion ................................................................................................................ 42
3.3.1. Siderophore concentrations and growth ........................................................ 42
3.3.2. Community interactions ................................................................................ 44
3.4. Conclusions .............................................................................................................. 46
vi
Chapter 4: The emperor’s new clothes ...................................................................................... 48
4.1. Methods.................................................................................................................... 49
4.2. Results ...................................................................................................................... 50
4.3. Discussion ................................................................................................................ 51
4.4. Conclusions .............................................................................................................. 53
Chapter 5: The hare and the tortoise .......................................................................................... 55
5.1. Materials and methods ............................................................................................. 56
5.2. Results ...................................................................................................................... 57
5.3. Discussion ................................................................................................................ 61
5.3.1. Dissolution rates ............................................................................................ 61
5.3.2. Nutrient uptake and dissolution stoichiometry ............................................. 63
5.4. Conclusions .............................................................................................................. 64
Chapter 6: Red Mars .................................................................................................................. 66
6.1. Materials and methods ............................................................................................. 68
6.1.1. Apparatus ...................................................................................................... 68
6.1.2. Culturing conditions ...................................................................................... 70
6.1.3. Microscopy and activity ................................................................................ 72
6.2. Results ...................................................................................................................... 73
6.3. Discussion ................................................................................................................ 76
6.4. Conclusions .............................................................................................................. 81
Chapter 7: Kosmos .................................................................................................................... 83
References ..................................................................................................................................... 87
Appendix A ................................................................................................................................. 109
S. oneidensis experiments ............................................................................................... 109
First set ..................................................................................................................... 114
Second set ................................................................................................................ 115
Third set ................................................................................................................... 116
P. aeruginosa experiment ............................................................................................... 116
Appendix B ................................................................................................................................. 118
Olivine surface area calculations .................................................................................... 118
Mineral dissolution data .................................................................................................. 119
First set ..................................................................................................................... 119
Second set ................................................................................................................ 121
Abiotic set ................................................................................................................ 124
Appendix C ................................................................................................................................. 128
Single chamber experiments ........................................................................................... 128
Mass and charge balance calculations in the two-chamber experiments ........................ 129
Thoughts on energetics of the bioelectrochemical reactor ............................................. 135
Conceptual design of a Mars-based bioelectrochemical ISRU system .......................... 138
vii
Water system............................................................................................................ 139
Nitrogen system ....................................................................................................... 139
Electrical system ...................................................................................................... 139
Ferric oxide minerals ............................................................................................... 139
Outputs ..................................................................................................................... 140
viii
List of Figures
Figure 1: Growth curves for S. oneidensis wild type (MR-1, dashed line) and siderophore
synthesis deletion mutant (MR-1Δ, solid lines). ........................................................... 29
Figure 2: SEM images of olivine grains from Shewanella MR-1 and deletion mutant MR-1Δ,
with and without MR-1 filtrate amendments. .............................................................. 30
Figure 3: Growth curves for P. aeruginosa wild type (PA-14, red lines) and siderophore
production deletion mutant (strain PA-14Δ, blue lines). ............................................. 32
Figure 4: SEM images of olivine grains from Pseudomonas PA-14 and abiotic control
experiments. ..................................................................................................................... 33
Figure 5: MR-1Δ grown with deferoxamine amendments. .................................................... 42
Figure 6: XPS spectra of iron on olivine materials of duplicate experiments. ...................... 51
Figure 7: Comparing olivine dissolution rates in biotic and abiotic experiments. ............... 59
Figure 8: The impact of microbial activity on olivine dissolution rates. ............................... 61
Figure 9: Design of the two-chamber bioelectrochemical reactor. ......................................... 69
Figure 10: Experimental results showing electric current and oxygen production. ............. 74
Figure 11: Microscope images of the working electrode in the biotic experiments. ............. 76
Figure 12: Conceptual diagrams showing possible pathways for electron transfer from the
working electrode to ferric oxides. ................................................................................ 78
ix
List of Abbreviations
AQDS Anthraquinone-2,6-Disulfonate
CNI Core Center of Excellence in Nano Imaging
DO Dissolved Oxygen
EET Extracellular Electron Transport
GOE Great Oxidation Event
HGT Horizontal Gene Transfer
HNLC High Nutrient Low Chlorophyll
ICP-OES Inductively Coupled Plasma - Optical Emission Spectrometry
ISRU In Situ Resource Utilization
ITO Indium Tin Oxide
MOXIE Mars Oxygen ISRU Experiment
OD Optical Density
PSM Proton Selective Membrane
RSG Redox Sensor Green
SEM Scanning Electron Microscope
USC University of Southern California
XPS X-ray photoelectron spectroscopy
XRF X-Ray Fluorescence
x
Abstract
The Earth has a unique and distinct history, as it is the only planet we know that harbours
life. Its evolution has been defined not only by cosmic and geologic processes, but also by
biological activity. Life most likely emerged from, and learned to interact intimately with its
mineral environment, not just to survive, but also to thrive, firmly altering Earth’s crust, oceans
and atmosphere in the process. This thesis explores the relationship between biology and geology,
and some of the effects interactions between bacteria and minerals can impart on planetary
systems. Specifically, the focus is on the mineralogy, speciation and bioavailability of iron as a
function of bacterial activity, and the corresponding impact these can have on environmental
geochemistry. A primary focus is the examination of iron limitation with respect to bacterial
siderophore synthesis. Iron is critical to fundamental biological processes such as respiration and
photosynthesis, and its bioavailability can impact primary and heterotrophic productivity. Yet in
oxic environments at near-neutral pH, iron is highly insoluble, rendering it, in principle,
unavailable as a nutrient for biological growth. We examined the impact of siderophores on the
speciation, mobility and bioavailability of iron from rock-forming silicate minerals, detailing the
mechanisms by which microbes directly use mineral substrates to support primary productivity, as
well as the consequent effects on silicate dissolution rates. Growth experiments were performed
with Shewanella oneidensis in an oxic, iron-depleted minimal medium, amended with olivine
minerals as the sole source of iron. Experiments included the wild-type strain MR-1, and a
siderophore synthesis gene deletion mutant strain (MR-1Δ). Relative to MR-1, MR-1Δ exhibited
a very pronounced growth penalty and an extended lag phase. However, substantial growth of MR-
1Δ, comparable to MR-1 growth, was observed when the mutant strain was provided with
siderophores in the form of either filtrate from a well-grown MR-1 culture, or commercially
xi
available siderophores (deferoxamine). These observations suggest that under aerobic conditions,
siderophores are critical for S. oneidensis to acquire iron from olivine. Growth-limiting
concentrations of deferoxamine amendments were observed to be ≤ 5 – 10 µM. X-Ray
photoelectron spectroscopy of the incubated olivine surfaces suggested that siderophores mobilize
ferric iron from the mineral surface. Dissolved silicon measured by optical emission spectrometry
revealed that while higher siderophore concentrations lead to higher olivine dissolution rates,
effective bacterial reuse of siderophores is a more important factor, enhancing dissolution rates by
an order of magnitude compared to similar abiotic experiments. Combined, these results
demonstrated that low (nM to µM) concentrations of siderophores can effectively mobilize iron
bound within silicate minerals, supporting very significant biological growth in limiting
environments. The specific mechanism likely involves siderophores removing a protective layer
of nano-meter thick iron oxides, enhancing both silicate dissolution rates and iron bioavailability.
This thesis also examined the use of bacterial extra cellular transport in cathode-oxidizing, iron
reducing, two-chambered bioelectric reactors. This reactor successfully produced molecular
oxygen as a by-product of microbial iron oxide reductive dissolution from hypothetical Martian
soils. These latter experiments served as a proof of concept for the development of potential in situ
resource utilization technologies for use on Mars, supporting its human exploration. Overall, this
work demonstrates how a detailed mechanistic understanding of bacteria-mineral interactions can
help understand the drivers of the evolution and dynamics of planetary processes — and can also
provide a means to develop effective tools in geoengineering applications.
1
Chapter 1: Gaia’s navel
We humans have serious limitations in our perception of our world, as our experiences,
understandings and imagination are heavily biased towards a circumstance made up of visible
light, in the millimeter to kilometer scale, and in the seconds to decades time frame. Some things
are therefore harder to perceive, and our limitations might make us pass over the reality of a much
more dynamic, complex and wondrous world: one in which entire regions of bedrock can dissolve
and reprecipitate, creating vast and rich underground mineral galleries; a world in which the whole
of the planet’s skin moves, slowly but relentlessly, directing the movements and collisions of entire
continents, and in the process sculpting mountain and volcanic chains; a world in which masses of
ocean and air exchange heat and gases with one another, moderating the planet’s elaborate climate
systems. And, perhaps most importantly, a world in which life not only thrives, but interacts
intimately with all aspects of this planet: mineral, water and gas, playing with them over eons. This
intimate relationship is certainly what makes our planet special far beyond any other world we
know of. The Earth molded life, and life shaped the Earth, both in concert so as to keep and
maintain our delicately balanced pale blue dot in just the right condition: mild, fertile and live
(Lovelock, 1989; Sagan, 1994).
1.1. Of life and minerals
Traditionally the silicate-based mineral, and carbon-based organic spheres have been
studied separately, as components of entirely different research disciplines. Yet mineral-organic
interactions offer a rich and complex interface, which at their core, exemplify the relationship,
dependency and the co-evolution of life with its host planet. This interface starts with an alluring
study in contrasts and the impacts of subtleties. Carbon sits above silicon, in group IV of the
2
periodic table, indicating that the two elements have 4 valence electrons in their elemental state,
highlighting a tendency for tetrahedral (sp3) hybridization and suggesting, at first glance, very
similar tendencies in chemical reactions. However, silicon is a bigger atom, with valence electrons
inhabiting the larger 3p shell. This simple fact affects such fundamental properties as elements’
coordination numbers and hybridization ranges, the length, angle, strength and energetics of bonds
with other elements, as well as their ability to form double or triple bonds (Petkowski et al., 2020).
Most importantly, silicon’s greater electropositivity tends to create more polar bonds, making these
significantly more susceptible to nucleophilic attack and hydrolysis than carbon compounds.
Additionally, Si-O bonds are disproportionally stronger (452 kJ/mol) than other Si- bonds, which,
in a reactive medium like water, easily leads to silicon polymerization into meshed and stiff Si-O
structures, rather than linear, flexible chains like hydrocarbons (Petkowski et al., 2020). Thus, on
a water-rich Earth, a simple difference in atomic radius ultimately relegated silicon to the relatively
rigid mineral world, forming the basis for not just rocks, but an entire planet; and carbon to the
dynamic organic world, the basis of life.
The bulk of the Earth is indeed made primarily of silicate, a very stable silicon oxide
tetrahedron that can survive the many extremes of the rock cycle, and which has been the core
component of this planet since original accretion. Life, however, was not always present on Earth,
and for an eon, our planet was sterile. And then, out of the blue, life emerged. Genesis, or at least
the spontaneous events that brought about self-organizing, autocatalytic, energy-gathering
reactions, remains perhaps the single most profound and important mystery ever faced by
humanity. Our understanding of life does however strongly suggest a few fundamental premises
on how it likely started: (1) it must have emerged in a geochemical environment rife with redox
potential, as redox reactions provide the energy to all life; and (2) these reactions must have
3
occurred in the presence of appropriate catalysts, required to effectively release this latent chemical
energy in an organized and contained manner (Sousa et al., 2013; Belmonte and Mansy, 2016;
Preiner et al., 2019).
Environments with such energetic imbalances might have been located in geological areas
of high chemical or heat transfer, such as alkaline hydrothermal vents, similar to modern “white
smokers.” One proposed process involves the exergonic serpentinization reaction and its
associated hydrogen gas production, which combined with high pre-Cambrian CO 2 (and HCO 3
-
)
concentrations, would have created an energetically fertile environment. Spontaneous reactions
between these species would have had the double effect of releasing energy, as well as generating
simple organic molecules (Russell and Martin, 2004; Russell et al., 2013; Sousa et al., 2013;
Preiner et al., 2019). As for catalysts, clues are laid out in biochemistry itself: life is completely
dependent on metal centers to power countless enzymatic reactions (Belmonte and Mansy, 2016).
Most notably, multivalent metal complexes such as iron-sulfur clusters in ferredoxins are critical
to fundamental metabolic processes such as electron transfer and biochemical energy generation,
both chemosynthetically and photosynthetically. Ferredoxins, which are nearly universally found
in living systems, are also the core component of hydrogenase enzymes, powering H 2-oxidizing
respiration: evidence consistent with a proposed primordial hydrogen-based metabolism. These
iron-sulfur clusters can further be thought of, in composition and structure, as individual crystal
units or greigite or mackinawite, minerals which, markedly, are also found in the mineral
assemblages of crustal hydrothermal system (Volbeda and Fontecilla-Camps, 2006). This
association between metal structures and fundamental metabolic processes ought not be thought
as mere coincidence, but rather highlights a deep-rooted relationship between the mineral and
organic worlds that dates since the emergence of life itself. In a broad sense, the critical role metals
4
and metal clusters play in the fundamental biochemical mechanics of life is evidence reflecting the
central part minerals likely played as catalysts in an early prebiotic Earth (Hall et al., 1971; Sousa
et al., 2013; Belmonte and Mansy, 2016; Camprubi et al., 2017; Cardenas-Rodriguez et al., 2018).
Accordingly, hypotheses for the emergence of life have evolved from Miller’s primordial
soup (Miller, 1953) to forms closer to that of a rock-based sandwich (Lane et al., 2010; Barge et
al., 2017). It has been proposed that phyllosilicates (such as mica, clays but also include
serpentinite), with their high surface charge, would have provided fertile and flexible environments
for polar prebiotic organic compounds to sorb and concentrate (Feng et al., 2005). A range of
reactions could have been catalyzed along mineral surfaces, as various metal ions centers are
known to enhance organic reactions of critical biological functionality. Specifically, iron-sulfur
clusters have been shown to promote the synthesis of various peptides as well as pyruvate, key
intermediates in metabolisms such as the citric acid cycle (Wächtershäuser, 1990, 2000; Huber,
1997, 2003; Cody, 2000, 2004). Phyllosilicates have been shown to effectively activate
compounds such as N 2, H 2 and CO 2, providing the catalysis required to generate biologically
relevant molecules (Preiner et al., 2019). They are also known to promote the homochiral
polymerization of nucleic acids into RNA, as well as accelerate the formation of fatty acid vesicles
(Hanczyc et al., 2003; Aldersley et al., 2011; Jheeta and Joshi, 2014). The added effects of fluid
flow, wet-dry cycles, and high cation concentrations in small pore spaces between phyllosilicate
sheets have also been proposed as mechanisms in creating fact-simile prebiotic intracellular
environments, supporting cofactor reactions with metals at the surface of minerals, and further
promoting the polymerization of nucleic and amino acids (Ertem et al., 2008; Hansma, 2010,
2013). It is thus clear that many key ingredients required in the spontaneous emergence of biotic
systems are closely related to, and catalyzed by, reactions at the organic-mineral interface.
5
If the sorption of organic compounds onto minerals surfaces played a critical role in the
emergence of life on Earth, the importance of this process on its continued evolution, and the
preservation of genetic information, seems to have attracted less scrutiny so far. Minerals have a
deeply stabilizing and protective influence on their sorbed organic guests, particularly providing
protection from photic, enzymatic and chemical degradation (Pietramellara et al., 2009; Poch et
al., 2015; Pedreira-Segade et al., 2016). The impact of this process is truly remarkable, as large
genomes have been found to be stable on geologically relevant time scales when sorbed onto
mineral matrices, with recent studies having found Neanderthal and Denisovan DNA bound to
sediments, preserved since the Pleistocene (Slon et al., 2017; Sheng et al., 2019). Recent findings
have further suggested that extracellular DNA (eDNA) bound to minerals could play a key role in
bacterial evolution through horizontal gene transfer (HGT) (Sand and Jelavić, 2018). eDNA is in
fact a critical component to the formation of bacterial biofilms (Whitchurch, 2002; Puigbò et al.,
2014) and codes abundantly for transposases: enzymes which function is specifically to catalyze
the transfer of genes across genomes. DNA of transposases has indeed recently been found to be
the most ubiquitous eDNA in nature, found in a wide range of environments, most notably within
biofilms of modern serpentinizing systems (Brazelton and Baross, 2009; Aziz et al., 2010).
Minerals are also known to provide the very energy required to sustain life, as they can act
as electron donors or acceptors, providing chemolithotrophic energy for microbial respiration. This
was a surprise to many at the time of discovery, since the architecture of living organisms does not
allow the absorption of larger mineral structures inside a cell (Madigan et al., 2014). The key to
this process involves the transfer of electrons extracellularly, promoting the reduction of ferric and
manganese oxides in anaerobic environments (Myers and Nealson, 1988; Nealson and Saffarini,
1994). While a canonical electron transport chain is typically described as an intracellular system,
6
extracellular electron transport (EET) involves electron transfer across cell membranes to external
metal acceptors. This process has been shown to be possible through a range of mechanisms,
including bacterial nanowires which can extend across micrometer-scale gaps, linking the cell with
a mineral electron acceptor via what amounts to a biological electric wire, or through intermediate
shuttles such as flavins that simply diffuse between cell receptor proteins and the terminal electron
acceptor (Shi et al., 2016). EET is now known to go so far as include interactions with electrodes,
and power engineered microbial fuel that can simultaneously breakdown organic contaminants
and generate electricity (Kim et al., 2004; Chen, 2004; Nealson, 2017). Importantly, more recent
findings show that EET seems to be not limited to unique bacteria with exotic genomes, but to be
in fact a common feature within generic microbial communities in marine sediment, with set
electric potential influencing the nature of bacterial assemblages that can grow in such
environments (Lam et al., 2019).
In summary, it is abundantly clear that mineral structures have played a critical role in the
emergence of life and continue to influence heavily its evolution and metabolism. Through
sorption and electron transfer, the mineral-organic interface allows life to acquire the energy it
needs through EET, promotes the long-term preservation of genetic information, and even
provides the conditions necessary for the very emergence of life. Organic-mineral interactions
have thus always been a central part of the story of life, its evolution, and relationship with the
Earth. A detailed and mechanistic exploration of nature of these interactions, and their global
impacts on biology and geosystems, is thus the focus of this thesis, with particular emphasis on
iron: its speciation, mobility, mineralogy and bioavailability to bacteria as both a critical nutrient
and a source of energy.
7
1.2. Bioavailability
It has been proposed that metal requirements in active biochemical machinery reflect the
geochemical conditions of the time when these systems developed (Saito et al., 2003; Belmonte
and Mansy, 2016). This hypothesis works particularly well in the case of iron, as this element is
crucial to fundamental biological processes such as photosynthesis, nitrogen fixation and EET (Shi
et al., 2016; Hutchins and Boyd, 2016), and is thus critical to living organisms and subject to high
biological demand. In an early world free of molecular oxygen, soluble ferrous iron would have
been widely available and easily acquired to build biomass and develop critical biological
functions (Saito et al., 2003). However, the appearance of oxygenic photosynthesis and the large-
scale production of molecular oxygen changed things dramatically, as iron in oxic environments
at near-neutral pH oxidizes very rapidly to iron(III) (Millero et al., 1987) and forms highly
insoluble oxy-hydroxide precipitates such as ferrihydrite, goethite or hematite (solubility constants
in the range of 10
-37
to 10
-44
; Schwertmann, 1991). Thus, in the world following the Great
Oxidation Event(s) (GOE), iron became scarcely available to meet biological requirements (Saito
et al., 2003). Considering that evolution does not backtrack (Madigan et al., 2014), organisms
remained thoroughly iron-dependent, and suddenly faced an existential problem: how to acquire
the much-needed iron, now that it had turned to rock? Life, then, did what it does well: it adapted.
It developed means to interact with minerals in order to extract the nutrients it needed from them
in order to thrive (Gadd, 2017).
Scientific knowledge to-date informs us of one major way in which microbes can
effectively acquire metal nutrients out of minerals: by enhancing metal mobilization out of
minerals through the sorption of organic compounds onto mineral surfaces. A mechanistic
understanding of this process brings us again to explore subtleties at the molecular level,
8
contrasting carbon with metal chemistry. While crystal lattices are inherently stable structures,
mineral surfaces are defined by elements left with open-ended, unfulfilled bonds, and are
inherently less stable (Brantley et al., 2008). In aqueous chemistry, the coordination environment
at the solid-water interface will determine surface-active processes. Metal centers in solid
substrates will commonly behave, due to generally low electronegativities, as electrophiles (or
Lewis acids), and make good sites for coordinated covalent bonds with dissolved nucleophiles
(Lewis bases), which can act as ligands or sorbents. One of the most common and effective
nucleophiles is oxygen, whether bound as water, hydroxyl ion, or as part of an organic compound
(Stumm, 1995). The formation of coordinate bonds between metal electrophiles and sorbent
nucleophiles can change the density, distribution and structure of electron clouds surrounding
metal atoms, even potentially changing their orbital hybridization, thus destabilizing metals out of
their crystal lattice (Otero et al., 2017). Sorbent denticity, or the number of coordinate bonds
formed with a metal, also significantly affects sorption strength, with multidentate sorbents having
greater metal-binding affinities, increasing thermodynamic as well as redox stability to the sorbent-
metal complex (Dhungana and Crumbliss, 2005).
Thus, a wide range of organic exudates have been characterized and identified in their
ability to sorb readily onto mineral surfaces and promote metal and nutrient mobility, a list that
includes many common metabolites such as oxalate, pyruvate, lactate, malonate, ascorbate, and
most importantly, siderophores (Kummert and Stumm, 1980; Olsen and Rimstidt, 2008; Ha et al.,
2008; Hider and Kong, 2010; Dehner et al., 2010). This sorption process can be enhanced via the
direct attachment of microbes to mineral surfaces, or the formation of biofilms, increasing
elemental mobility and bioavailability out of mineral phases (Aouad et al., 2006; Ahmed and
Holmström, 2015; Gerrits, 2019).
9
How and when such mechanisms of nutrient acquisition emerged along life’s history is
unknown. Many of the metabolites that have been identified to play a role in enhancing nutrient
mobility from minerals are primarily linked to metabolic pathways that have little or nothing to do
with nutrient acquisition (e.g., pyruvate and the Krebs cycle), raising the question whether such
elemental mobilization might have been a beneficial but accidental by-product of unrelated
biological activity. However, the existence of siderophores demands a different interpretation, as
they serve virtually no purpose other than chelating metals, most effectively ferric iron. They are
bi- to hexa-dentate compounds with extremely high binding affinities to a variety of metals, but
most markedly for ferric iron (logK f = 25 – 49), values significantly greater than those of other
common ligands such as transferrins or EDTA (logK f = 20 – 25) (Dhungana and Crumbliss, 2005;
Hider and Kong, 2010). Their synthesis has been found to be upregulated specifically in iron-
deficient conditions (Garibaldi and Neilands, 1956; Braun et al., 1998), and they require a
relatively extensive biochemical support systems (from metabolite synthesis to receptor proteins,
etc.) to be of use to an organism (Noinaj et al., 2010; Schalk and Guillon, 2013). Siderophore’s
exceptionally high binding affinity to oxidized, ferric iron, also coincides with the exceptionally
low solubility of the same element, hinting that siderophores, and more broadly, mechanisms to
extract nutrients from minerals, evolved during or after Earth’s GOE, when iron became much less
bioavailable.
This interpretation is consistent with recent molecular studies that, though uncalibrated on
a geological timescale, show that siderophore biosynthesis is an ancient process, ingrained deeply
into microbial genomes, and has undergone extensive vertical as well as horizontal gene transfer
across widely disparate clades (Thode et al., 2018). This is also consistent with the observation
that siderophore synthesis is non-ribosomal, and can include both D and L amino acids, further
10
suggesting that siderophore synthesis is not only ancient, but might have emerged through entirely
unique processes (Challis and Ravel, 2000; Bluhm et al., 2002; Raymond et al., 2003; Schwecke
et al., 2006; Abergel et al., 2009; Schalk and Guillon, 2013; Esmaeel et al., 2016). Further research
into siderophore phylogeny has also brought to light fascinating evolutionary dynamics, with
complex community relationships surrounding siderophore use, “sharing,” and “cheating,”
emphasizing that siderophores stand in the center of a long-standing struggle to acquire a scarce
nutrient (Amin et al., 2009a; Kümmerli et al., 2014; Butaitė et al., 2017). A mixed and complex
dynamic of antagonistic competition as well as mutualistic relationships across microbial
communities highlights a chronic selective pressure driving adaptation under nutrient limitation
for significant periods of time.
Long after the GOE, the problem of iron limitation has proven to be a chronic one, as iron
is, to this day, still a factor commonly limiting primary productivity in many parts of Earth’s
marine and terrestrial environments (Berman-Frank et al., 2001; Moore et al., 2013). In modern
oceans, measured dissolved iron concentrations are often in the rate-limiting concentrations of low
nanomolar to high picomolar range (Morel and Price, 2003). At these concentrations, ferric iron is
not abundant, soluble, or bioavailable enough to sustain life (Rich and Morel, 1990; Dhungana and
Crumbliss, 2005). Sources of iron are often limited to localized plumes in coastal areas (e.g., from
continental weathering via riverine and glacial inputs), near the ocean floor (from hydrothermal
vent activity), as well as through airborne dust deposition (mainly downwind of arid environments)
(Jickells et al., 2005). The first two sources generally do not contribute significantly to the total
oceanic iron inventory, or overall biological productivity, as the distinct oceanic geochemical
environment leads to rapid flocculation, aggregation and precipitation near these sources. Aeolian
inputs are however well known to contribute significantly to iron loading to the open oceans
11
(Sholkovitz, 1978; Sholkovitz et al., 1978; Jickells et al., 2005; Haase et al., 2009). Yet aeolian
inputs have been characterized as predominantly silicates (clay) and ferric oxyhydroxides,
particles still extremely insoluble in oceanic waters (Caquineau, 2002; Journet et al., 2008; Schroth
et al., 2009). Of the iron in seawater that is measured as being functionally dissolved (i.e. anything
smaller than 0.22 or 0.45 μm), most is either complexed to organic ligands or found in insoluble
colloidal forms, affecting its bioavailability (Rich and Morel, 1990; Rue and Bruland, 1995).
As a particularly revealing case study, this relationship was made clear through a series of
iron amendment experiments performed in various high nutrient low chlorophyll (HNLC) regions
of the oceans, where dissolved iron amendments to the ocean surface repeatedly produced, for a
time, phytoplankton blooms, which in turn impacted significantly carbon, nitrogen and silicon
fluxes in the study sites (Watson et al., 2000; Aumont and Bopp, 2006; Boyd et al., 2007).
However, these experiments have typically used highly soluble species of iron (such as ferrous
sulphate), which, as amendments, are far from representative of the natural inputs of iron to the
biosphere such as silicates and oxides. However, it has also been shown more recently that iron
bioavailability to marine photosynthetic eukaryotes in HNLC regions can be significantly
increased through monosaccharide amendments, as these created stable dissolved iron-saccharide
ligand complexes, more readily absorbed by phytoplankton (Hassler et al., 2011). Combined, these
results emphasize that organic-mineral complexation can have a profound influence on nutrient
mobility and bioavailability, and can thus regulate primary productivity and global biogeochemical
cycles.
With a very high degree of heterogeneity, soils offer a broader range of iron solubility,
subject in large part to porewater redox and pH (Knoepp et al., 2005). Yet, they can support very
large amounts of biomass, including highly iron-dependent photosynthetic plants. Iron’s
12
bioavailability in soils is commonly a function of its mobilization from primary and secondary
ferric minerals, and is highly subject to micrometer-scale mineral dissolution-precipitation
reactions, often mediated by organic exudates (Lindsay and Schwab, 1982; Lindsay, 1991;
Lemanceau et al., 2009; Krohling et al., 2016; Gadd, 2017).
In this context of pervasive iron limitation, siderophores serve as a prime example of how
sorption and chemical interactions at the organic-inorganic interface are central to life’s history
and evolution. By enhancing nutrient bioavailability, siderophores promote primary productivity
and highlight life’s impressive ability to adapt, and thrive in challenging and evolving planetary
environments. By doing so, microbes, and their interactions with minerals, can considerably
impact key global biogeochemical cycles, ocean chemistry and climate systems, highlighting the
deeply intertwined history and co-evolution of life and its host Earth.
1.3. Weathering
Some of life’s most significant biogeochemical processes on this planet revolve around
primary productivity and photosynthesis, as these create vast sinks for carbon, altering ocean
alkalinity and chemistry and reducing atmospheric CO 2 contents, changing global climate systems
(Anbar and Knoll, 2002; Lyons et al., 2014). Another less overt but still profound effect that life
can have on its planet is by altering the weathering of silicate rocks and Earth’s crust (Kump et al.,
2000).
The dissolution of silicate minerals releases silicate anions from a stable, structured crystal
lattice, to isolated anions in solution, which then hydrolyze readily and turn into the very weak
silicic acid. In generic conditions (near-neutral pH, livable temperatures and pressures) this process
generates significant amounts of alkalinity, which in turn affects ocean carbonate chemistry and
13
contributes to carbonate precipitation. Thus, silicate weathering plays a key role in controlling the
long-term carbon cycle, carbonate export to the geosphere, atmospheric CO2 contents, and thus
global climate systems (Berner et al., 1983; Ruddiman, 1997; Hilley and Porder, 2008; Hartmann
et al., 2009). Of particular interest are basaltic rocks as they form the bulk of the oceanic crust as
well as large components of the continental crust, and contain large amounts of olivine, a mineral
which dissolves at higher rates than other common silicates (Dessert et al., 2003; Dupré et al.,
2003; Wolff-Boenisch et al., 2011), making olivine dissolution an especially important component
of crustal weathering and carbonate chemistry. This process is in fact so pronounced that olivine
has been examined as a promising substrate for climate geo-engineering and managing
anthropogenic CO2 emissions (Köhler et al., 2010; Hartmann et al., 2013).
However, the environmental mechanisms controlling olivine dissolution kinetics are
poorly understood, as laboratory experiments have repeatedly failed to match observed
environmental dissolution rates (White and Brantley, 2003; Maher et al., 2006; Ganor et al., 2007).
This discrepancy has opened up questions about whether biological activity could play a
significant role in olivine dissolution. Documented experimental controls on olivine dissolution
rates include pH, temperature and solution composition, with dissolution rates increasing at non-
neutral pH conditions, high temperatures, and when exposed to high concentrations of organic
compounds such as oxalate and carboxylic acids, as these sorb onto metal cations in crystal lattices
and weaken oxygen bonds within the silicate structure (Pokrovsky and Schott, 2000; Olsen, 2007;
Olsen and Rimstidt, 2008; Pokrovsky et al., 2009; Wolff-Boenisch et al., 2011; Rimstidt et al.,
2012). More recent biotic experiments have suggested that dense microbial cultures (mainly
proteobacteria, actinobacteria and fungi) could enhance silicate dissolution rates, with carbon
oxidation through heterotrophic metabolism leading to localized decreases in pH, enhancing
14
dissolution (Wu et al., 2007, 2008; Ahmed and Holmström, 2015). However, other studies have
suggested the opposite, where microbial activity (also from a range of proteobacteria and fungi)
could inhibit silicate dissolution, with biofilm formation, or even microbial iron oxidation, leading
to the protective encapsulation of the mineral substrate (Santelli et al., 2001; Stockmann et al.,
2012; Oelkers et al., 2015). In both cases, the metabolic and geochemical mechanisms involved in
such bacteria-mineral interactions remained unclear.
All in all, it is evident that organic-mineral interactions stand at the interface linking the
bio- and geo- spheres, highlighting their deeply intertwined history. While life’s ability to access
iron as a nutrient is challenged in many environments on Earth ever since the GOE, the net source
of iron to the biosphere derives predominantly from the chemical weathering of crustal material,
mainly composed of silicate minerals. Thus, life’s ability to access metal nutrients out of mineral
phases impacts both primary productivity and crustal weathering dynamics, imparting a double,
biological as well as geochemical, effect on the global carbon cycle, ocean chemistry and climate
systems, and across much of Earth’s geological history.
1.4. Engineering
Contrasting with the Earth, Mars is a very static world that can be painted as a cold, dry
and oxidized desert. It has not supported any major active planetary systems (e.g., water, rock or
carbon cycles) in eons, nor has harboured any considerable amounts of life, if ever, in just as long
(Grotzinger et al., 2014). Accordingly, reactions at the organic-mineral interface, if existent, can
hardly be expected to play an important role in the planet’s modern geochemistry, or impact any
geochemical systems. Mars’ distinctive red color is in large part due to abundant and ubiquitous
ferric oxide minerals present in its topsoils. Though Martian regolith is generally composed of
15
detrital basaltic materials formed from the physical weathering of parent magmatic crustal material
(Downs and MSL Science Team, 2015; Ehlmann et al., 2017), topsoils on Mars contain copious
amounts of globally distributed, poorly crystalline nanophase ferric minerals, which include
hematite, maghemite, magnetite, goethite, jarosite and schwertmannite (Bell et al., 2000; Morris
et al., 2004; Hamilton et al., 2005; Goetz et al., 2005; Morris et al., 2006a). Furthermore, large,
coarsely crystalline hematite deposits have been found exposed in a number of surficial, water-
altered bedrock formations across the planet, with hematite accounting for up to 45 vol.% of some
outcrops (Christensen et al., 2000, 2001; Christensen, 2004; Fergason et al., 2014; Lanza et al.,
2016). In other words, Mars is an entirely rusted planet, with an iron oxide coat so expansive it
can be seen from Earth with the naked eye.
The origin of these oxides in Martian topsoils is unclear. One explanation might relate to
the constraints of early planetary formation. The relatively small size of Mars would have created
a shallower, cooler magmatic ocean during and after mantle-core separation, when compared to
Earth. With oxygen fugacity a function of temperature, the Martian core might have remained too
cool to promote oxygen diffusion out of the mantle, thus explaining a mantle persistently enriched
in metal-oxide content (e.g., FeO of ~ 18 wt.%) compared to that of Earth (~ 8 wt.%) (Rubie et al.,
2004). Accordingly, weathering of FeO-rich crustal material during a geologically active and wet
Mars might have contributed to a high abundance of oxide phases at the surface (Tuff et al., 2013).
The association of ferric oxides with surficial, water-altered formations further suggests chemical
weathering of crustal material in liquid, and more importantly, oxic environments. This has led to
the perhaps bold hypothesis of a Martian global oxidation event that might bear similarities with
that of the Earth, in wet and mild Noachian environments, leading to the precipitation of oxide
minerals (Christensen, 2004; Lanza et al., 2016). More recently, Martian atmospheric and soil
16
environments could still be considered oxidizing due in part to ample solar ultraviolet radiation,
the presence of photochemically-produced hydrogen peroxide in the atmosphere (Encrenaz et al.,
2012), or even by dust-storm derived triboelectricity (Delory et al., 2006). Thus, a range of
processes over Mars’ geological history and development have led to the continuous production
of environmental oxidants, causing the effective and thorough oxidation of its surface, relegating
iron to a globally abundant but mainly static mineral phase. By comparison, iron on Earth’s surface
has a much more complex fate, undergoing constant mobilization and cycling via a wide range of
redox and complexation reactions with the dynamic pool of organic carbon life generates.
Life depends on the transfer of electrons as well as the accumulation of mass in order to
acquire energy, grow and perpetuate itself (Madigan et al., 2014). The associated reactions produce
by-products that can transform an environment’s geochemistry and entropy. An economy of scale
has thus the potential to profoundly alter an entire planet’s biogeochemical systems, as has been
made plainly apparent on Earth with processes such as iron reduction, nitrogen fixation or
photosynthesis (Jickells et al., 2005), contrasting again with the much more static, inorganic Mars.
While it is easy to take such processes for granted as they occur every single day around us, life’s
ability to catalyze these reactions is truly astonishing. Breaking nitrogen-nitrogen triple bonds, or
powering an endergonic reaction and generating biochemical energy by utilizing freely available
photons, are spectacular achievements of biochemical innovation, which we humans have not been
able to emulate for our own purposes until very recently. And the means by which they are
mimicked rely, perhaps surprisingly, on manufactured systems that in no way resemble or overlap
with those that nature already offer (e.g., the Haber-Bosch process vs. the nitrogenase complex, or
photovoltaic cells vs. photosystems I & II). Broadly speaking, chemical engineering endeavors
have historically relied on abiotic, bulk reactions to achieve a desired output, paying little or no
17
attention to the fact that life might have already developed similarly outstanding capabilities to
catalyze beneficial geochemical reactions, and typically by much less energy-demanding means.
However, in our day and age, blessed with advanced genome editing and engineering capabilities
going so far as producing synthetic organisms (Gibson et al., 2010), it might behoove us to harness
life’s complex and self-sustaining biochemical machinery to improve and further human
endeavors. A few efforts of note do effectively make use of microbial capabilities, such as
applications in biomining and bioleaching of certain mineral ores, as well as efforts in contaminant
bioremediation (Schippers et al., 2014). More recently, striking advances in bioengineering
involve the simultaneous wastewater treatment of organic waste and electricity generation by
utilizing bacterial EET capabilities within bioelectric reactors (Chen, 2004; Bretschger et al., 2010;
Nealson, 2017). Yet, these latter systems have so far only relied on coupling carbon oxidation with
anode reduction reactions, overlooking metal oxide reduction, and thus, likely explore only a
fraction of the full potential of EET applications.
The prevalence of ferric oxides on Mars becomes particularly interesting in the context of
microbial ferric oxide reduction. Thus far, the bulk of studies on iron reduction have focused on
the speciation, mobility and bioavailability of iron, which clearly plays an important role on Earth
in modulating biological productivity and climactic systems (Martin, 1990; Konhauser et al., 2011;
Hutchins and Boyd, 2016). However, EET-driven ferric oxide reductive dissolution also implies
the congruent release of oxygen from the crystal lattice, a process that has attracted much less
scrutiny. Mars, covered in ferric oxide minerals as it is, contains an abundance of oxygen: if not
in its atmosphere, certainly within its soils. By utilizing microbial iron reduction, this oxygen could
be liberated from its crystal lattice, potentially providing a source of gaseous O 2: a resource critical
to support interplanetary transport, serving both as a propellant needed to refuel spacecrafts as well
18
as to sustain human life. Bacterial iron reduction could then prove to be a particularly valuable
process for engineering purposes, permitting the use of a globally abundant in-situ resource in the
form of iron oxides in order to pursue the exploration of Mars.
19
Chapter 2: Let them eat rocks!
Four words, associated with a Great Princess during a time of strife, did not help feed a
populace, nor could they even keep heads securely attached to their shoulders. But some creatures
might have taken such words closer to heart and learned not just to survive, but even thrive with
little more than rocks to eat.
Metals are a fundamental component of life and are required as the catalytic element in
numerous enzymatic systems (Belmonte and Mansy, 2016). In particular, iron is critical to primary
productivity and key to global biogeochemical processes such as photosynthesis, respiration,
nitrogen fixation and electron flow, including extracellular electron transfer (EET), and is thus
subject to high biological demand (Shi et al., 2016; Hutchins and Boyd, 2016). Yet, in many
modern natural environments on Earth (oxic, pH between 4 and 9), iron is found in its oxidized,
ferric state, in the form of insoluble oxide and hydroxide phases (Ksp values 10
-37
– 10
-44
;
Schwertmann, 1991). As a result of this low solubility, iron is often scarcely available to meet
biological requirements and often limits growth in many of Earth’s ecosystems, including major
parts of the marine environment.
In the oceans, dissolved iron concentrations have been measured in the concentrations of
low nanomolar to high picomolar range (Morel and Price, 2003), concentrations at which iron is
not abundant, soluble, or bioavailable enough to sustain life (Rich and Morel, 1990; Dhungana and
Crumbliss, 2005). Most of the iron in the oceans is in the form of phyllosilicate and ferric
oxyhydroxide particles, largely derived from aeolian inputs (Caquineau, 2002; Journet et al., 2008;
Schroth et al., 2009), as well as iron complexed to colloidal organic particles and ligands (Rue and
Bruland, 1995; Wu et al., 2001; Hassler et al., 2011; Sander and Koschinsky, 2011; Pinedo-
Gonzalez et al., 2014). Of the iron in seawater that is measured as being functionally dissolved
20
(i.e. anything smaller than 0.22 or 0.45 μm), most is either complexed to organic ligands or found
in insoluble colloidal forms, affecting its bioavailability (Rich and Morel, 1990). Nevertheless,
ocean primary productivity, photosynthesis and nitrogen fixation have been well characterized and
correlate spatially with iron loading into the oceans (primarily from aeolian origins), suggesting
that particulate iron can be rendered bioavailable (Berman-Frank et al., 2001; Jickells et al., 2005).
Soils offer a broader range of iron solubility, subject in large part to porewater redox and pH. Yet,
they can support very large amounts of biomass, including highly iron-dependent photosynthetic
plants. The bioavailability of iron in soils is highly subject to micrometer-scale mineral
dissolution-precipitation reactions, often mediated by organic exudates (Lindsay and Schwab,
1982; Lindsay, 1991; Lemanceau et al., 2009; Krohling et al., 2016; Gadd, 2017). Similarly,
seafloor sediments and rock-hosted ecosystems are characterized by highly variable iron solubility,
especially in ocean floor basalts and peridotites, where the available pool of soluble iron is limited
despite the abundance of iron-bearing minerals, contained primarily within common rock-forming
silicate minerals (Janecky and Seyfried, 1986).
In short, unfavorable geochemical conditions dramatically limit iron solubility in many of
Earth’s environments, and thus life’s ability to use it as a nutrient. Nonetheless, the sheer
abundance of primary productivity around the globe highlights that life has evolved pathways for
accessing nutrients such as iron that are otherwise scarcely available. While much has been learned
about these pathways in recent decades, questions remain about the nature of iron bioavailability
and the specific mechanisms involved in its acquisition directly from mineral phases. One of the
major unresolved questions concerns the role of microbial siderophore production in the
dissolution of silicate minerals.
21
Siderophores, organic compounds produced by bacteria and fungi that possess extreme
affinities to iron (logK f = 25 – 49), are known to act as key components of microbial iron
acquisition pathways (Dhungana and Crumbliss, 2005; Hider and Kong, 2010). Numerous
experiments have demonstrated that siderophores promote the dissolution of iron oxides under
conditions in which they would otherwise be insoluble (Kraemer et al., 1999; Cocozza et al., 2002;
Yoshida et al., 2002; Cheah et al., 2003). In tandem, live cultures of Pseudomonas sp. have been
shown to grow and produce abundant siderophores when provided with iron oxides and Fe(III)-
bearing clays as the sole iron source, suggesting the microbial use of siderophores for nutrient
acquisition (Hersman et al., 1996, 2000, 2001; Maurice et al., 2000, 2001; Ferret et al., 2014). The
effect of siderophore-mediated iron acquisition was isolated in experiments with P. mendocina:
whereas wild type strains were able to grow in the present of hematite, gene-deletion mutants that
lacked the ability to produce siderophore were not (Dehner et al., 2010). However, the deletion
mutant did grow in the presence of very small (<10 µm diameter) hematite particles (Dehner et
al., 2011), and when in direct contact with Fe(III)-bearing montmorillonite clays (Kuhn et al.,
2013), perhaps due to cell wall penetration and reductive biofilm formation, respectively.
Thus, despite understanding of (1) the importance of siderophores for microbial iron
acquisition from Fe-oxides and Fe(III)-bearing clays and (2) the role of both microbes and
siderophores in facilitating silicate mineral dissolution, the mechanistic role for siderophore
production in the microbial enhancement of silicate mineral dissolution remains to be resolved.
The distinction between the dissolution of silicates and the acquisition of iron from Fe-oxides is
important because, while Fe-oxides such as goethite are prevalent in many environments, they are
not ubiquitous, especially in rock-hosted settings — raising the question of nutrient acquisition for
the substantial microbial biomass that has been identified in these environments (Bailey et al.,
22
2009; Sudek et al., 2017; Stranghoener et al., 2018). Studying dissolution of Fe(II)-bearing
silicates also offers insight into the role of ligand-promoted dissolution in the absence of often-
coupled reductive processes associated with iron oxides and Fe(III)-bearing clays (Wang et al.,
2015). Moreover, silicate mineral dissolution plays a central role in the breakdown of rocks to
form soil and sediment (Anderson and Anderson, 2010), in the release of other rock-derived
nutrients including P, Ca, and Mg (Zaharescu et al., 2019), and in the regulation of the long-term
carbon cycle by production of alkalinity (Berner, 2004). This wide-ranging importance of silicate
dissolution motivates the present study in seeking a mechanistic understanding of the role of
microbial siderophore production.
Bacterial biofilms are dynamic, multi-layered microbial ecosystems that can grow to
several centimeters thickness, and can host strong chemical, nutrient and redox gradients, in the
process significantly improving bacterial survival and promoting active growth. They can be found
in almost all environments on Earth, including soils, sub-surface, aquatic and marine environments
(generally growing on mineral substrates), as well as in hosted environments such as within animal
tissues (Hall-Stoodley et al., 2004). As one important trigger to biofilm formation can in fact be
iron limitation (Stanley and Lazazzera, 2004), the metabolite-rich environment offered by biofilms
can lead to a wide array of mineral-organic interactions which may lead to mineral dissolution
without implicating siderophores.
In this study, we used the environmentally-relevant wild-type model bacterium Shewanella
oneidensis MR-1 (hereafter referred to simply as MR-1) to isolate siderophore production and
utilization as a specific biochemical process allowing iron acquisition directly from a primary,
rock-forming silicate mineral phase, supporting significant biological growth. We utilized MR-1
(Venkateswaran et al., 1999; Hau and Gralnick, 2007) and its corresponding targeted gene-deletion
23
mutant strain (hereafter referred to as MR-1), which is incapable of siderophore synthesis, but
which remains capable of siderophore uptake (ΔSO3031; Fennessey et al., 2010). We incubated
these Shewanella strains in an iron-depleted medium amended with olivine minerals as the sole
source of iron, and we monitored growth to test whether they could utilize silicate mineral
substrates in nutrient limited environments. We also compared the growth patterns of the
Shewanella strains with those of Pseudomonas aeruginosa strains grown in similar iron-deficient,
olivine amended conditions, particularly P. aeruginosa PA-14 wild type (hereafter referred to as
PA-14) and its corresponding targeted gene-deletion mutant strain (hereafter referred to as PA-
14), which is incapable of siderophore synthesis (both pyoverdine and pyochelin: ΔpvdA ΔpchE),
but which remains capable of siderophore uptake (Wang et al., 2011).
The choice of olivine as a mineral substrate was made as it is one of the most abundant
minerals that make up the bulk Earth. It is also composed of divalent iron, such that Shewanella
and Pseudomonas interactions with olivine-bound iron could not involve known dissimilatory iron
reduction processes (Wang and Newman, 2008; Coursolle and Gralnick, 2012). We analyzed the
incubated mineral surfaces with scanning electron microscopy (SEM) to track mineral dissolution
at the mineral-water interface. These observations allow us to identify a precise element
mobilization mechanism as a function of siderophore synthesis, and its impact on silicate
dissolution. Specifically, we tested how microbial growth depends on siderophore-mediated iron
acquisition from primary silicate minerals (as observed previously for Fe-oxides by Dehner et al.,
2010).
24
2.1. Materials and methods
2.1.1. Olivine
2.1.1.1. Preparation
Olivine grains were collected from the University of Southern California mineral collection
and were hand-crushed with a clean mortar and pestle. Crushed material was sieved to isolate size
fractions, with grains 150 – 300 µm diameter selected for these experiments. Prior to use in
experiments, all olivine grains were ultrasonicated and rinsed 7 times (5 minutes sonication per
rinse) in 200-proof ethanol to remove clay-sized particles and avoid potential abiotic dissolution
in the process, then air-dried in an oven at 130° C overnight. Immediately before use in
experiments, olivine grains were UV-sterilized for 30 minutes. Olivine grains were never
autoclaved in order to avoid potential iron oxidation and risks of secondary minerals coating the
grains.
2.1.1.2. Imaging and characterization
Secondary electron imaging by SEM was performed on olivine minerals following
experiments with a Nova NanoSEM 450 (FEI, US). Working distances were 5 – 10 mm and an
accelerating voltage of 10 kV was used. All samples were prepared using a 7-step ethanol
dehydration process, fixed with a Tousimis 815 critical point dryer (Rockville, US), and coated
with a Cressington 108 Pt/Pd sputter-coater (Watford, UK). Quantitative X-ray Fluorescence
(XRF) was performed on the fresh experimental olivine grains to determine iron mass and molar
ratios with an S8 Tiger (Bruker, US) in a separated vacuum, using scintillation and proportional
flow detectors. Both SEM and XRF work were performed at the Core Center of Excellence in
Nano Imaging (CNI) at the University of Southern California.
25
2.1.2. S. oneidensis experimental setup
2.1.2.1. Filtrate amendments
The pH 7 growth medium used was the M-1 minimal medium (Bretschger et al., 2007,
2010), slightly modified with a MOPS buffer (50 mM) and N-acetyl glucosamine (18 mM) as the
carbon source and electron donor, in order to avoid sorption from carboxylic acids (such as lactate)
onto mineral surfaces and potentially related abiotic enhancement of mineral dissolution (Olsen
and Rimstidt, 2008). All experiments were inoculated with 1.3x10
9
cells (for initial experimental
densities of 2.6x10
7
cells/mL in 50 mL medium) of the appropriate strain. These cells had first
been grown from individual isolated colonies and conditioned in the same minimal M-1 medium
(though iron-replete, with 3.6 µM FeSO 4, and containing no olivine), then extracted, triple-rinsed
and concentrated in iron-free medium prior to inoculation. All Shewanella experiments, as well as
media preparation, sterilization and preservation, were performed in acid-washed polycarbonate
or polypropylene containers and flasks to avoid potential metal contamination from biotically-
induced glass dissolution (Gorbushina and Palinska, 1999; Brehm et al., 2005; Aouad et al., 2006).
All growths were performed aerobically, incubated at 30⁰ C and shaken continuously at 120
revolutions per minute.
Batch experiments used a total volume of 50 mL of iron-deplete minimal medium. Flasks
were amended with 0.05 g of olivine such that bacterial growth would require acquisition of iron
directly from the amended minerals. Preceding these experiments, the MR-1 strain was grown up
to stationary phase in the same iron-deplete conditions with added olivine. Its filtrate (< 0.2 µm)
was recovered for use in later experiments and tested with the chrome-azurol S (CAS) reagent,
confirming the abundance of iron ligands in solution (Schwyn and Neilands, 1987). CAS assays
26
were also performed on experiment filtrate to verify that MR-1 were not producing siderophores
de novo. The filtrate amendment experiments were conducted in triplicate and consisted of (Fig.
1): (1) MR-1, (2) MR-1 provided with siderophores immediately (0 h) via 5 mL of MR-1 filtrate,
(3) MR-1 provided with siderophores after 24 h via 5 mL of MR-1 filtrate, and (4) MR-1
provided with no siderophores. Duplicate killed controls involved autoclaving flasks after having
inoculated them with MR-1 cells but prior to adding the olivine grains.
2.1.2.2. Cell counts
Cell counts were performed through optical density (OD, 600 nm) measurements of cell
cultures with a UV-2600 spectrophotometer (Shimadzu, Japan). Measurements of OD were not
affected by olivine minerals as these were 150 – 300 µm, a size fraction too large to remain
suspended under incubation conditions. The OD measurements were then correlated to pre-
established relationships of OD vs. cell counts. Cell counts were determined for MR-1 and MR-
1, using a Zeiss Axio fluorescence microscope. Briefly, cells were stained with Acridine Orange
(50 µg/mL), filtered onto a 25 mm diameter 0.2 µm filter (Whatman Nucleopore polycarbonate
membrane, GE Healthcare, US), and ten fields of view were counted. The total number of cells
per mL was calculated as a function of the average cell counts per filter, the volume filtered, the
area of the field of view, and the area of the filter.
2.1.3. P. aeruginosa experimental setup
2.1.3.1. Growth experiments
A standard pH 7.2, MOPS-based minimal growth medium was used for growth, made iron-
deplete and with 8 mM glucose added as source of carbon (Palmer et al., 2005). Pseudomonas
27
experiments were of a preliminary nature, and experiments were inoculated directly with
individual isolated colonies and not performed in duplicates. Batch experiments used a total
volume of 50 mL of iron-deplete minimal medium. Flasks were amended with 0.05 g of olivine
such that bacterial growth would require acquisition of iron directly from the amended minerals.
Additionally, precautions against the use of glassware were not taken as potential experimental
risks had not become clear yet. So contrary to Shewanella experiments, these were conducted in
acid-washed glass Erlenmeyer flasks, and media preparation was performed in acid-washed glass
containers. CAS assays were performed on experiment filtrate to verify that PA-14 were not
producing siderophores de novo. Pseudomonas experiments were performed aerobically,
incubated at 37⁰ C and were shaken continuously at 120 revolutions per minute.
2.1.3.2. Cell counts
Cell counts were performed through optical density (OD, 600 nm) measurements of cell
cultures with a UV-2600 spectrophotometer (Shimadzu, Japan). Measurements of OD were not
affected by olivine minerals as these were 150 – 300 µm, a size fraction too large to remain
suspended under incubation conditions.
2.2. Results
2.2.1. Mineral composition
Based on the XRF results, the olivine used in this study was determined to have a
composition of Mg 1.76Fe 0.24SiO 4 (i.e., Fo 88, predominantly forsterite with a secondary fayalite
component) – thus containing a substantial amount of mineral-bound iron.
28
2.2.2. S. oneidensis experiments
2.2.2.1. Growth patterns
MR-1 was able to grow to cell density > 10
9
cells/mL when provided with olivine as the
sole source of iron (Fig. 1). On the other hand, MR-1 was unable to grow to similar cell densities
under the same conditions and exhibited a prolonged lag phase. When the deletion mutant strain
was, however, provided with filtrate from a mature, siderophore-containing MR-1 culture that had
also been provided with olivine as a source of iron, MR-1 grew to cell densities in the 10
9
cells/mL range, similar to MR-1. The initial growth rate for MR-1 cultures amended with
siderophore-containing filtrate was faster than that for MR-1, though the MR-1 reached the same
cell density after about 24 hours. The addition of filtrate to a MR-1 culture after 24 hours of
effective dormancy also led to a pronounced growth response following addition of the filtrate,
with these experiments also reaching densities of 10
9
cells/mL. CAS assays of filtrate performed
throughout the experiments confirmed that MR-1Δ growths never produced iron ligands de novo.
29
Figure 1: Growth curves for S. oneidensis wild type (MR-1, dashed line) and siderophore
synthesis deletion mutant (MR-1Δ, solid lines). Compared to MR-1, MR-1Δ failed to grow
significantly in the presence of olivine as the sole source of iron. The addition of siderophore-
replete filtrate triggered growth and associated iron acquisition from olivine minerals in MR-1Δ
cultures. The experimental design is illustrated schematically at the top. Error bars represent the
standard deviation of experimental triplicates.
2.2.2.2. Olivine surface characterization
At the end of the experiments, the olivine grains exposed to siderophores, either from MR-
1 culture or from the siderophore-amended MR-1 culture, showed clear and common dissolution
features, resembling etch-pit scouring marks visible via SEM (Fig. 2A – 2C). These features were
30
absent on olivine grains from the abiotic controls and the mutant culture lacking siderophore
amendments (Fig. 2D – 2E).
Figure 2: SEM images of olivine grains from Shewanella MR-1 and deletion mutant MR-1Δ,
with and without MR-1 filtrate amendments. Images show etch pit-like scouring textures on
surfaces, inferred as mineral dissolution features, in MR-1 (A, B) and filtrate amended experiments
31
(C). These features are lacking in experiments without filtrate amendments (D, E). Scale bars
represent 5 µm.
2.2.3. P. aeruginosa experiments
2.2.3.1. Growth patterns
Contrary to the Shewanella experiments, both PA-14 and PA-14 were able to grow to
very high cell densities (OD > 1.4) when provided with olivine as the sole source of iron (Fig. 3).
When provided with no iron whatsoever (no olivine, no iron in the medium), both strains were
also able to grow substantially and to similar levels (OD ~ 0.8). It was also noted visually that both
Pseudomonas strains yielded very thick (up to 1 cm thick), dense biofilms in all experiments, and
in the case of olivine-amended experiments, encapsulated all mineral grains very effectively,
growing around them (biofilm noticeable under the SEM, Fig. 4A). Thus, it is clear that
experiments had very uneven cell distribution, with very different densities between the plankton
(inferred through OD), and biofilms (unknown densities, though expected to be orders of
magnitude greater than the plankton). CAS assays of growth filtrate throughout the experiments
confirmed that PA-14Δ growths never produced iron ligands de novo.
32
Figure 3: Growth curves for P. aeruginosa wild type (PA-14, red lines) and siderophore
production deletion mutant (strain PA-14Δ, blue lines). Both PA-14 and PA-14Δ, in the
presence of olivine, grew to comparably high densities. Both strains also grew substantially in the
absence of olivine, or any source of iron.
2.2.3.2. Olivine surface characterization
Olivine grains from heavy growth experiment were compared with grains from the abiotic
experiment under the SEM. Mineral surfaces exposed to PA-14 had very extensive and deep
scouring features, much more pronounced and severe still than those noticed in the Shewanella
experiments, features that were non-existent on mineral surfaces of the abiotic experiment (Fig.
4). Mineral grains from the PA-14Δ experiment were not recovered in sufficient quantity to
successfully image them. However, given the results from growth experiments (Fig. 3), it is
expected that mineral surfaces from PA-14Δ experiments would bear similar degrees of scouring
and dissolution textures.
33
Figure 4: SEM images of olivine grains from Pseudomonas PA-14 and abiotic control
experiments. Images show extensive etch pit-like scouring textures on olivine surfaces, inferred
as mineral dissolution features, associated with PA-14 biofilm (right side of the image; A),
compared to smooth mineral surfaces in abiotic controls (B). Scale bars represent 5 µm.
34
2.3. Discussion
Both Pseudomonas and Shewanella, wild types and deletion mutants, were able to grow to
very high cell densities when provided with olivine as sole source of nutrients, highlighting the
role silicate minerals can play as a direct source of nutrients, promoting microbial growth in
limiting environments. For the latter clade, MR-1Δ exhibited a very pronounced growth penalty
and an extended lag phase from an inability to produce siderophores. However, substantial growth
of MR-1Δ, comparable to MR-1 growth, was observed when the mutant strain was provided with
siderophores in the form of filtrate from a well-grown MR-1 culture, thus isolating siderophores
as single metabolites that have can enhance iron mobilization directly out of from mineral phases
into a bioavailable form.
When provided siderophores from the MR-1 filtrate, MR-1 seemed to have had a shorter
lag phase and showed measurable growth sooner than MR-1 (t = 0 – 20 h, Fig. 1). This difference
might be explained by relatively high siderophore concentrations compared to initial cell densities
(2.6x10
7
cells/mL) at time 0 h, demonstrating the quick response of MR-1 to begin upregulating
iron-siderophore uptake. Additionally, MR-1 would have been spared the energetic expense of
synthesizing siderophores, allowing these organisms to access iron and grow promptly, whereas
MR-1 would have had to first invest in siderophore synthesis in order to initiate growth.
As the experiments progressed, the growth phases of filtrate-amended MR-1 showed a
shallower, almost linear pattern, compared to the standard exponential growth of MR-1 (t = 20 –
40 h; Fig. 1). This difference might be explained by the constant siderophore concentrations
inhibiting exponential growth. In contrast, the exponential growth in MR-1 cell density
presumably occurred in tandem with an exponential increase in siderophore concentrations, since
35
these cultures were able to produce their own siderophores. This would have led to the more classic
microbial growth pattern observed (Fig. 1).
By comparison, the Pseudomonas deletion mutant strain PA-14Δ was not subject to a
growth penalty compared the wild type PA-14. These results suggest that siderophore synthesis is
not strictly necessary for iron mobilization and acquisition from mineral phases with Pseudomonas
strains, and that thick, multilayered biofilms can have abrasive qualities that lead to mineral
dissolution and nutrient release, supporting microbial growth even in the absence of siderophores.
It is currently unclear which metabolite(s) might be responsible for this iron mobilization and
mineral dissolution. Literature suggests some common respiration-related metabolites present in
biofilms, such as oxalate, lactate, pyruvate or ketogluconate, well known to sorb very effectively
onto mineral surfaces, likely present in very high concentrations within biofilm environments
(Kummert and Stumm, 1980; Stumm, 1995; Cheah et al., 2003; Olsen and Rimstidt, 2008; Ha et
al., 2008; Dehner et al., 2010). The abrasive quality of such metabolites might further be supported
by the observation that even in the complete absence of a source of iron (i.e., no olivine or iron in
the growth medium), both PA-14 and PA-14Δ grew significantly (Fig. 3). This suggests that
impurities in the glass Erlenmeyer flasks might have been a source of iron, implying that organic
compounds sorption might further have a destabilizing effect onto silicate glasses. Such a process
has already been documented, with bacterial colonization and biofilm formation enhancing the
dissolution of ocean floor basaltic glasses (Thorseth et al., 2003; Brehm et al., 2005; Aouad et al.,
2006; Bailey et al., 2009; Stockmann et al., 2012; Sudek et al., 2017). The precise mechanism of
glass dissolution, as well as the associated ligands and their concentrations within biofilms,
however still remain largely unconstrained.
36
Overall, the results presented here add to the literature, establishing that iron does not need
to be in a truly dissolved phase in order to be bioavailable, as minerals exposed to organic ligands,
particularly siderophores, can serve as a direct source of micronutrients supporting primary
productivity. Our study confirms that primary silicate minerals can act in this capacity, in tandem
with more widely-studied acquisition of iron from oxides and hydroxides. Thus common crust-
forming rocks such as basalt and granite, or even particulate silicates and dust, might all be
considered as part of the expanded pool of iron that is bioavailable through siderophore
complexing. This implies that the common quantification of iron available to the biosphere may
be biased, as standard protocols for the measurement of nutrients merely look at elements
contained within a specific size fraction (i.e., < 0.45 – 0.22 μm), and may therefore miss potentially
important sources of iron such as insoluble silicates in dust, soil or even crustal materials.
2.4. Conclusions
The results presented here shed important light on the capabilities of organic ligands,
especially siderophores and biofilms, in extracting iron directly out of silicate minerals and glasses.
These results demonstrate that primary silicate phases can be important sources of micronutrients
and should be included in calculating the pool of bioavailable iron. Thus, these insights have
important implications for our understanding of nutrient bioavailability, the biological controls on
mineral dissolution rates, and importantly, the limits and adaptability of life in limiting
environments. By enabling primary productivity in limiting environments, siderophores—and
their interactions with silicate minerals—have a direct impact on major biogeochemical cycles
associated with biological growth, such as carbon, nitrogen and oxygen. Such impacts can have a
37
profound influence on planetary evolution, influencing systems such as climate and ocean
dynamics as well as crustal weathering.
Let them eat brioche. If apocryphal, these words, pronounced in fact not by Marie
Antoinette but by Rousseau in his Confessions, still likely reflect 18
th
century royalty’s
expectations of a population’s ability to overcome starvation through adaption, finding new ways
to acquire food from exotic sources in times of dire need. Unfortunately for aristocrats and their
necks, such expectations seem to carry a greater degree of insight into broad scientific questions
regarding life’s profound ability to adapt to challenging and limiting environments, than that of
human societies. Since the Enlightenment, humanity has learned that animals hardly have such a
degree of metabolic flexibility, at least on short time scales. Life as a whole, particularly the
microbial kind, however, does. In the process of learning to eat rocks, life has found ways to thrive
where it otherwise could not. It has adapted to an evolving Earth by developing innovative
biochemical systems — allowing it to flourish when a planet was turning against it. The results, in
turn, further transformed ecosystems and wore down a planet’s crust.
38
Chapter 3: Regards to self interests
Adam Smith once asserted that butchers, brewers and bankers didn’t excel in the field of
benevolence, and that the best way to put food on a table, or for that matter attend to the wealth of
nations, was to look to one’s self interest. He also pointed out that if an object is found to be
desirable while its access is limited, it gains value. With value, trade can develop between those
desiring the object and those who control the means of production. Siderophores can solubilize
iron out of minerals such as ferric oxides and silicates in limiting environments, and provide a
resource critical to the survival of organisms. Those organisms that can produce siderophores
therefore have an edge over others in their ability to survive and reproduce. This however comes
at a cost: siderophores are relatively complex polypeptides, and their synthesis conveys an
energetic burden (Schalk and Guillon, 2013; Raymond et al., 2015). But in this expense,
siderophore producers gain a valued asset which non-producers will wish to acquire: the basis for
trade and an economy around a valuable resource. Thus, siderophores form the basis for dynamic
relationships between microbes, influencing the make and diversity of entire ecosystems.
While the full range of community dynamics surrounding siderophores is still not
completely constrained, certain relationships have become clear, and span the range from
mutualistic to antagonistic in nature. It has been found that within microbial communities, or even
within specific clades, some strains lack siderophore synthesis genes, but do have those encoding
for siderophore receptors and transporters (Thode et al., 2018). In the absence of any known
reciprocating service or metabolite, such strains have been labeled “cheaters”, as they benefit, free
of cost, from other organisms’ synthesis of siderophores and effort to access the limiting iron. This
cheating involves complex evolutionary relationships, driven by competition where producers
have a pressure to develop highly specific siderophores (i.e. such that non-producers can’t use),
39
and cheaters have a pressure to develop a biochemical cypher to keep utilizing siderophores (i.e.
having the right transporter for the right type of iron chelator). Such a dynamic has been best
observed in Pseudomonas communities, as illustrated by the unusually large range of complexity
of pyoverdine siderophores as their receptor complexes (Butaitė et al., 2017).
On the other hand, more mutualistic relationships also exist between siderophore producers
and non-producers. It is possible that non-producers might be providing other benefits to a
community so far unrecognized, a dynamic around auxotrophy already documented with vitamin
or amino acid synthesis (Sañudo-Wilhelmy et al., 2014). One notable mutualistic example with
siderophores includes plants and algae (generally speaking, photosynthesizing eukaryotes), as
these latter do not produce siderophores despite their photosystems requiring large amounts of
iron. Land plants rely on a symbiotic relationship with fungi (the only eukaryotes that do produce
siderophores) in the mycorrhizae to acquire inorganic nutrients, including iron (Bartholdy et al.,
2001). Similarly, marine algae have also developed a relationship with some bacteria (mainly
Marinobacter, commonly found in close association with coccolithophores and dinoflagellates) as
these produce light-sensitive siderophores, the photolysis of which reduces iron and cleaves the
chelate, in an particularly opportune location nearby the algae’s membrane, thus increasing iron
availability to these algae (Barbeau et al., 2001; Amin et al., 2009a, 2009b, 2012b). The nature of
siderophores can also influence the dynamics of siderophore use in a community, as diffusible
siderophores (and thus easily shared or stolen) have been most commonly found expressed in
highly structured habitats such as soils and hosted environments, while poorly diffusible
siderophores (amphiphiles) are generally more often found in low density, unstructured habitats
such as seawaters (Kümmerli et al., 2014).
40
Regardless of the nature of the relationship between siderophore producers and non-
producers, siderophores are a valued commodity developed by some organisms out of necessity to
access iron in times of limitations. Just like the butcher, brewer and banker, some microbes
developed an asset out of their own interest, and in the process provided a service to their
community that shaped their relationships and evolutionary dynamics. The economics of
siderophore production and use that are relevant to microbial growth, especially in relationship to
iron extraction out of minerals, are however poorly constrained. While it has been shown that
siderophores can significantly increase the dissolution rate of silicate mineral phases, mobilizing
iron from minerals (Rogers and Bennett, 2004; Hausrath et al., 2009; Torres et al., 2019), these
observations have all been made in abiotic experiments and at very high siderophore
concentrations (hundreds of micromolar to millimolar range), many orders of magnitude higher
than the low nanomolar to picomolar range measured environmentally (Kraemer, 2004; Mawji et
al., 2008; Velasquez et al., 2011; Boiteau et al., 2016). Meanwhile, a number of biotic experiments
have explored the role of live microbial cultures in increasing the dissolution rate of silicate
minerals and glasses (Liermann et al., 2000; Wu et al., 2007, 2008; Buss et al., 2007; Dehner et
al., 2011; Stockmann et al., 2012; Ferret et al., 2014; Ahmed and Holmström, 2015; Oelkers et al.,
2015; Perez et al., 2016; Harrold et al., 2018), though these have not isolated specific metabolic
processes. Thus, specific siderophore concentrations that are effective at promoting growth using
minerals as substrate are yet to be examined. Building on the experimental setup from Chapter 2
provides an opportunity to address this shortcoming; this approach and results are discussed here,
and the implications on microbial communities is further discussed.
41
3.1. Materials and methods
Additional batch reactor experiments were performed, identical in methods and setup as
those from Chapter 2 (section 2.1.2), though using MR-1 with amendments of the commercially
available siderophore deferoxamine mesylate (Sigma Aldrich, US), a mesylate salt of the
siderophore Desferrioxamine B. While this siderophore is a tris-hydroxamate, structurally
different from the cyclic di-hydroxamate siderophore produced by MR-1 (putrebactin; Ledyard
and Butler, 1997), desferrioxamine B was readily bioavailable to MR-1 and also readily available
commercially. A filter-sterilized deferoxamine stock solution was added to experimental flasks in
order to reach final concentrations of 0, 0.05, 0.2, 1, 5, 10, 25, 50 and 100 µM deferoxamine.
Experiments with concentrations of 0, 0.05, 1, 50 and 100 µM deferoxamine were performed in
triplicate and monitored over the entire growth. The other concentrations (0.2, 5, 10 and 25 µM
deferoxamine) were performed in duplicate and only measured for their final, stationary phase cell
densities (Fig. 5). Cell counts were performed as described in section 2.1.2.2.
3.2. Results
When MR-1 cultures were amended with the commercially available siderophore
deferoxamine at concentrations from 0.05 to 100 µM, they were able to grow, similar to the
cultures amended with MR-1 filtrate. In the experiments with commercial deferoxamine,
siderophore concentrations could be manipulated — and growth rate and stationary phase cell
density were found to be related to the concentration of siderophore added (Fig. 5). Providing the
deletion mutant with various concentrations of deferoxamine showed that at least ≥ 25 µM
deferoxamine was required to enable the deletion mutant to grow at rates and densities on par with
MR-1 (Figs. 1 & 5). In contrast, concentrations below 5 – 10 µM significantly limited growth
42
potential. Overall, stationary phase cell densities followed a logarithmic relationship with
siderophore concentrations (Fig. 5B).
Figure 5: MR-1Δ grown with deferoxamine amendments. (A) Growth curves for strain MR-
1Δ with increasing deferoxamine concentrations ranging from 0 – 100 µM, and (B) stationary
phase cell densities for cultures growing on olivine substrate, as a function of added deferoxamine
concentrations. Experiments with 0, 0.05, 1, 50 and 100 µM deferoxamine were performed in
triplicate and monitored over the entire growth phase (as shown in A, with stationary phase cell
density plotted in B), while the 0.2, 5, 10 and 25 µM deferoxamine experiments were performed
in duplicate and only measured for their final, stationary phase cell densities (additional data points
in B). Error bars represent one standard deviation on replicate analyses.
3.3. Discussion
3.3.1. Siderophore concentrations and growth
These results quantify siderophore concentrations that are biologically relevant to support
microbial growth when minerals act as sources of nutrients. Specifically, low- to mid-micromolar
concentrations of siderophores are, from a microbial perspective, sufficient to sustain very high
cell densities and intense microbial growth in what might otherwise be considered iron deplete
conditions. Siderophore concentrations ≥ 25 µM are substantially lower than those previously
reported as being effective in accelerating mineral dissolution by previous abiotic experiments,
43
typically in the range of hundreds to thousands of micromolar (Kraemer et al., 1999; Cocozza et
al., 2002; Yoshida et al., 2002; Cheah et al., 2003; Rogers and Bennett, 2004; Torres et al., 2019).
One reason for this difference likely relates to the repeated use and cycling of siderophore
molecules by actively growing microbial communities, making siderophores much more effective
at scavenging minerals than in abiotic environments. Additionally, lower siderophore
concentrations (˂ 10 µM) led to a muted growth of MR-1, and stationary phase densities were
significantly lower. Since the relationship between stationary phase cell densities and siderophore
concentration follows a logarithmic relationship (Fig. 5B), one could conceptually infer a notion
of “carrying capacity” from given siderophore concentrations, whereby at lower concentrations
(here, < 50 µM), the kinetics of siderophore cycling and iron acquisition from a mineral phase
would become an important factor limiting growth, and the activity of siderophores (i.e., their full
potential as iron harvesting systems) would be maximized.
Under this perspective, the spatial structure and emplacement of microbial communities
could prove very meaningful for understanding iron acquisition and primary productivity. In the
experiments conducted here, maximum cell densities were inhibited at a range of siderophore
concentrations (< 5 – 10 µM) that are still three to six orders of magnitude greater than siderophore
concentrations reported in field studies of ocean waters (Kraemer, 2004; Mawji et al., 2008;
Velasquez et al., 2011; Boiteau et al., 2016). Measurements of siderophore concentrations
dissolved in environmental media (i.e., ocean waters, ground waters, soil and sediment pore
waters) may therefore not be as meaningful as concentrations that might exist within micro- to
nanoscale environments along a mineral surface (such as within mineral-attached biofilms, or even
at the cell membrane-mineral interface along a bedrock or suspended particles). Such localized
environments likely provide a much more realistic setting for supporting micromolar siderophore
44
concentrations and, from the point of view of the bacteria, a more relevant scale of operations for
siderophore synthesis, cycling and uptake.
Indeed, it seems unlikely that microbial communities could thrive at environmental
densities (~ 10
5
cells/mL) merely on the dependence of passive diffusion of nano- to picomolar
concentrations of siderophores. In that sense, measurements of siderophore concentrations in
environmental media may more closely reflect siderophores “lost” from an active microbial
community from passive diffusion, rather than concentrations and scales relevant to effectively
support primary productivity. Numerical simulations have supported this idea, showing that the
effectiveness of planktonic siderophore diffusion as a means to acquire environmental iron was
mainly a function of parameters such as cell density, siderophore diffusion distances, binding
reaction rates, and siderophore excretion and decay rates. Considering all these conditions, the iron
acquisition through siderophore diffusion was found to be prohibitively inefficient, except for
microbial communities of relatively high densities (approx. 10
6
cells/mL and assuming all cells
can utilize all available siderophores) and siderophores had very low diffusion distances (high nm
to low µm ranges; (Völker and Wolf-Gladrow, 1999; Scholz and Greenberg, 2015; Leventhal et
al., 2019). These perspectives are also consistent with more recent work showing that the ability
of fungi to attach to olivine surfaces can heavily impact mineral weathering rates, compared to
fungi that are unable to attach to mineral surfaces (Gerrits, 2019). Together these observations
highlight the importance and relevant scales of distances in microbe-mineral interactions.
3.3.2. Community interactions
MR-1 was able to effectively utilize the amended deferoxamine (or in effect
Desferrioxamine B), a siderophore commonly associated with Actinobacteria (Roberts et al., 2012;
45
Ejje et al., 2013; Cruz-Morales et al., 2017), that is structurally different from putrebactin,
Shewanella’s endogenous siderophore. Putrebactin uptake by S. oneidensis occurs via a TonB-
dependent pathway encoded in the putA-putB gene cluster, a pathway shown to be critical in
putrebactin-mediated iron uptake. However, many uptake homologs for chelated-iron species have
been identified in S. oneidensis and are believed to promote the uptake of siderophores synthesized
by a range of bacteria (especially those from E. coli, B. subtilis, S. aureus and V. harveyi), as well
as ferric iron bound to α-hydroxyl acid ligands (such as lactate or citrate), independently of the
putrebactin uptake pathway (Liu et al., 2018). Structural differences between Shewanella’s
putrebactin (a cyclic di-hydroxamate) and the experimentally amended deferoxamine (a tris-
hydroxamate) can, under pH-neutral conditions, cause differences in ligand:metal ratios, whereby
di-hydroxamates have been documented to adopt a 3:2 ratio, and tris-hydroxamates a 1:1 ratio
(Ledyard and Butler, 1997; Dhungana and Crumbliss, 2005). Therefore, the siderophore
concentrations in deferoxamine amendment experiments may entail a slightly different dynamic
on iron mobility, compared to experiments with the endogenous siderophores of MR-1
(putrebactin).
These observations highlight that the range of chelated-iron receptor proteins is significant,
both across clades but also within single strains, which might be due to commonalities in iron
uptake pathways, often dependent on the generic TonB transport system, and ferric iron reductases
(Fennessey et al., 2010; Noinaj et al., 2010; Hartney et al., 2011). This insight furthers the
possibility that certain classes of siderophores might not strictly play exclusionary roles in iron
acquisition within microbial communities but might, in fact, be utilized and shared in a structured
relationship across various microbial phyla (Amin et al., 2012a; Kümmerli et al., 2014), potentially
involving beneficial associations (Amin et al., 2009a, 2012b). If, on the other hand, the shared
46
utilization of siderophores is based on a more antagonistic, “cheating” nature, this process might
prove to be very widespread across microbial communities, as S. oneidensis was shown to be able
to use a wide range of siderophores, even some of the strongest existing catechol siderophores
(i.e., E. coli’s enterobactin; Liu et al., 2018). This is further highlighted by the studies of
Pseudomonas communities which have shown strains incapable of producing siderophores,
though capable of their uptake, co-occurring with siderophore producing strains in natural
environments (Butaitė et al., 2017). In either case, models of siderophore diffusion and uptake
rates suggest that siderophore concentrations are only relevant to a cell within a radius of ~ 30 µm,
beyond which siderophores would be unavailable to that cell, lost to the environment—or other
opportunistic cells—through diffusion (Leventhal et al., 2019). Whether it be a cheating or sharing
relationship, the utilization of different siderophores across microbial clades is a well-documented
phenomenon. Therefore, by utilizing deferoxamine, an exogenous siderophore, as amendment to
MR-1Δ, the experimental siderophore concentrations documented here provide some first-order
quantitative constraints on siderophore concentrations that would be meaningful at the mid-
micrometer scale, within localized environments relevant to community interactions.
3.4. Conclusions
Some microbes looking to survive in a changing Earth developed unique abilities to acquire
a critical nutrient of limited access, and in the process, created not just ligands, but also value in
their synthesis: value that could benefit themselves as well as other organisms around them. Thus,
siderophore producers, whether willingly or not, provide a service to their community that can
define the nature of the interactions and associations between organisms, as well as evolutionary
dynamics in an ecosystem. To that effect, results from this study quantify some constraints on
47
biologically relevant siderophore concentrations that can impact mineral dissolution, microbial
growth, and offer further insights into the mechanics of how siderophores can influence microbial
community structure and dynamics.
48
Chapter 4: The emperor’s new clothes
In a prebiotic Earth, the terra-tons of carbon that are contained in the organic pool today
were present in the atmosphere as CO 2, and in the oceans as dissolved inorganic carbon. Thus, the
Hadean and Archean oceans would have been not only anaerobic but also of relatively low pH
(Lyons et al., 2014). Silicate materials are sensitive to pH and dissolve much more readily in acidic
or basic conditions (Brady and Walther, 1989), suggesting that Earth’s bulk crust was, in its early
eons, subject to higher weathering rates (Anbar and Knoll, 2002). This would have released large
quantities of metal cations commonly contained in silicate minerals into solution, creating
predominantly metal-replete ocean waters (Saito et al., 2003). Then came the GOE, imparting
major impacts on global CO 2 contents as well as ocean pH, alkalinity and redox (Anbar and Knoll,
2002; Lyons et al., 2014). Dissolved elements such as iron and manganese turned into minerals,
other minerals such as sulfides dissolved into sulfate, and microbial sulfate reduction turned that
back to sulfide. Thus metals that are more mobile under reducing conditions like iron were
sequestered from waters as oxides, and metals which might be more mobile under oxidizing
conditions like molybdenum were sequestered as sulfides. What followed would have been a
Proterozoic world both anoxic and sulfidic, starved of many metals critical to life functions, rich
in sulfide and lacking free molecular oxygen (Anbar and Knoll, 2002; Saito et al., 2003). All in
all, a very infertile environment for life to thrive in.
From a biological perspective, the loss of easy access to metals severely compromised the
ability to pursue fundamental functions such as photosynthesis, EET, nitrogen fixation, and in a
general sense, primary productivity (Lyons et al., 2014). From a mineral perspective, weathering
and dissolution rates likely decreased as ocean alkalinity and pH increased, and metal oxidation
promoted the formation of protective coatings on the surfaces of silicate minerals (Anbar and
49
Knoll, 2002). Specifically, iron oxidation documented in modern environments has been proposed
to have an impact on weathering rates by oxidizing the surface of silicate minerals and forming
insoluble hydrous ferric oxide coatings, inhibiting silicate dissolution (Schott and Berner, 1983).
Such coatings would need to be in the low nano-meter scale as they otherwise could not be detected
through SEM imaging (Figs. 2 & 4), occurring strictly at the mineral-water interface. If so, they
would be the component of minerals that interact directly with the aqueous solution, and thus at
the mineral-organic interface when sorbents are present. Here, we explored the fate, speciation and
mobility of iron at the surfaces of olivine minerals, and whether the microbial acquisition of iron
from silicates can be explained by siderophores removing Fe-oxides along mineral surfaces (as
previously proposed by Torres et al., 2019).
4.1. Methods
Analyses by X-Ray Photoelectron Spectroscopy (XPS) were used to characterize the
speciation of iron at the surface of mineral substrates. Experiments as laid out in Chapter 2 (section
2.1.2) were replicated, though focusing mainly on the wild type, the deletion mutant with no
amendments, and with abiotic controls as opposed to killed bacteria. Olivine grains were
recuperated after 50 hours incubation. Fresh olivine grains (prepared as in section 2.1.1.1) were
used as ferrous iron standard. Ferric iron standards were obtained by heating the same fresh olivine
grains in a furnace at 900⁰ C for 36 hours in order to completely oxidize its surface (Mackwell,
1992; Knafelc, 2010). These standards were specifically chosen over other available mineral
options as they would most closely match experimental silicate substrates, limiting potential
discrepancies in spectra due to differences in surrounding lattices. For consistency with SEM
imaging, all olivine grains analyzed under XPS underwent ethanol dehydration and critical point
50
drying (see section 2.1.1.2) before being mounted on carbon tape and a conductive silicon wafer.
XPS analyses were performed in an Axis Ultra DLD (KRATOS, USA), at CNI. A 45⁰ incident
monochromatic X-Ray beam (10 kV, 5 mA) scanned a target surface area of approx. 1 x 0.5 mm
in an analytical vacuum of approx. 2x10
-8
torr. Analyses were all performed in duplicate, with a
hybrid lens mode and charge stabilizer. Low resolution survey spectra were performed with a pass
energy of 160 eV, a step size of 1.0 eV, and a dwell time of 100 ms, with 10 sweeps. High
resolution elemental spectra were performed with a pass energy of 20 eV and a step size of 0.1 eV.
Fe 2p spectra underwent 35 sweeps with a dwell time of 158 ms, and O 1s spectra underwent 10
sweeps with a dwell time of 332 ms. Vamas data files were processed using CasaXPS software.
Due to the non-conductive nature of the olivine samples and the use of a charge neutralizer, spectra
calibration could not be performed effectively by the commonly used C 1s peak (Greczynski and
Hultman, 2017, 2020). Instead, the O 1s spectrum was used, with lattice peak set at 130.0 eV, a
method that has been consistent and effective for olivine substrates (Zakaznova-Herzog et al.,
2005, 2006, 2008; Yamashita and Hayes, 2008). Additionally, dissolved Fe(II) concentrations
were measured in experimental media from duplicate filtered samples (n = 2) of experiments with
MR-1 and MR-1, 45 h into the experiments, following an established ferrozine protocol (Myers
and Nealson, 1988) using a FLUOstar Optima (BMG Labtech, US) micro plate reader.
4.2. Results
In the XPS spectra, olivine minerals that had not been exposed to siderophores (i.e. abiotic
and deletion mutant samples) had a range of Fe 2p 1/2 peak position (709.37 – 710.30 eV), straddling
the ferrous and ferric iron binding energy end-members. On the other hand, olivine minerals that
had been exposed to very high concentrations of siderophores (i.e., MR-1 samples) consistently
51
showed spectra in the ferrous iron range. Furthermore, the spectra of samples from the MR-1
experiments displayed some of the poorest signal-to-noise ratio (Fig. 6).
Figure 6: XPS spectra of iron on olivine materials of duplicate experiments. Spectra from
minerals that were incubated abiotically and with MR-1Δ suggest a range of mixed iron valence
states, while minerals incubated with MR-1 clearly show ferrous iron spectra. “Olivine std” spectra
are fresh olivine surfaces as Fe(II) silicate standards, “Olivine Ox std” spectra are oxidized olivine
surfaces as Fe(III) standards.
4.3. Discussion
While microbial growth data presented in Chapters 2 and 3 clearly show that siderophores
enable microbial iron acquisition directly from a mineral phase, XPS results provide further
insights into the precise mechanism involved in iron mobilization. The range of mineral-surface
iron oxidation states displayed in the XPS spectra of samples from the abiotic and MR-1Δ
52
experiments suggests a significant degree of ferric oxide coating from abiotic weathering within
50 hours (Fig. 6). The formation of this coating would be consistent with previous literature
suggesting that the oxic weathering of fayalite could cause a nano-meter thin coating of insoluble
ferric hydroxides at the mineral surface, protecting it from further dissolution (Schott and Berner,
1983).
On the other hand, the minerals from the MR-1 experiments display an entirely ferrous
spectrum, as well as a poor signal-to-noise ratio (Fig. 6), suggesting a lack of iron oxide coating at
the mineral surface, and even potentially an iron-deplete silicate surface layer. Structurally,
siderophores act as excellent nucleophiles (Dhungana and Crumbliss, 2005), which predisposes
them to sorb onto metal cationic centers on mineral surfaces. Sorption of organic ligands can
destabilize the electronic structure of electrophiles (in our case metal cation centers) out of a crystal
lattice (Kraemer, 2004; Otero et al., 2017). Given the extreme affinity constants of siderophores
for ferric iron (logK f = 25 – 49) and their known ability to destabilize ferric oxide minerals
(Reichard et al., 2007; Hider and Kong, 2010), it follows that siderophores would destabilize the
protective ferric oxide coatings and in the process accelerate silicate dissolution. Such a coating is
believed to be low-nanometer scale in thickness as SEM images (best resolution of 0.5 – 1 µm)
collected in this study do not reveal any evidence of secondary oxide coatings on reacted olivine
surfaces, even though they clearly show scouring and weathering textures (Figs. 2 & 4). On the
other hand, XPS (surface penetration of 2 – 10 nm) analyses do detect changes in iron oxidation
between experiments. These results are consistent with recent transmission electron microscopy
analyses on olivine dissolution, showing nanometer layers of amorphous ferric oxides on
weathered olivine surfaces (Gerrits, 2019). Thus, siderophores can effectively scavenge protective
ferric oxide layers, mobilizing ferric iron into a soluble and bioavailable form, promoting the
53
abiotic oxidation of olivine minerals and concurrent dissolution. Such dissolution would likely
occur along and exacerbate weaknesses in the mineral lattice, further deepening etch pits (Figs. 2
& 4; Buss et al., 2007).
These experiments were all performed aerobically, conditions in which Shewanella’s Mtr
pathway is repressed (Fredrickson et al., 2008), indicating that microbial iron reduction likely did
not play a role in iron speciation in these experiments. Additionally, ferrozine assays performed
on experimental solutions yielded measurements below the detection limit, further supporting this
idea. This is consistent with previous studies showing that siderophores were not involved in
Shewanella’s dissimilatory iron reduction process (Fennessey et al., 2010). Similarly, if the Mtr
pathway played an important role, it might be expected that the MR-1Δ strain would also be able
to mobilize iron out of the oxidized coating, and thus acquire iron for growth despite a lack of
siderophore, a scenario contrary to the measured observations.
Lastly, it is still possible, at least hypothetically, that the simple sorption of siderophores
on the ferrous iron present in an olivine mineral could weaken the native crystal lattice. While
siderophores are best known for their affinities to ferric iron, their affinities for ferrous iron stand
in the same range as hydroxide and EDTA ligands (logK f = 14 and 23, respectively; Dhungana
and Crumbliss, 2005; Hider and Kong, 2010). It is thus conceivable that siderophores could still
destabilize ferrous iron directly out of a fayalite structure, exacerbating the process started with
the removal of the oxide coating.
4.4. Conclusions
These results indicate that secondary weathering phases, such as nanometer-scale metal
oxide layers, can form at mineral surfaces and play critical roles in controlling mineral dissolution,
54
metal mobility and microbial growth. Without the help of siderophores, bacteria are incapable of
acquiring enough iron to grow, as insoluble iron oxide coatings prevent the further release of iron
into solution. Siderophore synthesis, on the other hand, mobilizes the ferric iron into a bioavailable
form, removing the protective coat off silicates and promoting mineral dissolution.
With the GOE, life profoundly transformed the planet’s chemistry, impacting the mobility
of elements to the detriment of further biological growth, as nutrients like iron oxidized and
prevented the weathering of primary minerals. Ostensibly in response to this challenge, life
developed siderophores as a means to gain valuable metals directly out of minerals and acquire
nutrients to build biomass. In the process, life effectively gained the ability to denude silicates, and
generally crustal materials, by removing their coat and exposing them naked to the world. A naked
crust could then dissolve and erode faster, releasing not only metals to sustain life, but also silicate,
significantly raising alkalinity and further altering the planet’s chemistry, with impacts on the
carbon cycle, crustal weathering and climate systems.
55
Chapter 5: The hare and the tortoise
Crustal weathering is of particular importance in understanding Earth systems, as silicate
anions, when released from mineral phases into solution, act as a very effective proton sink. The
weathering of silicate-rich crustal materials therefore increases an environment’s alkalinity
considerably and drive atmospheric CO 2 sequestration. This is a process which, on geological time
scales, binds global carbon fluxes with crustal weathering, each acting as negative feedback
mechanism on the other. High CO 2 contents decrease pH and increase global temperatures, effects
leading to increased weathering rates and silicate dissolution, which in response sequesters carbon
out of the atmosphere, establishing a buffered equilibrium (Berner et al., 1983; Ruddiman, 1997).
This feedback system has profound implications in understanding Earth’s evolution and history.
In an early Earth, the onset the GOE would have not only led to metal oxide precipitates coating
silicates, hindering their weathering, but oxygenic photosynthesis would have decreased CO 2
contents, further reducing mineral dissolution rates and nutrient mobility and primary productivity.
In more modern settings, this very process has been proposed for use in geoengineering purposes
in an attempt to mitigate anthropogenic climate change. By disseminating olivine minerals in
environments that promote mineral dissolution (e.g., wet tropical basins), silicate weathering could
be enhanced, providing a means to sequester excess carbon from the atmosphere (Hartmann et al.,
2009, 2013; Köhler et al., 2010, p.; Montserrat et al., 2017).
Previous chapters of this thesis have clearly demonstrated that siderophores enhance iron
mobility and bioavailability out of olivine (Chapters 2 and 3), denuding it of a protective coating
(Chapter 4). It follows that these effects lead to an increase in mineral dissolution rates. This effect
has been documented in the literature, though mainly through abiotic experiments which document
increasing siderophore amendment concentrations leading to increases in mineral dissolution rates
56
(Kraemer et al., 1999; Cocozza et al., 2002; Yoshida et al., 2002; Cheah et al., 2003; Hausrath et
al., 2009; Torres et al., 2019). While such experiments isolate the effects of a single metabolite on
mineral dissolution rates, they do not consider the effect that live microbial iron acquisition and
growth has on siderophore activity. Abiotic experiments also often test for siderophores in the high
micromolar to low millimolar concentrations, many orders of magnitude greater than those
observed in natural environments (Mawji et al., 2008; Velasquez et al., 2011; Boiteau et al., 2016).
Conversely, biotic studies examining siderophore-driven mineral dissolution either have not
quantified the effects on dissolution rates (Hersman et al., 1996, 2000, 2001; Maurice et al., 2000,
2001; Ferret et al., 2014), or have not isolated a specific metabolite or biochemical process (Wu et
al., 2007, 2008; Stockmann et al., 2012; Ahmed and Holmström, 2015; Oelkers et al., 2015;
Harrold et al., 2018).
Here, we examine the effects of a single metabolite, siderophores, on the dissolution rates
of olivine in both biotic and abiotic experiments. This comparison sheds light on the effective use
of siderophores and their full potential for increasing mineral dissolution rates when utilized by
iron-limited bacteria — exploring the basic question of whether more haste does indeed mean less
speed: if siderophores have a greater effect in bulk quantity, or through quality in their effective
use.
5.1. Materials and methods
Batch reactor experiments were further performed, almost identical in methods and setup
as those from Chapters 3 (section 3.1), using MR-1 with amendments of the commercially
available siderophore deferoxamine mesylate (Sigma Aldrich, US), as well as abiotic experiments.
The differences were that these experiments were performed in 100 mL volumes, and with higher
57
initial cell amendments for biotic experiments of 5x10
10
cells (for a starting cell concentration of
5x10
8
cells/mL). These differences made it possible to accommodate larger sample volume
requirements needed for geochemical analysis, and to achieve maximum cell concentration as
rapidly as possible and minimize potential effects of exponential growth on mineral dissolution
rates — thus focusing on dissolution rates at constant cell densities. 1 mL samples were extracted
from biotic experiments and measured for cell concentration to monitor growth, and 4 - 5 mL
samples were extracted from the experiments for elemental analyses once exponential growth had
ended and cell counts were near constant in stationary phase. Abiotic experiments followed the
exact same setup, minus the addition of cells.
For biotic experiments, cell concentrations were measured via OD (as described in chapter
2, section 2.1.2). All samples collected for elemental analyses were filtered (0.2 µm), then acidified
with the addition of HNO 3 to a final concentration of ~ 0.5%, and preserved in the dark at 4° C
until analysis. Dissolved silicon concentrations were measured by Inductively Coupled Plasma -
Optical Emissions Spectrometry (ICP OES, model 5110, Agilent, US), at emission wavelengths
of 251.611 nm as well as 288.158 nm. Dissolved magnesium concentrations were also measured
by ICP OES, at emission wavelengths of 279.553 nm. Silicon concentrations during cell stationary
phases were used to infer mineral dissolution rates as silicon, unlike magnesium or iron, does not
have any biological function to Shewanella and is not a component of the minimal medium.
5.2. Results
As highlighted in chapter 3, MR-1Δ achieved a range of maximum cell densities given
varying deferoxamine amendments, fitting within the range of MR-1 growths (both live and killed,
Fig. 7A). Silicon concentrations in these biotic experiments increased at rates in direct correlation
58
to the deferoxamine amendments, and also within the limits of MR-1. Most notably, these
increases in silicon concentrations during stationary phase—and associated slopes from linear
regressions—were significantly greater than those from abiotic experiments at equal deferoxamine
concentrations (Fig. 7B & C).
Ratio of Mg/Si were calculated to track the stoichiometry of dissolution between biotic and
abiotic experiments. Specifically, ΔMg/Si ratio were considered, i.e.: the net change of Mg
concentrations, calculated by subtracting the original minimal medium Mg concentration (120
μM) from the measured experimental Mg concentration. These trends show decreasing ratio as a
function of increasing siderophore amendment. This was most pronounced with MR-1 and MR-
1Δ at 50 μM amendments experiments, while 0.05, 0 μM amendments and killed control
experiments showed overlapping results with abiotic experiments (Fig. 7D & E).
59
Figure 7: Comparing olivine dissolution rates in biotic and abiotic experiments. (A) Cell
counts for biotic experiments (MR-1 live and killed, and MR-1Δ with deferoxamine amendments).
(B & C) Si concentration measurements for biotic and abiotic experiments. (C & D) Ratio of net
Mg concentration changes over Si concentrations for biotic and abiotic experiments. Sampling and
measurements were performed upon onset of the experiment and during stationary phase of cell
growth in order to constrain mineral dissolution to constant cell densities. Dissolution rates were
inferred by the slope of the linear regressions. Error bars represent one standard deviation on
triplicate analyses.
60
Olivine dissolution rates were inferred from the slopes of regression lines on silicon
concentration and normalized to mineral surface area. Total surface area of olivine grains in these
experiments was calculated by assuming a perfect spherical shape of grains and a normal
distribution of grain sizes (more details in Appendix B). Normalized dissolution rates of MR-1Δ
at 50 μM amendments matched those from MR-1, consistent with previous observations of
maximal effective deferoxamine concentrations (Fig. 8, section 3.2). Normalized dissolution rates
were consistently greater in biotic experiments than abiotic ones. In biotic experiments that
displayed low or nonexistent cell growth (i.e., killed controls and MR-1Δ experiments of 0 and
0.05 μM deferoxamine amendments), dissolution rates were still close to those from abiotic
experiments. In experiments with pronounced cell growth (MR-1Δ experiments of 1, 50 μM
deferoxamine amendments, as well as MR-1 experiments), normalized dissolution rates were
greater than those of abiotic experiments by a factor of 3.0 to 6.3 (Fig. 8). The effective ratio of
dissolution rates between biotic and abiotic experiments was quantified by comparing the slopes
of linear regression lines through data of changing rates with siderophore concentrations, as this
approach compares net evolving rates avoids biases from y-intercepts. These revealed that biotic
experiments increased dissolution rates by an order of magnitude (Fig. 8).
61
Figure 8: The impact of microbial activity on olivine dissolution rates. Comparison of surface
area-normalized dissolution rates between biotic and abiotic experiments. Dotted lines are 95%
confidence levels of regression lines, vertical bars represent standard error associated with
regressions, and shaded areas are the standard errors associated with the dissolution rate values of
live and killed MR-1 experiments.
5.3. Discussion
5.3.1. Dissolution rates
The rates of silicon release clearly increase with increasing siderophore concentrations,
indicating that higher siderophore concentrations lead to higher dissolution rates, both in biotic
and abiotic experiments (Fig. 7B & C). While such conclusions for abiotic experiments are not
new, they do become meaningful when comparing between biotic and abiotic results, as they
quantify the impact of microbial activity on the dissolution rates. Comparing the slopes of
regression lines of dissolution rates as a function of siderophore concentration (Fig. 8) reveals that
dissolution rates are greater in live experiments than in abiotic ones by a ratio of 10.1 ± 1.5,
highlighting that at equal siderophore concentrations, bacteria can increase their impact on mineral
62
dissolution rates by an order of magnitude. This observation can be explained by the effective
recycling and reuse of siderophores by bacteria, an effect documented and recognized in the
biochemical literature (Raymond et al., 2003, 2015; Schalk and Guillon, 2013), but missing from
many studies that have explored siderophore-driven mineral dissolution rates with abiotic
experiments (Kraemer et al., 1999; Cocozza et al., 2002; Yoshida et al., 2002; Cheah et al., 2003;
Rogers and Bennett, 2004; Reichard et al., 2007; Torres et al., 2019). Such abiotic studies might
therefore provide results unrepresentative of active processes in natural environments, missing a
key component of siderophore dynamics.
Perhaps this increased dissolution rate value of 10.1 ± 1.5 due to microbial activity might
best understood as not only allowing the recycling and reuse of siderophores, but also improving
the efficiency and kinetics of the sorption of ligands onto mineral surfaces as well as the shuttle
transport of iron. As previously discussed in section 3.3, deferoxamine concentrations < 50 µM
have, in the experiments performed for this thesis, led to maximum cell densities lower than those
from MR-1 experiments, indicating that, at those concentrations, the full potential of siderophores
had been reached. Therefore, it seems likely that, at lower siderophore concentrations, the factor
limiting cell growth might relate to the kinetics of iron extraction from minerals (i.e., ligand
exchange reaction along the mineral surface), as well as diffusion between minerals and cells.
These reaction dynamics clearly happen in both biotic and abiotic experiments. Nonetheless, the
much-enhanced dissolution kinetics in live experiments (Fig. 8) tell us that these dynamics are
accelerated by an order of magnitude in the presence of viable cells compared to abiotic
experiments — demonstrating that cells do not merely depend on the passive diffusion and
reactions of siderophores, but in fact act as catalysts, enhancing their activity not merely in terms
of number of uses, but also in terms of their reaction and diffusion kinetics. This effect might relate
63
to bacteria location, whereby cells close to, or even bound to mineral surfaces, depend on much
lower diffusion distances, accelerating siderophore shuttling, as well as potentially enhancing
ligand exchange reaction rates along mineral surfaces through uncharacterized mechanisms.
This interpretation therefore calls for further caution in interpretation regarding
measurements of siderophores in natural environments. In open ocean waters, siderophore
concentrations have been measured at high picomolar to low nanomolar levels (Mawji et al., 2008;
Velasquez et al., 2011; Boiteau et al., 2016), concentrations that are certainly low in absolute terms.
However, their impact on mineral dissolution rates and elemental mobilization might still be
significant as biological activity will enhance their effects considerably by reusing siderophores
many times and increasing their activity significantly, giving them an effective activity much
greater than their measured concentrations might otherwise suggest. This holds particularly true if
we consider that siderophore concentrations in environmental media might in fact reflect
siderophores “lost” from active communities, and that concentrations within micro-environments
(such as biofilms) are likely higher, and more relevant to microbial activity. Such micro-
environments would be consistent with the modeled effectiveness of siderophore diffusion and use
within communities, with limits being higher cell densities and low micrometer distances (Völker
and Wolf-Gladrow, 1999). In these environments, the controls on mineral dissolution rates might
relate more to the kinetics of siderophore cycling and use by microbes, or proposed “carrying
capacity” (as discussed above, section 3.3) than their absolute concentrations.
5.3.2. Nutrient uptake and dissolution stoichiometry
Ratios of ΔMg/Si in biotic experiments correlate negatively with cell density (Fig. 7D), a
trend that can be attributed to cell growth, leading to both high Mg uptake and effective iron
64
acquisition causing mineral dissolution and the release of silicon. The lowest ΔMg/Si values seem
to be achieved around 25 – 30 hours into the experiments, after which the values start rising again,
a trend that could attributed to decreasing Mg uptake coupled with continuous mineral dissolution
and coupled Mg and Si release throughout the stationary phase. Biotic experiments with low cell
densities on the other hand show near constant ΔMg/Si ratio with values overlapping with those
from abiotic experiments (Fig. 7E). The range ΔMg/Si values for these latter experiments is 1.2 –
1.7, which are near the stoichiometric Mg:Si ratio of 1.76 measured on these olivine minerals
(Mg1.76Fe 0.24SiO 4, section 2.2.1). The killed control experiments do show the highest, and
relatively stable, ΔMg/Si values of 3 – 4 (Fig. 7D), which might be attributed to the release of cell-
bound Mg into solution after autoclaving the large number of cells (5x10
10
) injected into the
experiments.
5.4. Conclusions
While siderophores clearly accelerate silicate dissolution, bacterial activity can further
magnify their impact on dissolution rates by as much as an order of magnitude, increasing the
activity of siderophores and maximizing their potential for transferring iron from the mineral to
the biological pool. Old fables teach us that more haste might mean less speed. Assuredly, high
siderophore concentrations are to little avail if they are not used to effect. The recycling of
siderophores by microbes trades haste for effectiveness, such that even at seemingly low
concentrations, siderophores can still have high impacts on silicate dissolution. With an economy
of scale, such a process can define the weathering of Earth’s entire crust, and alter a planet’s
chemistry, carbon fluxes and climate systems. A thorough mechanistic understanding of processes
65
at the organic-inorganic interface could therefore be instrumental in helping design and implement
potent geoengineering practices, addressing modern planetary-scale challenges.
66
Chapter 6: Red Mars
It is worth contemplating that the Anthropocene might end up leaving a large imprint on
worlds far beyond the Earth. In fact, the same fundamental processes at the organic-inorganic
interface that could help address current man-made problems on this planet could also be harnessed
to allow the human exploration and settlement of other planets.
The logistical and time constraints inherent to interplanetary travel render any human visit
to Mars a deeply committing and isolating enterprise, punctuated at best by very sporadic resupply
events. The rapid development of self-sufficiency is therefore critical to any human presence, and
is highly dependent on effective in situ resource utilization (ISRU) systems. Oxygen will be one
of the most fundamental necessities for human presence on Mars, both to sustain human life and
provide a propellant for spacecrafts. Efforts to produce oxygen in a Martian setting have so far
mainly relied on purely abiotic, energy demanding engineered systems. Specifically, the Mars
oxygen ISRU experiment, or MOXIE, makes use of atmospheric carbon dioxide to generate
oxygen gas. This process not only has very high energy requirements, but also produces large
amounts of carbon monoxide as a by-product (Meyen et al., 2016), constraints that may not be
viable for supporting human crews, let alone entire communities. Alternative ISRU systems could
also potentially be used for oxygen, particularly with artificial photosynthesis, where light-
sensitive metal catalysts are used to produce desired by-products. Such systems have however so
far been mainly used to generate fuels such as hydrogen or carbohydrates using water or CO 2 as
substrates (McConnell et al., 2010; Li et al., 2014; Greenblatt et al., 2018). While water splitting
powered by artificial photosynthetic can generate molecular oxygen, the process consumes water,
using up another scarce resource in the context of Mars.
67
However, Martian topsoils contain copious amounts of globally distributed, poorly
crystalline nanophase ferric minerals, which include hematite, maghemite, magnetite, goethite,
jarosite and schwertmannite (Bell et al., 2000; Morris et al., 2004, 2006b). Large, coarsely
crystalline hematite deposits and spherules have also been found exposed in a number of surficial,
water-altered bedrock formations across the planet, with hematite accounting for up to 45 vol.%
of some outcrops (Christensen et al., 2000, 2001; Christensen, 2004; Fergason et al., 2014; Lanza
et al., 2016). Moreover, wind-borne magnetic dust, the matrix of global year-long Martian storms,
is in large part composed of magnetite, hematite and goethite, further contributing to a widespread
geographical distribution of nanocrystalline (~ 3 µm) ferric oxide particles (Hamilton et al., 2005;
Goetz et al., 2005). These mineral phases contain an abundance of oxygen which, if liberated and
transformed, could provide an immensely abundant source of O 2.
Microbial EET enables bacteria to fuel their metabolism by transferring electrons across
cell membranes to extra-cellular substrates, including highly insoluble mineral phases such as iron
or manganese oxides (Nealson and Saffarini, 1994). EET-driven metal oxide reductive dissolution
implies the congruent release of oxygen from the crystal lattice, the fate of which following mineral
dissolution is poorly characterized. Most EET-dependent bioelectric reactors have been developed
using anodes as the terminal electron acceptor and soluble organic carbon as the electron donor
(Nealson, 2017). Nevertheless, some bacteria have been shown to be able to use cathodes as
electron donors (using oxygen as an electron acceptor), through the reversal of electron flow in the
well-known Mtr respiratory pathway (Rabaey and Rozendal, 2010; Rosenbaum et al., 2011; Ross
et al., 2011; Rowe et al., 2015), further broadening potential applications in biotechnology. Some
research has been done in the application of microbial biocathode oxidation, though the focus has
been mainly on the production of hydrogen gas or organic molecules through the reduction of
68
protons, carbon dioxide, water or oxygen, processes generally referred to as microbial
electrosynthesis (Rabaey et al., 2008; Logan et al., 2008; Jeremiasse et al., 2010; Nevin et al.,
2010; Zhao et al., 2018). However, no work yet has studied the microbial coupling of two insoluble
substrates, specifically cathode oxidation and ferric oxide reduction. The objectives of this study
were therefore to (a) test the bacterial metabolic capacity and growth potential involved in the
electron transfer between two solid, extracellular substrates, and (b) in the process, develop and
substantiate a proof of concept for transformative ISRU biotechnologies that could greatly enhance
our capabilities of the human exploration of Mars by taking advantage of bacterial EET.
6.1. Materials and methods
6.1.1. Apparatus
We developed a custom-built, two-chambered, three-electrode bioelectrochemical reactor
in which a reductive chamber coupled microbially-mediated cathode oxidation and iron reduction,
and a reductive chamber oxidized water and generated molecular oxygen (Fig. 9). The chambers
were separated by a Nafion XL proton selective membrane (PSM, Fuel Cell Store, College Station,
TX). The reductive chamber contained a working electrode made of indium tin oxide (ITO)-coated
glass (~ 6.5 cm
2
, SPI supplies, West Chester, PA) and a custom-built reference electrode of Ag-
Ag/Cl in saturated KCl solution, both connected to a potentiostat (eDAQ Inc., Colorado Springs,
CO), with a working potential set at -351 mV vs. standard hydrogen electrode (SHE). The
reductive chamber further contained 0.1 g of crushed (< 63 µm), UV-sterilized red ochre hematite
(Ward’s Science, Rochester, NY). In biotic experiments, this chamber was populated by a Cyo-A
gene-deletion mutant strain of Shewanella oneidensis MR-1 ΔCyo-A, chosen for its EET
capabilities and reduced oxygen metabolism (Bretschger et al., 2007). The net reaction proposed
69
to have been carried in the reductive chamber is shown in equation 1, with electrons being
introduced into the reactor by the ITO cathode, creating a reductive and alkaline environment.
Fe 2O 3(s) + 2e + H 2O +2H
+
= 2Fe
2+
(aq) + 4OH
-
(1)
The oxidative chamber contained autoclaved 18.2 MΩ.cm water, a 100 µm dissolved
oxygen (DO) microsensor (Unisense, Aarhus, Denmark), and a custom-built counter electrode
made from titanium and platinum wires (also connected to the potentiostat). This way the oxidative
chamber provided an oxidative and acidic environment, creating molecular oxygen through the
proposed equation 2.
H 2O =
1
2
O 2 + 2e + 2H
+
(2)
Figure 9: Design of the two-chamber bioelectrochemical reactor. A proton-selective membrane
separates an oxidative abiotic chamber containing only water, from a reductive biotic chamber in
which microbially-catalyzed iron reduction is coupled with cathode oxidation. A proposed reaction
stoichiometry is highlighted. Chambers were custom-made with walls of borosilicate glass, with
ports made using rubber septa (see Methods).
Combined, these reactions created gradients in electric potential as well as alkalinity across
the two chambers of the bioreactor. This system promoted molecular oxygen production through
70
a stepped reaction, whereby microbially-catalyzed iron reduction through cathode oxidation was
balanced by the oxidation of water at the counter electrode. The PSM allowed the diffusion of free
protons while blocking the flow of reactants and microbes between chambers. All experimental
reactors were custom-built at the glass shop of USC from purchased borosilicate stock glass
components (Ace Glass Inc, Vineland, NJ). Moieties of the 2-chambered reactor were held
together by a clamped flange, sealed by a pair of custom-built Viton fluoroelastometer chemical-
resistant rubber sheet O-rings (McMaster-Carr, Santa Fe Springs, CA). The PSM was cleaned and
prepared following standard procedures prior to use (Chae et al., 2008), then autoclaved. Sampling
ports and the counter electrode were placed through crimped butyl rubber septa. To ensure airtight
connections, the working and reference electrodes were emplaced in a rubber stopper and sealed
with high temperature airtight silicone gasket as well as with high vacuum grease. All apparatuses
were confirmed airtight through continuous DO readings in static tests, confirming DO readings
were not affected by atmospheric contamination. All reactor chamber components were sonicated
in 10% liquinox, acid-washed, and autoclaved prior to use. Electrodes were washed in acetone and
70% ethanol prior to autoclaving and use. DO microsensors were acid washed and sterilized with
70% ethanol prior to use. Glass, rubber and electrode components were assembled and then
autoclaved for sterilization.
6.1.2. Culturing conditions
Preparing for biotic experiments, a single colony of ΔCyo-A was grown at 30º C in an
oxygenic, 18 mM lactate minimal medium (Bretschger et al., 2007, 2010) to an OD of ~ 1.0. To
induce a preliminary cell attachment to the working ITO electrode and promote electro-active
pathways, 1 mL of this culture was triple-rinsed and injected into a 3-electrode anode reducing
71
chamber containing the same 18 mM lactate minimal medium, kept anoxic through constant N 2
purging and a set potential of 699 mV vs. SHE.
After approximately 24 hours of lactate-oxidizing, anode-reducing growth, the working
and reference electrodes were removed from this chamber, rinsed with sterile experimental
minimal medium, and introduced into the biotic chamber of the 2-chamber reactor, thus
transferring biofilm-attached cells into the experimental reactor. Additionally, a set number of cells
were extracted from the high density, aerobic, 18 mM lactate medium growth flask, triple-rinsed
in sterile minimal medium, then injected in the biotic chamber of the bioelectrochemical reactor
in order to reach the desired experimental densities (10
7
and 10
8
cells/mL). The minimal medium
in the biotic chambers was the same as described above, but lacked both a carbon source and a
buffer, and had a starting pH of 8. These choices were designed to force cells to use the cathode
as an electron donor, promoted cross-PSM proton transfer, and removed a significant proton sink
(i.e., the buffer) that could otherwise impact alkalinity gradients. Anthraquinone-2,6-disulfonate
(AQDS, final concentration of 100 µM, below known toxic concentrations (Shyu et al., 2002;
Hong et al., 2007) and hematite were introduced into the biotic chamber, parts of the latter settling
directly onto the biofilm-supporting working electrode. Both chambers of the reactor were then
purged with filtered (0.2 µm) N 2 gas for up to three hours, with DO readings confirming stable
anoxic conditions. The experiments began when initial N 2 purging ended, and the cathodic
potential was set (-351 mV vs. SHE) on the working electrode. Abiotic experiments were
performed by simply introducing autoclaved electrodes into a sterile experimental reactor,
bypassing all preparatory cell growth and injection steps, and with both chambers containing 0.3%
formaldehyde (from a pH-neutral concentrate). The use of formaldehyde was deemed necessary
72
due to contamination risks previously encountered from the large number of manipulations
involved in assembling the experimental apparatus.
Purging of the biotic chamber with filtered high purity N 2 gas (~ 40 cc/min) was performed
~ 20 hours into the experiments as a means of mixing the medium in the biotic chambers, and to
test the added effects of increased medium flow across the working electrode and PSM, as AQDS
reaction rates are known to be affected by flow rates and microbial fuel cells have been shown to
be rate-limited by proton diffusion across the PSM (Liu et al., 2007; Bretschger et al., 2007, 2010).
Single-chamber experiments were prefaced by the same steps of lactate-oxidizing, anode-reducing
growth to induce biofilm attachment onto the ITO electrode, though no planktonic cells were later
added to achieve any desired cell density, and no AQDS was added to the experiments (Fig. C1,
appendix C).
6.1.3. Microscopy and activity
Samples were extracted from the biotic chamber during experiments, with aliquots
separated for total cell count using acridine orange (1 mg/mL stock) to quantify total planktonic
cells. Redox sensor green (RSG) stain, prepared as specified by the manufacturer (Molecular
Probes, Fisher Scientific., Waltham, MA) was also used to quantify cell counts based on activity
of the electron transport chain, both in the planktonic phase (throughout the experiments), and on
the working electrodes (at the end of the experiments). Images and counts were performed under
a Zeiss Axio optical microscope. Fe(II) concentrations were measured from duplicate sample
aliquots (n = 2) of the biotic chamber by ferrozine assays. pH measurements were performed at
the end of experiments using BDH narrow range strips (Prosource Scientific, Ottawa, ON,
Canada). Bioelectrochemical activity in the form of electrical current was monitored and recorded
73
at 60 second intervals with the eDAQ potentiostat. DO concentrations were measured every 60
seconds using a microsensor connected to a multichannel amplifier/multimeter (Unisense, Aarhus,
Denmark).
SEM secondary electron and EDAX imaging were performed on electrode materials
following experiments with a Nova NanoSEM 450 (FEI, Hilsboro, OR) at the Core Center for
Excellence in Nano Imaging at USC. Working distances of 5 – 10 mm and an accelerating voltage
of 10 kV were used. All samples were prepared using a 7-step ethanol dehydration process, fixed
with a critical point dryer (Tousimis, Rockville, MD), and coated with a Pt/Pd Cressington 108
sputter-coater.
6.2. Results
Cathode oxidation-hematite reduction experiments were performed with S. oneidensis cells
injected in the reductive chamber at concentrations of 0, 10
7
and 10
8
cells/mL. The minimal growth
medium was notably unbuffered, pH 8, and contained 100 µM anthraquinone-2,6-disulfonate
(AQDS), with the abiotic control experiments also containing 0.3% formaldehyde. Oxidative
chambers merely contained sterile 18.2 MΩ.cm water (i.e., even in the biotic experiments, the
oxidative chamber remained abiotic). All experiments showed stabilized electric current
production within the first few hours to 1 – 4 µA, with biotic experiments showing measurable
dissolved oxygen (DO) concentrations (2 – 5 µM) in the first 20 hours. Current production
increased by an order of magnitude (up to 40 µA) in all experiments as soon as mixing of the
reductive chamber began, corresponding with increased DO readings (5 – 25 µM) for the biotic
experiments. No DO was recorded in the abiotic experiments. Dissolved iron concentrations
74
increased significantly above background by the end of the experiment with 10
8
cells/mL (Fig.
10).
Figure 10: Experimental results showing electric current and oxygen production. Also shown
are cell counts and iron oxide dissolution. Experiments were injected with (A) 10
7
cells/mL, (B)
10
8
cells/mL, and (C) no cells in the biotic chamber. All experiments were performed at -351 mV
vs. standard hydrogen electrode (SHE). While mixing the reductive chambers increased both
75
electric current and oxygen production, the magnitude of oxygen production is a function of cell
density. This suggests that the presence of electro-active cells is critical in generating oxygen.
While the starting pH values were very different in the two chambers of the experiments
(growth medium pH = 8, 18.2 MΩ.cm water pH = 5.2), all experiments showed equal pH values
in both chambers by the end of the experiments, with the biotic experiment chambers ending at
pH 7.4 – 7.6, and the abiotic control chambers ending at pH 5.0 – 5.2. Planktonic cell densities
remained constant and close to the intended values, with RSG fluorescent cells accounting for 25
– 50% of total planktonic cell counts in both biotic experiments, confirming that bacteria had active
electron transport chains (Fig. 10). RSG-staining of the working electrode also showed pervasive
colonization of electro-active cells on hematite particles, with a density of ~ 10
5
cells/mm
2
by the
end of the biotic experiments. Scanning electron microscope (SEM) imaging further confirmed
working electrodes populated by ΔCyo-A with nanowire-like structures connecting ferric oxide
minerals and the electrode (Fig. 11). These results are consistent with a bioelectrochemical system
that supports the coupling of cathode oxidation and iron reduction facilitated by microbial electron
transport.
76
Figure 11: Microscope images of the working electrode in the biotic experiments. (A) SEM
image showing Shewanella cells (rod-shaped objects) connecting iron oxides (sharp-edged light
grey-white objects) with the working electrode (the flat grey background) via nanowire-like
structures. (B) Optical microscope image of cells stained with RSG, attached to the working
electrode, and (C) having colonized hematite particles. Each individual fluorescent point
represents a single redox-active cell. The latter shows the same area as (B) on the working
electrode, though with a focal point higher up the z axis. All scale bars are 5µm.
6.3. Discussion
Together, the results highlight a novel non-photosynthetic oxygen production pathway
catalyzed by the microbial transfer of electrons from a cathode to ferric oxide minerals. Cell
density seems to play an important role in the magnitude of electric current production, and more
importantly, strongly affects DO production. Keeping a well-mixed biotic chamber also
significantly increased both current and DO production. This is likely due to increased medium
77
flow across the working electrode and the PSM, as AQDS reaction rates are known to be affected
by flow rates, and proton diffusion rates across a PSM has been shown to be a dominant rate-
limiting step in bioelectric reactors (Liu et al., 2007; Bretschger et al., 2010). Mixing also likely
increases hematite particles’ reactivity by increasing their surface area exposure to a flowing
medium, consistent with a measurable increase in dissolved iron concentrations in the higher cell
density experiment after the onset of mixing (Fig. 10). Furthermore, since mixing involved purging
the biotic chamber with nitrogen, it is likely that it would have removed any DO that could have
diffused across the PSM, ensuring full Mtr metabolism and bacterial iron reduction. Indeed, Nafion
membranes are not perfectly impermeable to DO (Chae et al., 2008), and it is possible that DO
produced in the earlier, pre-mixing hours of the experiments could have diffused over to the biotic
chamber, decreasing iron reduction rates during this earlier time period. The onset of mixing
through purging would have removed any potential DO in the biotic chamber, removing this effect
and further enhancing reaction rates. Lastly, the abiotic control experiments also showed current
production, though no DO was measured. We attribute the current to abiotic electrochemical
reduction of AQDS, coupled with oxidation of formaldehyde that was inserted in the abiotic
controls. Formaldehyde oxidation is a known acidity-producing reaction, and this process would
be consistent with the notably lower pH measured at the end of the abiotic controls (5.0 – 5.2) than
at the end of the biotic experiments (7.4 – 7.6).
However, many processes at play in these experiments remain uncertain at this stage. While
details about specific, intermediate reactions are missing, some educated guesses can be explored
in light of the data available. Importantly, mass and charge balance calculations between electrical
current and DO production are broadly consistent with each other, along the stoichiometry of the
reaction described in Fig. 9 (further details discussed in Appendix C). However, some intermediate
78
reactions might still be cryptic, particularly with biotic experiments. Active bacterial electron
transfer can provide a range of additional pathways for electrons to flow from the working
electrode to the ferric oxides compared to abiotic processes (Fig. 12). The congruent AQDS redox
reactions and ferric oxide dissolution can further impact proton activity, which would promote the
flow of protons across the PSM to maintain mass and charge balance between the two chambers,
leading to oxygen production by driving the proposed water splitting in the oxidative chamber
(Fig. 9). Abiotic controls on the other hand offer fewer pathways to transfer electrons, and
associated proton flux, thus limiting reaction rates and oxygen production.
Figure 12: Conceptual diagrams showing possible pathways for electron transfer from the
working electrode to ferric oxides. (A) Abiotic electrochemical transfer through AQDS shuttles,
which may occur in both the biotic and abiotic experiments but which appears to be limited based
on the minimal observed dissolved O 2 production in the abiotic experiments (Fig. 2C); (B) direct
microbial solid-to-solid electron transfer; and (C) indirect solid-to-solid electron transfer via
AQDS electron shuttles, which appears to be the most likely candidate for DO production in the
biotic experiments (Fig. 10 A, B).
The use of AQDS and living organisms as catalysts opens up the possibility of a wide range
of intermediate compounds and by-products also being produced that could not be characterized
in this study. It is presumed that the microbial electron transfer pathway involved is the reversal
of the Mtr pathway, a process previously documented (Rabaey et al., 2008). Literature suggests
that in the context of these experiments, a likely mechanism involves Mtr complexes (namely
MtrC, B, A and CymA), which would allow the flow of electron across the cell membrane, both
79
in- and outbound, linked by intermediates such as NADH or quinone pools, transferring electrons
either within the cell or the periplasm (Ross et al., 2011; Rowe et al., 2017). Recent findings with
cable bacteria and the biochemical transfer of electrons along millimetre-scale distances highlight
the existence of a range of mechanisms for the transfer electrons between solid (in the case of cable
bacteria, living), extracellular structures (Shi et al., 2016; Bjerg et al., 2018; Cornelissen et al.,
2018). These findings further open the door to potentially novel or poorly understood biochemical
mechanisms for the transfer of electron transfer along cell membranes and between insoluble
substrates.
Another unanswered question relates to the amount of biochemical energy that can be
extracted through this transfer of electrons between solid extracellular substrates. While it has been
suggested that bacterial cathode oxidation offers scant metabolic energy and thus supports only
limited growth rates (Rosenbaum et al., 2011; Ross et al., 2011; Rowe et al., 2017), mere bacterial
maintenance, survival, or even presence may be enough in our experiments to catalyze the key
oxidizing geochemical reactions, as long as chemical activity is maintained within biological
structures. In these experiments however, no cell growth was observed over approximately 30
hours despite a relatively rich medium (Fig. 10), potentially suggesting that the amount of energy
available limits growth, albeit the lack of a good carbon source in the experimental medium would
also have also limited growth. The strong RSG signals however do suggest that bacteria were alive,
or at least their electron transport chains remained active. In this respect, a parallel can also be
drawn here with cable bacteria, for which case it remains unclear how exactly energy released
from redox couples separated by milli- to centimeters is distributed between cells situated along
the length of the cable. These cells merely transfer electrons between two elements that are in
effect extracellular to them (i.e., other cells situated up- and downstream) (Pfeffer et al., 2012; Shi
80
et al., 2016; Bjerg et al., 2018). Presumably, for cable bacteria as well as for bacteria studied here,
the flow of electrons—even if confined within cell membranes and periplasms—might still power
a proton motive force, which should promote the production of biochemical energy.
Lastly, it is unclear to what extent this oxidative process might have implications for natural
environments. This study suggests that steep redox and alkalinity gradients can have a strong effect
on environmental geochemistry, potentially addressing lingering questions around anaerobic
oxidative processes, such as pre-GOE “whiffs” of oxygen recorded in the geological record, or
pockets of oxygen found in deep, anaerobic sediments (Kepkay and Novitsky, 1980; Novitsky and
Kepkay, 1981; Anbar et al., 2007; Lyons et al., 2014). Molecular oxygen was produced here only
within a chamber containing very clean 18.2 MΩ.cm water, conditions which do not exist in the
environment. However, preliminary single chambered experiments also performed have yielded
significant amounts of manganese and iron oxide precipitates along the counter electrode, clearly
showing a strong oxidative pressure which was able to oxidize dissolved metals into mineral form
despite producing no detectable amounts of DO (Appendix C). Therefore, generating free O 2 might
not be strictly required to create a strong oxidative pressure which can precipitate mineral oxides
in a geologically stable form.
Overall, the reactor presented here successfully coupled microbial iron reduction with
molecular oxygen production. By utilizing ferric oxide minerals as an electron acceptor, the
energetic requirements involved in water splitting were thus minimized by a factor of ~ 3.5 when
compared to canonical hydrolysis (which otherwise uses protons as electron acceptor,
energetically much less favorable; see more discussion about energetics in Appendix C). With iron
oxide reduction generating significant amounts of alkalinity, iron reduction in the reductive
chamber would strongly promote the transfer of protons from the oxidative moiety across the PSM.
81
Microbial iron reduction thus acted as an effective catalyst, improving reaction rates by promoting
both the transfer of electrons and the transfer of protons across the two chambers. Separating the
reactor into two chambers isolated microbially-catalyzed reductive and oxidative reactions,
providing mechanistic insights into oxidative pathways for anaerobic environments. Combined,
these processes created an environment that has strong electric and alkalinity gradients, a catalyst,
and limited energetic requirements, driving an oxygen-generating reaction such as proposed in
Fig. 9. Future experiments could address many unanswered questions by including continuous
measurements of electric potential in the oxidative chamber and pH in both chambers, and by
implementing improved means of mixing. Cyclic voltammetry could also prove a powerful tool to
characterize the system’s energetics and optimize reaction rates. The use of mercuric chloride in
abiotic and killed controls would remove the effects of electrochemical aldehyde oxidation. Lastly,
experiments run with different electro-active microbes (e.g., Geobacter sp.) and metal oxides (e.g.,
ferric or manganese(IV) hydroxides) could offer depth to the current data set, as well as better
constraints on the role of mineral reactivity.
6.4. Conclusions
While a number of important uncertainties remain in this study, the results from the
experiments certainly reinforce the understanding that microbial activity can have a profound
impact on its environment, altering its redox, alkalinity, mineral dissolution as well as metal
mobility and bioavailability. Two-chambered bioelectric reactors can separate the oxidative and
reductive moieties of a reaction, and thus help characterize, quantify, and provide a mechanistic
understanding of microbial activity and its interactions with solid, mineral substrates. These results
also act as a proof of concept, supporting the viability of a Mars-based ISRU system that could use
82
such a bioelectrochemical system in order to derive molecular oxygen from microbial ferric oxide
mineral dissolution, in a system requiring minimal energy inputs, utilizing a virtually limitless
resource, consuming no water, and producing no harmful by-products (see further discussion on
the design of such an ISRU system in Appendix C). To that end, using bioelectrochemical reactors
offers significant advantages, as they could be powered directly by solar energy, removing the
need for consumable organic electron donor substrates, thus enhancing autonomous ISRU
capabilities of the system. Additionally, perchlorate is also a relatively abundant oxide present on
Mars, inviting the further use of similar bioelectrochemical reactors could also leverage microbial
perchlorate reduction, a known oxygen-generating biological process, with cathode oxidation.
This added step could not only reduce toxicity risks from elevated concentrations of a known
contaminant, but also further improve the recovery of oxygen out of Martian soils (Coates and
Achenbach, 2004; Wang and Coates, 2017). In this context, EET-capable bacteria can act as an
effective catalyst for the reductive dissolution of oxide minerals and the corresponding production
of molecular oxygen, laying the groundwork for the development of a viable source of oxygen
utilizing Martian soils, and thus have the potential to enable human planetary exploration.
83
Chapter 7: Kosmos
May the immeasurable diversity of phenomena which crowd into the picture of nature in
no way detract from that harmonious impression of rest and unity which is the ultimate object of
every literary or purely artistical composition.
Alexander von Humboldt
The data presented in this thesis tell us that siderophores can, in iron-limited environments,
effectively extract iron out of olivine, supporting significant cell growth and primary production.
They also provide quantified constraints on biologically relevant siderophore concentrations, and
the corresponding growth and cell densities these concentrations can support. The specific
mechanism proposed involves siderophores mobilizing ferric oxides, formed initially by the
abiotic oxidation of iron along the mineral surface, and further preventing the formation this nano-
meter thick protective layer. This process enhances not only nutrient bioavailability, but also
olivine dissolution rates; and the effective use and cycling of siderophores by active bacteria
magnifies siderophore activity and dissolution rates by an order of magnitude, compared to strictly
abiotic reactions.
These findings carry a number of significant implications. Mainly, the results emphasize
that primary silicate minerals can be important sources of micronutrients and should be included
when considering the pool of bioavailable iron. Indeed, the traditional and functional approach to
nutrient availability involves measurements of filtered aqueous media. This approach, however,
only takes into account elements already mobilized, and would miss their net sources. In
oligotrophic ocean waters, these are typically particulates of aeolian origin, characterized as
silicates (clay) and ferric oxyhydroxides particles, which are otherwise insoluble (Caquineau,
84
2002; Journet et al., 2008; Schroth et al., 2009). Soil and crustal environments are by nature
embedded within solid matrices, and iron mobility is thus a function of localized redox and pH
conditions that can define the dissolution or precipitation of elements from ubiquitous primary and
secondary minerals (Lindsay and Schwab, 1982; Lindsay, 1991; Lemanceau et al., 2009; Krohling
et al., 2016; Gadd, 2017).
Moreover, siderophore concentrations do not necessarily need to be exceedingly high to
have a significant impact on microbial growth, as low micromolar to mid nanomolar
concentrations can support significant cell densities (10
7
– 10
8
cells/mL). Siderophores are also a
valued commodity that can be shared across clades, meaning that they can support entire microbial
communities, as well as help define the relationships and dynamics, whether symbiotic or
antagonistic, across clades and even domains. This sharing is complemented by the effective reuse
and recycling of siderophores, quantified here, which indicates that seemingly low siderophore
concentrations might still have a larger impact on iron mobilization and silicate dissolution rates
than might otherwise be inferred from simple siderophore concentration measurements, as the net
activity of siderophores becomes intensified by microbial cycling. Combined, these conclusions
favor an interpretation of siderophore use to be most effective within dynamic, relatively dense
microbial communities located in close proximity to a mineral surface. Therefore, the general idea
of iron limitation might, in large part, be better understood as a function of (1) a community’s
access and distance to iron-bearing mineral phases, (2) the effective activity of siderophores
relating to the kinetics of their recycled use, and (3) the nature of interactions between siderophore-
producing and non-producing members of the community; rather than in the mere measurement
of iron concentrations in a filtered aqueous medium.
85
The extraction of iron from silicates by bacterial siderophores also has significant
implications for crustal and climate dynamics. Accelerated silicate dissolution point toward more
pronounced weathering processes, potentially increasing the rates of rock breakdown and soil
formation. It also implies the faster release of silicate anions to the aqueous environment,
significantly increasing the environment’s alkalinity, a process which generates a sink for
atmospheric CO 2, turning it to bicarbonate. At large scales, silicate weathering can therefore have
large impacts on global carbon cycles and climate systems, to such an extent that silicate,
especially olivine, dissolution has been proposed as a method to sequester carbon from the
atmosphere and mitigate anthropogenic climate change (Hartmann et al., 2009, 2013; Köhler et
al., 2010, p.; Montserrat et al., 2017). A detailed, mechanistic understanding of the controls on
silicate dissolution rates, such as the impact of microbial siderophore utilization, can therefore
provide the necessary tools in successfully implementing large-scale geoengineering efforts.
Similarly, bacteria-mineral interactions with EET can further lead to the rapid dissolution of oxide
minerals, producing steep redox as well as alkalinity gradients, a process that can drive
geochemical transformations such as the generation of molecular oxygen. A deep understanding
of this process can further help develop engineered bioelectric systems that can transform abundant
resources such as Martian soils, into valuable necessities such as oxygen, critical for the human
exploration of Mars. It therefore seems like utilizing microbial processes carries an enormous,
almost untapped potential to effectively use in engineered systems: offering the means to address
some of the most pressing issues of our times, climate change; or even promoting some of the
greatest achievements Mankind could strive for, the exploration of other worlds.
But beyond uses in geoengineering, bacteria-mineral interactions stand at the very core of
what made the Earth what it is, defining its history and evolution. The strong propensity of organic
86
molecules to sorb onto mineral surfaces, and the associated redox and alkalinity reactions that can
occur with specific metal catalysts, very likely played a critical role in the emergence of energy-
gathering autocatalytic systems: the very beginnings of life (Martin and Russell, 2007; Sousa et
al., 2013; Lane, 2017; Preiner et al., 2019). Once set in motion, life never stopped evolving, and
innovations like photosynthesis oxidized the world, redefining the way elements were distributed
between rocks and solution (Anbar and Knoll, 2002). Out of necessity, life adapted and developed
complex biochemical systems and entire metabolisms designed specifically to interact with
minerals, such as acquiring energy with EET, or nutrients with siderophores (Shi et al., 2016;
Thode et al., 2018), thus allowing primary productivity in what might otherwise have become
infertile conditions. These biological processes altered their environments further still by
dissolving minerals and mobilizing elements, defining the course of entire planetary systems such
as crustal weathering and climate. Yet mineral surfaces never stopped conditioning the
development of life. The impressive ubiquity and longevity of mineral-sorbed extra-cellular DNA
(Pietramellara et al., 2009; Brazelton and Baross, 2009; Aziz et al., 2010; Sand and Jelavić, 2018)
further suggest that minerals might act as a vector for lateral gene transfer, and perhaps most
importantly, help fulfill the central end-goal of life itself: fostering the preservation and
perpetuation of genetic information. In the end, the history and evolutions of the Earth, our home,
is one that cannot be understood in terms other than an intimate relationship between biology and
geology, a core component of which is driven by reactions at the organic-mineral interface.
87
References
Abergel, R.J., Zawadzka, A.M., Hoette, T.M., and Raymond, K.N., 2009, Enzymatic Hydrolysis
of Trilactone Siderophores: Where Chiral Recognition Occurs in Enterobactin and
Bacillibactin Iron Transport(1): Journal of the American Chemical Society, v. 131, p.
12682–12692, doi:10.1021/ja903051q.
Ahmed, E., and Holmström, S.J.M., 2015, Microbe–mineral interactions: The impact of surface
attachment on mineral weathering and element selectivity by microorganisms: Chemical
Geology, v. 403, p. 13–23, doi:10.1016/j.chemgeo.2015.03.009.
Aldersley, M.F., Joshi, P.C., Price, J.D., and Ferris, J.P., 2011, The role of montmorillonite in its
catalysis of RNA synthesis: Applied Clay Science, v. 54, p. 1–14,
doi:10.1016/j.clay.2011.06.011.
Amin, S.A., Green, D.H., Al Waheeb, D., Gärdes, A., and Carrano, C.J., 2012a, Iron transport in
the genus Marinobacter: BioMetals, v. 25, p. 135–147, doi:10.1007/s10534-011-9491-9.
Amin, S.A., Green, D.H., Gärdes, A., Romano, A., Trimble, L., and Carrano, C.J., 2012b,
Siderophore-mediated iron uptake in two clades of Marinobacter spp. associated with
phytoplankton: the role of light: BioMetals, v. 25, p. 181–192, doi:10.1007/s10534-011-
9495-5.
Amin, S.A., Green, D.H., Hart, M.C., Küpper, F.C., Sunda, W.G., and Carrano, C.J., 2009a,
Photolysis of iron–siderophore chelates promotes bacterial–algal mutualism: Proceedings
of the National Academy of Sciences, v. 106, p. 17071–17076.
Amin, S.A., Green, D.H., Küpper, F.C., and Carrano, C.J., 2009b, Vibrioferrin, an Unusual Marine
Siderophore: Iron Binding, Photochemistry, and Biological Implications: Inorganic
Chemistry, v. 48, p. 11451–11458, doi:10.1021/ic9016883.
Anbar, A.D. et al., 2007, A Whiff of Oxygen Before the Great Oxidation Event? Science, v. 317,
p. 1903–1906, doi:10.1126/science.1140325.
Anbar, A.D., and Knoll, A.H., 2002, Proterozoic ocean chemistry and evolution: a bioinorganic
bridge? science, v. 297, p. 1137–1142.
Anderson, R.S., and Anderson, S.P., 2010, Geomorphology: The Mechanics and Chemistry of
Landscapes: New York, NY, Cambridge University Press.
Aouad, G., Crovisier, J., Geoffroy, V., Meyer, J., and Stille, P., 2006, Microbially-mediated glass
dissolution and sorption of metals by Pseudomonas aeruginosa cells and biofilm: Journal
of Hazardous Materials, v. 136, p. 889–895, doi:10.1016/j.jhazmat.2006.01.026.
Atkins, P., and Jones, L., 2004, Chemical Principles: W. H. Freeman.
88
Aumont, O., and Bopp, L., 2006, Globalizing results from ocean in situ iron fertilization studies:
Global Biogeochemical Cycles, v. 20, doi:10.1029/2005GB002591.
Aziz, R.K., Breitbart, M., and Edwards, R.A., 2010, Transposases are the most abundant, most
ubiquitous genes in nature: Nucleic Acids Research, v. 38, p. 4207–4217,
doi:10.1093/nar/gkq140.
Bailey, B., Templeton, A., Staudigel, H., and Tebo, B.M., 2009, Utilization of Substrate
Components during Basaltic Glass Colonization by Pseudomonas and Shewanella Isolates:
Geomicrobiology Journal, v. 26, p. 648–656, doi:10.1080/01490450903263376.
Barbeau, K., A., Rue, E.L., Bruland, K.W., and Butler, A., 2001, Photochemical cycling of iron in
the surface ocean mediated by microbial iron (III)-binding ligands: Nature, v. 413, p. 409–
413.
Barge, L.M. et al., 2017, Thermodynamics, Disequilibrium, Evolution: Far-From-Equilibrium
Geological and Chemical Considerations for Origin-Of-Life Research: Origins of Life and
Evolution of Biospheres, v. 47, p. 39–56, doi:10.1007/s11084-016-9508-z.
Bartholdy, B.A., Berreck, M., and Haselwandter, K., 2001, Hydroxamate siderophore synthesis by
Phialocephala fortinii, a typical dark septate fungal root endophyte: Biometals, v. 14, p. 10.
Bell, J.F. et al., 2000, Mineralogic and compositional properties of Martian soil and dust: Results
from Mars Pathfinder: Journal of Geophysical Research: Planets, v. 105, p. 1721–1755,
doi:10.1029/1999JE001060.
Belmonte, L., and Mansy, S.S., 2016, Metal Catalysts and the Origin of Life: Elements, v. 12, p.
413–418, doi:10.2113/gselements.12.6.413.
Berman-Frank, I., Cullen, J.T., Shaked, Y., Sherrell, R.M., and Falkowski, P.G., 2001, Iron
availability, cellular iron quotas, and nitrogen fixation in Trichodesmium: Limnology and
Oceanography, v. 46, p. 1249–1260, doi:10.4319/lo.2001.46.6.1249.
Berner, R.A., 2004, The Phanerozoic Carbon Cycle: CO2 and O2: New York, NY, Oxford
University Press.
Berner, R.A., Lasaga, A.C., and Garrels, R.M., 1983, The Carbonate-Silicate Geochemical Cycle
and its Effect on Atmospheric Carbon Dioxide Over the Past 100 Million Years: American
Journal of Science, v. 283, p. 641–683.
Bjerg, J.T. et al., 2018, Long-distance electron transport in individual, living cable bacteria:
Proceedings of the National Academy of Sciences, v. 115, p. 5786–5791,
doi:10.1073/pnas.1800367115.
Bluhm, M.E., Kim, S.S., Dertz, E.A., and Raymond, K.N., 2002, Corynebactin and Enterobactin:
Related Siderophores of Opposite Chirality: Journal of the American Chemical Society, v.
124, p. 2436–2437, doi:10.1021/ja016651s.
89
Boiteau, R.M., Mende, D.R., Hawco, N.J., McIlvin, M.R., Fitzsimmons, J.N., Saito, M.A.,
Sedwick, P.N., DeLong, E.F., and Repeta, D.J., 2016, Siderophore-based microbial
adaptations to iron scarcity across the eastern Pacific Ocean: Proceedings of the National
Academy of Sciences, v. 113, p. 14237–14242.
Boyd, P.W. et al., 2007, Mesoscale Iron Enrichment Experiments 1993-2005: Synthesis and Future
Directions: Science, v. 315, p. 612–617.
Brady, P.V., and Walther, J.V., 1989, Controls on silicate dissolution rates in neutral and basic pH
solutions at 25°C: Geochimica et Cosmochimica Acta, v. 53, p. 2823–2830,
doi:10.1016/0016-7037(89)90160-9.
Brantley, S.L., Kubicki, J.D., and White, A.F. (Eds.), 2008, Kinetics of water-rock interaction:
New York, Springer Verlag, 833 p.
Braun, V., Hantke, K., and Köster, W., 1998, Bacterial iron transport: Mechanism, genetics and
regulation, in Sigel, A. and Sigel, H. eds., Metal Ions in Biological Systems, 35, p. 67–145.
Brazelton, W.J., and Baross, J.A., 2009, Abundant transposases encoded by the metagenome of a
hydrothermal chimney biofilm: The ISME Journal, v. 3, p. 1420–1424,
doi:10.1038/ismej.2009.79.
Brehm, U., Gorbushina, A., and Mottershead, D., 2005, The role of microorganisms and biofilms
in the breakdown and dissolution of quartz and glass: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 219, p. 117–129, doi:10.1016/j.palaeo.2004.10.017.
Bretschger, O. et al., 2007, Current Production and Metal Oxide Reduction by Shewanella
oneidensis MR-1 Wild Type and Mutants: Applied and Environmental Microbiology, v.
73, p. 7003–7012, doi:10.1128/AEM.01087-07.
Bretschger, O., Cheung, A.C.M., Mansfeld, F., and Nealson, K.H., 2010, Comparative Microbial
Fuel Cell Evaluations of Shewanella spp.: Electroanalysis, v. 22, p. 883–894,
doi:10.1002/elan.200800016.
Buss, H.L., Lüttge, A., and Brantley, S.L., 2007, Etch pit formation on iron silicate surfaces during
siderophore-promoted dissolution: Chemical Geology, v. 240, p. 326–342,
doi:10.1016/j.chemgeo.2007.03.003.
Butaitė, E., Baumgartner, M., Wyder, S., and Kümmerli, R., 2017, Siderophore cheating and
cheating resistance shape competition for iron in soil and freshwater Pseudomonas
communities: Nature Communications, v. 8, doi:10.1038/s41467-017-00509-4.
Camprubi, E., Jordan, S.F., Vasiliadou, R., and Lane, N., 2017, Iron catalysis at the origin of life:
IUBMB Life, v. 69, p. 373–381, doi:10.1002/iub.1632.
Caquineau, S., 2002, Mineralogy of Saharan dust transported over northwestern tropical Atlantic
Ocean in relation to source regions: Journal of Geophysical Research, v. 107,
doi:10.1029/2000JD000247.
90
Cardenas-Rodriguez, M., Chatzi, A., and Tokatlidis, K., 2018, Iron–sulfur clusters: from metals
through mitochondria biogenesis to disease: JBIC Journal of Biological Inorganic
Chemistry, v. 23, p. 509–520, doi:10.1007/s00775-018-1548-6.
Chae, K.J., Choi, M., Ajayi, F.F., Park, W., Chang, I.S., and Kim, I.S., 2008, Mass Transport
through a Proton Exchange Membrane (Nafion) in Microbial Fuel Cells
†
: Energy & Fuels,
v. 22, p. 169–176, doi:10.1021/ef700308u.
Challis, G.L., and Ravel, J., 2000, Coelichelin, a new peptide siderophore encoded by the
Streptomyces coelicolor genome: structure prediction from the sequence of its non-
ribosomal peptide synthetase: FEMS Microbiology Letters, v. 187, p. 111–114,
doi:10.1111/j.1574-6968.2000.tb09145.x.
Cheah, S.-F., Kraemer, S.M., Cervini-Silva, J., and Sposito, G., 2003, Steady-state dissolution
kinetics of goethite in the presence of desferrioxamine B and oxalate ligands: implications
for the microbial acquisition of iron: Chemical Geology, v. 198, p. 63–75,
doi:10.1016/S0009-2541(02)00421-7.
Chen, G., 2004, Electrochemical technologies in wastewater treatment: Separation and
Purification Technology, v. 38, p. 11–41, doi:10.1016/j.seppur.2003.10.006.
Christensen, P.R. et al., 2000, Detection of crystalline hematite mineralization on Mars by the
Thermal Emission Spectrometer: Evidence for near-surface water: Journal of Geophysical
Research: Planets, v. 105, p. 9623–9642, doi:10.1029/1999JE001093.
Christensen, P.R., 2004, Formation of the hematite-bearing unit in Meridiani Planum: Evidence
for deposition in standing water: Journal of Geophysical Research, v. 109,
doi:10.1029/2003JE002233.
Christensen, P.R., Morris, R.V., Lane, M.D., Bandfield, J.L., and Malin, M.C., 2001, Global
mapping of Martian hematite mineral deposits: Remnants of water-driven processes on
early Mars: Journal of Geophysical Research: Planets, v. 106, p. 23873–23885,
doi:10.1029/2000JE001415.
Coates, J.D., and Achenbach, L.A., 2004, Microbial perchlorate reduction: rocket-fuelled
metabolism: Nature Reviews Microbiology, v. 2, p. 569–580, doi:10.1038/nrmicro926.
Cocozza, C., Tsao, C.C., Cheah, S.-F., Kraemer, S.M., Raymond, K.N., Miano, T.M., and Sposito,
G., 2002, Temperature dependence of goethite dissolution promoted by trihydroxamate
siderophores: Geochimica et Cosmochimica Acta, v. 66, p. 431–438.
Cody, G.D., 2000, Primordial Carbonylated Iron-Sulfur Compounds and the Synthesis of
Pyruvate: Science, v. 289, p. 1337–1340, doi:10.1126/science.289.5483.1337.
Cody, G.D., 2004, Transition Metal Sulfides and the Origins of Metabolism: Annual Review of
Earth and Planetary Sciences, v. 32, p. 569–599,
doi:10.1146/annurev.earth.32.101802.120225.
91
Cornelissen, R. et al., 2018, The Cell Envelope Structure of Cable Bacteria: Frontiers in
Microbiology, v. 9, p. 3044, doi:10.3389/fmicb.2018.03044.
Coursolle, D., and Gralnick, J.A., 2012, Reconstruction of Extracellular Respiratory Pathways for
Iron(III) Reduction in Shewanella Oneidensis Strain MR-1: Frontiers in Microbiology, v.
3, doi:10.3389/fmicb.2012.00056.
Cruz-Morales, P., Ramos-Aboites, H.E., Licona-Cassani, C., Selem-Mójica, N., Mejía-Ponce,
P.M., Souza-Saldívar, V., and Barona-Gómez, F., 2017, Actinobacteria phylogenomics,
selective isolation from an iron oligotrophic environment and siderophore functional
characterization, unveil new desferrioxamine traits: FEMS Microbiology Ecology, v. 93,
doi:10.1093/femsec/fix086.
Dehner, C.A., Awaya, J.D., Maurice, P.A., and DuBois, J.L., 2010, Roles of Siderophores,
Oxalate, and Ascorbate in Mobilization of Iron from Hematite by the Aerobic Bacterium
Pseudomonas mendocina: Applied and Environmental Microbiology, v. 76, p. 2041–2048,
doi:10.1128/AEM.02349-09.
Dehner, C.A., Barton, L., Maurice, P.A., and DuBois, J.L., 2011, Size-Dependent Bioavailability
of Hematite (α-Fe 2O 3) Nanoparticles to a Common Aerobic Bacterium: Environmental
Science & Technology, v. 45, p. 977–983, doi:10.1021/es102922j.
Delory, G.T., Farrell, W.M., Atreya, S.K., Renno, N.O., Wong, A.-S., Cummer, S.A., Sentman,
D.D., Marshall, J.R., Rafkin, S.C.R., and Catling, D.C., 2006, Oxidant Enhancement in
Martian Dust Devils and Storms: Storm Electric Fields and Electron Dissociative
Attachment: Astrobiology, v. 6, p. 451–462, doi:10.1089/ast.2006.6.451.
Dessert, C., Dupré, B., Gaillardet, J., François, L.M., and Allègre, C.J., 2003, Basalt weathering
laws and the impact of basalt weathering on the global carbon cycle: Chemical Geology,
v. 202, p. 257–273, doi:10.1016/j.chemgeo.2002.10.001.
Dhungana, S., and Crumbliss, A.L., 2005, Coordination Chemistry and Redox Processes in
Siderophore-Mediated Iron Transport: Geomicrobiology Journal, v. 22, p. 87–98,
doi:10.1080/01490450590945870.
Downs, R.T. and MSL Science Team, 2015, Determining Mineralogy on Mars with the CheMin
X-Ray Diffractometer: Elements, v. 11, p. 45–50, doi:10.2113/gselements.11.1.45.
Dupré, B., Dessert, C., Oliva, P., Goddéris, Y., Viers, J., François, L., Millot, R., and Gaillardet,
J., 2003, Rivers, chemical weathering and Earth’s climate: Comptes Rendus Geoscience,
v. 335, p. 1141–1160, doi:10.1016/j.crte.2003.09.015.
Ehlmann, B.L. et al., 2017, Chemistry, mineralogy, and grain properties at Namib and High dunes,
Bagnold dune field, Gale crater, Mars: A synthesis of Curiosity rover observations:
Bagnold Dune Sands Composition: Journal of Geophysical Research: Planets, v. 122, p.
2510–2543, doi:10.1002/2017JE005267.
92
Ejje, N., Soe, C.Z., Gu, J., and Codd, R., 2013, The variable hydroxamic acid siderophore
metabolome of the marine actinomycete Salinispora tropica CNB-440: Metallomics, v. 5,
p. 1519, doi:10.1039/c3mt00230f.
Elgrishi, N., Rountree, K.J., McCarthy, B.D., Rountree, E.S., Eisenhart, T.T., and Dempsey, J.L.,
2018, A Practical Beginner’s Guide to Cyclic Voltammetry: Journal of Chemical
Education, v. 95, p. 197–206, doi:10.1021/acs.jchemed.7b00361.
Encrenaz, T., Greathouse, T.K., Lefèvre, F., and Atreya, S.K., 2012, Hydrogen peroxide on Mars:
Observations, interpretation and future plans: Planetary and Space Science, v. 68, p. 3–17,
doi:10.1016/j.pss.2011.03.019.
Ertem, G., Snellinger-O’Brien, A.M., Ertem, M.C., Rogoff, D.A., Dworkin, J.P., Johnston, M.V.,
and Hazen, R.M., 2008, Abiotic formation of RNA-like oligomers by montmorillonite
catalysis: part II: International Journal of Astrobiology, v. 7, p. 1–7,
doi:10.1017/S147355040700393X.
Esmaeel, Q., Pupin, M., Kieu, N.P., Chataigné, G., Béchet, M., Deravel, J., Krier, F., Höfte, M.,
Jacques, P., and Leclère, V., 2016, Burkholderia genome mining for nonribosomal peptide
synthetases reveals a great potential for novel siderophores and lipopeptides synthesis:
MicrobiologyOpen, v. 5, p. 512–526, doi:10.1002/mbo3.347.
Feng, X., Simpson, A.J., and Simpson, M.J., 2005, Chemical and mineralogical controls on humic
acid sorption to clay mineral surfaces: Organic Geochemistry, p. 14.
Fennessey, C.M., Jones, M.E., Taillefert, M., and DiChristina, T.J., 2010, Siderophores Are Not
Involved in Fe(III) Solubilization during Anaerobic Fe(III) Respiration by Shewanella
oneidensis MR-1: Applied and Environmental Microbiology, v. 76, p. 2425–2432,
doi:10.1128/AEM.03066-09.
Fergason, R.L., Gaddis, L.R., and Rogers, A.D., 2014, Hematite-bearing materials surrounding
Candor Mensa in Candor Chasma, Mars: Implications for hematite origin and post-
emplacement modification: Icarus, v. 237, p. 350–365, doi:10.1016/j.icarus.2014.04.038.
Ferret, C., Sterckeman, T., Cornu, J.-Y., Gangloff, S., Schalk, I.J., and Geoffroy, V.A., 2014,
Siderophore-promoted dissolution of smectite by fluorescent Pseudomonas: Smectite
alteration by bacteria: Environmental Microbiology Reports, v. 6, p. 459–467,
doi:10.1111/1758-2229.12146.
Fredrickson, J.K. et al., 2008, Towards environmental systems biology of Shewanella: Nature
Reviews Microbiology, v. 6, p. 592–603, doi:10.1038/nrmicro1947.
Gadd, G.M., 2017, Fungi, Rocks, and Minerals: Elements, v. 13, p. 171–176,
doi:10.2113/gselements.13.3.171.
Ganor, J., Lu, P., Zheng, Z., and Zhu, C., 2007, Bridging the gap between laboratory measurements
and field estimations of silicate weathering using simple calculations: Environmental
Geology, v. 53, p. 599–610, doi:10.1007/s00254-007-0675-0.
93
Garibaldi, J.A., and Neilands, J.B., 1956, Formation of Iron-binding Compounds by Micro-
organisms: Nature, v. 177, p. 526–527.
Gerrits, R., 2019, An experimental study of fungal olivine weathering [Ph.D.]: Freie Universität
Berlin, 190 p., https://refubium.fu-
berlin.de/bitstream/handle/fub188/25125/Dissertation_Gerrits.pdf?sequence=3&isAllowe
d=y.
Gibson, D.G. et al., 2010, Creation of a Bacterial Cell Controlled by a Chemically Synthesized
Genome: Science, v. 329, p. 52–56, doi:10.1126/science.1190719.
Goetz, W. et al., 2005, Indication of drier periods on Mars from the chemistry and mineralogy of
atmospheric dust: Nature, v. 436, p. 62–65, doi:10.1038/nature03807.
Gorbushina, A.A., and Palinska, K.A., 1999, Biodeteriorative processes on glass: experimental
proof of the role of fungi and cyanobacteria: Aerobiologia, v. 15, p. 183–192.
Greczynski, G., and Hultman, L., 2017, C 1s Peak of Adventitious Carbon Aligns to the Vacuum
Level: Dire Consequences for Material’s Bonding Assignment by Photoelectron
Spectroscopy: ChemPhysChem, v. 18, p. 1507–1512, doi:10.1002/cphc.201700126.
Greczynski, G., and Hultman, L., 2020, X-ray photoelectron spectroscopy: Towards reliable
binding energy referencing: Progress in Materials Science, v. 107, p. 100591,
doi:10.1016/j.pmatsci.2019.100591.
Greenblatt, J.B., Miller, D.J., Ager, J.W., Houle, F.A., and Sharp, I.D., 2018, The Technical and
Energetic Challenges of Separating (Photo)Electrochemical Carbon Dioxide Reduction
Products: Joule, v. 2, p. 381–420, doi:10.1016/j.joule.2018.01.014.
Grotzinger, J.P. et al., 2014, A Habitable Fluvio-Lacustrine Environment at Yellowknife Bay, Gale
Crater, Mars: Science, v. 343, p. 1242777–1242777, doi:10.1126/science.1242777.
Ha, J., Hyun Yoon, T., Wang, Y., Musgrave, C.B., and Brown, Jr., G.E., 2008, Adsorption of
Organic Matter at Mineral/Water Interfaces: 7. ATR-FTIR and Quantum Chemical Study
of Lactate Interactions with Hematite Nanoparticles: Langmuir, v. 24, p. 6683–6692,
doi:10.1021/la800122v.
Haase, K.M. et al., 2009, Diking, young volcanism and diffuse hydrothermal activity on the
southern Mid-Atlantic Ridge: The Lilliput field at 9°33′S: Marine Geology, v. 266, p. 52–
64, doi:10.1016/j.margeo.2009.07.012.
Hall, D.O., Cammack, R., and Rao, K.K., 1971, Role of Ferredoxins in the Origin of Life and
Biological Evolution: Nature, v. 233, p. 136–138.
Hall-Stoodley, L., Costerton, J.W., and Stoodley, P., 2004, Bacterial biofilms: from the Natural
environment to infectious diseases: Nature Reviews Microbiology, v. 2, p. 95–108,
doi:10.1038/nrmicro821.
94
Hamilton, V.E., McSween, H.Y., and Hapke, B., 2005, Mineralogy of Martian atmospheric dust
inferred from thermal infrared spectra of aerosols: Journal of Geophysical Research, v.
110, doi:10.1029/2005JE002501.
Hanczyc, M.M., Fujikawa, M., and Szostak, J.W., 2003, Experimental Models of Primitive
Cellular Compartments: Encapsulation, Growth, and Division: Science, v. 302, p. 618–
621, doi:10.1126/science.1089904.
Hansma, H.G., 2010, Possible origin of life between mica sheets: Journal of Theoretical Biology,
v. 266, p. 175–188, doi:10.1016/j.jtbi.2010.06.016.
Hansma, H.G., 2013, Possible origin of life between mica sheets: does life imitate mica? Journal
of Biomolecular Structure and Dynamics, v. 31, p. 888–895,
doi:10.1080/07391102.2012.718528.
Harrold, Z.R., Hausrath, E.M., Garcia, A.H., Murray, A.E., Tschauner, O., Raymond, J.A., and
Huang, S., 2018, Bioavailability of Mineral-Bound Iron to a Snow Algal-Bacterial
Coculture and Implications for Albedo-Altering Snow Algal Blooms (F. E. Löffler, Ed.):
Applied and Environmental Microbiology, v. 84, p. e02322-17, /aem/84/7/e02322-
17.atom, doi:10.1128/AEM.02322-17.
Hartmann, J., Jansen, N., Dürr, H.H., Kempe, S., and Köhler, P., 2009, Global CO2-consumption
by chemical weathering: What is the contribution of highly active weathering regions?
Global and Planetary Change, v. 69, p. 185–194, doi:10.1016/j.gloplacha.2009.07.007.
Hartmann, J., West, A.J., Renforth, P., Köhler, P., De La Rocha, C.L., Wolf-Gladrow, D.A., Dürr,
H.H., and Scheffran, J., 2013, Enhanced chemical weathering as a geoengineering strategy
to reduce atmospheric carbon dioxide, supply nutrients, and mitigate ocean acidification:
ENHANCED WEATHERING: Reviews of Geophysics, v. 51, p. 113–149,
doi:10.1002/rog.20004.
Hartney, S.L., Mazurier, S., Kidarsa, T.A., Quecine, M.C., Lemanceau, P., and Loper, J.E., 2011,
TonB-dependent outer-membrane proteins and siderophore utilization in Pseudomonas
fluorescens Pf-5: BioMetals, v. 24, p. 193–213, doi:10.1007/s10534-010-9385-2.
Hassler, C.S., Schoemann, V., Nichols, C.M., Butler, E.C.V., and Boyd, P.W., 2011, Saccharides
enhance iron bioavailability to Southern Ocean phytoplankton: Proceedings of the National
Academy of Sciences, v. 108, p. 1076–1081, doi:10.1073/pnas.1010963108.
Hau, H.H., and Gralnick, J.A., 2007, Ecology and Biotechnology of the Genus Shewanella: Annual
Review of Microbiology, v. 61, p. 237–258,
doi:10.1146/annurev.micro.61.080706.093257.
Hausrath, E.M., Neaman, A., and Brantley, S.L., 2009, Elemental release rates from dissolving
basalt and granite with and without organic ligands: American Journal of Science, v. 309,
p. 633–660, doi:10.2475/08.2009.01.
95
Hersman, L.E., Forsythe, J.H., Ticknor, L.O., and Maurice, P.A., 2001, Growth of Pseudomonas
mendocina on Fe(III) (Hydr)Oxides: Applied and Environmental Microbiology, v. 67, p.
4448–4453, doi:10.1128/AEM.67.10.4448-4453.2001.
Hersman, L., Huang, A., Maurice, P., and Forsythe, J.H., 2000, Siderophore Production and Iron
Reduction by Pseudomonas mendocina in Response to Iron Deprivation: Geomicrobiology
Journal, v. 17, p. 261–273, doi:10.1080/01490450050192965.
Hersman, L., Maurice, P., and Sposito, G., 1996, Iron acquisition from hydrous Fe(III)-oxides by
an aerobic Pseudomonas sp.: Chemical Geology, v. 132, p. 25–31, doi:10.1016/S0009-
2541(96)00038-1.
Hider, R.C., and Kong, X., 2010, Chemistry and biology of siderophores: Natural Product Reports,
v. 27, p. 637, doi:10.1039/b906679a.
Hilley, G.E., and Porder, S., 2008, A framework for predicting global silicate weathering and CO2
drawdown rates over geologic time-scales: Proceedings of the National Academy of
Sciences, v. 105, p. 16855–16859, doi:10.1073/pnas.0801462105.
Hong, Y., Guo, J., Xu, Z., Xu, M., and Sun, G., 2007, Humic substances act as electron acceptor
and redox mediator for microbial dissimilatory azoreduction by Shewanella decolorationis
S12: Journal of Microbiology and Biotechnology, v. 17, p. 428–437.
Huber, C., 2003, A Possible Primordial Peptide Cycle: Science, v. 301, p. 938–940,
doi:10.1126/science.1086501.
Huber, C., 1997, Activated Acetic Acid by Carbon Fixation on (Fe,Ni)S Under Primordial
Conditions: Science, v. 276, p. 245–247, doi:10.1126/science.276.5310.245.
Hutchins, D.A., and Boyd, P.W., 2016, Marine phytoplankton and the changing ocean iron cycle:
Nature Climate Change, v. 6, p. 1072–1079, doi:10.1038/nclimate3147.
Janecky, D.R., and Seyfried, W.E., 1986, Hydrothermal serpentinization of peridotite within the
oceanic crust: Experimental investigations of mineralogy and major element chemistry:
Geochimica et Cosmochimica Acta, v. 50, p. 1357–1378, doi:10.1016/0016-
7037(86)90311-X.
Jeon, B.-H., Dempsey, B.A., Burgos, W.D., and Royer, R.A., 2001, Reactions of ferrous iron with
hematite: Colloids and Surfaces A: Physicochemical and Engineering Aspects, v. 191, p.
41–55, doi:10.1016/S0927-7757(01)00762-2.
Jeremiasse, A.W., Hamelers, H.V.M., and Buisman, C.J.N., 2010, Microbial electrolysis cell with
a microbial biocathode: Bioelectrochemistry, v. 78, p. 39–43,
doi:10.1016/j.bioelechem.2009.05.005.
Jheeta, S., and Joshi, P., 2014, Prebiotic RNA Synthesis by Montmorillonite Catalysis: Life, v. 4,
p. 318–330, doi:10.3390/life4030318.
96
Jickells, T.D., An, Z.S., Andersen, K.K., Baker, A.R., Bergametti, G., Brooks, N., Cao, J.J., Boyd,
P.W., Duce, R.A., and Hunter, K.A., 2005, Global iron connections between desert dust,
ocean biogeochemistry, and climate: Science, v. 308, p. 67–71.
Journet, E., Desboeufs, K.V., Caquineau, S., and Colin, J.-L., 2008, Mineralogy as a critical factor
of dust iron solubility: Geophysical Research Letters, v. 35, doi:10.1029/2007GL031589.
Kepkay, P.E., and Novitsky, J.A., 1980, Microbial control of organic carbon in marine sediments:
Coupled chemoautotrophy and heterotrophy: Marine Biology, v. 55, p. 261–266,
doi:10.1007/BF00393778.
Kim, B.H., Park, H.S., Kim, H.J., Kim, G.T., Chang, I.S., Lee, J., and Phung, N.T., 2004,
Enrichment of microbial community generating electricity using a fuel-cell-type
electrochemical cell: Applied Microbiology and Biotechnology, v. 63, p. 672–681,
doi:10.1007/s00253-003-1412-6.
Knafelc, J., 2010, High temperature oxidation experiments of olivine with implications for mantle
magnetism [B.Sc. Hon.]: University of Wisconsin Oshkosh.
Knoepp, J.D., DeBano, L.F., and Neary, D.G., 2005, Chapter 3: Soil Chemistry, in Neary, D.G.,
Ryan, K.C., and DeBanno, L.F. eds., Wildland fire in ecosystems: effects of fire on soils
and water, Ogden, UT, U.S. Department of Agriculture, Forest Service, Rocky Mountain
Research Station, v. 4, p. 22.
Köhler, P., Hartmann, J., and Wolf-Gladrow, D.A., 2010, Geoengineering potential of artificially
enhanced silicate weathering of olivine: Proceedings of the National Academy of Sciences,
v. 107, p. 20228–20233.
Konhauser, K.O., Kappler, A., and Roden, E.E., 2011, Iron in Microbial Metabolisms: Elements,
v. 7, p. 89–93, doi:10.2113/gselements.7.2.89.
Kraemer, S.M., 2004, Iron oxide dissolution and solubility in the presence of siderophores:
Aquatic Sciences - Research Across Boundaries, v. 66, p. 3–18, doi:10.1007/s00027-003-
0690-5.
Kraemer, S.M., Cheah, S.-F., Zapf, R., Xu, J., Raymond, K.N., and Sposito, G., 1999, Effect of
hydroxamate siderophores on Fe release and Pb (II) adsorption by goethite: Geochimica et
Cosmochimica Acta, v. 63, p. 3003–3008.
Krohling, C.A., Eutrópio, F.J., Bertolazi, A.A., Dobbss, L.B., Campostrini, E., Dias, T., and
Ramos, A.C., 2016, Ecophysiology of iron homeostasis in plants: Soil Science and Plant
Nutrition, v. 62, p. 39–47, doi:10.1080/00380768.2015.1123116.
Kuhn, K.M., DuBois, J.L., and Maurice, P.A., 2013, Strategies of aerobic microbial Fe acquisition
from Fe-bearing montmorillonite clay: Geochimica et Cosmochimica Acta, v. 117, p. 191–
202, doi:10.1016/j.gca.2013.04.028.
97
Kümmerli, R., Schiessl, K.T., Waldvogel, T., McNeill, K., and Ackermann, M., 2014, Habitat
structure and the evolution of diffusible siderophores in bacteria (M. van Baalen, Ed.):
Ecology Letters, v. 17, p. 1536–1544, doi:10.1111/ele.12371.
Kummert, R., and Stumm, W., 1980, The surface complexation of organic acids on hydrous γ-
Al2O3: Journal of Colloid and Interface Science, v. 75, p. 373–385.
Kump, L.R., Brantley, S.L., and Arthur, M.A., 2000, Chemical weathering, atmospheric CO2, and
climate: Annual Review of Earth and Planetary Sciences, v. 28, p. 611–667.
Lam, B.R., Barr, C.R., Rowe, A.R., and Nealson, K.H., 2019, Differences in Applied Redox
Potential on Cathodes Enrich for Diverse Electrochemically Active Microbial Isolates
From a Marine Sediment: Frontiers in Microbiology, v. 10, p. 1979,
doi:10.3389/fmicb.2019.01979.
Lane, N., 2017, Proton gradients at the origin of life: BioEssays, v. 39, p. 1600217,
doi:10.1002/bies.201600217.
Lane, N., Allen, J.F., and Martin, W., 2010, How did LUCA make a living? Chemiosmosis in the
origin of life: BioEssays, v. 32, p. 271–280, doi:10.1002/bies.200900131.
Lanza, N.L. et al., 2016, Oxidation of manganese in an ancient aquifer, Kimberley formation, Gale
crater, Mars: Manganese Fracture Fills in Gale Crater: Geophysical Research Letters, v.
43, p. 7398–7407, doi:10.1002/2016GL069109.
Ledyard, K.M., and Butler, A., 1997, Structure of putrebactin, a new dihydroxamate siderophore
produced by Shewanella putrefaciens: JBIC Journal of Biological Inorganic Chemistry, v.
2, p. 93–97, doi:10.1007/s007750050110.
Lemanceau, P., Bauer, P., Kraemer, S., and Briat, J.-F., 2009, Iron dynamics in the rhizosphere as
a case study for analyzing interactions between soils, plants and microbes: Plant and Soil,
v. 321, p. 513–535, doi:10.1007/s11104-009-0039-5.
Leventhal, G.E., Ackermann, M., and Schiessl, K.T., 2019, Why microbes secrete molecules to
modify their environment: the case of iron-chelating siderophores: Journal of The Royal
Society Interface, v. 16, p. 20180674, doi:10.1098/rsif.2018.0674.
Li, K., An, X., Park, K.H., Khraisheh, M., and Tang, J., 2014, A critical review of CO2
photoconversion: Catalysts and reactors: Catalysis Today, v. 224, p. 3–12,
doi:10.1016/j.cattod.2013.12.006.
Lide, D.R. (Ed.), 2006, CRC Handbook of Chemistry and Physics: Boca Raton, FL, CRC Press.
Liermann, L.J., Kalinowski, B.E., Brantley, S.L., and Ferry, J.G., 2000, Role of bacterial
siderophores in dissolution of hornblende: Geochimica et Cosmochimica Acta, v. 64, p.
587–602, doi:10.1016/S0016-7037(99)00288-4.
98
Lindsay, W.L., 1991, Iron oxide solubilization by organic matter and its effect on iron availability,
in Iron nutrition and interactions in plants, p. 29–36.
Lindsay, W.L., and Schwab, A.P., 1982, The chemistry of iron in soils and its availability to plants:
Journal of Pland Nutrition, v. 5, doi:https://doi.org/10.1080/01904168209363012.
Liu, L., Li, S., Wang, S., Dong, Z., and Gao, H., 2018, Complex Iron Uptake by the Putrebactin-
Mediated and Feo Systems in Shewanella oneidensis (N.-Y. Zhou, Ed.): Applied and
Environmental Microbiology, v. 84, p. e01752-18, /aem/84/20/e01752- 18.atom,
doi:10.1128/AEM.01752-18.
Liu, H., and Logan, B.E., 2004, Electricity Generation Using an Air-Cathode Single Chamber
Microbial Fuel Cell in the Presence and Absence of a Proton Exchange Membrane:
Environmental Science & Technology, v. 38, p. 4040–4046, doi:10.1021/es0499344.
Liu, C., Zachara, J.M., Foster, N.S., and Strickland, J., 2007, Kinetics of Reductive Dissolution of
Hematite by Bioreduced Anthraquinone-2,6-disulfonate: Environmental Science &
Technology, v. 41, p. 7730–7735, doi:10.1021/es070768k.
Logan, B.E., Call, D., Cheng, S., Hamelers, H.V.M., Sleutels, T.H.J.A., Jeremiasse, A.W., and
Rozendal, R.A., 2008, Microbial Electrolysis Cells for High Yield Hydrogen Gas
Production from Organic Matter: Environmental Science & Technology, v. 42, p. 8630–
8640, doi:10.1021/es801553z.
Lovelock, J.E., 1989, Geophysiology, the science of Gaia: Reviews of Geophysics, v. 27, p. 215,
doi:10.1029/RG027i002p00215.
Lovley, D.R., Stolzt, J.F., and Phillips, E.J.P., 1987, Anaerobic production of magnetite by a
dissimilatory iron-reducing microorganism: , p. 3.
Lyons, T.W., Reinhard, C.T., and Planavsky, N.J., 2014, The rise of oxygen in Earth’s early ocean
and atmosphere: Nature, v. 506, p. 307–315, doi:10.1038/nature13068.
Mackwell, S.J., 1992, Oxidation Kinetics of Fayalite (Fe 2 SiO 4): Phys Chem Minerals, p. 9.
Madigan, M.T., Martinko, J.M., Bender, K.S., Buckley, D.H., and Stahl, D.A., 2014, Brock
biology of microorganisms: Bonston, MA, Benjamin Cummings.
Maher, K., Steefel, C.I., DePaolo, D.J., and Viani, B.E., 2006, The mineral dissolution rate
conundrum: Insights from reactive transport modeling of U isotopes and pore fluid
chemistry in marine sediments: Geochimica et Cosmochimica Acta, v. 70, p. 337–363,
doi:10.1016/j.gca.2005.09.001.
Martin, J.H., 1990, Glacial-interglacial CO 2 change: The Iron Hypothesis: Paleoceanography, v.
5, p. 1–13, doi:10.1029/PA005i001p00001.
99
Martin, W., and Russell, M.J., 2007, On the origin of biochemistry at an alkaline hydrothermal
vent: Philosophical Transactions of the Royal Society B: Biological Sciences, v. 362, p.
1887–1926, doi:10.1098/rstb.2006.1881.
Maurice, P.A., Lee, Y.-J., and Hersman, L.E., 2000, Dissolution of Al-substituted goethites by an
aerobic Pseudomonas mendocina var. bacteria: Geochimica et Cosmochimica Acta, v. 64,
p. 1363–1374, doi:10.1016/S0016-7037(99)00404-4.
Maurice, P., Vierkorn, M.A., Hersman, L.E., Fulghum, J.E., and Ferryman, A., 2001,
Enhancement of Kaolinite Dissolution by an Aerobic Pseudomonas mendocina Bacterium:
Geomicrobiology Journal, v. 18, p. 21–35, doi:10.1080/01490450151079752.
Mawji, E., Gledhill, M., Milton, J.A., Tarran, G.A., Ussher, S., Thompson, A., Wolff, G.A.,
Worsfold, P.J., and Achterberg, E.P., 2008, Hydroxamate Siderophores: Occurrence and
Importance in the Atlantic Ocean: Environmental Science & Technology, v. 42, p. 8675–
8680, doi:10.1021/es801884r.
McConnell, I., Li, G., and Brudvig, G.W., 2010, Energy Conversion in Natural and Artificial
Photosynthesis: Chemistry & Biology, v. 17, p. 434–447,
doi:10.1016/j.chembiol.2010.05.005.
Meyen, F.E., Hecht, M.H., and Hoffman, J.A., 2016, Thermodynamic model of Mars Oxygen
ISRU Experiment (MOXIE): Acta Astronautica, v. 129, p. 82–87,
doi:10.1016/j.actaastro.2016.06.005.
Miller, S.L., 1953, A Production of Amino Acids under Possible Primitive Earth Conditions:
Science, v. 117, p. 528–529.
Millero, F.J., Sotolongo, S., and Izaguirre, M., 1987, The oxidation kinetics of Fe (II) in seawater:
Geochimica et Cosmochimica Acta, v. 51, p. 793–801.
Montserrat, F., Renforth, P., Hartmann, J., Leermakers, M., Knops, P., and Meysman, F.J.R., 2017,
Olivine Dissolution in Seawater: Implications for CO
2
Sequestration through Enhanced
Weathering in Coastal Environments: Environmental Science & Technology, v. 51, p.
3960–3972, doi:10.1021/acs.est.6b05942.
Moore, C.M. et al., 2013, Processes and patterns of oceanic nutrient limitation: Nature Geoscience,
v. 6, p. 701–710, doi:10.1038/ngeo1765.
Morel, F.M.M., and Price, N.M., 2003, The Biogeochemical Cycles of Trace Metals in the Oceans:
Science, v. 300, p. 944–947.
Morris, R.V. et al., 2006a, Mössbauer mineralogy of rock, soil, and dust at Meridiani Planum,
Mars: Opportunity’s journey across sulfate-rich outcrop, basaltic sand and dust, and
hematite lag deposits: IRON MINERALOGY AT MERIDIANI PLANUM: Journal of
Geophysical Research: Planets, v. 111, p. n/a-n/a, doi:10.1029/2006JE002791.
100
Morris, R.V. et al., 2006b, Mössbauer mineralogy of rock, soil, and dust at Meridiani Planum,
Mars: Opportunity’s journey across sulfate-rich outcrop, basaltic sand and dust, and
hematite lag deposits: IRON MINERALOGY AT MERIDIANI PLANUM: Journal of
Geophysical Research: Planets, v. 111, p. n/a-n/a, doi:10.1029/2006JE002791.
Morris, R.V., Klingelhoefer, G., Bernhardt, B., Schröder, C., Rodionov, D.S., De Souza, P.A.,
Yen, A., Gellert, R., Evlanov, E.N., and Foh, J., 2004, Mineralogy at Gusev Crater from
the Mössbauer spectrometer on the Spirit Rover: Science, v. 305, p. 833–836.
Myers, C.R., and Nealson, K.H., 1988, Microbial reduction of manganese oxides: Interactions with
iron and sulfur: Geochimica et Cosmochimica Acta, v. 52, p. 2727–2732,
doi:10.1016/0016-7037(88)90041-5.
Nano, G.V., and Strathmann, T.J., 2006, Ferrous iron sorption by hydrous metal oxides: Journal
of Colloid and Interface Science, v. 297, p. 443–454, doi:10.1016/j.jcis.2005.11.030.
Nealson, K.H., 2017, Bioelectricity (electromicrobiology) and sustainability: Microbial
Biotechnology, v. 10, p. 1114–1119, doi:10.1111/1751-7915.12834.
Nealson, K.H., and Saffarini, D., 1994, Iron and manganese in anaerobic respiration:
environmental significance, physiology, and regulation: Annual Reviews in Microbiology,
v. 48, p. 311–343.
Nevin, K.P., Woodard, T.L., Franks, A.E., Summers, Z.M., and Lovley, D.R., 2010, Microbial
Electrosynthesis: Feeding Microbes Electricity To Convert Carbon Dioxide and Water to
Multicarbon Extracellular Organic Compounds (R. R. Colwell, Ed.): mBio, v. 1,
doi:10.1128/mBio.00103-10.
Noinaj, N., Guillier, M., Barnard, T.J., and Buchanan, S.K., 2010, TonB-Dependent Transporters:
Regulation, Structure, and Function: Annual Review of Microbiology, v. 64, p. 43–60,
doi:10.1146/annurev.micro.112408.134247.
Novitsky, J., and Kepkay, P., 1981, Patterns of Microbial Heterotrophy Through Changing
Environments in a Marine Sediment: Marine Ecology Progress Series, v. 4, p. 1–7,
doi:10.3354/meps004001.
Oelkers, E.H., Benning, L.G., Lutz, S., Mavromatis, V., Pearce, C.R., and Plümper, O., 2015, The
efficient long-term inhibition of forsterite dissolution by common soil bacteria and fungi
at Earth surface conditions: Geochimica et Cosmochimica Acta, v. 168, p. 222–235,
doi:10.1016/j.gca.2015.06.004.
Olsen, A.A., 2007, Forsterite dissolution kinetics: Applications and implications for chemical
weathering [Ph.D.]: Virginia Institute of Technology, 69 p.
Olsen, A.A., and Rimstidt, D.J., 2008, Oxalate-promoted forsterite dissolution at low pH:
Geochimica et Cosmochimica Acta, v. 72, p. 1758–1766, doi:10.1016/j.gca.2007.12.026.
101
Orsetti, S., Laskov, C., and Haderlein, S.B., 2013, Electron Transfer between Iron Minerals and
Quinones: Estimating the Reduction Potential of the Fe(II)-Goethite Surface from AQDS
Speciation: Environmental Science & Technology, v. 47, p. 14161–14168,
doi:10.1021/es403658g.
Otero, R., Vázquez de Parga, A.L., and Gallego, J.M., 2017, Electronic, structural and chemical
effects of charge-transfer at organic/inorganic interfaces: Surface Science Reports, v. 72,
p. 105–145, doi:10.1016/j.surfrep.2017.03.001.
Palmer, K.L., Mashburn, L.M., Singh, P.K., and Whiteley, M., 2005, Cystic Fibrosis Sputum
Supports Growth and Cues Key Aspects of Pseudomonas aeruginosa Physiology: Journal
of Bacteriology, v. 187, p. 5267–5277, doi:10.1128/JB.187.15.5267-5277.2005.
Pedreira-Segade, U., Feuillie, C., Pelletier, M., Michot, L.J., and Daniel, I., 2016, Adsorption of
nucleotides onto ferromagnesian phyllosilicates: Significance for the origin of life:
Geochimica et Cosmochimica Acta, v. 176, p. 81–95, doi:10.1016/j.gca.2015.12.025.
Perez, A., Rossano, S., Trcera, N., Huguenot, D., Fourdrin, C., Verney-Carron, A., van
Hullebusch, E.D., and Guyot, F., 2016, Bioalteration of synthetic Fe(III)-, Fe(II)-bearing
basaltic glasses and Fe-free glass in the presence of the heterotrophic bacteria strain
Pseudomonas aeruginosa: Impact of siderophores: Geochimica et Cosmochimica Acta, v.
188, p. 147–162, doi:10.1016/j.gca.2016.05.028.
Petkowski, J.J., Bains, W., and Seager, S., 2020, On the Potential of Silicon as a Building Block
for Life: Life, v. 10, p. 84, doi:10.3390/life10060084.
Pfeffer, C. et al., 2012, Filamentous bacteria transport electrons over centimetre distances: Nature,
v. 491, p. 218–221, doi:10.1038/nature11586.
Pietramellara, G., Ascher, J., Borgogni, F., Ceccherini, M.T., Guerri, G., and Nannipieri, P., 2009,
Extracellular DNA in soil and sediment: fate and ecological relevance: Biology and
Fertility of Soils, v. 45, p. 219–235, doi:10.1007/s00374-008-0345-8.
Pinedo-Gonzalez, P., West, A.J., Rivera-Duarte, I., and Sañudo-Wilhelmy, S.A., 2014, Diel
Changes in Trace Metal Concentration and Distribution in Coastal Waters: Catalina Island
As a Study Case: Environmental Science & Technology, v. 48, p. 7730–7737,
doi:10.1021/es5019515.
Poch, O., Jaber, M., Stalport, F., Nowak, S., Georgelin, T., Lambert, J.-F., Szopa, C., and Coll, P.,
2015, Effect of Nontronite Smectite Clay on the Chemical Evolution of Several Organic
Molecules under Simulated Martian Surface Ultraviolet Radiation Conditions:
Astrobiology, v. 15, p. 221–237, doi:10.1089/ast.2014.1230.
Pokrovsky, O.S., and Schott, J., 2000, Kinetics and mechanism of forsterite dissolution at 25 C
and pH from 1 to 12: Geochimica et Cosmochimica Acta, v. 64, p. 3313–3325.
102
Pokrovsky, O.S., Shirokova, L.S., Bénézeth, P., Schott, J., and Golubev, S.V., 2009, Effect of
Organic Ligands in Heterotrophic Bacteria on Wollastonite Dissolution Kinetics:
American Journal of Science,.
Preiner, M., Xavier, J.C., Vieira, A. do N., Kleinermanns, K., Allen, J.F., and Martin, W.F., 2019,
Catalysts, autocatalysis and the origin of metabolism: Interface Focus, v. 9, p. 20190072,
doi:10.1098/rsfs.2019.0072.
Puigbò, P., Lobkovsky, A.E., Kristensen, D.M., Wolf, Y.I., and Koonin, E.V., 2014, Genomes in
turmoil: quantification of genome dynamics in prokaryote supergenomes: BCM Biology,
p. 19.
Rabaey, K., Read, S.T., Clauwaert, P., Freguia, S., Bond, P.L., Blackall, L.L., and Keller, J., 2008,
Cathodic oxygen reduction catalyzed by bacteria in microbial fuel cells: The ISME Journal,
v. 2, p. 519–527, doi:10.1038/ismej.2008.1.
Rabaey, K., and Rozendal, R.A., 2010, Microbial electrosynthesis — revisiting the electrical route
for microbial production: Nature Reviews Microbiology, v. 8, p. 706–716,
doi:10.1038/nrmicro2422.
Raymond, K.N., Allred, B.E., and Sia, A.K., 2015, Coordination Chemistry of Microbial Iron
Transport: Accounts of Chemical Research, v. 48, p. 2496–2505,
doi:10.1021/acs.accounts.5b00301.
Raymond, K.N., Dertz, E.A., and Kim, S.S., 2003, Enterobactin: An archetype for microbial iron
transport: Proceedings of the National Academy of Sciences, v. 100, p. 3584–3588,
doi:10.1073/pnas.0630018100.
Reichard, P.U., Kretzschmar, R., and Kraemer, S.M., 2007, Dissolution mechanisms of goethite
in the presence of siderophores and organic acids: Geochimica et Cosmochimica Acta, v.
71, p. 5635–5650, doi:10.1016/j.gca.2006.12.022.
Rich, H.W., and Morel, F.M.M., 1990, Availability of Well-Defined Iron Colloids to the Marine
Diatom Thalassiosira weissflogii: Limnology and Oceanography, v. 35, p. 652–662.
Rimstidt, J.D., Brantley, S.L., and Olsen, A.A., 2012, Systematic review of forsterite dissolution
rate data: Geochimica et Cosmochimica Acta, v. 99, p. 159–178,
doi:10.1016/j.gca.2012.09.019.
Roberts, A.A., Schultz, A.W., Kersten, R.D., Dorrestein, P.C., and Moore, B.S., 2012, Iron
acquisition in the marine actinomycete genus Salinispora is controlled by the
desferrioxamine family of siderophores: FEMS Microbiology Letters, v. 335, p. 95–103,
doi:10.1111/j.1574-6968.2012.02641.x.
Rogers, J.R., and Bennett, P.C., 2004, Mineral stimulation of subsurface microorganisms: release
of limiting nutrients from silicates: Chemical Geology, v. 203, p. 91–108,
doi:10.1016/j.chemgeo.2003.09.001.
103
Rosenbaum, M., Aulenta, F., Villano, M., and Angenent, L.T., 2011, Cathodes as electron donors
for microbial metabolism: Which extracellular electron transfer mechanisms are involved?
Bioresource Technology, v. 102, p. 324–333, doi:10.1016/j.biortech.2010.07.008.
Ross, D.E., Flynn, J.M., Baron, D.B., Gralnick, J.A., and Bond, D.R., 2011, Towards
Electrosynthesis in Shewanella: Energetics of Reversing the Mtr Pathway for Reductive
Metabolism (S. Xu, Ed.): PLoS ONE, v. 6, p. e16649, doi:10.1371/journal.pone.0016649.
Rowe, A.R., Chellamuthu, P., Lam, B., Okamoto, A., and Nealson, K.H., 2015, Marine sediments
microbes capable of electrode oxidation as a surrogate for lithotrophic insoluble substrate
metabolism: Frontiers in Microbiology, v. 5, doi:10.3389/fmicb.2014.00784.
Rowe, A.R., Rajeev, P., Jain, A., Pirbadian, S., Okamotao, A., Gralnick, J.A., El-Naggar, M.Y.,
and Nealson, K., 2017, Tracking electron uptake from a cathode into Shewanella cells:
implications for generating maintenance energy from solid substrates.: bioRxiv, p. 116475.
Rubie, D.C., Gessmann, C.K., and Frost, D.J., 2004, Partitioning of oxygen during core formation
on the Earth and Mars: Nature, v. 429, doi:10.1038/nature02473.
Ruddiman, W.F. (Ed.), 1997, Tectonic Uplift and Climate Change: Charlottesville, Virginia,
University of Virginia, http://dx.doi.org/10.1007/978-1-4615-5935-1 (accessed April
2018).
Rue, E.L., and Bruland, K.W., 1995, Complexation of iron(III) by natural organic ligands in the
Central North Pacific as determined by a new competitive ligand equilibration/adsorptive
cathodic stripping voltammetric method: Marine Chemistry, v. 50, p. 117–138.
Russell, M.J., and Martin, W., 2004, The rocky roots of the acetyl-CoA pathway: Trends in
Biochemical Sciences, v. 29, p. 358–363, doi:10.1016/j.tibs.2004.05.007.
Russell, M.J., Nitschke, W., and Branscomb, E., 2013, The inevitable journey to being:
Philosophical Transactions of the Royal Society B: Biological Sciences, v. 368, p.
20120254–20120254, doi:10.1098/rstb.2012.0254.
Sagan, C., 1994, Pale blue dot: a vision of the human future in space: New York, Random House.
Saito, M.A., Sigman, D.M., and Morel, F.M.M., 2003, The bioinorganic chemistry of the ancient
ocean: the co-evolution of cyanobacterial metal requirements and biogeochemical cycles
at the Archean–Proterozoic boundary? Inorganica Chimica Acta, v. 356, p. 308–318,
doi:10.1016/S0020-1693(03)00442-0.
Sand, K.K., and Jelavić, S., 2018, Mineral Facilitated Horizontal Gene Transfer: A New Principle
for Evolution of Life? Frontiers in Microbiology, v. 9, p. 2217,
doi:10.3389/fmicb.2018.02217.
Sander, S.G., and Koschinsky, A., 2011, Metal flux from hydrothermal vents increased by organic
complexation: Nature Geoscience, v. 4, p. 145–150, doi:10.1038/ngeo1088.
104
Santelli, C.M., Welch, S.A., Westrich, H.R., and Banfield, J.F., 2001, The effect of Fe-oxidizing
bacteria on Fe-silicate mineral dissolution: Chemical Geology, v. 180, p. 99–115.
Sañudo-Wilhelmy, S.A., Gómez-Consarnau, L., Suffridge, C., and Webb, E.A., 2014, The Role of
B Vitamins in Marine Biogeochemistry: Annual Review of Marine Science, v. 6, p. 339–
367, doi:10.1146/annurev-marine-120710-100912.
Schalk, I.J., and Guillon, L., 2013, Pyoverdine biosynthesis and secretion in Pseudomonas
aeruginosa : implications for metal homeostasis: Pyoverdine biosynthesis: Environmental
Microbiology, v. 15, p. 1661–1673, doi:10.1111/1462-2920.12013.
Schippers, A., Sand, W., and Glombitza, F. (Eds.), 2014, Geobiotechnology I: New York,
Springer.
Scholz, R.L., and Greenberg, E.P., 2015, Sociality in Escherichia coli: Enterochelin Is a Private
Good at Low Cell Density and Can Be Shared at High Cell Density (G. A. O’Toole, Ed.):
Journal of Bacteriology, v. 197, p. 2122–2128, doi:10.1128/JB.02596-14.
Schott, J., and Berner, R.A., 1983, X-ray photoelectron studies of the mechanism of iron silicate
dissolution during weathering: Geochimica et Cosmochimica Acta, v. 47, p. 2233–2240,
doi:10.1016/0016-7037(83)90046-7.
Schroth, A.W., Crusius, J., Sholkovitz, E.R., and Bostick, B.C., 2009, Iron solubility driven by
speciation in dust sources to the ocean: Nature Geoscience, v. 2, p. 337–340,
doi:10.1038/ngeo501.
Schwecke, T., Göttling, K., Durek, P., Dueñas, I., Käufer, N.F., Zock-Emmenthal, S., Staub, E.,
Neuhof, T., Dieckmann, R., and von Döhren, H., 2006, Nonribosomal Peptide Synthesis
in Schizosaccharomyces pombe and the Architectures of Ferrichrome-Type Siderophore
Synthetases in Fungi: ChemBioChem, v. 7, p. 612–622, doi:10.1002/cbic.200500301.
Schwertmann, U., 1991a, Solubility and dissolution of iron oxides, in Chen, Y. and Hadar, Y. eds.,
Iron Nutrition and Interactions in Plants: “Proceedings of the Fifth International
Symposium on Iron Nutrition and Interactions in Plants”, 11–17 June 1989, Jerusalem,
Israel, 1989, Dordrecht, Springer Netherlands, p. 3–27, doi:10.1007/978-94-011-3294-
7_1.
Schwertmann, U., 1991b, Solubility and dissolution of iron oxides, in Chen, Y. and Hadar, Y. eds.,
Iron Nutrition and Interactions in Plants, Dordrecht, Springer, v. 43, p. 25.
Schwyn, B., and Neilands, J.B., 1987, Universal Chemical Assay for the Detection and
Determination of Siderophores: Analytical Biochemistry, v. 160, p. 47–56.
Sheng, X., Qin, C., Yang, B., Hu, X., Liu, C., Waigi, M.G., Li, X., and Ling, W., 2019, Metal
cation saturation on montmorillonites facilitates the adsorption of DNA via cation bridging:
Chemosphere, v. 235, p. 670–678, doi:10.1016/j.chemosphere.2019.06.159.
105
Shi, L., Dong, H., Reguera, G., Beyenal, H., Lu, A., Liu, J., Yu, H.-Q., and Fredrickson, J.K.,
2016, Extracellular electron transfer mechanisms between microorganisms and minerals:
Nature Reviews Microbiology, v. 14, p. 651–662, doi:10.1038/nrmicro.2016.93.
Sholkovitz, E.R., 1978, The flocculation of dissolved Fe, Mn, Al, Cu, NI, Co and Cd during
estuarine mixing: Earth and Planetary Science Letters, v. 41, p. 77–86.
Sholkovitz, E.R., Boyle, E.A., and Price, N.B., 1978, The removal of dissolved humic acids and
iron during estuarine mixing: Earth and Planetary Science Letters, v. 40, p. 130–136.
Shyu, J.B.H., Lies, D.P., and Newman, D.K., 2002, Protective Role of tolC in Efflux of the
Electron Shuttle Anthraquinone-2,6-Disulfonate: Journal of Bacteriology, v. 184, p. 1806–
1810, doi:10.1128/JB.184.6.1806-1810.2002.
Slon, V. et al., 2017, Neandertal and Denisovan DNA from Pleistocene sediments: Science, v. 356,
p. 605–608, doi:10.1126/science.aam9695.
Sousa, F.L., Thiergart, T., Landan, G., Nelson-Sathi, S., Pereira, I.A.C., Allen, J.F., Lane, N., and
Martin, W.F., 2013, Early bioenergetic evolution: Philosophical Transactions of the Royal
Society B: Biological Sciences, v. 368, p. 20130088–20130088,
doi:10.1098/rstb.2013.0088.
Stanley, N.R., and Lazazzera, B.A., 2004, Environmental signals and regulatory pathways that
influence biofilm formation: Signalling in biofilm formation: Molecular Microbiology, v.
52, p. 917–924, doi:10.1111/j.1365-2958.2004.04036.x.
Stockmann, G.J., Shirokova, L.S., Pokrovsky, O.S., Bénézeth, P., Bovet, N., Gislason, S.R., and
Oelkers, E.H., 2012, Does the presence of heterotrophic bacterium Pseudomonas reactans
affect basaltic glass dissolution rates? Chemical Geology, v. 296–297, p. 1–18,
doi:10.1016/j.chemgeo.2011.12.011.
Stranghoener, M., Schippers, A., Dultz, S., and Behrens, H., 2018, Experimental Microbial
Alteration and Fe Mobilization From Basaltic Rocks of the ICDP HSDP2 Drill Core, Hilo,
Hawaii: Frontiers in Microbiology, v. 9, p. 1252, doi:10.3389/fmicb.2018.01252.
Stumm, W., 1995, The Inner-Sphere Surface Complex: A Key to Understanding Surface
Reactivity, in Huang, C.P., O’Melia, C.R., and Morgan, J.J. eds., Aquatic Chemistry,
Washington, DC, American Chemical Society, v. 244, p. 1–32, doi:10.1021/ba-1995-
0244.ch001.
Sudek, L.A., Wanger, G., Templeton, A.S., Staudigel, H., and Tebo, B.M., 2017, Submarine
Basaltic Glass Colonization by the Heterotrophic Fe(II)-Oxidizing and Siderophore-
Producing Deep-Sea Bacterium Pseudomonas stutzeri VS-10: The Potential Role of Basalt
in Enhancing Growth: Frontiers in Microbiology, v. 8, doi:10.3389/fmicb.2017.00363.
106
Thode, S.K., Rojek, E., Kozlowski, M., Ahmad, R., and Haugen, P., 2018, Distribution of
siderophore gene systems on a Vibrionaceae phylogeny: Database searches, phylogenetic
analyses and evolutionary perspectives (C.-H. Kuo, Ed.): PLOS ONE, v. 13, p. e0191860,
doi:10.1371/journal.pone.0191860.
Thorseth, I.H., Pedersen, R.B., and Christie, D.M., 2003, Microbial alteration of a 0-30Ma seafloor
and sub-seafloor basaltic glasses from the Australian Antarctic Discordance: Earth and
Planetary Science Letters, v. 215, p. 237–247.
Torres, M.A., Dong, S., Nealson, K.H., and West, A.J., 2019, The kinetics of siderophore-mediated
olivine dissolution: Geobiology, doi:10.1111/gbi.12332.
Tovar-Sanchez, A., and Sañudo-Wilhelmy, S.A., 2011, Influence of the Amazon River on
dissolved and intra-cellular metal concentrations in <i>Trichodesmium</i>
colonies along the western boundary of the sub-tropical North Atlantic Ocean:
Biogeosciences, v. 8, p. 217–225, doi:10.5194/bg-8-217-2011.
Tuff, J., Wade, J., and Wood, B.J., 2013, Volcanism on Mars controlled by early oxidation of the
upper mantle: Nature, v. 498, p. 342–345, doi:10.1038/nature12225.
Velasquez, I., Nunn, B.L., Ibisanmi, E., Goodlett, D.R., Hunter, K.A., and Sander, S.G., 2011,
Detection of hydroxamate siderophores in coastal and Sub-Antarctic waters off the South
Eastern Coast of New Zealand: Marine Chemistry, v. 126, p. 97–107,
doi:10.1016/j.marchem.2011.04.003.
Venkateswaran, K. et al., 1999, Polyphasic taxonomy of the genus Shewanella and description of
Shewanella oneidensis sp. nov.: International Journal of Systematic Bacteriology, v. 49, p.
705–724.
Volbeda, A., and Fontecilla-Camps, J.C., 2006, Catalytic Nickel–Iron–Sulfur Clusters: From
Minerals to Enzymes, in Simonneaux, G. ed., Bioorganometallic Chemistry,
Berlin/Heidelberg, Springer-Verlag, Topics in Organometallic Chemistry, v. 17, p. 57–82,
doi:10.1007/3418_003.
Völker, C., and Wolf-Gladrow, D.A., 1999, Physical limits on iron uptake mediated by
siderophores or surface reductases: Marine Chemistry, v. 65, p. 227–244,
doi:10.1016/S0304-4203(99)00004-3.
Wächtershäuser, G., 1990, Evolution of the first metabolic cycles.: Proceedings of the National
Academy of Sciences, v. 87, p. 200–204, doi:10.1073/pnas.87.1.200.
Wächtershäuser, G., 2000, Life as We Don’t Know It: Science, v. 289, p. 1307–1908,
doi:10.1126/science.289.5483.1307.
Wang, O., and Coates, J., 2017, Biotechnological Applications of Microbial (Per)chlorate
Reduction: Microorganisms, v. 5, p. 76, doi:10.3390/microorganisms5040076.
107
Wang, Y., and Newman, D.K., 2008, Redox Reactions of Phenazine Antibiotics with Ferric
(Hydr)oxides and Molecular Oxygen: Environmental Science & Technology, v. 42, p.
2380–2386, doi:10.1021/es702290a.
Wang, Z., Schenkeveld, W.D.C., Kraemer, S.M., and Giammar, D.E., 2015, Synergistic Effect of
Reductive and Ligand-Promoted Dissolution of Goethite: Environmental Science &
Technology, v. 49, p. 7236–7244, doi:10.1021/acs.est.5b01191.
Wang, Y., Wilks, J.C., Danhorn, T., Ramos, I., Croal, L., and Newman, D.K., 2011, Phenazine-1-
Carboxylic Acid Promotes Bacterial Biofilm Development via Ferrous Iron Acquisition:
Journal of Bacteriology, v. 193, p. 3606–3617, doi:10.1128/JB.00396-11.
Watson, A.J., Bakker, D.C.E., Ridgwell, A.J., Boyd, P.W., and Law, C.S., 2000, Effect of Iron
supply on Southern Ocean CO2 uptake and Implications for glacial atmospheric CO2:
Nature, v. 407, p. 730–733.
Whitchurch, C.B., 2002, Extracellular DNA Required for Bacterial Biofilm Formation: Science,
v. 295, p. 1487–1487, doi:10.1126/science.295.5559.1487.
White, A.F., and Brantley, S.L., 2003, The effect of time on the weathering of silicate minerals:
why do weathering rates differ in the laboratory and field? Chemical Geology, v. 202, p.
479–506, doi:10.1016/j.chemgeo.2003.03.001.
Wolff-Boenisch, D., Wenau, S., Gislason, S.R., and Oelkers, E.H., 2011, Dissolution of basalts
and peridotite in seawater, in the presence of ligands, and CO2: Implications for mineral
sequestration of carbon dioxide: Geochimica et Cosmochimica Acta, v. 75,
doi:10.1016/j.gca.2011.07.004.
Wu, J., Boyle, E., Sunda, W., and Wen, L.-S., 2001, Soluble and colloidal iron in the oligotrophic
North Atlantic and North Pacific: Science, v. 293, p. 847–849.
Wu, L., Jacobson, A.D., Chen, H.-C., and Hausner, M., 2007, Characterization of elemental release
during microbe–basalt interactions at T=28°C: Geochimica et Cosmochimica Acta, v. 71,
p. 2224–2239, doi:10.1016/j.gca.2007.02.017.
Wu, L., Jacobson, A.D., and Hausner, M., 2008, Characterization of elemental release during
microbe–granite interactions at T=28°C: Geochimica et Cosmochimica Acta, v. 72, p.
1076–1095, doi:10.1016/j.gca.2007.11.025.
Yamashita, T., and Hayes, P., 2008, Analysis of XPS spectra of Fe2+ and Fe3+ ions in oxide
materials: Applied Surface Science, v. 254, p. 2441–2449,
doi:10.1016/j.apsusc.2007.09.063.
Yoshida, T., Hayashi, K., and Ohmoto, H., 2002, Dissolution of iron hydroxides by marine
bacterial siderophore: Chemical geology, v. 184, p. 1–9.
108
Zaharescu, D.G. et al., 2019, Ecosystem-bedrock interaction changes nutrient
compartmentalization during early oxidative weathering: Scientific Reports, v. 9, p. 15006,
doi:10.1038/s41598-019-51274-x.
Zakaznova-Herzog, V.P., Nesbitt, H.W., Bancroft, G.M., and Tse, J.S., 2008, Characterization of
leached layers on olivine and pyroxenes using high-resolution XPS and density functional
calculations: Geochimica et Cosmochimica Acta, v. 72, p. 69–86,
doi:10.1016/j.gca.2007.09.031.
Zakaznova-Herzog, V.P., Nesbitt, H.W., Bancroft, G.M., and Tse, J.S., 2006, High resolution core
and valence band XPS spectra of non-conductor pyroxenes: Surface Science, v. 600, p.
3175–3186, doi:10.1016/j.susc.2006.06.012.
Zakaznova-Herzog, V.P., Nesbitt, H.W., Bancroft, G.M., Tse, J.S., Gao, X., and Skinner, W.,
2005, High-resolution valence-band XPS spectra of the nonconductors quartz and olivine:
Physical Review B, v. 72, p. 205113, doi:10.1103/PhysRevB.72.205113.
Zhao, J., Zhang, C., Sun, C., Li, W., Zhang, S., Li, S., and Zhang, D., 2018, Electron transfer
mechanism of biocathode in a bioelectrochemical system coupled with chemical
absorption for NO removal: Bioresource Technology, v. 254, p. 16–22,
doi:10.1016/j.biortech.2018.01.066.
109
Appendix A
Data from cell growth experiments from olivine iron acquisition.
S. oneidensis experiments
Cell growth from filtrate amendment experiments (Fig. 1).
Volumes (mL)
Exp # Condition
Net # of cells
added
MR-1
Filtrate Medium
Ol added
(300-150um) Total
1 MR-1Δ control 1.31E+09 -- 49.5 0.05 g 50
2 MR-1Δ control 1.31E+09 -- 49.5 0.05 g 50
3 MR-1Δ control 1.31E+09 -- 49.5 0.05 g 50
4 MR-1 control 1.41E+09 -- 48.5
0.05 g
50
5 MR-1 control 1.41E+09 -- 48.5 0.05 g 50
6 MR-1 control 1.41E+09 -- 48.5 0.05 g 50
7 MR-1Δ time 0 1.31E+09 5 44.5 0.05 g 50
8 MR-1Δ time 0 1.31E+09 5 44.5 0.05 g 50
9 MR-1Δ time 0 1.31E+09 5 44.5 0.05 g 50
10 MR-1Δ time 24 1.31E+09 5 44.5 0.05 g 45
11 MR-1Δ time 24 1.31E+09 5 44.5 0.05 g 45
12 MR-1Δ time 24 1.31E+09 5 44.5 0.05 g 45
13 Killed control 1.31E+09 -- 49.5 0.05 g 50
14 Killed control 1.31E+09 -- 49.5 0.05 g 50
Time
(h) Exp # 1 2 3 4 5 6 7 8 9 10 11 12 13 14
0
OD 600
-0.001
-0.001
0
0.002
0
-0.002
0
0.001
0.001
0
0
-0.001
0.003
0.001
Cells/mL
-5.00E+06
-5.00E+06
0.00E+00
1.00E+07
0.00E+00
-1.00E+07
0.00E+00
5.00E+06
5.00E+06
0.00E+00
0.00E+00
-5.00E+06
1.50E+07
5.00E+06
Ave
-3.33E+06
0.00E+00
3.33E+06
-1.67E+06
1.00E+07
Stdev
2.89E+06
1.00E+07
2.89E+06
2.89E+06
7.07E+06
13.75
OD 600
0.004
0.005
0.004
0.061
0.064
0.078
0.187
0.19
0.186
0.011
0.013
0.005
0.005
0.003
110
Cells/mL
2.00E+07
2.50E+07
2.00E+07
3.05E+08
3.20E+08
3.90E+08
9.35E+08
9.50E+08
9.30E+08
5.50E+07
6.50E+07
2.50E+07
2.50E+07
1.50E+07
Ave
2.17E+07
3.38E+08
9.38E+08
4.83E+07
2.00E+07
Stdev
2.89E+06
4.54E+07
1.04E+07
2.08E+07
7.07E+06
15
OD 600
0.012
0.01
0.007
0.087
0.087
0.101
0.229
0.24
0.233
0.007
0.009
0.008
0.004
0.004
Cells/mL
6.00E+07
5.00E+07
3.50E+07
4.35E+08
4.35E+08
5.05E+08
1.15E+09
1.20E+09
1.17E+09
3.50E+07
4.50E+07
4.00E+07
2.00E+07
2.00E+07
Ave
4.83E+07
4.58E+08
1.17E+09
4.00E+07
2.00E+07
Stdev
1.26E+07
4.04E+07
2.78E+07
5.00E+06
0.00E+00
17.5
OD 600
0.009
0.007
0.01
0.166
0.161
0.21
0.347
0.346
0.336
0.011
0.012
0.014
0.005
0.003
Cells/mL
4.50E+07
3.50E+07
5.00E+07
8.30E+08
8.05E+08
1.05E+09
1.74E+09
1.73E+09
1.68E+09
5.50E+07
6.00E+07
7.00E+07
2.50E+07
1.50E+07
Ave
4.33E+07
8.95E+08
1.72E+09
6.17E+07
2.00E+07
Stdev
7.64E+06
1.35E+08
3.04E+07
7.64E+06
7.07E+06
111
20
OD 600
0.007
0.007
0.007
0.222
0.215
0.214
0.37
0.342
0.336
0.009
0.011
0.011
0.007
0.004
Cells/mL
3.50E+07
3.50E+07
3.50E+07
1.11E+09
1.08E+09
1.07E+09
1.85E+09
1.71E+09
1.68E+09
4.50E+07
5.50E+07
5.50E+07
3.50E+07
2.00E+07
Ave
3.50E+07
1.09E+09
1.75E+09
5.17E+07
2.75E+07
Stdev
0.00E+00
2.18E+07
9.07E+07
5.77E+06
1.06E+07
24
OD 600
0.013
0.01
0.012
0.43
0.473
0.441
0.455
0.438
0.416
0.015
0.012
0.01
0.014
0.014
Cells/mL
6.50E+07
5.00E+07
6.00E+07
2.15E+09
2.37E+09
2.21E+09
2.28E+09
2.19E+09
2.08E+09
7.50E+07
6.00E+07
5.00E+07
7.00E+07
7.00E+07
Ave
5.83E+07
2.24E+09
2.18E+09
6.17E+07
7.00E+07
Stdev
7.64E+06
1.12E+08
9.78E+07
1.26E+07
0.00E+00
27.5
OD 600
0.011
0.009
0.007
0.702
0.81
0.796
0.604
0.59
0.512
0.023
0.022
0.022
0.011
0.001
Cells/mL
5.50E+07
4.50E+07
3.50E+07
3.51E+09
4.05E+09
3.98E+09
3.02E+09
2.95E+09
2.56E+09
1.15E+08
1.10E+08
1.10E+08
5.50E+07
5.00E+06
Ave
4.50E+07
3.85E+09
2.84E+09
1.12E+08
3.00E+07
112
Stdev
1.00E+07
2.94E+08
2.48E+08
2.89E+06
3.54E+07
31
OD 600
0.013
0.008
0.009
0.76
0.916
0.916
0.706
0.666
0.596
0.039
0.037
0.037
0.009
0.001
Cells/mL
6.50E+07
4.00E+07
4.50E+07
3.80E+09
4.58E+09
4.58E+09
3.53E+09
3.33E+09
2.98E+09
1.95E+08
1.85E+08
1.85E+08
4.50E+07
5.00E+06
Ave
5.00E+07
4.32E+09
3.28E+09
1.88E+08
2.50E+07
Stdev
1.32E+07
4.50E+08
2.78E+08
5.77E+06
2.83E+07
37.75
OD 600
0.014
0.012
0.012
1.097
1.19
1.128
0.77
0.707
0.653
0.153
0.148
0.155
0.009
0.001
Cells/mL
7.00E+07
6.00E+07
6.00E+07
5.49E+09
5.95E+09
5.64E+09
3.85E+09
3.54E+09
3.27E+09
7.65E+08
7.40E+08
7.75E+08
4.50E+07
5.00E+06
Ave
6.33E+07
5.69E+09
3.55E+09
7.60E+08
2.50E+07
Stdev
5.77E+06
2.37E+08
2.93E+08
1.80E+07
2.83E+07
41.5
OD 600
0.015
0.013
0.013
1.141
1.219
1.183
0.8
0.703
0.64
0.325
0.324
0.321
0.016
0
Cells/mL
7.50E+07
6.50E+07
6.50E+07
5.71E+09
6.10E+09
5.92E+09
4.00E+09
3.52E+09
3.20E+09
1.63E+09
1.62E+09
1.61E+09
8.00E+07
0.00E+00
113
Ave
6.83E+07
5.91E+09
3.57E+09
1.62E+09
4.00E+07
Stdev
5.77E+06
1.95E+08
4.03E+08
1.04E+07
5.66E+07
45
OD 600
0.011
0.012
0.011
1.094
1.116
1.234
0.918
0.74
0.705
0.48
0.46
0.412
0.011
-0.001
Cells/mL
5.50E+07
6.00E+07
5.50E+07
5.47E+09
5.58E+09
6.17E+09
4.59E+09
3.70E+09
3.53E+09
2.40E+09
2.30E+09
2.06E+09
5.50E+07
-5.00E+06
Ave
5.67E+07
5.74E+09
3.94E+09
2.25E+09
2.50E+07
Stdev
2.89E+06
3.76E+08
5.71E+08
1.75E+08
4.24E+07
50.5
OD 600
0.012
0.011
0.01
1.102
1.15
1.086
1.095
0.818
0.78
0.667
0.62
0.567
0.008
-0.002
Cells/mL
6.00E+07
5.50E+07
5.00E+07
5.51E+09
5.75E+09
5.43E+09
5.48E+09
4.09E+09
3.90E+09
3.34E+09
3.10E+09
2.84E+09
4.00E+07
-1.00E+07
Ave
5.50E+07
5.56E+09
4.49E+09
3.09E+09
1.50E+07
Stdev
5.00E+06
1.67E+08
8.60E+08
2.50E+08
3.54E+07
Cell growth from Deferoxamine amendment experiments (Fig. 5):
114
First set
Volumes (mL)
Exp
# Condition
Net # of cells
added DFx added
DFx
concentration
Medium
(no Fe)
Ol added
(300-
150um) Total
1
MR-1Δ +
DFx 1.31E+09
2.5 umol (179uL
of 14.0mM) 50 uM 49.3 0.05 g 50
2
MR-1Δ +
DFx 1.31E+09
2.5 umol (179uL
of 14.0mM) 50 uM 49.3 0.05 g 50
3
MR-1Δ +
DFx 1.31E+09
2.5 umol (179uL
of 14.0mM) 50 uM 49.3 0.05 g 50
4
MR-1Δ +
DFx 1.31E+09
5.0 umol (172uL
of 29.1mM) 100 uM 49.3 0.05 g 50
5
MR-1Δ +
DFx 1.31E+09
5.0 umol (172uL
of 29.1mM) 100 uM 49.3 0.05 g 50
6
MR-1Δ +
DFx 1.31E+09
5.0 umol (172uL
of 29.1mM) 100 uM 49.3 0.05 g 50
7 MR-1Δ 1.31E+09 0 0 49.5 0.05 g 50
8 MR-1Δ 1.31E+09 0 0 49.5 0.05 g 50
9 MR-1Δ 1.31E+09 0 0 49.5 0.05 g 50
Time (h) Exp # 1 2 3 4 5 6 7 8 9
1
OD600 0.002 0.002 0.001 0.001 0.001 0.002 0.002 0.001 0.001
cells/mL 1.00E+07 1.00E+07 5.00E+06 5.00E+06 5.00E+06 1.00E+07 1.00E+07 5.00E+06 5.00E+06
Ave 8.33E+06 6.67E+06 6.67E+06
Stdev 2.89E+06 2.89E+06 2.89E+06
13.5
OD600 0.118 0.102 0.105 0.080 0.069 0.091 0.004 0.005 0.008
cells/mL 5.90E+08 5.10E+08 5.25E+08 4.00E+08 3.45E+08 4.55E+08 2.00E+07 2.50E+07 4.00E+07
Ave 5.42E+08 4.00E+08 2.83E+07
Stdev 4.25E+07 5.50E+07 1.04E+07
17
OD600 0.248 0.233 0.244 0.217 0.202 0.215 0.004 0.002 0.002
cells/mL 1.24E+09 1.17E+09 1.22E+09 1.09E+09 1.01E+09 1.08E+09 2.00E+07 1.00E+07 1.00E+07
Ave 1.21E+09 1.06E+09 1.33E+07
Stdev 3.88E+07 4.07E+07 5.77E+06
20
OD600 0.464 0.405 0.418 0.377 0.341 0.363 0.002 0.002 0.003
cells/mL 2.32E+09 2.03E+09 2.09E+09 1.89E+09 1.71E+09 1.82E+09 1.00E+07 1.00E+07 1.50E+07
Ave 2.15E+09 1.80E+09 1.17E+07
Stdev 1.55E+08 9.07E+07 2.89E+06
OD600 0.61 0.58 0.608 0.546 0.518 0.531 -0.001 0.005 0.003
22
cells/mL 3.05E+09 2.90E+09 3.04E+09 2.73E+09 2.59E+09 2.66E+09 -5.00E+06 2.50E+07 1.50E+07
Ave 3.00E+09 2.66E+09 1.17E+07
Stdev 8.39E+07 7.01E+07 1.53E+07
24.5
OD600 0.712 0.704 0.75 0.646 0.678 0.669 0.003 0.003 0.003
cells/mL 3.56E+09 3.52E+09 3.75E+09 3.23E+09 3.39E+09 3.35E+09 1.50E+07 1.50E+07 1.50E+07
Ave 3.61E+09 3.32E+09 1.50E+07
Stdev 1.23E+08 8.25E+07 0.00E+00
27
OD600 0.79 0.797 0.824 0.706 0.786 0.766 0.007 0.003 0.01
cells/mL 3.95E+09 3.99E+09 4.12E+09 3.53E+09 3.93E+09 3.83E+09 3.50E+07 1.50E+07 5.00E+07
Ave 4.02E+09 3.76E+09 3.33E+07
Stdev 8.98E+07 2.08E+08 1.76E+07
31
OD600 0.879 0.886 0.931 0.789 0.889 0.88 0.002 0.003 0.002
cells/mL 4.40E+09 4.43E+09 4.66E+09 3.95E+09 4.45E+09 4.40E+09 1.00E+07 1.50E+07 1.00E+07
Ave 4.49E+09 4.26E+09 1.17E+07
Stdev 1.41E+08 2.77E+08 2.89E+06
36.5
OD600 0.932 0.938 0.988 0.845 0.976 0.929 0.003 0.002 0.001
cells/mL 4.66E+09 4.69E+09 4.94E+09 4.23E+09 4.88E+09 4.65E+09 1.50E+07 1.00E+07 5.00E+06
Ave 4.76E+09 4.58E+09 1.00E+07
Stdev 1.54E+08 3.32E+08 5.00E+06
44 OD600 0.974 0.956 1.008 0.896 0.992 0.976 0.004 0.006 0
115
cells/mL 4.87E+09 4.78E+09 5.04E+09 4.48E+09 4.96E+09 4.88E+09 2.00E+07 3.00E+07 0.00E+00
Ave 4.90E+09 4.77E+09 1.67E+07
Stdev 1.32E+08 2.57E+08 1.53E+07
Second set
Volumes (mL)
Exp # Condition
Net # of cells
added DFx added
DFx
concentration
Medium
(no Fe)
Ol added
(300-150um) Total
1
MR-1Δ +
DFx 1.31E+09
50 nmol (36uL of
1.4mM) 1uM 49.6 0.05 g 50
2
MR-1Δ +
DFx 1.31E+09
50 nmol (36uL of
1.4mM) 1uM 49.6 0.05 g 50
3
MR-1Δ +
DFx 1.31E+09
50 nmol (36uL of
1.4mM) 1uM 49.6 0.05 g 50
4
MR-1Δ +
DFx 1.31E+09
2.5 nmol (1.8uL of
1.4mM) 50nM 49.6 0.05 g 50
5
MR-1Δ +
DFx 1.31E+09
2.5 nmol (1.8uL of
1.4mM) 50nM 49.6 0.05 g 50
6
MR-1Δ +
DFx 1.31E+09
2.5 nmol (1.8uL of
1.4mM) 50nM 49.6
0.05 g
50
7 MR-1Δ 1.31E+09 0 0 49.6 0.05 g 50
8 MR-1Δ 1.31E+09 0 0 49.6 0.05 g 50
9 MR-1Δ 1.31E+09 0 0 49.6 0.05 g 50
Time (h) Exp # 1 2 3 4 5 6 7 8 9
1
OD600 0.002 0.002 0.001 0.001 0.001 0.002 0.002 0.001 0.001
cells/mL 1.00E+07 1.00E+07 5.00E+06 5.00E+06 5.00E+06 1.00E+07 1.00E+07 5.00E+06 5.00E+06
Ave 8.33E+06 6.67E+06 6.67E+06
Stdev 2.89E+06 2.89E+06 2.89E+06
15.5
OD600 0.116 0.173 0.185 0.041 0.050 0.050 -0.001 -0.001 0.004
cells/mL 5.80E+08 8.65E+08 9.25E+08 2.05E+08 2.50E+08 2.50E+08 -5.00E+06 -5.00E+06 2.00E+07
Ave 7.90E+08 2.35E+08 3.33E+06
Stdev 1.84E+08 2.60E+07 1.44E+07
18.5
OD600 0.213 0.216 0.216 0.048 0.069 0.056 0.002 0.003 0.004
cells/mL 1.07E+09 1.08E+09 1.08E+09 2.40E+08 3.45E+08 2.80E+08 1.00E+07 1.50E+07 2.00E+07
Ave 1.08E+09 2.88E+08 1.50E+07
Stdev 8.66E+06 5.30E+07 5.00E+06
22.5
OD600 0.256 0.263 0.263 0.048 0.075 0.058 0.003 0.004 0.005
cells/mL 1.28E+09 1.32E+09 1.32E+09 2.40E+08 3.75E+08 2.90E+08 1.50E+07 2.00E+07 2.50E+07
Ave 1.30E+09 3.02E+08 2.00E+07
Stdev 2.02E+07 6.83E+07 5.00E+06
28.5
OD600 0.305 0.335 0.326 0.055 0.07 0.067 0.007 0.004 0.004
cells/mL 1.53E+09 1.68E+09 1.63E+09 2.75E+08 3.50E+08 3.35E+08 3.50E+07 2.00E+07 2.00E+07
Ave 1.61E+09 3.20E+08 2.50E+07
Stdev 7.70E+07 3.97E+07 8.66E+06
39.5
OD600 0.364 0.384 0.372 0.059 0.085 0.092 0.004 0.001 0.003
cells/mL 1.82E+09 1.92E+09 1.86E+09 2.95E+08 4.25E+08 4.60E+08 2.00E+07 5.00E+06 1.50E+07
Ave 1.87E+09 3.93E+08 1.33E+07
Stdev 5.03E+07 8.69E+07 7.64E+06
44
OD600 0.377 0.392 0.388 0.059 0.085 0.077 0.003 0 0.002
cells/mL 1.89E+09 1.96E+09 1.94E+09 2.95E+08 4.25E+08 3.85E+08 1.50E+07 0.00E+00 1.00E+07
Ave 1.93E+09 3.68E+08 8.33E+06
Stdev 3.88E+07 6.66E+07 7.64E+06
116
Third set
Volumes (mL)
Exp # Condition
Net # of
cells added DFx added
DFx
concentration
Medium
(no Fe)
Ol added (300-
150um) Total
1
MR-1Δ +
DFx 1.31E+09
10nmol (7.2uL of
1.4mM DFx) 200nM 49.9
0.05 g
50
2
MR-1Δ +
DFx 1.31E+09
10nmol (7.2uL of
1.4mM DFx) 200nM 49.9
0.05 g
50
3
MR-1Δ +
DFx 1.31E+09
250nmol (18uL
of 14mM DFx) 5uM 49.9
0.05 g
50
4
MR-1Δ +
DFx 1.31E+09
250nmol (18uL
of 14mM DFx) 5uM 49.9
0.05 g
50
5
MR-1Δ +
DFx 1.31E+09
500nmol (17uL
of 29.08mM
DFx) 10uM 49.9
0.05 g
50
6
MR-1Δ +
DFx 1.31E+09
500nmol (17uL
of 29.08mM
DFx) 10uM 49.9
0.05 g
50
7
MR-1Δ +
DFx 1.31E+09
1.25umol (43uL
of 29.08mM) 25uM 49.9
0.05 g
50
8
MR-1Δ +
DFx 1.31E+09
1.25umol (43uL
of 29.08mM) 25uM 49.9
0.05 g
50
Exp # 1 2 3 4 5 6 7 8
Time (h)
OD600 0.202 0.193 0.713 0.731 0.730 0.836 0.895 0.861
43
cells/mL 1.01E+09 9.65E+08 3.57E+09 3.66E+09 3.65E+09 4.18E+09 4.48E+09 4.31E+09
Ave
9.88E+08 3.61E+09 3.92E+09 4.39E+09
Stdev
3.18E+07 6.36E+07 3.75E+08 1.20E+08
P. aeruginosa experiment
Cell growth from olivine experiments (Fig. 3).
Volumes (mL)
Exp # Condition cells added
Medium
(no Fe)
Ol added (300-
150um) Total
1 PA-14Δ 1 colony 49.9
0.05 g
50
2 PA-14Δ + Ol
1 colony
49.9
0.05 g
50
3 PA-14 1 colony 49.9 0.05 g 50
4 PA-14 + Ol
1 colony
49.9
0.05 g
50
117
5
Abiotic
control
0
49.9
0.05 g
50
Exp # time (h) 7 23 31 47 55 71 79
1 OD -0.001 0.24 0.383 0.634 0.7 0.803 0.843
2 OD 0.005 0.228 0.452 1.033 1.512 1.212 1.2
3 OD -0.011 0.076 0.278 0.57 0.701 0.885 0.905
4 OD -0.011 0.064 0.35 1.039 1.484 1.408 1.146
5 OD 0.003 0.002 0.003 0.005 0.001 -0.003 0.001
118
Appendix B
Olivine surface area calculations
The surface area of olivine minerals was calculated, assuming a perfect spherical shape of
grains and by breaking down a 6-component normal distribution of median grain sizes: 162.5 μm,
187.5 μm, 212.5 μm, 237.5 μm, 262.5 and 287.5 μm, following a normal distribution (5, 15 and
30% grain fractions). Mineral density was interpolated from measured molar fraction (88%
forsterite, section 2.2.1), using known densities of fayalite and forsterite end members. Values for
calculation details are laid out below.
Mineral grain diameter (um) 150 - 300 um
Total mass (g) 0.1
Total experimental volume (L) 0.1
Olivine composition
Mg 1.76Fe 0.24SiO
4
Mg Fe Si O
Total
Stoichiometry 1.76 0.24 1 4
7.00
Mass (g/mol) 24.305 55.845 28.085 15.999
Mineral molar mass (g/mol) 42.7768 13.4028 28.085 63.996
148.26
Forsterite Fayalite
Experimental minerals
Mineral density (g/mm3) 0.00327 0.00439
0.0034044
Normal distribution 150 - 175 175 - 200 200 - 225 225 - 250 250 - 275 275 - 300
Sum
% 5% 15% 30% 30% 15% 5%
Median diameter of grain
(mm) 0.1625 0.1875 0.2125 0.2375 0.2625 0.2875
Median Radius of grain (mm) 0.08125 0.09375 0.10625 0.11875 0.13125 0.14375
Surface area/grain (mm2) 0.082957875
0.11044687
5
0.14186287
5
0.17720587
5
0.2164758
8
0.2596728
8
Vol/grain (mm3) 0.001685082
0.00258859
9
0.00376823
3
0.00526079
9
0.0071031
1
0.0093319
9
Mass/grain (g) 5.73669E-06 8.81263E-06 1.28286E-05 1.79099E-05 2.4182E-05 3.177E-05
Mass of fraction (g) 0.005 0.015 0.03 0.03 0.015 0.005
0.1
# of grains in fraction 872 1702 2339 1675 620 157
Total surface area of grains
(mm2) 72.30461936
187.992010
3
331.750606
5 296.82949
134.28000
7
40.867828
3
1064.0
2
119
Mineral dissolution data
First set
Exp
# Condition
Net # of cells
added DFx added
DFx
concentration
Ol added
(300-
150um)
Total
starting
volume (mL)
1 MR-1Δ 5.0E+10 0 0uM 0.1 g
100
2 MR-1Δ 5.0E+10 0 0uM 0.1 g 100
3 MR-1Δ 5.0E+10 0 0uM 0.1 g 100
4
MR-1Δ +
Deferoxamine 5.0E+10
0.005umol (33uL of
0.15mM sid solution) 0.05uM 0.1 g
100
5
MR-1Δ +
Deferoxamine 5.0E+10
0.005umol (33uL of
0.15mM sid solution) 0.05uM 0.1 g
100
6
MR-1Δ +
Deferoxamine 5.0E+10
0.005umol (33uL of
0.15mM sid solution) 0.05uM 0.1 g
100
7
MR-1Δ +
Deferoxamine 5.0E+10
0.1umol (67uL of 1.5mM
sid solution) 1uM 0.1 g
100
8
MR-1Δ +
Deferoxamine 5.0E+10
0.1umol (67uL of 1.5mM
sid solution) 1uM 0.1 g
100
9
MR-1Δ +
Deferoxamine 5.0E+10
0.1umol (67uL of 1.5mM
sid solution) 1uM 0.1 g
100
Time
(h)
Sample
# 1 2 3 4 5 6 7 8 9
1
OD600 0.088 0.091 0.088 0.086 0.080 0.077 0.096 0.062 0.076
cells/mL 4.40E+08 4.55E+08 3.00E+00 4.30E+08 4.00E+08 3.85E+08 4.80E+08 3.10E+08 3.80E+08
Ave 2.98E+08 4.05E+08 3.90E+08
Stdev 2.58E+08 2.29E+07 8.54E+07
20
OD600 0.062 0.063 0.083 0.066 0.065 0.073 0.180 0.177 0.170
cells/mL 3.10E+08 3.15E+08 4.15E+08 3.30E+08 3.25E+08 3.65E+08 9.00E+08 8.85E+08 8.50E+08
Ave 3.47E+08 3.40E+08 8.78E+08
Stdev 5.92E+07 2.18E+07 2.57E+07
22.5
OD600 0.073 0.075 0.073 0.084 0.070 0.090 0.168 0.202 0.192
cells/mL 3.65E+08 3.75E+08 3.65E+08 4.20E+08 3.50E+08 4.50E+08 8.40E+08 1.01E+09 9.60E+08
Ave 3.68E+08 4.07E+08 9.37E+08
Stdev 5.77E+06 5.13E+07 8.74E+07
25.5
OD600 0.069 0.075 0.077 0.070 0.085 0.080 0.184 0.196 0.197
cells/mL 3.45E+08 3.75E+08 3.85E+08 3.50E+08 4.25E+08 4.00E+08 9.20E+08 9.80E+08 9.85E+08
Ave 3.68E+08 3.92E+08 9.62E+08
Stdev 2.08E+07 3.82E+07 3.62E+07
28
OD600 0.067 0.079 0.089 0.071 0.070 0.080 0.197 0.229 0.200
cells/mL 3.35E+08 3.95E+08 4.45E+08 3.55E+08 3.50E+08 4.00E+08 9.85E+08 1.15E+09 1.00E+09
Ave 3.92E+08 3.68E+08 1.04E+09
Stdev 5.51E+07 2.75E+07 8.84E+07
53
OD600 0.055 0.086 0.061 0.057 0.061 0.065 0.195 0.205 0.221
cells/mL 2.75E+08 4.30E+08 3.05E+08 2.85E+08 3.05E+08 3.25E+08 9.75E+08 1.03E+09 1.11E+09
Ave 3.37E+08 3.05E+08 1.04E+09
Stdev 8.22E+07 2.00E+07 6.56E+07
Time (h) 1
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
120
1 2.99 123.02
124.53 2.27
0.11 3.92
3.68 0.21
2 3.00 123.43 0.10 3.56
3 3.09 127.13 0.10 3.56
4 3.00 123.43
125.90 2.14
0.09 3.20
3.68 0.41
5 3.09 127.13 0.11 3.92
6 3.09 127.13 0.11 3.92
7 3.13 128.78
126.31 2.29
0.04 1.42
1.54 0.21
8 3.06 125.90 0.04 1.42
9 3.02 124.25 0.05 1.78
Time (h) 20
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.09 127.13
128.09 2.03
0.13 4.63
4.63 0.36
2 3.08 126.72 0.12 4.27
3 3.17 130.43 0.14 4.98
4 3.07 126.31
127.82 1.44
0.13 4.63
4.51 0.54
5 3.11 127.96 0.11 3.92
6 3.14 129.19 0.14 4.98
7 2.98 122.61
123.71 1.04
0.13 4.63
5.10 0.54
8 3.01 123.84 0.14 4.98
9 3.03 124.67 0.16 5.70
Time (h) 22.5
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.13 128.78
129.05 0.86
0.07 2.49
2.61 0.21
2 3.12 128.37 0.07 2.49
3 3.16 130.01 0.08 2.85
4 3.09 127.13
126.72 1.09
0.07 2.49
3.09 1.03
5 3.05 125.49 0.07 2.49
6 3.10 127.55 0.12 4.27
7 2.98 122.61
122.75 0.24
0.14 4.98
4.75 1.44
8 2.98 122.61 0.17 6.05
9 2.99 123.02 0.09 3.20
Time (h) 25.5
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.08 126.72
127.96 1.48
0.13 4.63
4.75 0.21
2 3.10 127.55 0.14 4.98
121
3 3.15 129.60 0.13 4.63
4 3.11 127.96
127.82 0.24
0.12 4.27
4.87 0.54
5 3.10 127.55 0.14 4.98
6 3.11 127.96 0.15 5.34
7 2.97 122.20
122.33 0.63
0.14 4.98
5.58 0.54
8 2.99 123.02 0.16 5.70
9 2.96 121.79 0.17 6.05
Time (h) 28
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.11 127.96
128.23 1.26
0.12 4.27
4.51 0.21
2 3.09 127.13 0.13 4.63
3 3.15 129.60 0.13 4.63
4 3.10 127.55
127.55 0.00
0.15 5.34
5.22 0.21
5 3.10 127.55 0.14 4.98
6 3.10 127.55 0.15 5.34
7 2.98 122.61
122.75 0.24
0.16 5.70
5.93 0.41
8 2.98 122.61 0.16 5.70
9 2.99 123.02 0.18 6.41
Time (h) 53
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.14 129.19
129.19 0.41
0.15 5.34
5.34 0.36
2 3.13 128.78 0.16 5.70
3 3.15 129.60 0.14 4.98
4 3.08 126.72
128.23 1.32
0.15 5.34
5.34 0.00
5 3.14 129.19 0.15 5.34
6 3.13 128.78 0.15 5.34
7 3.12 128.37
127.82 0.95
0.21 7.48
7.48 0.36
8 3.08 126.72 0.20 7.12
9 3.12 128.37 0.22 7.83
Second set
Exp # Condition
Net # of cells
added DFx added
DFx
concentration
Ol added (300-
150um)
Total starting
volume (mL)
1
MR-1Δ +
Deferoxamine 5.00E+10
5umol (167uL of
30mM stock) 50uM 0.1g
100
2
MR-1Δ +
Deferoxamine 5.00E+10
5umol (167uL of
30mM stock) 50uM 0.1g
100
122
3
MR-1Δ +
Deferoxamine 5.00E+10
5umol (167uL of
30mM stock) 50uM 0.1g
100
4 MR-1 5.00E+10 0 0 0.1g 100
5 MR-1 5.00E+10 0 0 0.1g 100
6 MR-1 5.00E+10 0 0 0.1g 100
7 MR-1 killed 5.00E+10 0 0 0.1g 100
8 MR-1 killed 5.00E+10 0 0 0.1g 100
9 MR-1 killed 5.00E+10 0 0 0.1g 100
Time
(h) Exp # 1 2 3 4 5 6 7 8 9
1
OD600 0.069 0.088 0.069 0.054 0.042 0.056 0.059 0.054 0.061
cells/mL 3.45E+08 4.40E+08 3.45E+08 2.70E+08 2.10E+08 2.80E+08 2.95E+08 2.70E+08 3.05E+08
Ave
3.77E+08 2.53E+08 2.90E+08
Stdev
5.48E+07 3.79E+07 1.80E+07
24
OD600 0.552 0.525 0.554 1.278 1.461 1.402 0.060 0.038 0.046
cells/mL 2.76E+09 2.63E+09 2.77E+09 6.39E+09 7.31E+09 7.01E+09 3.00E+08 1.90E+08 2.30E+08
Ave
2.72E+09 6.90E+09 2.40E+08
Stdev
8.10E+07 4.67E+08 5.57E+07
28
OD600 0.624 0.628 0.656 1.458 1.766 1.644 0.052 0.05 0.058
cells/mL 3.12E+09 3.14E+09 3.28E+09 7.29E+09 8.83E+09 8.22E+09 2.60E+08 2.50E+08 2.90E+08
Ave
3.18E+09 8.11E+09 2.67E+08
Stdev
8.72E+07 7.76E+08 2.08E+07
30.5
OD600 0.679 0.672 0.714 1.436 1.584 1.572 0.045 0.058 0.05
cells/mL 3.40E+09 3.36E+09 3.57E+09 7.18E+09 7.92E+09 7.86E+09 2.25E+08 2.90E+08 2.50E+08
Ave
3.44E+09 7.65E+09 2.55E+08
Stdev
1.13E+08 4.11E+08 3.28E+07
33
OD600 0.722 0.699 0.748 1.484 1.652 1.608 0.050 0.048 0.061
cells/mL 3.61E+09 3.50E+09 3.74E+09 7.42E+09 8.26E+09 8.04E+09 2.50E+08 2.40E+08 3.05E+08
Ave
3.62E+09 7.91E+09 2.65E+08
Stdev
1.23E+08 4.36E+08 3.50E+07
46
OD600 0.838 0.796 0.821 1.386 1.678 1.680 0.041 0.027 0.055
cells/mL 4.19E+09 3.98E+09 4.11E+09 6.93E+09 8.39E+09 8.40E+09 2.05E+08 1.35E+08 2.75E+08
Ave
4.09E+09 7.91E+09 2.05E+08
Stdev
1.06E+08 8.46E+08 7.00E+07
Time (h) 1
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.05 125.49
125.08 0.71
0.11 3.92
3.92 0.36
2
3.05 125.49 0.10 3.56
3 3.02 124.25 0.12 4.27
4 3.01 123.84 124.67 0.71 0.11 3.92 3.68 0.21
123
5 3.04 125.08 0.10 3.56
6 3.04 125.08 0.10 3.56
7 3.31 136.19
135.64 0.48
0.12 4.27
4.39 0.54
8 3.29 135.36 0.14 4.98
9 3.29 135.36 0.11 3.92
Time (h) 24
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 2.67 109.85
110.40 0.63
0.22 7.83
8.19 0.94
2
2.7 111.09 0.21 7.48
3 2.68 110.27 0.26 9.26
4 1.81 74.47
68.98 4.91
0.18 6.41
7.48 0.94
5 1.58 65.01 0.23 8.19
6 1.64 67.48 0.22 7.83
7 3.5 144.00
138.52 4.75
0.13 4.63
4.27 0.36
8 3.3 135.77 0.11 3.92
9 3.3 135.77 0.12 4.27
Time (h) 28
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 2.71 111.50
110.95 0.63
0.27 9.61
9.02 0.74
2 2.68 110.27 0.23 8.19
3 2.7 111.09 0.26 9.26
4 1.66 68.30
64.32 3.50
0.20 7.12
8.07 0.90
5 1.5 61.72 0.25 8.90
6 1.53 62.95 0.23 8.19
7 3.23 132.89
134.13 1.23
0.12 4.27
4.51 0.41
8 3.29 135.36 0.12 4.27
9 3.26 134.13 0.14 4.98
Time (h) 30.5
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 2.73 112.32
111.36 1.04
0.28 9.97
9.49 0.82
2 2.68 110.27 0.24 8.55
3 2.71 111.50 0.28 9.97
124
4 1.58 65.01
63.09 1.71
0.23 8.19
9.02 0.74
5 1.5 61.72 0.27 9.61
6 1.52 62.54 0.26 9.26
7 3.28 134.95
135.50 0.95
0.14 4.98
5.10 0.21
8 3.28 134.95 0.14 4.98
9 3.32 136.60 0.15 5.34
Time (h) 33
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1
2.75 113.15
111.64 1.44
0.31 11.04
10.80 0.74
2 2.71 111.50 0.28 9.97
3 2.68 110.27 0.32 11.39
4 1.66 68.30
64.05 3.69
0.23 8.19
10.80 2.29
5 1.5 61.72 0.35 12.46
6 1.51 62.13 0.33 11.75
7 3.3 135.77
135.36 1.09
0.14 4.98
4.75 0.74
8 3.26 134.13 0.11 3.92
9 3.31 136.19 0.15 5.34
Time (h) 46
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 2.9 119.32
119.18 0.63
0.43 15.31
15.43 0.21
2 2.88 118.49 0.43 15.31
3 2.91 119.73 0.44 15.67
4 2.04 83.93
81.46 2.70
0.38 13.53
14.12 0.74
5 1.99 81.88 0.42 14.95
6 1.91 78.58 0.39 13.89
7 3.29 135.36
135.36 1.65
0.12 4.27
4.51 0.21
8 3.25 133.72 0.13 4.63
9 3.33 137.01 0.13 4.63
Abiotic set
Exp #
Net # of cells
added DFx added
DFx
concentration
Ol added (300-
150um)
Total starting
volume (mL)
1 0 0 0 0.1g
100
2 0 0 0 0.1g
100
125
3 0 0 0 0.1g
100
4 0
0.005umol (33uL of
0.15mM sid solution) 50nM 0.1g
100
5 0
0.005umol (33uL of
0.15mM sid solution) 50nM 0.1g
100
6 0
0.005umol (33uL of
0.15mM sid solution) 50nM 0.1g
100
7 0
0.1umol (67uL of
1.5mM sid solution) 1uM 0.1g
100
8 0
0.1umol (67uL of
1.5mM sid solution) 1uM 0.1g
100
9 0
0.1umol (67uL of
1.5mM sid solution) 1uM 0.1g
100
Time (h) 1
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 2.95 121.37
121.51 0.24
0.06 2.14
2.02 0.21
2 2.95 121.37 0.05 1.78
3 2.96 121.79 0.06 2.14
4 2.95 121.37
121.79 0.71
0.06 2.14
1.90 0.21
5 2.95 121.37 0.05 1.78
6 2.98 122.61 0.05 1.78
7 2.98 122.61
122.47 0.24
0.05 1.78
1.78 0.00
8 2.98 122.61 0.05 1.78
9 2.97 122.20 0.05 1.78
10 2.97 122.20
122.61 0.71
0.05 1.78
1.66 0.21
11 3.00 123.43 0.04 1.42
12 2.97 122.20 0.05 1.78
Time (h) 22
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 2.99 123.02
123.57 0.95
0.07 2.49
2.49 0.00
2 3.03 124.67 0.07 2.49
3 2.99 123.02 0.07 2.49
4 2.98 122.61
123.57 1.32
0.08 2.85
2.73 0.21
5 2.99 123.02 0.07 2.49
6 3.04 125.08 0.08 2.85
7 3.05 125.49
125.35 0.24
0.08 2.85
2.61 0.21
8 3.04 125.08 0.07 2.49
9 3.05 125.49 0.07 2.49
10 3.05 125.49
126.45 1.04
0.09 3.20
3.20 0.00
11 3.07 126.31 0.09 3.20
126
12 3.10 127.55 0.09 3.20
Time (h) 25
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 2.99 123.02
123.29 0.24
0.07 2.49
2.49 0.00
2 3.00 123.43 0.07 2.49
3 3.00 123.43 0.07 2.49
4 3.01 123.84
123.84 0.00
0.08 2.85
2.73 0.21
5 3.01 123.84 0.07 2.49
6 3.01 123.84 0.08 2.85
7 3.07 126.31
125.21 1.04
0.08 2.85
2.85 0.00
8 3.02 124.25 0.08 2.85
9 3.04 125.08 0.08 2.85
10
3.10 127.55
126.17 1.44
0.09 3.20
3.44 0.41
11 3.03 124.67 0.09 3.20
12 3.07 126.31 0.11 3.92
Time (h) 28
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.02 124.25
123.98 0.24
0.07 2.49
2.49 0.00
2 3.01 123.84 0.07 2.49
3 3.01 123.84 0.07 2.49
4 3.04 125.08
124.67 1.09
0.08 2.85
2.73 0.21
5 3.00 123.43 0.07 2.49
6 3.05 125.49 0.08 2.85
7 3.05 125.49
125.35 0.24
0.08 2.85
2.73 0.21
8 3.04 125.08 0.07 2.49
9 3.05 125.49 0.08 2.85
10 3.05 125.49
126.59 1.26
0.09 3.20
3.44 0.41
11 3.07 126.31 0.09 3.20
12 3.11 127.96 0.11 3.92
Time (h) 31.5
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.00 123.43 123.84 0.71 0.08 2.85 2.73 0.21
127
2 3.03 124.67 0.07 2.49
3 3.00 123.43 0.08 2.85
4 2.99 123.02
124.25 1.48
0.08 2.85
3.09 0.21
5 3.01 123.84 0.09 3.20
6 3.06 125.90 0.09 3.20
7 3.06 125.90
125.76 0.63
0.08 2.85
2.73 0.21
8 3.04 125.08 0.07 2.49
9 3.07 126.31 0.08 2.85
10 3.06 125.90
126.72 0.82
0.10 3.56
3.68 0.21
11 3.08 126.72 0.10 3.56
12 3.10 127.55 0.11 3.92
Time (h) 48
Mg-279 Si-288
Exp # ppm uM Ave (uM) SD ppm uM Ave (uM) SD
1 3.02 124.25
124.80 0.48
0.09 3.20
3.09 0.21
2 3.04 125.08 0.08 2.85
3
3.04 125.08 0.09 3.20
4 3.05 125.49
127.00 2.27
0.09 3.20
3.44 0.41
5 3.06 125.90 0.09 3.20
6 3.15 129.60 0.11 3.92
7 3.10 127.55
126.31 1.48
0.10 3.56
3.32 0.41
8 3.03 124.67 0.08 2.85
9 3.08 126.72 0.10 3.56
10
3.13 128.78
128.92 1.04
0.13 4.63
4.63 0.36
11 3.11 127.96 0.12 4.27
12 3.16 130.01 0.14 4.98
128
Appendix C
Single chamber experiments
In addition to the two-chamber bioelectric experiments reported in Chapter 6 of the main
text, a series of additional experiments were conducted in a single chamber with similar design,
except that the oxidative and reductive portions of the system were contained within a single
chamber, with no membrane separations (Fig. C1). In these experiments, ΔCyo-A cells were only
added to the chamber as a biofilm attached to the working electrode, with no cells injected to the
plankton. Also, AQDS was never added to the medium. In other respects, these experiments were
identical to the two-chamber versions.
Results from these single-chamber experiments are presented in Figs. S2 & S3, showing
current and DO production, as well as images of metal oxide precipitation at the counter electrodes.
In these experiments, planktonic cell counts increased from non-detectable at the onset of
experiments, to ~ 5x10
7
total cells/mL, and ~ 1x10
7
RSG-fluorescent cells/mL, suggesting that
some of the biofilm-cultured, ITO-attached cells had reverted to a planktonic state during these
experiments. Dissolved Fe(II) was never measured above the detection limits of the ferrozine assay
(~ 2 µM), likely due to a rapid cycling, i.e., microbial reductive dissolution followed quickly by
oxidation and precipitation at the counter electrode. The low current production (~ 100 nA, Fig.
S2) most likely reflects the lack of AQDS added to the experiments, decreasing the number of
electron transfer pathways (see Fig. 4). Overall, these single chamber experiments show that
despite low current production (1 – 2 orders of magnitude lower than in the 2-chamber
experiments), and no detectable DO signal, a strong oxidative pressure was still generated at the
oxidative counter electrode, effectively oxidizing and precipitating iron and manganese oxides on
this electrode (Fig. C3).
129
Mass and charge balance calculations in the two-chamber experiments
The measurement of electric current (and thus total electron transfer) along with Fe(II) and
DO concentrations throughout each experiment allow for mass and charge balance calculations
following the proposed equations 1 & 2 and in Fig. 9. These calculations are separated between
the pre-mixing and mixing phases of each experiment. In all experiments current production in the
first 19 – 20 hours (pre-mixing) was ~ 1 – 3 µA. This amount of current would translate to DO
increases of 0.8 – 2.4 µM in the oxidizing chamber (Fig. 10). Such predictions are consistent with
DO readings during that time period in the biotic experiments (both 10
7
and 10
8
cells/mL), but not
in the abiotic controls which saw no measurable DO above background. During the 6 – 10 hours
of mixing, electric current increased by up to an order of magnitude (5 – 40 µA). Measured DO
concentrations increased to 5 – 28 µM in the biotic experiments but did not increase in the abiotic
controls. In the experiments with 10
7
cells/mL (Fig. 10A), 10 hours of current at 5 – 20 µA would
translate to DO increases of 2 – 8 µM, predictions consistent with measurements of 5 µM (Fig.
10A). In the experiments with 10
8
cells/mL, 6 hours of current at 20 – 40 µA would translate to
DO increases of 5 – 11 µM, lower than the measured 20 – 28 µM (Fig. 10B). In the abiotic controls,
6 hours of current 10 – 15 µA would theoretically translate to DO increases of 2.5 – 3.7 µM, greater
than the DO measurements which showed no increase at all (Fig. 10C). Thus, with a few
exceptions, most of the experiments yielded a balance between current and DO production, at least
within the variability in the signal of both. The exceptions are worth further consideration.
The most notable of these is that, in the abiotic controls, no DO was measured despite
current observed in both the pre-mixing and mixing phases. The current likely reflects abiotic
AQDS electrochemical reduction coupled with formaldehyde oxidation, as discussed in the main
130
text. This reaction however would not have strongly promoted hematite reduction, as evidenced in
the lack of reduced iron readings from the ferrozine assays. At 100 µM, AQDS could have
absorbed electrons from a current of 3 µA for up to 400 hours, or from 40 µA for up to 30 hours,
until all AQDS would have been reduced to AH 2DS. Thus, AQDS could have acted as a significant
pool of electrons, buffering electric and hematite reduction activity, and therefore alkalinity
gradients. Importantly, this AQDS effect cannot explain the DO production in the biotic
experiments since the abiotic experiments had AQDS but no DO was measured, emphasizing that
the electron shuttling by ΔCyo-A was critical to producing strong alkalinity gradients to promote
water splitting in the oxidative chamber. Lastly, DO measurements being greater than the
stoichiometric predictions in the mixed 10
8
cells/mL experiments might be the reflection of a
transient state caused by mixing. If AQDS acted as a sink of electrons during the unmixed portion
of the experiments due to stagnating medium waters, sudden mixing might cause the latent
accumulation of unreacted AH 2DS to drive iron reduction and oxygen-generating reactions in a
manner that might not be an immediate reflection of stoichiometric equilibrium.
Measurements in all of the two-chamber experiments could have been affected by
limitations in the Nafion PSM material, known to be susceptible to biofouling, cation exchange
reactions affecting proton transfer efficiency, and more importantly permeability to oxygen gas.
The latter effect has been well characterized, with measurable diffusion effects at the micromolar
range over periods of hours to days (Liu and Logan, 2004; Chae et al., 2008). These effects could
have further contributed to some of the discrepancies observed between charge balance
calculations and direct measurements.
Dissolved iron concentration measurements can also be compared to predictions based on
electrical current generation and the proposed reaction stoichiometry. In all experiments, current
131
production in the first 19 – 20 hours (before mixing started) was ~ 1 – 3 µA, an amount of current
which would translate to Fe(II) increases of 3 – 9 µM in the reducing, biotic chamber (Fig. 2).
During the 6 – 10 hours of mixing that followed, electric current increased by up to an order of
magnitude (5 – 40 µA). Predicted Fe(II) could then be expected to further increase by 5 – 60 µM
in the reducing chamber. These estimates are substantially higher than measurements done with
the ferrozine assay, which peak at ~ 5 µM by the end of the 10
8
cells/mL experiments. These
discrepancies between measurements and calculated predictions can likely be explained by active
processes of iron cycling. Ferrous iron is known to sorb effectively onto organic matter, including
cell membranes, as well as onto metal oxide particles (including hematite) (Jeon et al., 2001; Nano
and Strathmann, 2006; Tovar-Sanchez and Sañudo-Wilhelmy, 2011). Fe(II) has also been shown
to form secondary precipitates such as magnetite, a process identified with ΔCyo-A strains as a
byproduct of iron reduction, and which could be promoted by AQDS electrochemistry (Lovley et
al., 1987; Orsetti et al., 2013). Lastly, extra cellular electron transfer is a metabolism that requires
large amounts of iron, and significant amounts of reduced iron could be absorbed by cells. Such
processes would immobilize Fe(II) and sequester it from the aqueous phase, and impact
measurements by ferrozine assays.
132
Fig. C1. Design of the single-chamber bioelectric reactor, with all three electrodes contained in
the same experimental compartment, along with minimal growth medium, electrode-attached
ΔCyo-A cells, and hematite. No AQDS or planktonic cells were added to these single-chambered
experiments.
133
Fig. C2. DO and electric current results from single-chamber experiments. No AQDS or
planktonic cells were added to the experiments. Fe(II) concentrations were not detectable (< 2 µM)
throughout these experiments, likely due to rapid re-oxidation and precipitation on the cathode
(Fig. C3). Total cell densities in the planktonic phase were 3.5 – 7.5x10
7
cells/mL. Negative DO
values reflect the effects of electric potential on DO sensors, something difficult to include into
the calibration process.
134
Fig. C3. SEM image (A) and EDAX elemental maps (B-D) of the counter electrode in single-
chamber experiments, showing the ubiquitous precipitation of manganese and iron oxide mineral
precipitates covering the Pt counter electrode, demonstrating strong in-situ oxidation activity.
Manganese precipitates are believed to originate from the contents of the minimal growth medium
(29 µM MnSO 4). Iron precipitates are believed to originate from both the contents of the minimal
growth medium (3.6 µM FeSO 4), and microbial reduction of hematite, followed by its re-
oxidation.
135
Thoughts on energetics of the bioelectrochemical reactor
In this study, the proposed reactions can be described as electrochemical half reactions,
which standard potentials are known (Lide, 2006):
E
0
2H 2O = 4H
+
+ 4e + O 2 -1.23 V
4H
+
+ 4e + 2Fe 2O 3 + 2H 2O = 4Fe
2+
+ 8OH
-
0.77 V
For a net reaction:
2H 2O + 2Fe 2O 3 = 4Fe
2+
+ 8OH
-
+ O 2 -0.46 V
Gibbs free energy can be linked to cell potential via (Atkins and Jones, 2004):
∆𝐺 0
= −𝑛 ∗ 𝐹 ∗ 𝐸 0
Where E
0
is the standard potential of a reaction (V, or J/C), F is Faraday’s constant (96,485 C/mol),
and n is the stoichiometry of electron flow of the reaction (mol). Examining standard conditions,
our net reaction implies an endergonic reaction with Gibbs free energy of 177 kJ, or 89 kJ/mol of
water split in the abiotic chamber.
However, standard potential measurements imply pH conditions of 0, not representative of
experimental conditions. More realistic conditions can be calculated with experimental conditions,
following:
∆𝐺 𝑐𝑒𝑙𝑙 = −𝑛 ∗ 𝐹 ∗ 𝐸 𝑐𝑒𝑙𝑙
With experimental E set with the potentiostat at -0.351 V, the reaction would have a Gibbs free
energy of 135 kJ, or 68 kJ per mol of water split. This is certainly still an endergonic reaction, but
the ability to use ferric iron as an electron acceptor lowers the energetic requirements by a factor
136
of ~ 3.5 when compared to canonical hydrolysis, which otherwise uses protons as an electron
acceptor and requires 475 kJ, or 237 kJ per mol of water split, with H 2 and O 2 as by-products.
These numbers are consistent with calculations made from empirical data available from
this study. For example, if a current of 30 μA were sustained for 8 hours (as representative of the
high cell density (10
8
cells/mL) experiment during mixing), an expected increase of DO of ~ 10
μM might be expected in the oxidative chamber from charge balance calculations (topic further
discussed in this appendix, above). This would imply a flow of 5.39x10
18
electrons, or a charge of
0.863 C. With this information, electrical energy requirements can be measured with:
𝐸𝑛𝑒𝑟𝑔𝑦 = 𝑉 ∗ 𝑄
Where E is expressed in joules, V is the potential of a system (in volts or J/C), and Q is the charge
transferred (C). With a potential of 0.35 V, the bioelectric reactor would require 0.303 J to power
such an increase in O 2 concentrations in our reactor, or 68 kJ/mol of water split. On the other hand,
standard hydrolysis, with a potential of 1.23 V, would require 1.07 J to power such an increase in
O 2 concentrations, or 237 kJ/mol of water split.
These calculations are straightforward and revealing, though they have some limits since
they use single datum points. Cyclic voltammetry (CV), on the other hand, is a powerful tool which
can further help characterize the system’s energetics. Deriving from the canonical Gibbs free
energy equation, the energetics of bioelectric reactors are commonly described and characterized
with continuous CV data, and using the Nernst equation (Elgrishi et al., 2018):
𝐸 = 𝐸 0
+
𝑅𝑇
n𝐹 𝑙𝑛𝑄
Where E is the potential in the reactor, E
0
is the standard potential of a species, F is Faraday’s
constant, n is the stoichiometry of electron flow of the reaction, R is the universal gas constant, T
137
is the temperature, and Q the reaction coefficient. This equation, with all the data CV experiments
can provide, could add a much more precise understanding of the energetics of the bioelectric
reactor.
In our experiments, and with the data currently available:
E = -0.351 V
E
0
= -0.459 V
R = 8.3145 J/molK
T = 298 K
n = 4
Q =
[𝐹𝑒
2+
]
4
∗[𝑂𝐻
−
]
8
∗[𝑂 2
]
[Fe2O3]∗[H2O]
When the reaction is at equilibrium (Q = 1), then E = E
0
. In our case there is clearly a disparity of
~ 100 mV between these two values, which most likely represents the differences in chemical
conditions between standard potentials and our experiments (i.e., pH, ionic strength of medium,
etc.) Therefore, in experimental settings, E
0
is replaced with the formal potential, E
0’
, which is
experimentally determined with cyclic voltammetry (Elgrishi et al., 2018).
𝐸 = 𝐸 0′
+
𝑅𝑇
n𝐹 𝑙𝑛𝑄
For Q, measurements of ferrous iron concentrations, DO and pH can help constrain the numerator
throughout CV experiments as a function of potential, and the denominator is 1, as defined by the
standard activities of water and solids. A complete CV curve could thus characterize current
production as a function of potential, helping to optimize the system for maximum effect. This
would be particularly revealing if the potential in the oxidative chamber were measured
continually, allowing the comparison of the reactions in their respective, separate chambers.
138
Conceptual design of a Mars-based bioelectrochemical ISRU system
The bioelectrochemical reactor used in experiments presented here were static, batch
reactors with no water or medium inflows or outflows, and containing finite amounts of
consumable ferric oxide minerals. Such a configuration is not efficient or sustainable for the larger-
scale implementation of this reactor in an ISRU context. Certainly, this reactor will need to be
further developed as a two-chambered, parallel flow-through system, in which fresh microbial
media and ferric oxides can be slowly brought into the reductive chamber, and the aqueous ferrous
iron waste can be disposed of. Similarly for the abiotic chamber, pure and anaerobic water will
need to slowly enter the chamber, allowing the generated oxygen to be transported to an extraction
system before the water can be recycled. A schematic of this comprehensive system is outlined in
Fig. C4, and discussed below.
Fig. C4. Schematic of ISRU system application. Core bioelectrochemical reactor is highlighted
in the center, with supporting systems such as water, nitrogen, electrical, as well as oxygen and
waste products
139
Water system
While the bioelectrochemical reactor presented here did not create any net water mass
losses, it did lead to a transfer of water mass from the oxidative to the reductive chamber.
Therefore, a water management system will be needed to provide very clean and anaerobic water
(i.e. nitrogen-purged 18.2 MΩ.cm water) to the reactor, as well as regulate the balance of water
between both sides of the reactor through a buffer tank. Similarly, a separate tank holding
microbial medium will be necessary to ensure the appropriate conditions (such as pH, nutrient
concentrations, etc.) are maintained to sustain optimal microbial activity in the biotic chamber.
Nitrogen system
To ensure waters flowing into the reactor are perfectly anaerobic, a source of high purity
gas (typically nitrogen, though other options such as argon could work just as well) will be needed
to purge waters of any potential oxygen they might carryover from their sources.
Electrical system
The bioelectrochemical reactor depended on an external energy source to produce the
electric potential and sustain electrical current. Any standard 110 – 220 V outlet will suffice, such
that the power source could depend on any power system established on a field location. Solar
power is likely to prove valuable in that respect.
Ferric oxide minerals
The consumable material which did get lost through the reaction of this system was ferric
oxide minerals. Accordingly, these minerals will need to be extracted from Martian regolith and
prepared as necessary (i.e., separated from other mineral phases, salts removed, etc.) before being
injected into the system.
140
Outputs
The bioelectrochemical reactor presented here used electricity and ferric oxides as input,
and produced ferrous iron and oxygen as outputs. It will be important to continually remove ferrous
iron from the biotic chamber as it can become toxic to bacteria at high concentrations, as well as
inhibit the core chemical reaction. This will be done as part of the flow-through design, as water
slowly flowing out of the biotic chamber will be processed to remove the ferrous iron. While the
water will be recycled back as microbial growth medium, the iron will be disposed of or recycled
as needed. On the other side of the equation, the flow-through design will cycle oxygen-generating
water to an extraction process, separating out oxygen gas from water, after which the water will
be reused as part of the water system.
Abstract (if available)
Abstract
The Earth has a unique and distinct history, as it is the only planet we know that harbours life. Its evolution has been defined not only by cosmic and geologic processes, but also by biological activity. Life most likely emerged from, and learned to interact intimately with its mineral environment, not just to survive, but also to thrive, firmly altering Earth’s crust, oceans and atmosphere in the process. This thesis explores the relationship between biology and geology, and some of the effects interactions between bacteria and minerals can impart on planetary systems. Specifically, the focus is on the mineralogy, speciation and bioavailability of iron as a function of bacterial activity, and the corresponding impact these can have on environmental geochemistry. A primary focus is the examination of iron limitation with respect to bacterial siderophore synthesis. Iron is critical to fundamental biological processes such as respiration and photosynthesis, and its bioavailability can impact primary and heterotrophic productivity. Yet in oxic environments at near-neutral pH, iron is highly insoluble, rendering it, in principle, unavailable as a nutrient for biological growth. We examined the impact of siderophores on the speciation, mobility and bioavailability of iron from rock-forming silicate minerals, detailing the mechanisms by which microbes directly use mineral substrates to support primary productivity, as well as the consequent effects on silicate dissolution rates. Growth experiments were performed with Shewanella oneidensis in an oxic, iron-depleted minimal medium, amended with olivine minerals as the sole source of iron. Experiments included the wild-type strain MR-1, and a siderophore synthesis gene deletion mutant strain (MR-1Δ). Relative to MR-1, MR-1Δ exhibited a very pronounced growth penalty and an extended lag phase. However, substantial growth of MR-1Δ, comparable to MR-1 growth, was observed when the mutant strain was provided with siderophores in the form of either filtrate from a well-grown MR-1 culture, or commercially available siderophores (deferoxamine). These observations suggest that under aerobic conditions, siderophores are critical for S. oneidensis to acquire iron from olivine. Growth-limiting concentrations of deferoxamine amendments were observed to be ≤ 5 − 10 μM. X-Ray photoelectron spectroscopy of the incubated olivine surfaces suggested that siderophores mobilize ferric iron from the mineral surface. Dissolved silicon measured by optical emission spectrometry revealed that while higher siderophore concentrations lead to higher olivine dissolution rates, effective bacterial reuse of siderophores is a more important factor, enhancing dissolution rates by an order of magnitude compared to similar abiotic experiments. Combined, these results demonstrated that low (nM to μM) concentrations of siderophores can effectively mobilize iron bound within silicate minerals, supporting very significant biological growth in limiting environments. The specific mechanism likely involves siderophores removing a protective layer of nano-meter thick iron oxides, enhancing both silicate dissolution rates and iron bioavailability. This thesis also examined the use of bacterial extra cellular transport in cathode-oxidizing, iron reducing, two-chambered bioelectric reactors. This reactor successfully produced molecular oxygen as a by-product of microbial iron oxide reductive dissolution from hypothetical Martian soils. These latter experiments served as a proof of concept for the development of potential in situ resource utilization technologies for use on Mars, supporting its human exploration. Overall, this work demonstrates how a detailed mechanistic understanding of bacteria-mineral interactions can help understand the drivers of the evolution and dynamics of planetary processesㅡand can also provide a means to develop effective tools in geoengineering applications.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Environmental controls on alkalinity generation from mineral dissolution: from the mineral surface to the global ocean
PDF
Calcite and aragonite dissolution in seawater: kinetics, mechanisms, and fluxes in the North Pacific
PDF
Germanium and silicon isotope geochemistry in terrestrial and marine low-temperature environments
PDF
Chemical weathering across spatial and temporal scales: from laboratory experiments to global models
PDF
Lithium isotopes as a carbon cycle proxy: examining the effects of high-temperature weathering, understanding brachiopod archives, and validating Cenozoic records
PDF
The marine, neutrophilic, and chemolithoautotrophic iron-oxidizing bacteria: insights into the physiology of Zetaproteobacteria and the discovery of novel iron-oxidizing Gammaproteobacteria
PDF
Quorum sensing interaction networks in bacterial communities
PDF
Electrochemical investigations and imaging tools for understanding extracellular electron transfer in phylogenetically diverse bacteria
PDF
An investigation into low current density and microbially mediated arsenic electrocoagulation kinetics
PDF
Genetic diversity and bacterial death in the context of adaptive evolution
PDF
Microbial interaction networks: from single cells to collective behavior
PDF
Anaerobic iron cycling in an oxygen deficient zone
PDF
Great Salt Lake ooids: insights into rate of formation, potential as paleoenvironmental archives, and biogenicity
PDF
Isotopic insights to human-impacted hydrologic systems: from urban watersheds to hydraulically fractured basins
PDF
Aerobic respiration rates of bacteria: the influence of nutrient-limitation, cell density, and nutrient pulsing
PDF
Stromatolites as biosignatures and paleoenvironmental records: experiments with modern mats and examples from the Eocene Green River Formation
PDF
Essay on monetary policy, macroprudential policy, and financial integration
PDF
Studies on the influence of bacteria and carbon source on the products of dissimilatory iron reduction
PDF
Exploring temporal changes in surface species on weathered feldspar mineral surfaces using solid-state NMR spectroscopy
PDF
Genomic and physiological characterization of Escherichia coli evolving in long-term batch culture
Asset Metadata
Creator
Van Den Berghe, Martin (author)
Core Title
Exploring bacteria-mineral interactions
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Geological Sciences
Publication Date
02/22/2021
Defense Date
12/18/2020
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
bacteria,interactions,Mineral,OAI-PMH Harvest
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
El-Naggar, Moh (
committee member
), Nealson, Kenneth (
committee member
), West, A. Joshua (
committee member
)
Creator Email
martin.vandenberghe@gmail.com,mdvanden@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-420649
Unique identifier
UC11668584
Identifier
etd-VanDenBerg-9278.pdf (filename),usctheses-c89-420649 (legacy record id)
Legacy Identifier
etd-VanDenBerg-9278.pdf
Dmrecord
420649
Document Type
Dissertation
Rights
Van Den Berghe, Martin
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
Tags
interactions