TECHNICAL REPORTS: HEAVY METALS IN THE ENVIRONMENT
Ecology of Sulfate-Reducing Bacteria in an Iron-Dominated, Mining-Impacted
Freshwater Sediment
Srivi Ramamoorthy, Jeffrey S. Piotrowski, Heiko W. Langner, and William E. Holben University of Montana
Matthew J. Morra University of Idaho
R. Frank Rosenzweig* University of Montana
A legacy of lead and silver mining in its headwaters left Lake
Coeur d’Alene, Idaho with a sediment body that is highly
reduced and contains up to 100 g kg−1 iron and a smaller fraction
of chemically active sulfide phases. The dynamic character
of these sulfides and their importance for the sequestering
of contaminating trace elements prompted this study of the
sulfate-reducing bacteria (SRB) involved in their production.
We estimated parameters indicative of the distribution and
activity of SRB in relation to season, site, and depth. Most
probable number estimates and quantitative PCR assays of an
SRB-specific functional gene, α-adenosine 5´-phosphosulfate
reductase, indicated 103 to 106 cultivable cells and 105 to 107
gene copy numbers g−1 dry wt sediment, respectively. Although
culture-based estimates of SRB abundance correlated poorly
with site, season, depth, total S, or pore water SO4, non–
culture-based estimates of SRB abundance were markedly
higher at contaminated sites and positively correlated with
pore water SO4. Ex situ estimates of 35SO4 respiration and
acid volatile sulfides abundance also showed strong among-site
effects, indicating elevated sulfidogenesis at contaminated sites.
These observations support the view that biogenic sulfides may
act in concert with reduced iron to retain soluble metal(loid)s
in the solid phase.
Copyright © 2009 by the American Society of Agronomy, Crop Science
Society of America, and Soil Science Society of America. All rights
reserved. No part of this periodical may be reproduced or transmitted
in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system,
without permission in writing from the publisher.
Published in J. Environ. Qual. 38:1–10 (2009).
doi:10.2134/jeq2007.0577
Received 31 Oct. 2007.
*Corresponding author ([email protected]).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA
S
 1895, approximately 115 million metric tonnes of mine
tailings have been produced in the Coeur d’Alene (CDA)
mining district, the so-called Silver Valley of northern Idaho (USA).
Roughly 60% of these materials are believed to have entered the
CDA River system (Fig. 1) (Mink, 1971). Although metal loading
of the CDA River was restricted in 1968 by the construction of
tailing ponds (Javorka, 1991), contaminated stream-bed and
stream-bar sediments continue to be resuspended by spring runoff
and secondarily transported into Lake CDA (Javorka, 1991).
Historically, the main sources for heavy metal input into Lake CDA
are the South Fork and main stem of the CDA River (Funk et al.,
1975; Galbraith, 1992; La Force et al., 1999; Reece et al., 1978;
Horowitz and Elrick, 1993). Total abundance of Pb, Zn, and As
in Lake CDA sediments has been estimated at 3820 ± 241, 2995
± 125, and 201 ± 11 mg kg1, respectively, while concentrations in
the contiguous, pristine St. Joe River (SJ) delta are 78 ± 15, 751 ±
40, and 12 ± 3 mg kg1, respectively (Harrington et al., 1998b).
Analysis of sediment pore water shows that iron, lead, arsenic,
manganese, and other trace element concentrations can greatly
exceed federal drinking water limits even though surface waters
remain in compliance (Harrington et al., 1998b). Combined
radiometric (137Cs) and stratigraphic data (e.g., location of Mt. St.
Helens ash) indicate that sedimentation rate in Lake CDA ranges
from 1.4 to 2.1 cm yr1 (Horowitz and Elrick, 1993).
The dominant reactive element in the CDA sediment system
is iron, which ranges from approximately 83,000 mg kg1 in the
CDA River delta to approximately 38,000 mg kg1 in the pristine
SJ delta. In reduced sediments below the sediment–water interface, soluble Fe(II) in pore water increases with depth from 0.1 to
approximately 0.7 mol L1 (Cummings et al., 2000; Toevs et al.,
2006). X-ray absorption near-edge structure spectroscopy and Xray diffraction demonstrate a preponderance of Fe(II) minerals such
as siderite (25–60% total iron); sulfidic minerals are present in relatively low abundance (2–2.5% of total iron) (Toevs et al., 2006).
We have previously reported the presence of abundant and diverse
S. Ramamoorthy, J.S. Piotrowski, W.E. Holben, and R.F. Rosenzweig, Division of
Biological Sciences, Univ. of Montana, Missoula, MT 59812-4824; H.W. Langner, Dep.
of Geology, Univ. of Montana, Missoula, MT 59812; M.J. Morra, Dep. of Plant Soil &
Entomological Sciences, Univ. of Idaho, Moscow, ID 83844.
Abbreviations: APS, adenosine 5-phosphosulfate reductase; AVS, acid volatile
sulfides; CDA, Coeur d’Alene; HP, Harlow Point; MPN, most probable number; PP,
Peaceful Point; qPCR, quantitative polymerase chain reaction; SJ, Saint Joe River; SOX,
sulfur oxidizers; SRB, sulfate-reducing bacteria; SRP, sulfate reduction potential.
1
For proofing purposes only
Fig. 1. Map showing Lake Coeur d’Alene and (inset) sampling sites. The Bunker Hill EPA Superfund site is centered on Kellogg, ID. Sampling sites
PP (Peaceful Point) and HP (Harlow Point) are in the highly contaminated Coeur d’Alene River delta, whereas site SJ (St. Joe) is in the delta of
the pristine St. Joe River.
populations of dissimilatory iron-reducing bacteria (Cummings
et al., 2000; Cummings et al., 2003). These include novel species (Cummings et al., 1999b), some of which are capable of
dissimilatory arsenic reduction (Niggemyer et al., 2001). We
have hypothesized that iron-reducing microbes play a significant
role in sediment diagenesis and contribute to the accumulation
of arsenic near the redox boundary (Harrington et al., 1998a;
Cummings et al., 2000). Recent molecular spectroscopic studies
support this hypothesis (Toevs et al., 2006).
Bacterial sulfate respiration in anoxic soils and sediments
generates bisulfide, which at circumneutral pH can lead to
the precipitation of metal sulfides. The present investigation
explores the ecology of sulfate-reducing bacteria (SRB) in the
sedimentary system of Lake CDA with a view toward better understanding their contribution to sediment diagenesis.
Three observations suggest that bacterial sulfidogenesis could
play a role in contaminant sequestration. First, compared with
the nearby pristine SJ delta, sediments in the CDA delta are
many-fold enriched in acid volatile sulfides (AVS) commonly
attributed to SRB activity (Harrington et al., 1999). Second,
culture-based estimates indicate the presence of up to 106 SRB
g1 wet weight sediment (Harrington et al., 1998a). Third,
XANES molecular spectroscopy confirms the presence of metal (di)sulfides, such as FeS2 (Toevs et al., 2006).
To understand better the role of bacteria in sulfidogenesis,
we tested the following hypotheses: (i) If sulfidogenesis occurs
in Lake CDA sediments, then SRB should be abundant and
active; (ii) if, owing to their characteristic metabolism, SRB are
intrinsically metal resistant, then SRB abundance should be
independent of metal concentration; and (ii) the rate of SO42
reduction should covary positively with SRB abundance, the
concentration of aqueous-phase SO42, and the amount of
AVS. To test these hypotheses, we estimated the following parameters over an annual cycle: abundance, distribution, and activity of SRB; sediment depth profiles of Eh and pH; aqueous
phase distribution and abundance of SO42; and solid-phase
distribution and abundance of metal(loid)s, sulfur, and AVS.
Material and Methods
Study Site and Sampling Regime
Lake CDA is a natural lake of glacial origin located in the
northern panhandle of Idaho (USA) (Fig. 1). It covers 129.5
km2, has a maximum depth of 61 m, and has an average retention time of 0.48 yr (Woods, 1989). Temperature profiles indicate that Lake CDA is monomictic with Fall overturn. Although
shallow regions can be dimictic and mesotrophic, the rest of
the lake is classified as oligotrophic (Minter, 1971; Woods and
Beckwith, 1996). Samples were collected in Spring 2001, Summer 2001, Fall 2001, and Spring 2002 from three locales: two
highly contaminated sites—Peaceful Point (PP) (N47° 28.287’,
W116° 50.672) and Harlow Point (HP) (N47° 25.835’, W116°
46.603’)—and one uncontaminated site near the SJ delta (N47°
22.855’, W116° 45.0785’). For all 2001 samplings, sediment
cores (25–30 cm in length) were obtained with the assistance of
divers; Spring 2002 cores were obtained using a gravity-coring
device (Mudroch and MacKnight, 1994). Polycarbonate tubing
was used to collect and confine sediments, which were sealed
and stored upright during transport in a 4°C cooler flushed with
N2(g). Once in the laboratory, intact cores were extruded from
bottom to top onto a sterile glass tray and brought into an anaerobic chamber containing an atmosphere of N2:CO2:H2 gas
(85:10:5) for further processing.
© ASA, CSSA, SSSA
+PVSOBMPG&OWJSPONFOUBM2VBMJUZr7PMVNFr9m9
pH and Eh
All measurements were performed in the laboratory under
anaerobic conditions within 24 h of sampling. Pore water pH
was measured for each 5-cm interval. Alkalinity was determined
by Gran-Plot titration to pH 4 (Stumm and Morgan, 1996).
Sediment Eh was determined for each 5-cm interval using an
Orion platinum combination electrode, which was periodically
checked against an iodine standard solution.
Elemental Analyses
Sediment cores were sectioned every 5 cm, and each section was homogenized before subsampling for total metal
analysis. Subsamples were dried at 80°C before being shipped
to ACME Analytical Laboratories Ltd. (Vancouver, British Columbia, Canada) for elemental analysis by inductively coupled
plasma emission spectrometry after aqua regia digest (USEPA
Method 3050B). Analysis of metals in pore water samples was
performed by inductively coupled plasma emission spectrometry using an in-house Thermo Jarrell Ash instrument (Thermo
Fisher Scientific, Waltham, MA). Total carbon and nitrogen
in the solid phase were determined using a Fisons Instruments
CE1110 Elemental Analyzer (Fisons Instruments, San Carlos,
CA) using methods prescribed by the manufacturer.
fide concentration estimated by spectrophotometric assay of
methylene blue at 660 nm (Cline, 1969).
Total Reduced Inorganic Sulfide
Serum vials and sulfide traps were prepared as described
previously. Two to three grams of sediment from each 5-cm
depth interval were added to each apparatus. Serum vials filled
with water instead of sediment served as negative controls, and
serum vials filled with sodium sulfide served as positive controls. Five milliliters of anoxic 12 mol L1 HCl (5 mL) and 10
mL of 1MCr (II) solution were added to each sediment sample
(Fossing and Jorgensen, 1989; Ulrich et al., 1997). Serum vials
were placed in a rotary shaker at 120 rpm and incubated for 7
d at 30°C. Sediment oxidation by Cr-HCl releases H2S, which
in turn is trapped by zinc acetate according to the equation
(CH3COO)2 Zn + H2S (gas) ZnS (trapped) + 2CH3COOH
(Ulrich et al., 1997). After incubation, the trap was removed,
and sulfide concentration was estimated by spectrophotometric
assay using methylene blue (Cline, 1969).
For proofing purposes only
Analysis of Sulfur and Sulfur Species
Total Sulfur in Sediment
Air-dried sediments from each 5-cm interval were sent to
ACME Analytical Laboratories Ltd. (Vancouver, Canada) for analysis using a LECO carbon and sulfur analyzer (Gravel, 1998).
Sulfate
Pore water from every 5-cm section was obtained by centrifuging a known quantity of sediment. Extracted pore water was
stored at 4°C after the addition of 5% zinc acetate to preserve
sulfides present until analysis. Pore water sulfate concentration
was determined by ion chromatography using a Dionex 500
IC (Dionex, Sunnyvale, CA) equipped with an AS4 anion exchange column (Dionex).
Sulfate Reduction Assay
Sulfate reduction potential (SRP) was estimated radiometrically in homogenized, undiluted sediment microcosms by following conversion of 35SO4(aq) to 35S2(g), after the method of
Ulrich et al. (1997). Serum vials with sulfide traps were prepared as described previously with known additions of 35SO4
(5–10 µCi) and incubated for 7 d at 25°C. Each assay was
performed in triplicate. Autoclaved sediment and sediment
amended with 5 mL of 50 mol L1 sodium molybdate served
as negative controls. The fraction of SO4 reduced (fSR) was
calculated by dividing 35S recovered from the sulfide traps by
the total 35SO4 concentrations added (all determined via liquid scintillation counting). Sulfate reduction potential was calculated according to the equation SRP = (fSR [SO4])/(mst),
where [SO4] is the total sulfate concentration, ms is mass of
sediment in microcosm, and t is time.
Estimating Abundance of Sulfate-Reducing Bacteria
Culture-based Estimates Using Most Probable Number Analysis
© ASA, CSSA, SSSA
Acid Volatile Sulfide
The acid volatile sulfide (AVS) extraction apparatus consisted of sealed 160-mL serum vials containing zinc acetate–
filled 12 × 75 mm culture tubes that served as sulfide traps.
Vials and tubes were left overnight inside a controlled, reduced
atmosphere (N2:CO2:H2, 85:10:5) before initiating the experiment. Two to three grams of sediment from each 5-cm depth
interval were added to each serum vial, after which the sulfide
trap containing 4 mL of 10% anoxic zinc acetate ((CH3COO)2
Zn·2H2O) was inserted. Serum vials were capped with butyl
rubber stoppers and crimp sealed. Negative controls consisted
of vials filled with 3 mL deionized H2O instead of sediment,
and positive controls consisted of serum vials filled with 3 mL
of 10 mol L1 sodium sulfide. A total of 10 mL of anoxic 6 mol
L1 HCl was added to all serum vials (Allen and Fu, 1991), and
vials were placed on a rotary shaker at 120 rpm and incubated
for 7 d at 30°C. After incubation, traps were removed and sulRamamoorthy et al.: SRB in Iron-Dominated, Mining-Impacted Sediment
Abundance of cultivable sulfate reducers was estimated by
replicate terminal dilution enrichments. Most probable number (MPN) data were collected on two cores per site for each
sampling event. Samples were processed in an anaerobic chamber (N2:CO2:H2, 85:10:5 ratio), and anaerobic technique was
used during all media preparation and culture manipulations.
The medium was the same as that described by Widdel (1980).
Medium was boiled and cooled under O2–free N2:CO2 (90:10)
and dispensed into degassed Balch tubes in 9-mL aliquots, sealed
with butyl rubber stoppers, and autoclaved. Medium was reduced after autoclaving using 0.1 mL of freshly prepared sodium
dithionate (3 g per 100 ml stock). Electron donor (sodium lactate [10 mol L1]) and electron acceptor (Na2SO4 [10 mol L1])
were added separately from sterile anaerobic 1 mol L1 stocks.
One gram of the homogenized sediment from each 5-cm depth
interval was added to 9 mL of media. After thorough mixing,
0.1 mL of the suspension was added to serum vials in triplicate.
3
Suspensions were taken through seven successive tenfold serial
dilutions so that the final serum vial represented a 108 dilution.
Sediment dilutions were incubated in the dark for 70 d at room
temperature before being scored for blackening indicative of sulfidogenesis. All MPNs were calculated using the techniques and
tables published by APHA (1998).
Non–culture-based Estimates Using Quantitative Polymerase
Chain Reaction
Sediment SRB abundance was also estimated by a quantitative polymerase chain reaction (qPCR) based assay targeting the
gene encoding the -subunit of adenosine 5-phosphosulfate reductase (APS) (EC 1.8.99.2). DNA template for quantification
of -APS copy number was extracted from frozen sediments using the Bio101 DNA extraction kit for soils (Qbiogene, Carlsbad,
CA). DNA extraction efficiency (82 ± 18%) was determined by
estimating the recovery of a known number of APS gene copies
from whole sediment. The primer pair (APS forward: 5-CGC
TGA AAT GAC CAT GA W GG-3; APS reverse: 5-GGG
CCG TAA CCG TCC TTG AA-3) was generated using consensus sequences created from aligned sequences available in GenBank for the APS reductase -subunit and designed to amplify
APS from dissimilatory SRB (see Supplemental Fig. 1). Real-time
qPCR reactions were performed with a Bio-Rad iCycler (BioRad Laboratories, Hercules, CA) using the SYBRGreen I detection method (BioWhittaker Molecular Applications, Rockland,
ME). Each 25-µL PCR reaction contained a 1 concentration
of Promega PCR universal master mix (Promega, Madison WI),
0.15 mol L1 MgCl2, 1:10,000 dilution of SYBRGreen I, and
0.8 mg mL1 bovine serum albumin. To generate standards, APS
from pure cultures of the sulfate reducer Desulfovibrio idahoensis
(DSMZ 15450; Sass et al., in press) was amplified using the same
primers. The amplified product was purified and cloned using
a Promega cloning kit (Kit No. A1380), and the clones restriction digested with Not1 (Promega, Madison, WI). Plasmid copy
number per microliter was determined for linearized clones in the
concentration range of 2.5 × 107 to 2.5 × 102 copies mL1. These
were used to relate threshold cycle number to copy number of
target sequences and to generate standard curves for quantification
in unknown samples. The correlation coefficient for the dilution
series of the qPCR standard was 0.938. The standard thermal profile for amplification was 2 min at 50°C; 10 min at 95°C; and then
30 cycles of 15 sec at 95°C, 1 min at 62°C, 1 min at 72°C, and
1 min at 82°C. Fluorescence was measured in each cycle after the
82°C step to avoid measuring nonspecific dye binding. Error due
to PCR bias was estimated by replicate amplification of the same
DNA and determined to be ±10%. Escherichia coli genomic DNA
served as a negative control for our assays.
contaminant sequestration. The relationship between SRB activity and factors likely to influence this activity (e.g., abundance of
SRB, depth, spatial distribution, pH, Eh, concentration of heavy
metals, and concentration of SO4) was analyzed by ANOVA using a factorial design (4 seasons × 3 sites × 5 depths). Analysis of
variance was also used to test for significant differences in metal
concentration, pH, Eh, and sulfur speciation among sites, seasons,
and depths. We used Spearman’s rank correlation of unaveraged
samples (rs) to test for relationships between SRP, sulfate, and
depth because these data were not normally distributed. Analyses
were performed using the statistical software package SPSS (SPSS
Inc., Chicago, IL). The parameters AVS, SRP (ex situ sulfate reduction potential), and SRB abundance were determined not to
be autocorrelated across seasons or sites. We compared qPCR data
across sites and depths using the Kruskal-Wallis one-way ANOVA
with a Bonferroni corrected multiple comparisons Z-test to determine differences between treatments because these data did not
meet the assumptions of normality. When we compared qPCR
data across sites at specific depths, we were able to use ANOVA
and Tukey’s test for multiple comparisons. We calculated these statistics using NCSS 2000 (NCSS, Kaysville, UT).
For proofing purposes only
Results and Discussion
pH and Eh
Sediment pH and Eh were measured at 5-cm intervals to a
depth of 25 cm below the sediment–water interface. Lake CDA
sediments are characterized by pH and Eh gradients that range
from 7.5 to 6.5 and from 50 to 250 mV, respectively (see
Supplemental Fig. 2). Compact, fine-grained, contaminantladen sediments at PP and HP are much more highly reduced
(100 mV) near the sediment–water interface than the loose
organic matter–rich sediments of SJ. Evidence of bioturbation
is conspicuously absent at PP and HP. These observations are
consistent with previous reports (Harrington et al., 1998a,
1998b; Cummings et al., 2000; Toevs et al., 2006).
Elemental Analyses
© ASA, CSSA, SSSA
Statistical Analysis
Estimates of the potential for biotic sulfidogenesis and the
abundance of acid volatile sulfide (AVS) enable direct and indirect
inferences, respectively, concerning SRB activity. Such inferences
could provide insight into how these organisms influence trace
element fate and stability. Likewise, changes in the distribution of
SRB in relation to season, site, or depth could indicate their role in
Elemental analyses (Fig. 2) confirm that sediments at sites PP
and HP are many-fold enriched in heavy metals and metalloids,
relative to site SJ, as previously observed (Harrington et al., 1998a,
1998b; Cummings et al., 2000). The sediment pool of iron is markedly elevated at all three sites compared with that of trace elements
(Fig. 2). Vertical distribution of iron and trace elements is uniform
at SJ, whereas it is variable and element specific at sites PP and HP.
The abundance of redox-active elements such as iron and arsenic
is complex; for example, at PP in the CDA River delta, we observe
peaks for arsenic and iron near the sediment–water interface as
well as in deeper sediment. By contrast, at contaminated sites the
abundance of redox-inactive elements, such as lead and zinc, tend
to increase monotonically with depth. Lake CDA sediments are
circumneutral and highly reduced, consistently ranging between
50 and 75 mV near the sediment–water interface to between
150 and 250 meV at 25 cm depth. Superimposing the metal
abundances onto sediment Eh profiles invites the interpretation
that iron and arsenic are redox active and undergoing diagenesis,
+PVSOBMPG&OWJSPONFOUBM2VBMJUZr7PMVNFr9m9
For proofing purposes only
Fig. 2. Abundance of Fe, As, Pb, S, and Zn as a function of distance from the sediment–water interface in Lake Coeur d’Alene sediments (mean ± SE).
HP, Harlow Point; SJ, St. Joe; PP, Peaceful Point.
whereas lead and zinc are redox inactive and distributed according
to historical patterns of contaminant deposition (also see Cummings et al., 1999a, 2000). These inferences are partly supported
by statistical analyses confirming that concentrations of all metals
vary significantly across sites (P = 0.01) and that concentrations of
iron, arsenic, manganese, and copper vary significantly with depth
(P = 0.01). With the exceptions of manganese and copper (see
Supplemental Fig. 3), total metal abundance also seems to vary
with each sampling event (P = 0.01), an observation that most
likely reflects site-specific variation.
Together, these observations suggest sediment diagenesis involving transformation of ester-bound sulfates to sulfide.
Abundance of Sulfate-Reducing Bacteria
Total sulfur abundance at contaminated sites PP and HP exceeds that observed at SJ by 2.5- to 6-fold (e.g., 5400 mg kg1
PP, 2700 mg kg1 HP, and 800 mg kg1 SJ; Fig. 2). Like iron
and other metals, sulfur is uniformly distributed in the uncontaminated sediment, an observation explained by regular input
of background concentrations of these elements from SJ sediments. However, in contaminated sediments, sulfur abundance
increases as a function of depth; varies with season, site, and
depth (P = 0.01); and positively covaries with total metal abundance (R2 = 0.969). These results are consistent with our understanding of the mineralogy of the sulfidic ore bodies from which
silver, zinc, and lead were extracted, as well as historical patterns
of contaminant deposition (Horowitz and Elrick, 1993).
Previous data based on sequential selective extractions indicate that regardless of depth, >50% of metal(loid)s in Lake
CDA sediments are associated with an operationally defined
sulfidic phase (Harrington et al., 1998b). These results are contradicted by recent spectroscopic analyses. Toevs et al. (2006)
report that 2 to 2.5% of iron in this environment is pyritic.
Interestingly, 20 to 50% of the total sulfur is pyritic; the proportion of pyritic minerals increases with depth and is inversely
proportional to the abundance of organic ester-bound sulfates.
A limited MPN analysis indicated that the Lake CDA environment supported 104 to 106 cultivable SRB g1 wet weight
sediment (Harrington et al., 1998a). Because SRB are known
to play a role in sediment diagenesis and trace element sequestration, we sought to understand better their abundance, distribution, and activity.
Comprehensive MPN analysis of cultivable SRB across three
sites over an annual cycle reveals that densities of cultivable, lactate-utilizing SRB range from 106 cells g1 dry weight sediment
(Fig. 3a). These values are lower than previous estimates, a finding we attribute to the use of a single electron donor (lactate)
rather than the mixture (of lactate, formate, and acetate) used
by Harrington et al. (1998a). Most probable number–based estimates of lactate-utilizing SRB are unrelated to depth (Fig. 3a)
or to site, season, metal concentration, and pore water sulfate
(ANOVA, not shown), even though pore water sulfate concentrations are generally highest near the sediment–water interface
in Lake CDA sediments, especially at contaminated sites (Fig.
4b). These findings contrast with other aquatic systems where
cultivable SRB abundance is greatest near the sediment–water
interface but decreases with depth as pore water sulfate diminishes (Teske et al., 1998; Sahm et al., 1999). Indigenous lactateutilizing SRB may be evenly distributed because they are able
to use electron acceptors other than sulfate for growth. Indeed,
we have isolated such microbes from Lake CDA sediments and
found that they can use Fe(III) and fumarate as alternative electron acceptors (Ramamoorthy et al., 2006).
We also estimated SRB abundance using non–culture-based
methods. Specifically, we developed and used primers that amplify a 64-bp region of the gene encoding the -subunit of APS
Ramamoorthy et al.: SRB in Iron-Dominated, Mining-Impacted Sediment
5
Total Sulfur
© ASA, CSSA, SSSA
For proofing purposes only
Fig. 3. (a) Most probable number (MPN) estimates of cultivable sulfate-reducing bacteria (SRB) g−1 dry weight sediment in relation to distance from
the sediment–water interface (circles, mean ± SEM between cores). (b) Copy number g−1 dry weight sediment of the eubacterial SRB gene
α-APS reductase (EC 1.8.99.2) estimated by quantitative polymerase chain reaction (triangles, mean ± SE, Fall 2001 only). HP, Harlow Point;
SJ, St. Joe; PP, Peaceful Point.
reductase (EC 1.8.99.2) to perform quantitative PCR analyses of DNA isolated from whole sediment (see Supplemental
Fig. 1). –Adenosine 5-phosphosulfate reductase is a Fe-S
flavoprotein that catalyzes reduction of APS (adenosine-5phosphosulfate) to sulfite and AMP (Friedrich, 2002). Primers
APS-FW and AP-SR amplify a variety of SRB, including members of the families Desulfovibrionaceae and Desulfobacteriaceae, as well as genera Desulfotomaculum and Desulfoporosinus
(see Supplemental Fig. 1). To the extent that -APS is present
in single- or low-copy number, qPCR approximates the total
number of SRB genomes present.
Quantitative PCR using -APS–specific primers was performed on community DNA isolated from sediments collected
in Fall 2001 (Fig. 3b), a time when we observed a sharp increase
in pore water sulfate (Fig. 4b) and sulfate-reduction potential (Fig.
4a) in contaminated sediments. denosine 5-phosphosulfate reductase gene abundance ranged from 4.6  105 to 1.9 
107 copies g1 dry weight sediment, values that are consistently
higher than our culture-based MPN estimates (Fig. 3a). The most
likely explanation for this discrepancy, which has been noted in
other studies (e.g., Teske et al., 1996; Miller, 2002), is that our
media and culture conditions failed to support growth of the most
abundant species of sulfate reducers living in these sediments. The
distribution data obtained by qPCR also differ qualitatively from
data obtained by MPN analysis. When averaged over all depths,
we found -APS copy number to be significantly higher (P <
0.05) at contaminated sites PP and HP than at pristine site SJ (Fig.
3b). At all sites, copy number was significantly higher (P < 0.05)
in surficial (2.5 cm) than in deep (12.5 cm) sediments. We also
compared -APS copy number at sites from the 2.5- and 7.5-cm
depths only. Again, copy number was significantly higher at sites
PP and HP than at site SJ (P < 0.05). We detected no difference in
copy number between sites PP and HP at these depths.
The distribution of SRB estimated by non–culture-based
methods agrees with previous reports that SRB are most abundant near the sediment–water interface where pore water sulfate
concentrations tend to be higher (Teske et al., 1998; Sahm et
al., 1999) (Fig. 4b). –Adenosine 5-phosphosulfate reductase
copy number is an order of magnitude higher at contaminated
sites than at our pristine control, especially in surficial sediments
where pore water sulfate is an order of magnitude higher at contaminated sites HP and PP than at site SJ (Fig. 4b, Fall 2001).
Also noteworthy is our observation that -APS abundance (>107
copies g1 dry weight sediment) is greatest overall in the 0- to
5-cm interval at site PP. Here, mean sediment arsenic, lead, and
zinc values approach 0.3, 4, and 3.5 g kg1, respectively.
Contaminated sediments evidently support appreciable numbers of sulfate reducers (Ramamoorthy et al., 2006; Sass et al., in
press), and we speculate that their characteristic metabolism may
contribute to their success in the Lake CDA environment. There
may also be niche overlap between SRB and dissimilatory metal
reducers. For example, arsenic reducers OREX-4 and BEN-RB
are able to support growth using sulfate as a sole terminal electron acceptor (Macy et al., 2000; Newman et al., 1997a, 1997b).
Furthermore, we may be underestimating the amount of sulfate
available to support such organisms. Certain SRB isolated from
freshwater and marine sediments have been shown to disproportionate sulfur (4So + 4H2O  SO42 + 3HS + 5H+) (Bak,
1993; Finster et al., 1998; Jackson and McInerney, 2000). This
transformation promotes rapid sulfate turnover and results in
© ASA, CSSA, SSSA
+PVSOBMPG&OWJSPONFOUBM2VBMJUZr7PMVNFr9m9
For proofing purposes only
© ASA, CSSA, SSSA
Fig. 4. (a) Ex situ estimate of sulfate reduction potential by addition of 35SO4 to sediment microcosms (circles, mean ± SE); units are ng 35SO4 reduced
g−1 d−1. (b) Distribution of pore water SO4 in mg S L−1 (ppm) as a function of sediment depth (triangles, mean ± SE). HP, Harlow Point; SJ, St.
Joe; PP, Peaceful Point.
SO42 concentrations that are below the detection limit of most
standard techniques (Ulrich et al., 1997).
Although -APS has been viewed as diagnostic of sulfate
reducers, Meyer and Kuever (2007) found dissimilatory APS
reductase gene (aprBA) homologs in sulfur oxidizers (SOX) of
the genera Chlorobia and Thiobacillus. Many SOX, including
the best-known representative, Thiobacillus ferrooxidans, are restricted to aerobic, low-pH environments quite unlike our system. The incidence of aprBA in certain sulfur oxidizers is best
explained by lateral gene transfer, and its distribution seems
to be limited to photo- and chemolithotrophs that are strict
anaerobes or at least facultative anaerobes (Meyer and Kuever
2007). Sulfur oxidizers containing aprAB have been found in
freshwater hot springs and in marine environments as picoplankton and invertebrate symbionts. It is likely therefore that
some -APS copies detected by our qPCR primers are present
in SOX genomes. However, based on the facts that Lake CDA
is an oligotrophic freshwater system and that its sediments are
reduced and circumneutral, contain appreciable acid volatile
sulfide, and at contaminated sites exhibit a fine-grained, gunmetal gray appearance characteristic of mine tailings, we feel
that the majority of -APS copies detected in this system are
Ramamoorthy et al.: SRB in Iron-Dominated, Mining-Impacted Sediment
7
associated with SRB. Further study is needed to test for the
possibilities that sulfate oxidizers and/or sulfur-disproportionating bacteria are present and active in Lake CDA sediments.
Such organisms could enter in syntrophic associations with
SRB, providing an endogenous pool of sulfate.
Sulfate Reduction Potential
We evaluated the potential for microbial sulfate reduction
by amending sediment microcosms with radioactive sulfate
(35SO4). Ex situ estimates indicate that the distribution of sulfate reduction activity strongly and positively correlates with
pore water sulfate concentration (rs = 0.90; P < 0.0001) (Fig.
4b). Because pore water SO4 was markedly higher during
spring and fall, near the sediment water interface, and at contaminated sites, ANOVA revealed highly significant differences
in estimated sulfate reduction rates with respect to season (P
< 0.001), depth (P < 0.001), and site (P < 0.05). The observation of elevated SRP during fall and spring can be attributed to lake turnover and to increased nutrient loading during
spring runoff, while the negative correlation (rs = 0.34; P <
0.001) between depth and SRP is most easily explained by the
diminishing availability of pore water sulfate with depth. Our
observation that SRP was significantly higher at sites PP and
HP is consistent with findings that the abundance of all sulfur
species was significantly higher at these sites relative to SJ (Fig.
5) and that TRIS/total sulfur ratios were up to 5 times higher
at the PP and HP sites than at the pristine SJ site (Fig. 2 and
5). Absolute rate estimates derived from these ex situ experiments should be interpreted with caution because they were
conducted in suspensions (rather than in packed sediments)
and at temperatures (25°C) that exceeded those commonly
found in the lake benthos (8–12°C).
terface (0–5 cm) at all three sites. However, although AVS concentration modestly increased with depth at SJ, maximal AVS
abundance occurred between 5 and 15 cm in contaminated
sediments. This agrees well with the work of other researchers
who found negligible AVS near the sediment–water interface
and an AVS peak at depths of 10 to 15 cm (Herlihy and Mills,
1985). Although depth-specific AVS abundance was seasonally
invariant at the pristine site, it underwent as much as a sevenfold seasonal variation at sites PP and HP. These values greatly
exceed those observed in other freshwater sediments (Smith
and Klug, 1981; Herlihy and Mills, 1985).
Elevated sulfide production at contaminated sites may be
attributed to higher pore water SO42 concentrations. Sulfate
levels ranged up to approximately 90 µmol L1 at PP and HP,
whereas the highest concentration measured at SJ was approximately 20 µmol L1 (Fig. 4b). Also, uncontaminated site SJ is
relatively high in carbon (mean SJ, 3.14% [SD, 0.41]; mean
PP, 2.49% [SD, 0.34]) and nitrogen (mean SJ, 0.25% [SD,
0.04]; mean PP, 0.09% [SD, 0.03]) but low in total sulfur (Fig.
2), sulfate (Fig. 4b), and TRIS (Fig. 5). These data invite the
interpretation that SRB activity in pristine sediments may be
sulfate limited, although in certain low-sulfate hot springs, active sulfur cycling can promote high rates of sulfate reduction
(Dillon et al., 2007).
For proofing purposes only
Acid Volatile Sulfides
Although microcosm studies assess the potential for sulfate
reduction, the abundance of recently deposited acid volatile
sulfide (AVS) can be viewed as indicative of actual S2 biogenesis. Although this parameter is operationally defined, many
researchers find it reasonable to assume that AVS represents
recent sulfate reduction (Kennedy et al., 1998).
Our estimates of AVS varied significantly in relation to season, site, and depth (P = 0.001) (Fig. 5). Acid volatile sulfide
concentrations were generally higher during spring and summer; these observations are attributable to increased in situ sulfate reduction and/or spatial heterogeneity. Other workers have
reported similar changes in in situ sulfate reduction during
summer and have related these findings to increased temperature and increased nutrient loading (Smith and Klug, 1981).
Concentrations of AVS were markedly higher at the contaminated sites (Fig. 5). At SJ, the AVS concentration ranges from
0 to 40 mg sulfide L1 (mean, 13; SD, 12.1), whereas at PP
and HP, AVS concentrations range from 0 to 178 (mean, 52;
SD, 40.3) and 0 to 186 mg sulfide L1 (mean, 71; SD, 57.8),
respectively. Vertical profiles of AVS also differed between contaminated and uncontaminated sites. The level of AVS was
close to or below detection limits near the sediment–water in-
Conclusions
Our study was aimed at testing three hypotheses that address the question of whether SRB provide sulfide equivalents
that help stabilize metal contaminants in a mining-impacted
freshwater ecosystem. We reasoned that if sulfidogenesis occurs in this environment, then SRB should be numerous and
metabolically active. We also reasoned that their characteristic activity should render SRB intrinsically metal resistant and
make their distribution independent of contaminant distribution. Finally, we hypothesized that the rate of SO42 reduction
determined in microcosms would positively covary with SRB,
SO42, and AVS content of whole sediments used to inoculate
these microcosms. Our data clearly support the first hypothesis. Previous studies, in which acridine orange-stained sediments were examined by epiflourescence microscopy, indicated
that total bacterial abundance at contaminated sites ranged
from 105 to 108 cells per gram wet weight sediment, of which 1
to 10% were cultivable SRB (Harrington et al., 1998a, 1998b).
Our qPCR estimates of SRB abundance generally agree with
these data. Support for our second hypothesis is found in
culture-based estimates that indicate that (lactate-utilizing)
SRB abundance correlates poorly with sites and depths and,
therefore, metal concentration. On the other hand, non–culture-based estimates suggest that SRB abundance is an order
of magnitude or higher at contaminated sites, especially near
the sediment–water interface where pore water sulfate concentrations and arsenic concentrations are highest. However,
because SRB abundance is lowest where the sediment content
of other heavy metals is highest, the distribution pattern based
on non–culture-based estimates does not support our second
© ASA, CSSA, SSSA
+PVSOBMPG&OWJSPONFOUBM2VBMJUZr7PMVNFr9m9
For proofing purposes only
Fig. 5. Concentration of acid volatile sulfide (AVS) (closed circles, mean ± SE) and total reduced inorganic sulfide (TRIS) (triangles, mean ± SE) as a
function of distance from the sediment–water interface. HP, Harlow Point; SJ, St. Joe; PP, Peaceful Point.
hypothesis. Finally, ex situ estimates of sediment sulfate reduction potential, viewed in the contexts of AVS and SO4 distribution and qPCR-based estimates of SRB abundance, invite
the interpretation that contaminated CDA sediments support
greater sulfate respiration than sediments in the nearby pristine
SJ delta, supporting our third hypothesis.
The rate of sulfate reduction in lake sediments depends on
several factors other than those investigated in this study. These
factors include supply of organic matter and competition among
microbial guilds for different electron donors and acceptors, including sulfate arising from the metabolic activities of sulfur-oxidizing and sulfur-disproportionating bacteria. Our data reveal
highly significant differences among sites, seasons, and depths in
the abundance and distribution of pore water sulfate and in the
total abundance of sulfur, carbon, iron, and contaminating trace
elements. We lack many of the data needed to completely specify
which factors control the rate of sulfate reduction in these environments and to predict the relative contributions made by biogenic sulfide versus iron and carbonates to contaminating trace
element sequestration. Further investigation is needed because
the geochemistry of Lake CDA sediments is dominated by iron,
and this element is likely to play important roles in trace element
sequestration as a reductant (Fe2+) and as a sorbent [FeO(OH)]
(Shaked et al., 2004). Lake CDA is undergoing rapid development as a regional center for outdoor recreation and serves as the
primary water supply for six municipalities. Given the level of
metal(loid) contamination in these sediments and the sensitivity
of metal(loid) biogeochemistry to fluctuations in redox and pH,
further studies are urgently needed to provide a rational basis for
managing this system.
Acknowledgments
The authors gratefully acknowledge grant support from
the US Geological Survey (99-HQGR-0218) (MM and
RFR), the National Science Foundation (EPS-00-91995),
and the US Dep. of Energy (DE-FG02-00ER63036) (RFR).
The writing of this manuscript was facilitated by NASA
(NNX07AJ28G) (RFR). We also wish to express our thanks to
Leigh Winoweicki for assistance in organizing sampling efforts
and to Dr. Evgeny Kroll for providing analyses of APS probe
specificity. The manuscript was significantly improved by the
critical commentary of Johnnie Moore, Scott Fendorf, David
Nicholas, Kathryn Socie, and two anonymous reviewers.
© ASA, CSSA, SSSA
Ramamoorthy et al.: SRB in Iron-Dominated, Mining-Impacted Sediment
Supplemental Information
Supplemental Figures 1 to 3 are available online at http://
jeq.scijournals.org.
References
Allen, H.E., and G. Fu. 1991. Determination of acid volatile sulfide and
selected simultaneously extractable metals in sediment. Environmental
Protection Agency, Office of Water, Washington, DC.
APHA. 1998. 9221 C. Estimation of bacterial density, p. 9-51–9-52. In L.
S. Clesceri, A.E. Greenberg, and A.D. Eaton (eds.) Standard methods
for the examination of water and wastewater. American Public Health
Association, Washington, DC.
Bak, F. 1993. Fermentation of inorganic sulfur compounds by sulfate-reducing
bacteria. p. 75–78. In R. Guerrero, C. Pedrós-Alió (ed.) Trends Microbial
Ecol. Spanish Society for Microbiology, Barcelona, Spain.
Cline, J.D. 1969. Spectrophotometric determination of hydrogen sulfide in
natural waters. Limnol. Oceanogr. 14:454–458.
Cummings, D.E., F. Caccavo, S.E. Fendorf, and R.F. Rosenzweig. 1999a.
Arsenic mobilization by the dissimilatory Fe(III)-reducing bacterium
9
Shewanella alga BrY. Environ. Sci. Technol. 33:723–729.
Cummings, D.E., F. Caccavo, S. Spring, and R.F. Rosenzweig. 1999b.
Ferribacterium limneticum gen. Nov., sp. Nov., an Fe(III)-reducing
microorganism isolated from mining-impacted freshwater lake sediments.
Arch. Microbiol. 171:183–188.
Cummings, D.E., A.W. March, B. Bostick, S. Spring, F. Caccavo, S.E. Fendorf,
and R.F. Rosenzweig. 2000. Evidence for microbial Fe(III) reduction in
anoxic, mining-impacted lake sediments (Lake Coeur d’Alene, USA).
Appl. Environ. Microbiol. 66:154–162.
Cummings, D.E., O. Snoeyenbos-West, D.T. Newby, A.M. Niggemyer,
D.R. Lovley, L.A. Achenbach, and R.F. Rosenzweig. 2003. Diversity of
Geobacteraceae species inhabiting metal-polluted freshwater lake sediments
ascertained by 16S rDNA analyses. Microb. Ecol. 46:257–269.
Dillon, J.G., S. Fishbain, S.R. Miller, B.M. Bebout, K.S. Habicht, S.M. Webb,
and D.A. Stahl. 2007. High rates of sulfate reduction in a low sulfate
hot spring microbial mat driven by a low diversity of sulfate-respiring
microorganisms. Appl. Environ. Microbiol. 73:5218–5226.
Finster, K., W. Liesack, and B. Thamdrup. 1998. Elemental sulfur and
thiosulfate disproportionation by Desulfocapsa sulfoexigens sp. nov., a
new anaerobic bacterium isolated from marine surface sediment. Appl.
Environ. Microbiol. 64:19–25.
Fossing, H., and B.B. Jorgensen. 1989. Measurement of bacterial sulfate
reduction in sediments: Evaluation of single-step chromium reduction
method. Biogeochemistry 8:205–222.
Friedrich, M.W. 2002. Phylogenetic analysis reveals multiple lateral transfers
of adenosine-5’-phosphosulfate reductase genes among sulfate-reducing
microorganisms. J. Bacteriol. 184:278–289.
Funk, W.H., F.W. Rabe, F. Royston, G. Bailey, P. Bennett, K. Shah, J.C. Sheppard,
N. Savage, S.B. Bauer, A. Bourg, G. Bannon, G. Edwards, D. Anderson, P.
Syms, J. Rothert, and A. Seamster. 1975. An integrated study on the impact
of metallic trace element pollution in the Coeur d’Alene-Spokane Rivers
and lake drainage system, Pullman and Moscow. Joint Project Completion
report, Washington State Univ.– Univ. of Idaho, Moscow, ID.
Galbraith, M.J. 1992. An assessment of mining wastes using the EPA regulatory
leaching tests. Univ. of Idaho, Moscow, ID.
Gravel, J. 1998. Method and specifications for Group2 Total C&S. ACME
Analytical Labs, Vancouver, Briish Columbia, Canada.
Harrington, J.M., S.E. Fendorf, and R.F. Rosenzweig. 1998a. Biotic generation
of arsenic (III) in heavy metal contaminated lake sediments. Environ.
Sci. Technol. 32:2425–2430.
Harrington, J.M., S.E. Fendorf, B.W. Wielinga, and R.F. Rosenzweig. 1999.
Response to comments on “Phase associations and mobilization of iron
and trace metals in sediments of Lake Coeur d’Alene, ID.” Environ. Sci.
Technol. 33:203–204.
Harrington, J.M., M.J. Laforce, W.C. Rember, S.E. Fendorf, and R.F. Rosenzweig.
1998b. Phase associations and mobilization of iron and trace elements in
Coeur d’Alene Lake, Idaho. Environ. Sci. Technol. 32:650–656.
Herlihy, A.T., and A.L. Mills. 1985. Sulfate reduction in freshwater sediments
receiving acid mine drainage. Appl. Environ. Microbiol. 49:179–186.
Horowitz, A.J., and K.A. Elrick. 1993. Effect of mining and related activities
on the sediment trace element geochemistry of Lake Coeur d’Alene,
Idaho, USA. Part I: Surface sediments. Hydrol. Processes 7:403–423.
Jackson, B.E., and M.J. McInerney. 2000. Thiosulfate disproportionation
by Desulfotomaculum thermobenzoicum. Appl. Environ. Microbiol.
66:3650–3653.
Javorka, E. 1991. Lake water quality assessment, Coeur d’Alene Lake, Benewah
and Kootenai counties. Idaho Coeur d’Alene Subagency, Plummer, ID.
Kennedy, L.G., J.W. Everett, K.J. Ware, R. Parsons, and V. Green. 1998. Iron
and sulfur mineral analysis methods for natural attenuation assessments.
Bioremediation J. 2:259–276.
La Force, M.J., S. Fendorf, L. Guangchao, and R.F. Rosenzweig. 1999.
Redistribution of trace elements from contaminated sediments of Lake
Coeur d’Alene during oxygenation. J. Environ. Qual. 28:1195–1200.
Macy, J.M., J.M. Santini, B.V. Pauling, A.H. O’Neill, and L.I. Sly. 2000.
Two new arsenate/sulfate-reducing bacteria: Mechanisms of arsenate
reduction. Arch. Microbiol. 173:49–57.
Meyer, B., and J. Kuever. 2007. Phylogeny of the alpha and beta subunits
of the dissimilatory adenosine-5’-phosphosulfate (APS) reductase from
sulfate-reducing prokaryotes: Origin and evolution of the dissimilatory
sulfate-reduction pathway. Microbiol. 153:2026–2044.
Miller, S.C.M., J.J. LiPuma, and J.L. Parke. 2002. Culture-based and nongrowth-dependent detection of the Burkholderia cepacia complex in soil
environments. Appl. Environ. Microbiol. 68:3750–3758.
Mink, L.L. 1971. Water quality of the Coeur d’Alene river basin. Univ. of
Idaho, Moscow, ID.
Minter, R.F. 1971. Plankton population structure in the lower Coeur d’Alene
river delta and lake. M.S. thesis. Univ. of Idaho, Moscow, ID.
Mudroch, A., and S. MacKnight. 1994. Handbook of techniques for aquatic
sediments sampling, 2nd ed. Lewis Publishers, Boca Raton, FL.
Newman, D.K., T.J. Beveridge, and F.M.M. Morel. 1997a. Precipitation
As2S3 by Desulfotomaculum aripigmentum. Appl. Environ. Microbiol.
63:2022–2028.
Newman, D.K., E.K. Kennedy, J.D. Coates, D. Ahmann, D.J. Ellis, D.R. Lovlely,
and F.M.M. Morel. 1997b. Dissimilatory arsenate and sulfate reduction in
Desulfotomaculum auripigmentum sp. nov. Arch. Microbiol. 63:380–388.
Niggemyer, A., S. Spring, E. Stackebrandt, and R.F. Rosenzweig. 2001.
Isolation and characterization of a novel As(V) reducing bacterium:
Implications for arsenic mobilization and the genus Desulfitobacterium.
Appl. Environ. Microbiol. 67:5568–5580.
Ramamoorthy, S., H. Sass, H. Langner, P. Schumann, R.M. Kroppenstedt, S.
Spring, J. Overmann, and R.F. Rosenzweig. 2006. Desulfosporosinus lacus
sp. nov., a sulfate-reducing bacterium isolated from pristine freshwater
lake sediments. Int. J. Syst. Evol. Microbiol. 56:2729–2736.
Reece, D.E., J.R. Felkey, and C.M. Wai. 1978. Heavy metal pollution in the
sediments of the Coeur d’Alene River, ID. Environ. Geol. 2:289–293.
Sass, H., Ramamoorthy, S., Yarwood, C., Langner, H., Schumann,
P., Kroppenstedt, R.M., Spring, S. and R.F. Rosenzweig. 2009.
Characterization of novel sulfate-reducing bacteria from a metal(loid)contaminated freshwater sediment: Description of Desulfovibrio
idahonensis sp. nov. Int J Syst Evol Microbiol (in press).
Sahm, K., B.J. MacGregor, B.B. Jørgensen, and D.A. Stahl. 1999. Sulphate
reduction and vertical distribution of sulphate-reducing bacteria
quantified by rRNA slot-blot hybridization in a coastal marine sediment.
Environ. Microbiol. 1:65.
Shaked, Y., Y. Erel, and A. Sukenik. 2004. The biogeochemical cycle of iron
and associated elements in Lake Kinneret. Geochim. Cosmochim. Acta
68:1439–1451.
Smith, R.L., and M.J. Klug. 1981. Reduction of sulfur compounds in the sediments
of a eutrophic lake basin. Appl. Environ. Microbiol. 41:1230–1237.
Stumm, W. and J.J. Morgan. 1996. Aquatic chemistry, chemical equilibria and
rates in natural waters, 3rd ed. John Wiley & Sons, Inc., New York.
Teske, A., C. Niels, B. Ramsing, K. Habicht, M. Fukui, J. Küver, B.B.
Jørgensen, and Y. Cohen. 1998. Sulfate-reducing bacteria and their
activities in cyanobacterial mats of Solar Lake (Sinai, Egypt). Appl.
Environ. Microbiol. 4:2943–2951.
Teske, A., C. Wawer, G. Muyzer, and N.B. Ramsing. 1996. Distribution of
sulfate-reducing bacteria in a stratified fjord (Mariager Fjord, Denmark)
as evaluated by most-probable-number counts and denaturing gradient
gel electrophoresis of PCR-amplified ribosomal DNA fragments. Appl.
Environ. Microbiol. 62:1405–1415.
Toevs, G.R., M.J. Morra, M.L. Polizzotto, D.G. Strawn, B.C. Bostick, and S.
Fendorf. 2006. Metal(loid) diagenesis in mine-impacted sediments of
Lake Coeur d’Alene, Idaho. Environ. Sci. Technol. 40:2537–2543.
Ulrich, G.A., L.R. Krumholz, and J.M. Suflita. 1997. A rapid and simple
method for estimating sulfate reduction activity and quantifying
inorganic sulfides. Appl. Environ. Microbiol. 63:1627–1630.
Widdel, F. 1980. Anaerober abbau von fettsauren und benzoesaure durch neu
isolierte arten sulfate-reduzierender bakterian. Ph.D. thesis. Univ. of
Gottingen, Gottingen, Federal Republic of Germany.
Woods, P.F. 1989. Hypolimnetic concentrations of dissolved oxygen, nutrients,
and trace elements in Coeur d’Alene Lake, Idaho. U.S. Geological Survey
Water-Resources Investigations Report 89-4032. USGS, Washington, DC.
Woods, P., and M. Beckwith. 1996. USGS Open-file Report 95-740. USGS,
Washington, DC.
For proofing purposes only
© ASA, CSSA, SSSA
10
+PVSOBMPG&OWJSPONFOUBM2VBMJUZr7PMVNFr9m9