Low oxygen and chemical kinetic constraints on the geochemical
niche of neutrophilic iron(II) oxidizing microorganisms
Gregory K. Druschel1, David Emerson2*, R. Sutka2**, P. Suchecki3 and George W. Luther, III4
1 – University of Vermont, Department of Geology, 180 Colchester Ave. Burlington, VT 05405
USA. gdrusche@uvm.edu
2 – American Type Culture Collection
3 – Oakland, CA
4 – College of Marine and Earth Studies, University of Delaware, Lewes, DE 19958, USA
* Present address: Bigelow Laboratory for Ocean Sciences, West Boothbay Harbor, ME 04575,
USA
** Present address: GV Instruments, Wythenshawe, United Kingdom
2
ABSTRACT
Neutrophilic iron oxidizing bacteria (FeOB) must actively compete with rapid abiotic processes
governing Fe(II) oxidation and as a result have adapted to primarily inhabit low-O2 environments
where they can more effectively couple Fe(II) oxidation and O2-reduction. The distribution of
these microorganisms can be observed through the chemical gradients they affect, as measured
using in situ voltammetric analysis for Fe(II), Fe(III), O2, and FeS(aq). Field and laboratory
determination of the chemical environments inhabited by the FeOB were coupled with detailed
kinetic competition studies for abiotic and biotic processes using a pure culture of FeOB to
quantify the nature of the geochemical niche these organisms inhabit. Results show FeOB
inhabit geochemical niches where O2 concentrations are below approximately 50 M. This is
supported by a series of kinetic measurements made on Sideroxydans lithotrophicus (ES-1, a
novel FeOB) that compared biotic/abiotic(killed control) iron oxidation rates; these experiments
demonstrated that abiotic processes are favored above 50 µM O2. The microbial habitat is thus
largely controlled by the kinetics governing neutrophilic iron oxidation in microaerophilic
environments, which is dependent on Fe(II) concentration, PO2, temperature and pH in addition
to the surface area of iron oxyhydroxides and the cell density/activity of FeOB . Additional field
and lab culture observations suggest a potentially important role for the iron-sulfide aqueous
molecular cluster, FeS(aq), in the overall cycling of iron associated with the environments these
microorganisms inhabit.
3
1. INTRODUCTION
Central to addressing questions about the role microorganisms play in the cycling of
elements on any scale are the specific geochemical settings in which organisms actively
metabolize redox species. Iron cycling is a process of intense interest that has implications for
deciphering large changes in ocean and atmospheric chemistry through deep time, the
identification of microbial activity on other planets such as Mars, and the mobility of a host of
contaminants in modern earth settings (Straub et al., 2001; Benison and LaClair, 2003; Edwards
et al., 2004; Emerson and Weiss, 2004; Kappler and Newman, 2004; Roden et al., 2004; Ferris,
2005; Kump and Seyfried, 2005; Rouxel et al., 2005; Rouxel et al., 2006; Yamaguchi and
Ohmoto, 2006; Stucki et al., 2007; Neubauer et al.,in press). In any of these environments
geochemical control on microbial ecology can be manifested in diverse ways, but one key
element is to consider the rates at which organisms can utilize existing substrates including
electron donors and acceptors to garner energy for growth. Conversely, the ecology and
physiology of iron-utilizing microbes may significantly impact the geochemistry as
microorganisms themselves are responsible for controlling the rates of different processes, which
influence the gradients of elements coincident to their individual niche. Competition between
microorganisms and/or competition between microbial metabolic reactions and abiotic reactions
are thus an important part of deciphering iron cycling and can be investigated in terms of the
relative kinetics of individual processes.
In any natural system, the cycling of elements is controlled not only by the microorganisms
that can catalyze reactions, but also by the formation of specific aqueous complexes and minerals
that can affect what form these elements are present in. Iron and sulfur are quite commonly
associated with each other in different environments – and the settings neutrophlic iron oxidizers
4
are found in are no different. Reduced iron and sulfur strongly interact, and the solubility
product (log K) for the first Fe-S mineral to form in these environments (mackinawite) via:
Fe2+ + HS-  FeSmackinawite + H+
(1)
is reported from various experiments as 2.13±0.27 (at pH between 6.5 and 8; Wolthers et al.,
2005), 3.00±0.12 (Davison et al., 1999), and 3.88 – 3.98 (Benning et al., 2000, recomputed by
Rickard, 2006) while a recent report by Chen and Liu (2005) tabulate 46 field measurements
between 2.20 and 3.83. It is also well established that in a number of environments metastable
concentrations of polynuclear clusters can exist in solutions associated with mineral dissolution
and precipitation (Luther et al., 1999, 2002; Rozan et al., 2000; Furrer et al., 2004; Navrotsky,
2004). In the iron-sulfide system, iron sulfide clusters are often abbreviated FeS(aq) (after
Theberge and Luther, 1997), although it is important to realize that this notation likely represents
a continuum of polynuclear species of differing stoichiometry and charge (i.e. some combination
of different FexSy species). Iron sulfide clusters are known to exist in a number of marine and
freshwater environments (DeVitre et al., 1988; Theberge and Luther, 1997; Davison et al., 1999;
Luther et al., 2003; Druschel et al., 2004; Luther et al., 2005; Roesler et al., 2007), and are
thought to play a significant role in the precipitation of iron sulfide minerals (Rickard and
Luther, 1997; Butler et al., 2004; Rickard, 2006; Roesler et al., 2007). Rickard (2006) recently
showed that FeS(aq) establishes an equilibrium with the Fe-S mineral mackinawite which is the
first Fe-S mineral formed in low temperature reducing environments (Rickard, 2006).
Historically, microorganisms that utilize ferrous iron as a substrate are well known, in part as
a few species have distinctive morphologies,for example, stalk forming Gallionella spp.and
sheath forming Leptothrix spp. In freshwater, these organisms grow as dense communities in
mat-like structures where anoxic water bearing Fe(II) comes into contact with air, such as often
5
happens in wetlands or springs. The mat structure and density is dependent upon the flow
conditions, when flow is rapid (>0.5 m/s) the mats will be quite dense; however, under slow
flow, the mats may exist as loose aggregations of floculant iron oxyhydroxides (hydrous ferric
oxides, HFO). It is in this context that neutrophilic iron oxidizing bacteria (FeOB) provide an
interesting example of a group of microbes that are restricted in their ecological niche due to
kinetic constraints on their energy source (Kirby et al., 1999; Burke and Banwart, 2002;
Neubauer et al., 2002; Edwards et al., 2004; James and Ferris, 2004; Ferris, 2005;). These
organisms must outcompete abiotic reactions which consume Fe(II) in oxic and suboxic settings
at circumneutral pH conditions (Emerson and Revsbech, 1994; Edwards et al., 2004; James and
Ferris, 2004; Ferris, 2005). This competition is sensitive to pH as the kinetics of Fe(II) oxidation
with O2 are described by the rate law:
d [ Fe 2 ]
 k[ Fe 2 ][O2 ][OH  ]2
dt
(2)
where k= 8.0 x 1013 L2 mol-2 atm-1 at 25ºC (Singer and Stumm, 1970). It is these chemical
realities, which restrict neutrophilic FeOB to inhabit suboxic microhabitats, or niches, where low
[O2] allow biotic oxidation rates to further outpace the abiotic rates of Fe(II) oxidation (James
and Ferris, 2004; Roden et al., 2004; Ferris, 2005).
Several studies have investigated different aspects of the kinetic competition surrounding
neutrophilic iron oxidation both in the field and in the laboratory, and these have provided
valuable insight to these processes (Emerson and Revsbech, 1994; Neubauer et al., 2003; James
and Ferris, 2004; Roden et al., 2004; Rentz et al, 2007). However, previous studies have not
attempted to understand chemical dynamics in natural microbial mat communities of FeOB and
relate these back to kinetics of Fe-oxidation in pure cultures of a lithotrophic Fe-oxidizing
bacterium. To do this necessitates measuring both profiles of chemical species and their reaction
6
kinetics. This study used voltammetry to make detailed field and laboratory measurements in an
effort to decipher the specific environmental niche of neutrophilic FeOB, and the specific
chemical conditions that an isolate, Sideroxydans lithotrophicus, grows best in when cultured in
gradients that mimic natural conditions, and finally, to determine the kinetics of biotic vs. abiotic
rates in the context of field and culture results.
2. METHODS
2.1. Study Site - Contrary Creek Wetland
Contrary Creek is a small creek located within the Virginia Piedmont gold-pyrite belt
near the town of Mineral, Virginia. Areas around Contrary Creek were mined extensively until
about 80 years ago, which has left a legacy of low pH metal-contaminated water. Contrary
Creek is also fed by circumneutral seeps, and adjacent wetlands that have been the subject of
microbial studies on Fe-cycling (Anderson and Robbins, 1998; Emerson et al., 2004; Weiss et
al., 2005). The groundwater seep-fed wetland chosen for this study is located approximately 50
meters away from the main drainage of Contrary Creek. The study site was accessible by foot,
about 0.5 miles away from the nearest road (County Road 208), along an established footpath.
Studies in the wetland area describe this setting as one of typically low flow (<0.5 m/sec), pH
between 5.8 and 6.4, and Fe(II) between 30 and 300 M (Emerson and Weiss, 2004).
We visited this site on 2 separate occasions, November 2002 and August 2003, and
performed a battery of voltammetric profiles to describe the iron, manganese, sulfur, and oxygen
chemistry in space at this site.
2.2. Field Voltammetry Measurements
7
Voltammetric equipment was set up on a small wooden platform installed at the study
site over a selected portion of the wetland where lateral flow was minimal, and floculant iron
oxides, indicating significant microbial activity, were observed. Voltammetry allowed direct, in
situ, measurement of the chemical species present in profiles with minimal perturbation during
analysis. This system has proven to be very useful for analyzing a wide variety of redox species
in a number of environments, including O2, H2O2, Fe2+, Fe3+, FeS(aq), Mn2+, H2S, Sxn-, S8, S2O32-,
S4O62-, and HSO3- (Luther et al., 1991; Brendel and Luther, 1995; Xu et al., 1998; Dollhopf et
al., 1999; Luther et al., 1998; Luther et al., 1999; Taillefert et al., 2000; Luther et al., 2001;
Druschel et al., 2003; Druschel et al., 2004; Glazer et al., 2004; Glazer et al., 2006). For
analyses in the field, a DLK-100A Potentiostat (Analytical Instrument Systems, Flushing, NJ)
was employed with a computer controller and software. A standard three-electrode system was
employed for all experiments. The working electrode was 100 m gold amalgam (Au/Hg) made
in a 5 mm glass tube drawn out to a 0.2-0.3 mm tip. The electrode was constructed and prepared
after standard practices (Brendel and Luther, 1995). A Ag/AgCl reference electrode and a Pt
counter electrode were placed in the water near the measurement site. The working electrode
was mounted on a three-axis micromanipulator (CHPT manufacturing, Georgetown, DE) which
was operated by hand to descend in increments between 0.2 and 2 mm for each sampling point.
Electrochemical measurements began when the working electrode was carefully lowered to the
point where the water surface tension was broken and the tip was as close to the surface as
possible (defined as 0 depth). Cyclic voltammetry was performed in triplicate at each sampling
point in the profile at 1000 mV/second between -0.1 and -1.8 V (vs. Ag/AgCl) with an initial
potential of -0.1 V held for 2 seconds. In order to keep the working electrode surface clean, the
electrode was held at -0.9 V between sampling scans.
8
Calibration of the electrodes was accomplished by standard addition methods using
waters collected at the site and filtered with a 0.2 m nucleopore filter spiked with stock
solutions of FeCl2, MnCl2, and Na2S (Sigma reagents).
The water and stock solutions were
purged with ultra high-purity (UHP) argon before analysis. Sulfide standards were amended to
pH 10, and Fe(II) stock was prepared in 0.01 M HCl soln before addition to a purged water
containing excess hydroxylamine hydrochloride as a reductant.
Samples for total reduced iron were also collected at the site, immediately filtered using
0.2 m filters, and analyzed by the ferrozine method (Stookey, 1970). pH was measured in the
field using a standard combination electrode calibrated with pH 4.0 and 7.0 buffers.
Temperature was measured using a YSI thermistor.
2.3 Reagents and standards
Reagents for calibration of the electrodes were prepared from their respective salts in 18
MΩ water. Water was purged for at least 20 minutes with UHP N2 to eliminate O2 in the case of
preparing standards of reduced species. Fe(II) solutions were prepared from ferrous ammonium
sulfate in 0.01 M HCl, Mn(II) standards prepared from Mn(II)Cl2*4H2O and sulfide standards
prepared from Na2S*9H2O. Before weighing the Na2S*9H2O salts were rinsed first with purged
water and dried with kimwipes® Sulfide standards were periodically checked by iodometric
titration.
2.4 Field microbiology
Samples of the Fe mat material were collected in close vicinity to the locations where
voltametric profiles were obtained using a 10 ml pipet to obtain approximately 40 ml of sample
9
from within the top 3 cm of the mat material. These samples were transported on ice back to the
laboratory. A 2 ml subsample was fixed with 2% (w/v) glutaraldehyde and used for obtaining
direct counts of the total cell number (Emerson & Moyer, 1997) and evaluating the dominant
morphology of the iron oxides. From the live sample, serial 10-fold dilutions were prepared in
sterile MWMM medium and these were used to inoculate a 3 tube series of most probable
number (MPN) tubes, ranging from 10-2 to 10-7 using FeS gradient tubes described below The
MPN tubes were incubuated in the dark at room temperature and checked periodically for at least
3 weeks. Tubes that exhibited formation of a sharply defined horizontal band of Fe-oxidation in
proximity to the oxic-anoxic interface were considered positive for growth of FeOB. An aliquot
of the Fe-oxide containing band was removed and stained with Syto 13 (Molecular Probes) to
confirm that copious numbers of bacterial cells were associated with the oxides from
presumptive-positive tubes. Cell numbers were determined using a standard MPN table.
2.5 Laboratory gradient culture tube analyses
The isolate Sideroxydans lithotrophicus, strain ES-1, which was isolated from
groundwater and has been described previously, (Emerson and Moyer 1997; Emerson et al., in
press) was used for laboratory studies. S. lithotrophicus was grown in opposing gradients of
Fe(II) and O2 established in 60x15mm screw-cap glass tubes as previously described (Emerson
and Floyd, 2005). The Fe(II) gradient was established by including a 1% agarose-stabilized plug
of synthesized FeS, or mackinawite, at the bottom overlain by a 0.15% agarose-stabilized
mineral salts-bicarbonate buffered medium (using Wolfe’s mineral medium, MWMM). To
properly establish gradients, the tubes must be inoculated within 24h of preparation, this was
accomplished by removing the top and inserting a pipette tip with 10 l of inoculant into the gel
10
almost to the FeS plug and injecting the inoculant as the tip was withdrawn. As the bacteria grow
they form a well-defined band of rust-colored iron oxides at the oxide-anoxic interface. Several
sets of gradient culture tubes were prepared and inoculated at different times so that different age
samples could be measured conterminously with the electrodes. Duplicate control tubes which
were not inoculated were established at the same time. These gradient culture tubes, both live
and control sets, were examined after 2 and 5 days of growth, using the same voltammetric
electrodes, potentiostat, and micromanipulator setup described above. The counter and reference
electrodes were placed in the top of the gradient culture tube and the working electrode inserted
using a micromanipulator to quantify the iron, sulfur, and oxygen species with depth.
2.6. Kinetic experiments with S. lithotrophicus isolates
Kinetic experiments to define rate constants associated with microbial oxidation and
abiotic oxidation of Fe(II) by the isolate (S. lithotrophicus) were performed using a dropping
mercury electrode system (DME) controlled with an Analytical Instrument Systems DLK-100A.
The Princeton Applied Research model 303A DME has a 10ml glass electrochemical cell which
contains the end of the capillary used to distribute fresh mercury drops for each analysis, a Pt
counter electrode, and a saturated calomel (SCE) reference electrode connected to the cell
through a saturated KCl-filled glass tube terminated with a Vycor frit. The electrochemical cell
on the DME can be purged and continually flushed with gas, to establish and maintain specific
PO2 conditions throughout an experiment. The gas input to the DME cell was controlled with a
Matheson gas mixer which was used to control PO2 by adjusting the flow rates of N2 and air
(pressurized using a small air pump run through a 500 ml suction flask apparatus as a pressure
pulse dampener). For each experiment, the PO2 was adjusted and equilibrated prior to the
11
addition of microbial cells and Fe(II), with the corresponding O2 concentration directly measured
by voltammetry.
Kinetic experiments were carried out in the DME cell filled with 10 ml of 0.1 M KCl
purged and equilibrated with a set amount of O2. For these experiments, the ES-1 cells were
grown in liquid phase MWMM medium gradient culture plates using FeS as the Fe(II) source
(Emerson and Floyd, 2005). The cultures were harvested after 4 to 5 days, or the equivalent of
the late log phase of growth. The cells were concentrated by centrifugation and resuspended in
fresh MWMM medium. An aliquot of the cell suspension was removed and fixed with
glutaraldehyde for direct counts to determine the cell number. The living cell suspension was
placed in a sterile glass tube with a butyl rubber stopper and the headspace was degassed with
sterile N2 to reduce any possible effects of O2 toxicity on these microaerophilic bacteria. All
experiments were performed within 5 h of harvesting the initial culture. For the experimental
procedure, 0.5 ml of cell suspension was added to the DME cell followed by an addition of 100
M Fe(II) stock (pH was measured after addition to the buffered solution and changes were
found to be minimal) to begin the biotic experiment. A sequential killed control experiment was
performed by adding an aliquot of sodium azide from a 0.1 M stock solution to the DME cell
that contained the microbes from the biotic experiment. The final azide concentration was 1.0
mM. An aliquot of FeCl2 solution was added to the cell to bring the solution Fe(II) concentration
back to 100 M and to set time zero for the kill control experiment. Voltammetric analyses
measured Fe(II) and O2 in triplicate every 2 minutes in the reaction cells for up to 20 minutes
(depending on the experiment and observed reaction progress) for these biotic and killed control
(abiotic) experiments. The addition of microbial cells from the gradient plates included
significant amounts of HFO, which is also a catalyst for the oxidation of Fe(II) (Rentz et al,
12
2007). While the biotic oxidation of Fe(II) in the initial biotic experiments certainly forms
additional HFO, the amounts of HFO that would be formed in the course of the experiment
would be a vanishingly small percentage of the total amount of HFO in the system and we
assume would have a negligible effect.
2.7 Effects of azide on S. lithotrophicus
It has been shown previously that azide is an effective poison in inhibiting respiration in natural
samples of Fe-oxidizing bacteria without altering Fe chemistry (Rosson et al, 1984; Emerson and
Revsbech, 1994); however we wanted to confirm its efficacy against the pure culture at even
lower concentrations than have previously been used. To do this, a culture of S. lithotrophicus
was grown in gradient plates for 20h to early log phase (1.9 x 107 cells. ml-1). Subsamples of
these cultures were removed for direct counts and then sodium azide was added to the plates at
concentrations of 0.1, 0.5, and 1.0 mM. The plates were incubated for an additional 72h,
corresponding to the late log phase of growth, and then cell counts were done again.
Even 0.1 mM azide caused nearly complete cessation of growth with a growth yield of 2.4 x 107
cells.ml-1 after 72h, compared to 1.0 x 108 cells.ml-1 for the control, while 0.5 and 1.0 mM azide
resulted in complete killing of S. lithotrophicus.
2.8 Geochemical speciation and statistical calculations
Geochemical speciation calculations were made using the Geochemist’s Workbench
version 3 suite of programs. Additional thermodynamic data was used for iron and sulfur species
from Luther et al. (1996), Rickard (2006), and Rickard and Luther (2006). Statistical
calculations (best fit lines, standard error, p-values) were made using Minitab 14.
13
3. RESULTS
3.1 Field Results
The site was visited in November of 2002 and August of 2003, mean water temperature
in November was 10.1°C and in August was 20.5°C. Bulk Fe(II) concentrations in mat samples
collected within 0.5m of where profiles were done measured 171 µM and 235 µM, respectively,
using the ferrozine assay. These values are all consistent with long-term records (13 months) of
Fe(II) and T(°C) that have been recorded previously (Emerson & Weiss, 2004). During both
visits, flow rates were minimal; however in 2002 there was more water present in the wetland. In
both cases abundant precipitates of iron oxyhydroxides were present that formed a loosely
aggregated microbial iron mat between 5 and 10 cm thick. The mat was overlaid with several cm
of water and rested upon a loose gravel bottom. The iron mats were composed primarily of
Leptotrhix ochracea- like sheaths, and the stalks of Gallionella ferruginea-like bacteria were also
observed. Total cell counts from mat material collected in November, 2002, revealed 6.5 x 107
cells/cc mat, and MPNs for FeOB from this same material indicated there were approximately
4.3 x 106 cells/cc mat of putative Fe-oxidizers. From the August 2003 samples the numbers for
direct counts were 9.6 x 106 cells/cc mat and FeOB MPNs yielded 8.25 x 105 cells/cc mat. These
results suggest that FeOB could account for between 5 and 10% of the population.
Figure 1 represents 4 of the hundreds of voltammograms collected at the site in 2002.
Broad peaks for organically complexed Fe(III) (Taillefert et al., 2000) are present on the forward
scan at approximately -0.65V corresponding to the reduction of Fe3+(org) to Fe(II).
Fe(III)(org) + e-  Fe2+
(3)
14
The 32.0 mm scan in Figure 1 shows a small peak at -0.75 attributed to H2S, with a forward peak
(in the negative direction) due to the reaction:
HgS + 2 H+ + 2 e-  H2S + Hg0
(4)
as the HgS was deposited during the holding of the initial potential at -0.1 for 2 seconds and is
re-formed in the return scan by the reaction:
H2S + Hg0  HgS + 2H+ + 2 e-
(5)
The forward peak at -1.15 V is due to the reduction of FeS(aq):
FeS(aq) + 2 e- + 2 H+  Fe(Hg) + H2S
(6)
The return peak near -0.75 V is primarily due to the sulfide released in eq. 3 reforming HgS. We
note that in 2002 the predominant form of reduced iron at this location was FeS(aq), not
Fe2+(H2O)6 (Fe2+ hereafter) or another simple ion pair of Fe(II).
Figure 2 represents 4 of hundreds of voltammograms collected in late 2003 which were
compiled to construct the Profile in Figure 3. The surface voltammogram illustrates an oxic
environment where the only discernible redox species reacting correspond to a 2-step reduction
of O2 at the electrode surface:
O2 + 2e- + 2 H+  H2O2
(7)
H2O2 + 2 e- + 2 H+  2 H2O
(8)
Subsequent voltammograms represent reactions involving Fe(III) species and FeS(aq) species as
described above in addition to the presence of Fe(II) and Mn(II) species per the forward scan
reactions:
Fe2+ + 2 e- + Hg0  Fe(Hg)
(9)
Mn2+ + 2 e- + Hg0  Mn(Hg)
(10)
15
The scans in Figure 2 also show a peak that has been assigned to a FeS+Mn(aq) species (Druschel
et al., 2004). We note that in 2003 reduced iron existed in several forms, including Fe(II) and
different cluster forms.
From voltammetric scans taken at discreet depths at intervals through the water column
of the wetland study site, several vertical profiles of the water column were compiled. A
representative profile of the oxygen, iron, manganese, and sulfide concentrations with depth
(Figures 3A and 3B) indicate the system becomes suboxic/anoxic within a few millimeters of the
surface, and the formation of iron organic complexes (which are electroactive, indicated as Fe3+
on Figure 3A and 3B) occurs as a result of the microaerophilic oxidation of Fe(II). The lowest
measurable O2 concentration detected in the microaerophilic region, which sees a very sharp
concommittant increase in Fe(III), is 20 M. FeS(aq) occurs entirely within the water column,
and for the most part is the sole form of dissolved sulfide in the system. We also note that
several profiles were taken at different locations in the water column of the Contrary Creek
wetland study site, within centimeters of each other. In all of these, waters became
suboxic/anoxic within the first couple of millimeters and in all of these the oxidation of Fe(II) to
Fe(III) likely occurred, based on the shape of the profiles. However, there was heterogeneity in
the system with respect to sulfide. Profiles collected 5 cm away from the one represented in
Figure 3 had no dissolved sulfide present as free H2S or as metal sulfide clusters.
3.2 Laboratory culture experiments
Gradients of iron, oxygen, and sulfur were measured with voltammetric microelectrodes
at 0.5 to 1 mm vertical increments after 2 and 5 days of growth of S. lithotrophicus in laboratory
gradient tubes. The inoculated tubes with active bands of FeOB were compared to uninoculated
16
control tubes (Figures 4A-4D). Cultures of ES-1 established a significantly sharper contact
where Fe(II) and O2 interact than the control samples. Cultures established a zone of Fe(II)
oxidation at a lower O2 concentration than control samples, and regulated that zone for longer
times by balancing the rates of Fe(II) oxidation and O2 diffusion. Maximum oxygen levels
associated with Fe(II) oxidation in the upper parts of the growth band in these culture tubes were
between approximately 15 and 50 M O2, while O2 levels at the bottom of the growth band were
below detection (approximately 5 uM). A separate experiment showed that the bands of actively
growing ES-1 move downward toward the solid FeSmackinawite agarose plug at the bottom of the
tube, suggesting that growth is occurring in this very low O2 region of the gradient (see
supplemental movie file EA-1). In a control tube that contained Fe(II) but was not inoculated, a
band of iron oxidation formed, but the band remained in place, becoming denser, but not
migrating in response to either Fe(II) or O2.
Interestingly, FeS(aq) was also found to diffuse from the bottom FeSmackinawite agarose
plug, consistent with recent observations from Rickard (2006) that the anoxic (non-oxidative)
dissolution of nanocrystalline mackinawite comes to an equilibrium with FeS(aq) and Fe(II)
depending on pH and the following dissolution reactions:
FeSmackinawite = FeS(aq)
(11)
FeSmackinawite (or FeS(aq)) + 2 H+ = Fe2+ + H2S
(12)
3.3 Laboratory Kinetic Experiments
A suite of kinetic experiments was carried out to quantify the kinetic constraints
governing the competition between abiotic and biotic (ES-1) Fe(II) oxidation rates as a function
of O2 concentrations. The axenic ES-1 cultures, grown on gradient culture plates, were added to
17
the electrochemical cell at a final cell concentration of 2.1 x 107 cells/ml. The cultures were
active and contained HFO, abiotic control experiments were run on the same solution by adding
azide and Fe+ to the electrochemical cell and repeating the measurements. This normalized any
catalytic activity of the HFO between biotic and subsequent abiotic experimental sets. Raw data
associated with these experiments can be found in the supporting materials (Druschel et al., EA2). The kinetic experiments were treated as pseudo-first order reactions, and plotted as log[Fe]
vs time where the slope of the line is used to derive the rate constant (equal to k1=2.3*(slope);
Brezonik, 1994). Table 1 gives the experimental O2 concentrations, calculated rate constants
(min-1), the abiotic:biotic ratio from rate constants for each experimental set, slope data (log
[Fe2+] vs. time), 1-sigma standard error and % error for the fitted line, and p-values calculated
comparing abiotic to biotic reaction slopes. Figures 5A-5H present the pseudo-first order fits for
the sets of kinetic experiments at each O2 concentration.
To compare abiotic vs. biotic rate data, the p-value comparing 2 different time vs. log
[Fe2+] slopes was calculated to assess the degree of difference or similarity between the 2 slopes;
p values less than 0.05 indicate the slopes are statistically different from each other to 95%
confidence. All rate comparisons except the highest O2 concentration (Experiment 9) were
statistically different at this level (Table 1). Differences of reaction rate constants between
separate sets of experiments are largely due to differences in HFO surface area and cell activities
(cell densities should have been pretty similar between experiments) which were not possible to
constrain and control between experiment sets, though the comparison of abiotic to biotic rates
remains consistent.
4. DISCUSSION
18
4.1 Geochemical niche of neutrophilic iron oxidizing microbes
Understanding the details of a microbe’s environmental niche is an important step in
developing a more complete elucidation of the functioning and dynamics of a microbial
ecosystem. For FeOB this is especially true, since their survival and optimal growth at
circumneutral pH is dependent upon growth in a very restricted niche that provides low
concentrations of Fe(II) and O2 at a steady flux. That these organisms are successful is evident
from their ubiquity and abundance in the habitats where they grow; however, specific
biogeochemical details of their ecosystem have been lacking. This work presents one of the first
high resolution field profiles of both oxygen and different iron species that have been done in an
iron microbial mat ecosystem where Fe(II) oxidizing microbes dominate. In addition, similar
profiles were done in a pure culture of an iron-oxidizing bacterium growing in a gradient that the
microbes influenced, as well as measurements of the iron oxidation rate of this organism in
response to varying O2 concentrations. Together these results enhance our understanding of the
dynamics of circumneutral lithotrophic microbial ecosystems that utilize Fe(II) as an energy
source
Geochemical profiles at the field site (Figures 3A and 3B) indicate that O2 is consumed in
the first few mm of the iron mat, and that Mn and Fe cycling predominate once low or
undetectable O2 concentrations prevail. The Fe(II) concentration increases with depth in the mat
indicating that Fe(II) is upwelling from Fe(II)-rich anoxic groundwater. It is also likely that Fereduction is occurring in the deeper, anaerobic parts of the mat, and that some of the Fe(II) is the
result of activity of Fe-reducing bacteria. The soluble Fe(III) distribution also mirrors the activity
associated with lower O2 iron(II) oxidation in that Fe(III) increases in the low-O2 zone. The
Fe(III) is distributed through to the bottom of the profile, likely due to some combination of
19
further Fe(II)/ FeS(aq)/FeMnS(aq) oxidation and/or the settling/diffusion of solid or nanoparticulate
Fe(III) forms. These field results from the Contrary Creek wetland study site clearly indicate
that iron oxidizing microbes in circumneutral waters are predominantly active in low oxygen
geochemical niches, a result which is consistent with a number of previous studies (Emerson and
Weiss, 1997; Emerson et al., 1999; Sobolev and Roden, 2001, 2002; Neubauer et al., 2002;
Emerson and Moyer, 2004; Ferris, 2004; Roden et al., 2004; James and Ferris, 2005;). Our field
observations are supported by voltammetric evidence describing the geochemical niche these
same organisms occupy in gradient culture tubes, a result found in two other studies (Emerson
and Moyer, 1997; Roden et al., 2004). It should be noted that this work stands in sharp contrast
to another recent study using voltammetric electrodes to investigate the iron dynamics of the
Chocolate Pots thermal spring in Yellowstone National Park (Trowburst et al., in press). The
anoxic water feeding Chocolate Pots is circumneutral and contains similar Fe(II) concentrations
as at Contrary Creek; however the water is warmer (around 50°) and supports a robust
cyanobacterial mat. At Chocolate Pots, it is the oxygen produced by the cyanobacterial
community that tightly controls the Fe-oxidation rates, and there is little or no evidence that
FeOB are able to compete with the rapid abiotic Fe-oxidation that results. At Contrary Creek,
and at similar iron seeps, cyanobacteria and eukaryotic green algae are normally present at low
numbers, and are not known to influence the iron dynamics.
The additional observation in our study of the possible role of FeS(aq) in iron cycling for
these systems is of considerable note, as it is consistent in both the field and gradient culture tube
results. As a result of this observation, we did attempt to prepare a medium in which FeS(aq)
could be the predominant substrate for the ES-1 isolate, as was observed during one of our
sampling visits to Contrary Creek (Figure 1). However, we were unable to prepare a medium in
20
the absence of Fe(II) which would be required to validate the idea of substrate uptake.
Subsequent experiments (ongoing) in the Druschel lab at the University of Vermont are
indicating the development of FeS(aq) as a result of abiotic non-oxidative mackinawite dissolution
in agarose gradient tubes, but either the concurrent dissolution of mackinawite with H+ to form
Fe(II) (equation 12) or the reversible dissociation of FeS(aq) also yields Fe(II) in those solutions,
consistent with the recent results of Rickard (2006). Calculations of the thermodynamic
equilibrium expected between FeSmackinawite, Fe(II), and FeS(aq) in these solutions using the
Geochemist’s workbench indicate that FeS(aq) and Fe(II) should both be present if the system is
at equilibrium (though the FeS(aq) is calculated as the monomer and is as such a mathematical
construct, but its presence is supported by voltammetric evidence). The observation at Contrary
Creek of a system with predominant FeS(aq) without significant Fe(II) (Figure 1) when similar
abiotic experiments and calculations indicate Fe(II) should be present, suggest the neutrophilic
iron oxidizing microbial community may preferentially utilize the Fe(II) ion over the cluster
form. The FeS(aq) cluster may thus serve an important environmental role in the transport of iron
in these systems if not as a directly utilized metabolic component, and certainly is a critical
component describing the solubility of iron associated with mackinawite dissolution (Luther et
al., 2003; Rickard, 2006; Rickard and Luther, 2006)
4.2 Relating the kinetic results to the geochemical niche of neutrophilic iron oxidizing
microbes
While the importance of a low oxygen geochemical niche for neutrophilic iron oxidizing
organisms is well established (Emerson and Weiss, 1997; Emerson et al., 1999; Sobolev and
Roden, 2001, 2002; Neubauer et al., 2002; Emerson and Moyer, 2004; James and Ferris, 2004;
21
Roden et al., 2004; Ferris, 2005), quantifying the kinetic limits of this geochemical niche and the
role of the organisms in developing and maintaining it are much less understood. Kinetic
experiments for FeOB grown in gradient culture and inoculated into an electrochemical cell in
the lab yielded pseudo-first order rate constants between 0.2 and 0.02 min-1, which based on
measured cell densities (2.1x107 cells ml-1), translates to between 1 x10-9 and 1x10-10 min-1 cell1
. These overall rate constants, which include the rate associated with the organisms, as well as
with autocatalytic and aqueous chemical oxidation reactions, compare reasonably well with rate
constants from natural mat suspensions determined by Rentz et al. (2007) of .029 ± 0.004 to
0.249 ± 0.042 min-1. These data also compare favorably with data in Neubauer et al. (2003) who
also worked with pure cultures and calculated an average cell-normalized rate in bioreactors of
1.4x10-1 to 8.3x10-3 (average of 4.2x10-2) pmol cell-1 h-1, compared to our rates at 100 M Fe(II)
concentration of 5x10-2 to 5x10-3 pmol cell-1 h-1 . Rentz et al. (2007) were working with
environmental samples and did not attempt to determine cell densities of FeOB in their
experiments, but based on our cell-normalized rates, one may roughly estimate that the cell
density associated with the overall rate they saw using field samples in miniature bioreactors is
on the order of 10-7 cells ml-1. This is, in turn, comparable to MPN determinations of Feoxidizers found at this site (4 x 106 cells ml-1).
Iron(II) oxidation in both the field and gradient culture tubes occurred in a low-O2
window that had a maximum concentration near 50 M O2. Kinetic experiments (Figure 5A-5I)
indicate that the relative rates of iron(II) oxidation (abiotic:biotic rate constants, Table 1)
approach a 1:1 ratio around these higher O2 values. The abiotic:biotic rate constant ratios for the
experiments at 45 and 50 M O2 (Experiments 5 and 8, respectively, Table 1) are still less than
1.0, but it is important to note that the biotic rate may include the abiotic rate. Given the
22
experimental design, as the rate of abiotic iron(II) oxidation becomes faster than the biotic rate,
the overall rate measured will then be a measure of that abiotic rate and not the biotic signal.
Therefore it is the approach of the abiotic:biotic rate constant ratio to 1:1 that indicates an
approaching condition where the biotic reaction has ceased to favorably compete with the abiotic
reaction. If one subtracts out the biotic rate from the background abiotic reaction rate (possible
as the same solution with almost exactly the same amount of HFO in the cell was used to
determine the abiotic control rate in each experimental set), the % of total reaction attributed to
the biotic component falls quickly (Figure 6, Table 2). The results of the experiments presented
here quantitatively demonstrate that the competition between abiotic and biotic rates swings in
favor of the abiotic rate above approximately 50 M O2 which then limits the geochemical niche
in which circumneutral iron oxidizing microorganisms thrive.
4.3 Variability between kinetic datasets
Experimental sets were run at different times for this study; the number of cells, their
activity, and the surface area of HFO were likely quite different between experimental sets (but
not for aiotic:biotic comparisons where experimental design dictates equivalent HFO and cell
density for the biotic and subsequent abiotic (kill control) experiments) . The significant
differences in measured biotic reaction rates for the low O2 experiments performed in this study
exemplifies the potential importance of cell density and activity (Table 2). For example, the
biotic rate was 88% of the total for an experiment run in 15 M O2 (Experiment 6), while it was
20% of the total for another experiment run in 16 M O2 (Experiment 3). The significant
differences in between abiotic rates for different experimental sets demonstrate the significant
differences between experimental sets based on the amount of HFO transferred with each
23
inoculation (as cell densities and microbial activity should not be part of the abiotic rates and
variability at very similar O2, Fe(II), and pH conditions see different rates). We note that any
attempt to derive quantitative models of iron oxidation must therefore consider cell density,
activity, and HFO surface area as critical parameters.
5. CONCLUSIONS
The geochemical niche that neutrophilic iron(II) oxidizing microorganisms are restricted
to is principally constrained by the O2 conditions where biotic rates of Fe(II) oxidation
outcompete abiotic rates. At this study site and in axenic culture gradient tubes of the isolate ES1, the habitat of neutrophilic iron(II) oxidizing microorganism activity was restricted to
environments where the O2 levels were less than approximately 50 M. Kinetic experiments
successfully described the competitive kinetics of biotic vs. abiotic Fe(II) oxidation as a key
element to controlling the niche in which these organisms thrive. Cell density, activity, and HFO
surface area are all key parameters affecting these competing rates in the environment. Ironsulfur molecular clusters, FeS(aq), are also potentially important intermediates and Fe(II) sources
in these types of systems, and may additionally provide insight to links between sulfur cycling
and iron cycling as FeS(aq) may be a key source of dissolved Fe(II) for microorganisms in
microaerophilic settings.
ACKNOWLEDGEMENTS
This work was supported by a subcontract to the University of Delaware and to the ATCC from
the NASA Astrobiology grant to University of California, Berkeley. GKD gratefully
24
acknowledges support from the American Chemical Society Petroleum Research Fund (43356GB2) and NSF-EPSCoR-VT (NSF EPS Grant 0236976). We thank Charoenkwan Kraiya and
Brian Glazer for their assistance in the field measurements and Doug Nixon for making the
borosilicate glass portion of the solid state voltammetric microelectrodes. Melissa Merrill Floyd
was instrumental in setting up the gradient tubes and managing the time-lapse image collection.
REFERENCES
Anderson, J.E., and Robbins, E.I. (1998) Spectral reflectance and detection of iron-oxide precipitates
associated with acidic mine drainage. Photogramm Eng Rem S 64 (12), 1201-1208.
Benison, K.C., and LaClair, D.A. (2003) Modern and ancient extremely acid saline deposits: Terrestrial
analogs for martian environments? Astrobiology 3 (3), 609-618.
Benning, L.G., Wilkin, R.T., and Barnes, H.L. (2000) Reaction pathways in the Fe-S system below
100ºC. Chem Geol. 167, 25-51.
Brendel, P.J., and Luther, G.W. (1995) Development of a Gold Amalgam Voltammetric Microelectrode
for the Determination of Dissolved Fe, Mn, O-2, and S(-II) in Porewaters of Marine and FreshWater Sediments. Envirol Sci Technol. 29 (3), 751-761.
Burke, S.P. and Banwart, S.A. (2002) A geochemical model for removal of iron(II)(aq) from mine water
discharges. Appl Geochem. 17(4), 431-443.
Brezonik, P.L. (1994) Chemical Kinetics and Process Dynamics in Aquatic Systems. CRC Press, Lewis
Publishers, Boca Raton. 754 p.
Butler I.B., Boettcher M.E., Rickard D. and Oldroyd A. (2004) Sulfur isotope partitioning during
experimental formation of pyrite via the polysulfide and hydrogen sulfide pathways; implications
for the interpretation of sedimentary and hydrothermal pyrite isotope records. Earth Planet Sci
Lett 228, 495-509.
25
Chen W. F. and Liu T. K. (2005) Ion activity products of iron sulfides in groundwaters: Implications from
the Choshui fan-delta, Western Taiwan. Geochim. Cosmochim. Ac 69, 3535-3544.
Davison W., Phillips N., and Tabner B.J. (1999) Soluble iron sulfide species in natural waters: reappraisal
of their stoichiometry and stability constants. Aquat Sci 61, 23-43.
De Vitre, R.R., Buffle, J., Perret, D. amd Baudat, R. (1988) A study of iron and manganese
transformations at the O2/S(-II) transition layer in a eutrophic lake (Lake Bret, Switzerland): A
multimethod approach. Geochim. Cosmochim. Ac. 52, 1601-1613.
Dollhopf M. E., Nealson K. H., and Luther G. W. (1999) In situ solid state Au/Hg voltammetric
microelectrodes to analyze microbial metal reduction. Abstr Pap Am Chem S 217, U853-U853.
Druschel, G.K., Hamers, R.J., Luther, G.W., and Banfield, J.F. (2003) Kinetics and mechanism of
trithionate and tetrathionate oxidation at low pH by hydroxyl radicals. Aquat Geochem, 9(2), 145164.
Druschel G. K., Sutka R., Emerson D., Luther G. W., Kraiya C., and Glazer. B. (2004) Voltammetric
investigation of Fe-Mn-S species in a microbially active wetland. In Eleventh Internat. Symp.
Water-Rock Interaction WRI-11, Vol. 2 (ed. R. B. Wanty and R. R. Seal), pp. 1191-1194.
Edwards, K.J., Bach, W., McCollom, T.M., and Rogers, D.R. (2004) Neutrophilic iron-oxidizing bacteria
in the ocean: Their habitats, diversity, and roles in mineral deposition, rock alteration, and
biomass production in the deep-sea. Geomicrobiol Jl 21, 393-404.
Emerson, D. and Moyer, C. (1997) Isolation and characterization of novel iron-oxidizing bacteria that
grow at circumneutral pH. Appl. Environ. Microbiol. 63, 4784–4792.
Emerson, D. and Floyd, M.M. (2005) Enrichment and isolation of iron-oxidizing bacteria at neutral pH.
Method Enzymol. 397, 112-124.
Emerson, D. and Revsbech, N.P. (1994) Investigation of an iron-oxidizing microbial mat community
located near Aarhus Denmark: Field studies. Appl. Environ. Microbiol 60, 4022-4031.
Emerson, D. and Weiss, J. V. (2004) Bacterial iron oxidation in circumneutral freshwater habitats:
findings from the field and the laboratory. Geomicrobiol. J. 21, 405-414.
26
Emerson, D., Weiss, J. V., and Megonigal, J. P. (1999) Iron-oxidizing bacteria are associated with ferric
hydroxide precipitates (Fe-plaque) on the roots of wetland plants. Appl. Environ. Microbiol. 65,
2758-2761.
Emerson, D., Rentz, J.A., and Plaia, T. (in press) Sideroxydans lithotrophicus, gen. nov., sp. nov. and
Gallionella capsiferriformans sp.nov., oxygen-dependent ferrous iron-oxidizing bacteria that
grow at circumneutral pH. Int J Syst Evol Micr.
Ferris, F.G. (2005) Biogeochemical properties of bacteriogenic iron oxides. Geomicrobiology J 22, 7985.
Furrer G., Phillips B.L., Ulrich K.U., Pothig R., Casey W.H. (2002) The origin of aluminum flocs in
polluted streams. Science, 297, 2245-2247.
Glazer, B.T., Marsh, A.G., Stierhoff, K., and Luther, G.W. (2004) The dynamic response of optical
oxygen sensors and voltammetric electrodes to temporal changes in dissolved oxygen
concentrations. Anal Chim Acta, 518(1-2), 93-100.
Glazer, B.T., Luther, G.W., Konovalov, S.K., Friederich, G.E., Nuzzio, D.B., Trouwborst, R.E., Tebo,
B.M., Clement, B., Murray, K., Romanov, A.S. (2006) Documenting the suboxic zone of the
Black Sea via high-resolution real-time redox profiling. Deep-Sea Res Pt II 53 (17-19), 17401755.
James, R.E. and Ferris, F.G. (2004) Evidence for microbial-mediated iron oxidation at a neutrophilic
groundwater spring. Chem Geol 212, 301-311.
Kappler, A. and Newman, D.K. (2004) Formation of Fe(III)-minerals by Fe(II)-oxidizing
photoautotrophic bacteria. Geochim Cosmochim Ac, 68(6), 1217-1226.
Kirby, C.S., Thomas, H.M., Southam, G., and Donald, R. (1999) Relative contributions of abiotic and
biological factors in Fe(II) oxidation in mine drainage. Appl Geochem, 14(4), 511-530.
Kump, L.R. and Seyfried, W.E. (2005) Hydrothermal fluxes during the Precambrian: Effect of low
oceanic sulfate concentrations and low hydrostatic pressure on the composition of black smokers.
Earth Planet Sc Lett 235, 654-662.
27
Luther, G.W., Ferdelman, T.G., Kostka, J.E., Tsamakis, E.J., Church, T.M. (1991) Temporal and Spatial
variability of reduced sulfur species (FeS2, S2O32-) and porewater parameters in salt-march
sediments. Biogeochemistry 14(1), 57-88.
Luther G.W., Rickard D., Theberge S.M., and Oldroyd A (1996) Determination of metal (bisulfide)
stability constants of Mn, Fe, Co, Ni, Cu, and Zn by voltammetric methods. Environ Sci Tech 30,
671-679.
Luther G.W., Brendel P.J., Lewis B.L., Sundby B., Lefrancois L., Silverberg N., and Nuzzio D.B.,
(1998): Simultaneous measurement of O2-, Mn, Fe, I-, and S2- in marine pore waters with a solidstate voltammetric microelectrode. Limnol Oceanogr 43, 325-333.
Luther GW, Theberge SM, and Rickard DT (1999) Evidence for aqueous clusters as intermediates during
zinc sulfide formation. Geochim Cosmochim Ac 63, 3159-3169.
Luther G.W., Glazer B.T., Hohmann L., Popp J.I., Taillefert M., Rozan T.F., Brendel , P.J., Theberge
S.M., and Nuzzio D.B. (2001) Sulfur speciation monitored in situ with solid state gold amalgam
voltammetric microelectrodes: polysulfide as a special case in sediments, microbial mats and
hydrothermal vent waters. J Environ Monitor 3, 61-66.
Luther G.W., Theberge S.M., Rozan T.F., Rickard D., and Oldroyd A. (2002) Aqueous copper sulfide
clusters as intermediates during copper sulfide formation. Environ Sci Tech 36, 394-402.
Luther G.W., Glazer B., Ma S., Trouwborst R., Shultz B., Druschel G.K., and Kraiya C. (2003) Iron and
sulfur chemistry in a stratified lake: Evidence for iron-rich sulfide complexes. Aquatic Geochem
9:87-110.
Luther G.W. and Rickard D. (2005) Metal sulfide cluster complexes and their biogeochemical importance
in the environment. J Nano Res 7, 389-407.
Navrotsky, A. (2004) Energetic clues to pathways to biomineralization: Precursors, clusters, and
nanoparticles. Proc Natl Acad Sci , 12096-12101.
28
Neubaeur, S.C., Emerson, D., and Megonigal, J.P. (2002) Life at the energetic edge: kinetics of
circumneutral iron oxidation by lithotrophic iron-oxidizing bacteria isolated from the wetland
plant rhizosphere. Appl. Environ. Microbiol 68, 3988-3995.
Neubaeur, S.C., Emerson, D., and Megonigal, J.P. (in press) Microbial oxidation and reduction of iron in
the root zone and influences on metal mobility. In: Violante, A., Huang, P.M., and Gadd, G.M.
(eds.) Biophysico-Chemical Processes of heavy metals and metalloids in soil environments.
International Union of Pure and Applied Chemistry.
Rentz, J.A., Kraiya, C., Luther, G.W., Emerson, D. (2007) Control of ferrous iron oxidation within
circumneutral microbial iron mats by cellular activity and autocatalysis. Environ Sci Technol. 41,
6084-6089.
Rickard, D. (2006) The solubility of FeS. Geochim Cosmochim Ac 70, 5779-5789.
Rickard, D. and Luther GW (1997) Kinetics of pyrite formation by the H2S oxidation of iron (II)
monosulfide in aqueous solutions between 25 and 125oC: the mechanism. Geochim Cosmochim
Ac 6, 135-147.
Rickard, D. and Luther G.W. (2006) Metal sulfide complexes and clusters. In: Sulfide Mineralogy and
Geochemistry (eds by Vaughan DJ, Washington DC). Mineral Soc Am. Rev Mineral Geochem
61, 421-504.
Roden, E.E., Sobolev, D., Glazer, B., and Luther, G.W. (2004) Potential for microscale bacterial Fe redox
cycling at the aerobic-anaerobic interface. Geomicrobiology J 21, 379-391.
Roesler A.J., Gammons C.H., Druschel G.K., Oduro H.D., and Poulson S.R. (2007) Geochemistry of
flooded underground mine workings influenced by bacterial sulfate reduction. Aquat Geochem,
currently online.
Rouxel, O.J., Bekker, A., and Edwards, K.J. (2005) Iron isotope constraints on the archean and
Paleoproterozoic ocean redox state. Science 307, 1088-1091.
Rouxel, O.J., Bekker, A., and Edwards, K.J. (2006) Response to Comment on “Iron isotope constraints on
the Archean and Paleoproterozoic ocean redox state”. Science, 311, 177b.
29
Rosson, R. A.; Tebo, B. M.; Nealson, K. H. (1984) Use of Poisons in Determination of Microbial
Manganese Binding Rates in Seawater. Appl. Environ. Microbiol, 47, 740-745.
Rozan T.F., Lassman M.E., Ridge D.P., Luther G.W. (2000) Evidence for iron, copper and zinc
complexation as multinuclear sulfide clusters in oxic rivers. Nature 406, 879-882.
Singer P.C. and Stumm W. (1970) Acidic mine drainage: The rate-determining step. Science 167, 11211123.
Sobolev, D. and Roden, E.E. (2001) Suboxic deposition of ferric iron by bacteria in opposing gradients of
Fe(II) and oxygen at circumneutral pH. Appl. Environ. Microbiol. 67, 1328–1332.
Sobolev, D., and Roden, E.E. (2002) Evidence for rapid microscale bacterial redox cycling of iron in
circumneutral environments. Anton Leeuw Int J G, 81 (1-4), 587-597.
Stookey, L.L. (1970) Ferrozine – a new spectrophotometric reagent for iron. Anal Chem, 42, 779-781.
Straub, K.L., Benz, M., and Schink, B. (2001) Iron metabolism in anoxic environments at near neutral
pH. FEMS Microbiol Ecol, 34, 181-186.
Stucki, J.W., Lee, W., Goodman, B.A., and Kostka, J.E. (2007) Effects of in situ biostimulation on iron
mineral speciation in a sub-surface soil. Geochim Cosmochim Ac, 71(4), 835-843.
Theberge S. M. and Luther G. W. (1997) Determination of the Electrochemical Properties of a soluble
aqueous FeS species present in sulfidic solutions. Aquat. Geochem. 3, 191-211.
Taillefert M., Bono A.B., and Luther G.W. (2000) Reactivity of freshly Fe(III) in synthetic solutions and
marine (pore)waters: voltammetric evidence of an aging process. Environ Sci Technol 34, 21692177.
Trouwborst, R. E., Johnston, A., Koch, G., Luther, G. W. III , and Pierson, B. K. (in press)
Biogeochemistry of Fe(II) oxidation in a photosynthetic microbial mat: Implications for
Precambrian Fe(II) oxidation. Geochim Cosmochim Ac.
30
Weiss, J.V., Emerson, D., Backer, S., and Megonigal, P. (2003) Enumeration of Fe(II)-oxidizing and
Fe(III)-reducing bacteria in the root zone of wetland plants: Implications for a rhizosphere iron
cycle. Biogeochemistry 64, 77-96.
Weiss, J.V., Emerson, D., and Megonigal, P. (2005) Rhizosphere iron(III) deposition and reduction in a
Juncus effuses-dominated wetland. Soil Biol Biochem, 69, 1861-1870.
Wolthers M., Charlet L., van der Linde P.R., Rickard D., and van der Weidjen C.H. (2005) Surface
chemistry of disordered mackinawite. Geochim Cosmochim Ac 69, 3469-3481.
Xu K., Dexter S. C., and Luther G. W. (1998) Voltammetric microelectrodes for biocorrosion studies.
Corrosion 54(10), 814-823.
Yamaguchi, K.E. and Ohmoto, H. (2006) Comment on “Iron isotope constraints on the archaen and
Paleoproterozoic ocean redox state. Science 311, 177a.
31
Figure 1 – Voltammetric scans taken in the Contrary Creek Wetland Study site at 4 different
depths during 2002. Note the predominance of the peak for FeS(aq) and the lack of detectable
Fe(II) in this sample.
32
Figure 2 – Selected voltammograms from 2003 sampling trip at selected depths. These scans are
representative of the data used to construct the profile in Figure 3.
33
Profile - Contrary Creek
Profile - Contrary Creek
0
0
O2
O2
5
1
Fe(III)
Fe(III)
Mn(II)
FeS(aq)
15
FeMnS(aq)
20
Depth (mm)
Depth (mm)
(FeII)
Fe(II)
10
Mn(II)
2
FeS(aq)
FeMnS(aq)
3
4
25
5
30
35
6
0
200
400
600
Conc. (M) or current (nA)
0
50
100 150 200 250 300 350
Conc. (M) or current (nA)
Figure 3 – Profiles generated from voltammetric scans performed in the contrary Creek wetland
study site. 2A is profile through top 40 mm while 2B is a close-up view of the top 6 mm. The
FeS(aq) and Fe(III) forms are presented in current output (nA) and cannot be calibrated and
presented in concentration units.
34
Control - 2 days
0
0
2
2
4
4
6
6
O2
8
10
Fe(II)
12
FeS(aq)
(nA*10)
Depth (mm)
Depth (mm)
ES-1 Culture - 2 days
8
O2
10
Fe(II)
12
14
14
16
16
18
18
20
FeS(aq)
(nA*10)
20
0
100
200
Concentration (M)
300
0
0
0
2.5
2.5
5
5
7.5
7.5
O2
10
Fe(II)
12.5
FeS(aq)
(nA*10)
15
300
Control - 5 days
Depth (mm)
Depth (mm)
ES-1 culture - 5 days
100
200
Concentration (M)
Fe(II)
12.5
FeS(aq)
(nA*10)
15
17.5
17.5
20
20
22.5
22.5
25
O2
10
25
0
200
400
600
Concentration (M )
800
0
200
400
600
Concentration (M)
800
35
Figure 4A-D – Vertical profiles compiled from voltammetric scans at 0.5 to 1 mm increments in
gradient culture tubes inoculated with ES-1 after 2 and 5 days, respectively with associated
abiotic controls. Note the FeS(aq) signal is presented in current (nA) directly from the
voltammograms as there is no way to calibrate for this moiety.
36
Experiment 1 - 10  M O2
Experiment 2 - 9  M O2
-4.0
-4.0
-4.1
Biotic
-4.2
Azide
control
-4.3
Linear
(Azide
control)
Linear
(Biotic)
-4.4
-4.5
0
5
10
15
log Fe(II) (M)
log Fe(II) (M)
-4.1
Biotic
-4.2
-4.3
Azide
control
-4.4
Linear
(Azide
control)
Linear
(Biotic)
-4.5
-4.6
20
0
time (minutes)
Experiment 3 - 16  M O2
10
time (minutes)
15
20
Experiment 4 - 25  M O2
-4
-4.00
-4.1
-4.20
-4.2
Biotic
-4.3
Azide
control
-4.4
-4.5
-4.6
-4.7
0
5
10
time (minutes)
15
Linear
(Azide
control)
Linear
(Biotic)
log Fe(II) (M)
log Fe(II) (M)
5
-4.40
biotic
-4.60
Azide
control
-4.80
-5.00
-5.20
0
5
time (minutes)
10
Linear
(Azide
control)
Linear
(biotic)
37
Experiment 6 - 15  M O2
Experiment 5 - 45  M O2
-4.0
-4.0
Biotic
Azide
Control
-4.2
Linear
(Azide
Control)
Linear
(Biotic)
-4.3
-4.4
0
5
10
log Fe(II) (M)
log Fe(II) (M)
-4.0
-4.1
-4.1
Azide
control
-4.2
-4.2
Linear
(Azide
control
20)
Linear
(Biotic)
-4.3
0
15
5
10
15
time (minutes)
time (minutes)
Experiment 7 -25  M O2
Experiment 8 -50  M O2
-4.0
-4.0
-4.1
Biotic
-4.1
-4.2
Azide
control
-4.3
Linear
(Azide
control
)
20 Linear
(Biotic)
-4.4
0
5
10
15
time (minutes)
log Fe(II) (M)
log Fe(II) (M)
Biotic
-4.1
Biotic
-4.2
Azide
control
-4.3
-4.4
-4.5
0
5
10
time (minutes)
15
Linear
(Azide
control
)
20 Linear
(Biotic)
Exp 9 - 275  M O2
-4.4
-4.6
log Fe(II) (M)
-4.8
Biotic
-5.0
Azide
control
-5.2
-5.4
Linear
(Biotic)
-5.6
-5.8
-6.0
0
2
4
6
8
10
Linear
(Azide
control)
12
time (min)
Figure 5A-I – Compiled pseudo-first-order rate plots for sets of biotic and abiotic control
experiments at each O2 concentration investigated in this study. Rate data and statistical
38
analyses for each set is tabulated in Table 1 and raw data are available in the online materials
(Appendix EA-1; web address).
39
% biotic rate of the total
measured rate
Biotic % of reaction rate
100
90
80
70
60
50
40
30
20
10
0
0
100
200
300
O2 (M)
Figure 6 – Plot of the calculated % of reaction rate attributed to the biotic oxidation of iron in
each experimental set.
40
Experiment
1B
1A
ratio
rate constant, abiotic:biotic
O2 (M)
k (min -1 )
rate constant
10
0.046
10
0.026
0.6
slope
-0.020
-0.011
SE for
slope
0.00068
0.00080
%SE
-3.4%
-7.1%
0.7
-0.028
-0.021
0.0007
0.0005
-2.3%
-2.5%
0.0018
0.0011
p-value
to biotic
slope
0.000
Notes
biotic
abiotic, azide control
0.000
biotic
abiotic, azide control
-3.9%
-3.0%
0.000
biotic
abiotic, azide control
2B
2A
9
9
0.065
0.047
3B
3A
16
16
0.105
0.084
0.8
-0.0457
-0.0367
4B
4A
25
25
0.223
0.143
0.6
-0.097
-0.062
0.0021
0.0016
-2.6%
-2.2%
0.000
biotic
abiotic, azide control
5B
5A
45
45
0.055
0.047
0.9
-0.024
-0.020
0.00020
0.00055
-0.8%
-2.7%
0.000
biotic
abiotic, azide control
6B
6A
15
15
0.030
0.004
0.1
-0.0129
-0.0016
0.00049
0.00024
-3.8%
-15.2%
0.000
biotic
abiotic, azide control
7B
7A
25
25
0.040
0.025
0.6
-0.017
-0.011
0.00037
0.00054
-2.1%
-5.0%
0.000
biotic
abiotic, azide control
8B
8A
50
50
0.045
0.034
0.8
-0.019
-0.015
0.00058
0.00043
-3.0%
-2.9%
0.000
biotic
abiotic, azide control
9B
9A
275
275
0.197
0.188
1.0
-0.086
-0.082
0.0039
0.0035
-4.6%
-4.3%
0.611
biotic
abiotic, azide control
Table 1 – Tabulated experimental O2 concentrations, calculated rate constants (min-1), the
abiotic:biotic ratio from rate constants for each experimental set, slope data (log [Fe2+] vs. time),
1-sigma standard error and % error for the fitted line, and p-values calculated comparing abiotic
to biotic reaction slopes from Minitab calculations (p>0.05 indicates no statistical difference
between biotic and abiotic rates).
41
O2
(M)
9
10
15
16
25
25
45
50
275
% biotic rate
of total rate
27
43
88
20
36
38
15
23
4
Table 2 – Calculated percentage of the biotic rate out of the total rate for experiment sets 1-9.