MICROBIAL AND GEOCHEMICAL PROCESSES CONTROLLING THE OXIDATION
AND REDUCTION OF ARSENIC IN SOILS
by
Deanne Christine Masur
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Master of Science
in
Land Rehabilitation
MONTANA STATE UNIVERSITY
Bozeman, Montana
April 2007
© COPYRIGHT
by
Deanne Christine Masur
2007
All Rights Reserved
ii
APPROVAL
of a thesis submitted by
Deanne Christine Masur
This thesis has been read by each member of the thesis committee and has been found to
be satisfactory regarding content, English usage, format, citations, bibliographic style, and
consistency, and is ready for submission to the Division of Graduate Education.
Dr. William P. Inskeep
Approved for the Department of Land Resources and Environmental Science
Dr. Jon M. Wraith
Approved for the Division of Graduate Education
Dr. Carl A. Fox
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the requirements of a master’s degree at
Montana State University - Bozeman, I agree that the Library shall make it available to
borrow under the rules of the Library.
If I have indicated my intention to copyright this thesis by including a copyright notice
page, copying is allowable only for scholarly purposes, consistent with the “fair use” as
prescribed in the U.S. Copyright Law. Requests for permission for extended quotation from
or reproduction of this thesis in whole or in parts may be granted only by the copyright
holder.
Deanne Christine Masur
April 2007
iv
The work herein is dedicated to
my future children.
“Hope deferred makes the heart sick,
but when dreams come true there is life and joy.”
- Proverbs 13:12
v
ACKNOWLEDGEMENTS
I thank foremost my funding sources: the National Science Foundation, the United States
Department of Agriculture-National Research Initiative and the Thermal Biology Institute for
making my work possible and keeping change in my pocket. I also thank Dr. Bill Inskeep,
Dr. Tim McDermott and Dennis Neuman for their support and guidance as committee
members. For their advice, assistance in obtaining data or keeping instrumentation running,
many thanks go to: Galena Ackerman, Kim Ackerson, Julie Armstrong, Mary Bateson, Seth
D’Imperio, Natsuko Hamamura, John Heine, Sarah Korf, Mark Kozubal, Dr. Corrine Lehr,
Dr. Rich Macur, Joe Martin, Erica Miller, Amanda Nagy, Katie Schultz, Peyton Taylor, and
Stacie Zellin.
I thank my wonderful friends and family for their comfort and willing ears throughout
this endeavor, particularly my fiancé, Andrew, who also contributed countless proofreads and
“whip cracks”. Also thanks to my parents, though I am fairly certain they did not fully
understand my thesis topic to the bitter end, they nonetheless always supported my work and
were proud of my every accomplishment. Lastly, I am thankful for the fortitude afforded me
by our heavenly Father to continue the pursuit of this achievement.
vi
TABLE OF CONTENTS
LIST OF TABLES................................................................................................................. viii
LIST OF FIGURES ................................................................................................................. ix
ABSTRACT............................................................................................................................. xi
1. REVIEW OF PERTINENT LITERATURE .........................................................................1
History and Usage................................................................................................................. 1
Human Toxicity .................................................................................................................... 2
Arsenic Chemistry ................................................................................................................ 6
Microbial Interactions........................................................................................................... 8
Arsenite Oxidation ............................................................................................................ 9
Arsenate Reduction......................................................................................................... 10
Multiplicity and Ubiquity of Arsenic Transforming Mechanisms ................................. 14
Soil Arsenic Contamination................................................................................................ 17
Summary and Project Goals ............................................................................................... 20
2. EFFECTS OF ARSENIC PRESSURE ON MICROBIAL DIVERSITY
AND AS-REDOX CAPABILITY..................................................................................23
Introduction......................................................................................................................... 23
Materials and Methods ....................................................................................................... 26
Soil Collection and Chemical Characterization.............................................................. 26
Column Experiments ...................................................................................................... 27
Cultivation and Characterization of Pure-culture Isolates .............................................. 31
Molecular Analysis of Column Soils and Isolate DNA Extracts.................................... 33
Amplification of Arsenite Oxidase Gene Fragments...................................................... 35
Results and Discussion ....................................................................................................... 35
Column Experiments ...................................................................................................... 35
Isolate Cultivation........................................................................................................... 39
Molecular Analysis ......................................................................................................... 45
Arsenite Oxidase Gene Amplification ............................................................................ 48
Conclusions and Implications............................................................................................. 50
3. INHIBITION OF MICROBIAL ARSENATE REDUCTION BY
PHOSPHATE .................................................................................................................53
Introduction......................................................................................................................... 53
Materials and Methods ....................................................................................................... 59
Isolate Selection and Preparation.................................................................................... 59
Liquid Culture Experiments............................................................................................ 60
vii
TABLE OF CONTENTS - CONTINUED
Isolate Confirmation through Full-Length Sequencing .................................................. 61
Amplification of arsC mRNA transcripts....................................................................... 62
Results and Discussion ....................................................................................................... 62
Effects of Phosphate on Microbial Oxidation of Arsenite.............................................. 62
Effects of Phosphate on Microbial Reduction of Arsenate............................................. 63
Effects of Phosphate on Cell Growth in the Presence of High Arsenic.......................... 66
Possible Mechanisms Controlling Phosphate Inhibition of Arsenate Reduction ........... 71
Conclusions and Implications............................................................................................. 76
4. SUMMARY AND CONCLUSIONS .................................................................................79
Summary of Problem.......................................................................................................... 79
Summary of Objectives and Conclusions........................................................................... 80
Implications of the Current Work....................................................................................... 82
LITERATURE CITED ............................................................................................................86
APPENDICES .........................................................................................................................95
APPENDIX A: SUPPLEMENTARY MATERIAL FOR CHAPTER 2.................................96
APPENDIX B: SUPPLEMENTARY MATERIAL FOR CHAPTER 3 ...............................124
APPENDIX C: PRELIMINARY ANALYSIS OF SOIL SAMPLES
COLLECTED FROM ANACONDA-DEER LODGE COUNTY, MONTANA .............158
viii
LIST OF TABLES
Table
Page
1.1.
Various primary gene products of known ars operons ..................................................11
2.1.
Physical and chemical characteristics of Amsterdam soil used in column
experiments. ..................................................................................................................27
2.2. Chemical composition of soil solution media (SSM) used in column
experiments. ..................................................................................................................28
2.3A. Pure cultures of bacteria isolated from columns receiving 2 mg L-1
(26.7 µM) of either arsenite or arsenate. .......................................................................41
2.3B. Pure cultures of bacteria isolated from columns receiving 20 mg L-1
(267 µM) of either arsenite or arsenate. ........................................................................42
2.3C. Pure cultures of bacteria isolated from columns receiving 200 mg L-1
(2.7 mM) of either arsenite or arsenate. ........................................................................43
2.3D. Pure cultures of bacteria isolated from columns withstanding no As
pressure..........................................................................................................................44
2.4.
Potentially dominant microbial populations within column treatments
based on 16S rRNA DGGE banding patterns and co-migrating
cultivars. ........................................................................................................................47
3.1. Concentrations of phosphate and arsenic, and corresponding P:As ratios
used in experiments to examine effects of phosphate on either the
oxidation of arsenite or the reduction of arsenate. ........................................................61
3.2. Calculated kinetic values for three known arsenate-reducing isolates
(Agrobacterium str. 5B, Arthrobacter str. S6 and Bacillus str. S18)
based on Michaelis-Menten modeling. .........................................................................76
ix
LIST OF FIGURES
Figure
Page
1.1. Patient with hyperkeratosis on feet from arsenic exposure in West
Bengal, India. .....................................................................................................................3
1.2. Degree of arsenic contamination in groundwater and where affected
citizens have been identified in districts of West Bengal and Bangladesh. .......................4
1.3. Degree of arsenic contamination in groundwater of the United States. .............................5
1.4. Modes of cell arsenic uptake and extrusion using E. coli as an example. ..........................7
1.5. Microbiological components of the global arsenic cycle. ................................................17
2.1. Polycarbonate column apparatus used in As transformation experiments
showing placement of two ceramic lysimeters (sampling ports). ....................................30
2.2. Arrangement of column experiment. ................................................................................31
2.3. Percent of total soluble arsenic present as arsenite (gray) or arsenate
(black) as a function of time in columns amended with 2 mg L-1 (26.7
µM) of either arsenite or arsenate.....................................................................................36
2.4. Percent of total soluble arsenic present as arsenite (gray) or arsenate
(black) as a function of time in columns amended with 20 mg L-1 (267
µM) of either arsenite or arsenate.....................................................................................38
2.5. Percent of total soluble arsenic present as arsenite (gray) or arsenate
(black) as a function of time in columns amended with 200 mg L-1 (2.7
mM) of either arsenite or arsenate....................................................................................39
2.6. Representative co-migration using DGGE of short fragment 16S rDNA
sequences from two isolates compared to environmental DGGE bands
from the 200 mg L-1 (2.7 mM) arsenate-treated column from which they
were isolated.....................................................................................................................48
2.7. Phylogenetic tree of selected, deduced prokaryotic amino acid sequences
of the large subunit of the aerobic bacterial arsenite oxidase (AroA-like). .....................50
3.1. Percent of arsenite relative to total As plotted as a function of time in
experiments containing arsenite-oxidizing strains (A) Agrobacterium
tumefaciens str. 5A, and (B) Variovorax sp. str. RM1, each subjected to
five phosphate:arsenite ratios. ..........................................................................................64
x
LIST OF FIGURES - CONTINUED
Figure
Page
3.2. Percent of arsenate relative to total As plotted as a function of time in
experiments containing arsenate–reducing strains (A) Agrobacterium
tumefaciens str. 5B, (B) Arthrobacter sp. str. S6 and (C) Bacillus sp. str.
S18, each subjected to five phosphate:arsenate ratios......................................................65
3.3. Oxidation of arsenite by arsenite-oxidizing strains (A) Agrobacterium
tumefaciens str. 5A and (B) Variovorax sp. str. RM1 in the presence of
1000 µM As......................................................................................................................68
3.4. Reduction of arsenate by arsenate-reducing isolates (A) Agrobacterium
tumefaciens str. 5B, (B) Arthrobacter sp. str. S6 and (C) Bacillus sp. str.
S18 at high levels of arsenate (1000 µM) and the effects of elevated
phosphate (1000 µM) on arsenate reduction. ...................................................................69
3.5. Effect of 1000 µM As on cell vitality and ameliorating effects of elevated
phosphate (1000 µM). ......................................................................................................70
3.6. Expression of the arsenate reductase gene, arsC, in Agrobacterium
tumefaciens str. 5B as a function of time (0 – 24 hours) and phosphate
concentration (50 and 1000 µM). .....................................................................................71
xi
ABSTRACT
Arsenic (As) is a common contaminant in soil-water systems, where it exists
predominately as arsenate (AsV) or arsenite (AsIII), the latter of which is considered to be the
more mobile and toxic form. The amount of arsenite or arsenate in natural water systems is
influenced by geochemical conditions and the presence of As transforming microorganisms.
Consequently, the goals of this study were to evaluate the effects of: (i) arsenic concentration
on microbial populations responsible for As oxidation-reduction in a previously
uncontaminated soil, and (ii) phosphate:arsenic ratio on the oxidation or reduction of arsenic.
Laboratory column experiments were conducted to evaluate the influence of soil
arsenic concentration on microbial community composition and to identify microorganisms
and mechanisms responsible for As transformations occurring under aerobic conditions.
Indigenous microorganisms within a previously uncontaminated agricultural soil were
exposed to arsenite or arsenate at three concentrations (2, 20 and 200 mg As L-1) for
approximately 30 days. Near complete biotic oxidation of arsenite (>96%) was observed in 2
and 20 mg As L-1 treatment columns. Results indicated that the net transformation in this
soil was arsenite-oxidation; however, the addition of 200 mg arsenite L-1 inhibited oxidation.
Sixty-two microorganisms were isolated from the columns; however, 43 of these were
arsenate-reducers, and only 1 organism was capable of arsenite oxidation. Results of this
study suggest that As perturbation of a previously uncontaminated soil does not significantly
decrease microbial diversity and that cultivation techniques may be biased toward arsenatereducing microorganisms.
Phosphate is a chemical analog to arsenate and may inhibit microbial uptake of
arsenate, thus preventing its reduction to arsenite. Five selected microorganisms isolated
from As-treated soil columns or from As-impacted soils near Anaconda, MT were used to
evaluate the effects of phosphate on arsenate-reduction and arsenite-oxidation. Cultures were
initially spiked with various P:As ratios, incubated for approximately 48 hours, and analyzed
periodically for arsenate and arsenite. Arsenate reduction was inhibited at high P:As ratios,
but only at elevated levels of phosphate (500 and 1000 µM). This work supports that land
application of phosphate could minimize microbiological reduction of arsenate to arsenite,
thus reducing As bioavailability.
1
1. REVIEW OF PERTINENT LITERATURE
Arsenic (As) is a naturally occurring and ubiquitous element in soils, sediments and
natural waters, ultimately originating from igneous rock (Cullen and Reimer, 1989). Arsenic
is the twentieth most abundant element in the earth’s crust with an average concentration of
2-3 mg kg-1 (Francesconi and Kuehnelt, 2002).
Typical background concentrations of
arsenic in surface soils vary worldwide (0.1 – 1000 mg kg-1), and concentrations exceeding
100 mg kg-1 are generally associated with anthropogenic inputs from a variety of possible
sources including irrigation with high-As containing surface or ground water, combustion of
coal, mining and smelting processes, and the application of arsenic-based pesticides
(Cervantes et al., 1994; Mukhopadhyay et al., 2002; Belluck et al., 2003).
History and Usage
The use of arsenic in art, medicine, and chemistry has been documented since as early as
the Bronze Age (Nriagu, 2002). The brilliant white, yellow and orange colors of various
arsenic minerals were often used as pigments in paints, and its mica-like luster made arsenic
mixtures desirable for coating mirrors and statues and producing gold-colored paintings and
texts. Arsenic was adopted in cultural medicine as early as 1550 B.C. when it was used to
treat maladies ranging from sore throat to leprosy. Its use continued to grow in scope
through its peak in the nineteenth century when Nriagu (2002) documents that in Western
cultures “every major disease known was being subjected to arsenotherapy.” By this era,
arsenic also became prevalent as a pesticide in agriculture and livestock operations and its
2
use as a wood preservative became common a few decades later. Even today, arsenic is in
use globally, including the United States, in medicine, agriculture and commercial enterprise.
Medicinally, an arsenic trioxide derivative called Trisenox is an intravenous therapeutic
currently recommended for treating some forms of cancer (Silver et al., 2002).
In
agriculture, organoarsenicals (primarily roxarsone) remain a common feed additive for
broilers to enhance chicken growth and to moderate common diseases such as coccidiosis
(Silver et al., 2002; Oremland and Stolz, 2003; Stolz et al., 2007). In industry, chromated
copper arsenic (CCA) is currently a primary wood preservative in the U.S., and a 2003
estimate suggested it could be found in 90% of all outdoor wooden structures (Nriagu, 2002;
Belluck et al., 2003). However, according to the U.S. EPA (2006b), CCA treated lumber has
been phased out for residential use since 2004.
Human Toxicity
Since antiquity, arsenic has also been a known lethal toxin; the word itself is a derivative
of the Greek term arsenikon, which literally means “potent” (Nriagu, 2002). According to
the U.S. EPA (2006a), symptoms of chronic arsenic poisoning range from debilitating
(nausea, diarrhea, skin discoloration, wart-like skin lesions [Figure 1.1], partial paralysis and
loss of vision) to fatal (cancer of the bladder, lungs, skin, kidney, nasal passages, liver, and
prostate). Conversely, arsenic has been suggested to have “an essential or beneficial function
in ultra trace amounts,” that is between 12 and 50 µg kg-1, in some animal species, namely
rats, chickens, and goats (NRC, 2005). However, arsenic has been used for centuries as a
human poison; in fact, before 1836 when adequate low-level arsenic detection methods were
3
available, arsenic trioxide acquired the nickname “inheritance powder” - a reference to the
extent to which it was used homicidally by greedy heirs (Nriagu, 2002; Oremland and Stolz,
2003). More recently, environmental health issues related to arsenic have been associated
primarily with soil and water contamination.
Figure 1.1. Patient with hyperkeratosis on feet from arsenic exposure in West Bengal,
India. (Reproduced from Chowdhury et al., 2000.)
In the last decade, arsenic has gained heightened global attention as a result of high
concentrations in drinking water supplies of many countries including Japan, China, India,
Bangladesh, Thailand, Mongolia, Taiwan, Argentina, Chile, Mexico and the United States
(Fazal et al., 2001; Nordstrom, 2002). The most extensive problems have occurred in
Bangladesh and West Bengal (Figure 1.2). Conservative estimates indicate that 21 million
citizens have been exposed to concentrations above Bangladesh’s current As standard (0.05
mg L-1), although the number of exposed may be closer to 40 million (Fazal et al., 2001).
This “arsenic epidemic” began in the 1970s when the predominant source of drinking water
for rural Bangladesh switched from surface ponds to groundwater tubewells in an attempt to
4
mitigate high infant mortality rates caused by water-borne diseases such as cholera.
However, in 1993 the Department of Public Heath Engineering discovered high
concentrations of As in the well-drawn water, and by the time the discovery was made,
thousands of citizens were already suffering from arsenic-related illnesses (World Bank,
1999; Fazal et al., 2001). Although it is widely accepted that the source of arsenic is from
naturally occurring sedimentary deposits, significant research has explored the chemical and
hydrologic processes controlling As fate and transport in these systems, and numerous relief
efforts have focused on low-cost technologies for removal of As from family and community
drinking water sources (World Bank, 1999; Hoque et al., 2000; Fazal et al., 2001).
West
Bengal
Bangladesh
Figure 1.2. Degree of arsenic contamination in groundwater and where affected citizens
have been identified in districts of West Bengal and Bangladesh. (Modified from
Chowdhury et al., 2000.)
5
The visibility of the As water quality crisis in southeast Asia stimulated renewed calls by
the World Health Organization (WHO) for the U.S. to lower its drinking water standard and
in January 2001 the United States officially lowered the standard for arsenic from 50 µg L-1
to 10 µg L-1 (Smith et al., 2002). The new drinking water criterion took effect in February
2006 (US EPA, 2006a) and has serious financial implications for many industries, including
agricultural and mining enterprises and municipalities responsible for water treatment as U.S.
groundwater arsenic concentrations often exceed 10 µg L-1 (Figure 1.3) (Ryker, 2001). The
potential detriment to environmental and human health and the financial implications of
arsenic contamination provide impetus for understanding factors controlling the speciation
and transport of arsenic in soil and natural waters.
Figure 1.3. Degree of arsenic contamination in groundwater of the United States.
(Reproduced from Ryker, 2001.)
6
Arsenic Chemistry
Arsenic is a group V metalloid that naturally exists in approximately 150 minerals, is
generally associated with mineralized deposits high in sulfide, and as a minor component in
other metal sulfides of Cu, Au, Ag, Zn, Sb and Fe. The three predominant arsenic sulfides
are realgar (arsenic disulfide, As2S2), orpiment (arsenic tri-sulfide, As2S3) and arsenopyrite
(ferrous arsenic sulfide, FeAsS) (Fazal et al., 2001; Oremland and Stolz, 2003). Arsenic is
observed predominantly in four oxidation states: As-III (arsine), As0 (elemental arsenic), AsIII
(arsenite), and AsV (arsenate). The latter two are the most commonly observed oxidation
states in soil environments. Of these, arsenate is predicted to be the most common under
aerobic conditions, whereas arsenite predominates in anaerobic environments (Sadiq, 1997;
Oremland and Stolz, 2003). Most soil environments typically have pH values between 5.5
and 8 (Winegardner, 1995). This range coupled with known pKa values of various arsenic
species suggest that, in soil environments, H3AsO30 (pKa = 9.2) is the primary arsenite
species observed, whereas H2AsO4- (pKa = 7.0) and HAsO4-2 (pKa = 11.5) are both present
as predominant arsenate species (Cullen and Reimer, 1989). Arsenate and arsenite have
extremely different chemical properties and exhibit different toxicological and environmental
properties. Consequently, chemical speciation can play an important role controlling the fate
and transport of As in soil and water environments.
The inorganic chemistry of arsenate is analogous to phosphate, thus microorganisms have
difficulty distinguishing the toxin from the structurally-related essential nutrient.
Consequently, arsenate is generally taken up by cells via phosphate transporters within the
cell membrane (Figure 1.4) (Mukhopadhyay et al., 2002; Oremland and Stolz, 2003).
7
Arsenate can prove lethal when substituted for phosphate in metabolic processes such as
oxidative phosphorylation, the primary method of energy generation for most
microorganisms (Oremland and Stolz, 2003). In soil environments, arsenate generally sorbs
more strongly and to a greater variety of minerals making it less mobile and less bioavailable
than arsenite.
Figure 1.4. Modes of cell arsenic uptake and extrusion using E. coli as an example. Pit
and the three-component PstB, PstC, and PhoS are phosphate transporters capable of
arsenate uptake. GlpF is a glycerol transport protein capable of arsenite transport. ArsC
(cytoplasmic arsenate reductase) and ArsAB (membrane-associated arsenite efflux
proteins) are employed in arsenic detoxification (reproduced from Mukhopadhyay et al.,
2002).
Arsenite is considerably more toxic to microorganisms than arsenate (Tamaki and
Frankenberger, 1992; Qin et al., 2006). One study indicates that arsenite is also five times
more likely to cause chromosome damage to human cells than arsenate, and exposure to
arsenite reduced human cell growth by a factor of approximately ten over the pentavalent
form (Nakamuro and Sayato, 1981). Arsenite is uncharged at pH values below 9.2 and is
able to enter cells through universal aqua-glyceroporins embedded in the cell membrane
8
(Rosen, 2002) (Figure 1.4). The mechanism of arsenite toxicity involves binding with the
sulfhydryl groups of many cell proteins, often incapacitating their function (Mukhopadhyay
et al., 2002; Oremland and Stolz, 2003).
Microbial Interactions
Early Earth (the first 50 million years of existence) was thought to be such an intensely
volcanic environment that outgassing is thought to have caused the formation of the
atmosphere an estimated 200 million years after Earth’s creation (Allègre et al., 1986).
Volcanic gaseous emissions are known to contain noteworthy concentrations of arsenic
(Stoiber, 1995); consequently, this trace element was likely an important component of
Earth’s early environment. In the reducing atmosphere of prehistoric Earth, arsenite would
have likely been the predominant form of arsenic, thus arsenite pumps are thought to have
developed early in the evolution of microorganisms as a detoxification mechanism
(Mukhopadhyay et al., 2002). Later in the evolutionary history of prokaryotes, the first
cyanobacteria are thought to have introduced oxygen into the atmosphere approximately 2.7
billion years ago (Kasting and Catling, 2003), allowing for possible abiotic oxidation of
arsenite to arsenate. In response, microorganisms are thought to have adapted arsenate
reductases (Silver et al., 2002; Mukhopadhyay et al., 2002) as a mechanism for reducing
arsenate to arsenite, which could then be extruded by arsenite efflux pumps.
The concept of a bacterial arsenic resistance mechanism was first proposed based on
work regarding plasmid-encoded Staphylococcus aureus arsenic resistance (Novick and
Roth, 1968).
Subsequent research has shown that numerous phylogenetically distinct
9
microorganisms possess an arsenic resistance mechanism encoded either chromosomally or
in a plasmid. Currently, the best understood mechanisms of microbial As transformation
include detoxification (one for AsIII oxidation and one for AsV reduction), chemolithotrophic
AsIII oxidation, and dissimilatory AsV reduction.
Arsenite Oxidation
The first concrete evidence of bacterial arsenite oxidation dates to 1918 when the first
arsenite-oxidizing bacterium, Bacillus arsenoxidans, was isolated from a cattle-dipping
solution in South Africa (Green, 1918). Since that discovery, at least 22 additional genera
(including members of Bacteria and Archaea) have been shown to oxidize arsenite (Ehrlich,
2002; Oremland and Stolz, 2003; Macur et al., 2004; Inskeep et al., 2007). Although the
oxidation of arsenite with oxygen as an electron acceptor is highly exergonic (Rxn 1.1;
Santini et al., 2000; Inskeep et al., 2005), there have only been a few microorganisms
characterized that are capable of growth using arsenite as a sole electron donor (Santini et al.,
2000; Ehrlich, 2002; Oremland et al., 2002; Santini and vanden Hoven, 2004).
2H3AsO3 + O2 Æ HAsO42- + H2AsO4- + 3H+ (∆G0 = -256 kJ mol-1)
(Rxn 1.1)
Though the majority of arsenite-oxidizing microorganisms do not appear to gain energy
from arsenite oxidation, evidence has been obtained for several heterotrophic organisms that
arsenite oxidases play a role in increased As resistance via detoxification (Muller et al., 2003;
Macur, et al., 2004). All currently known aerobic arsenite oxidases appear to share a similar
protein structure containing two sub-units: a Mo-pterin protein and a Rieske Fe-S protein
(Ellis et al., 2001; Muller et al., 2003; Santini and vanden Hoven, 2004; Inskeep et al., 2007).
10
The only arsenite oxidase crystal structure characterized to date is from the heterotrophic
organism, Alcaligenes faecalis (Anderson et al., 1992; Ellis et al., 2001). This protein is
located in the periplasm between the inner and outer membranes and is composed of one
large (825 amino acid) molybdopterin (Mo-pterin) subunit consisting of a single Mo held
between two pterin cofactors and an iron-sulfur (3Fe-4S) cluster and one small (134 amino
acid) Rieske-like subunit also composed of an iron-sulfur (2Fe-2S) cluster (Ellis et al., 2001).
The oxidation reaction is speculated to occur as arsenite binds to the Mo center of the large
subunit, where arsenite transfers its two lone electrons to MoVI, temporarily forming MoIV
before the electrons are passed through the two iron-sulfur clusters to the periplasmic
proteins cytochrome c and azurin, which act as terminal electron acceptors. In the process,
the arsenate produced dissociates from the Mo group; although arsenic is still present in the
periplasm, it is in a less detrimental form.
Arsenate Reduction
Microbial detoxification of arsenate is mediated by arsenate reductases (ArsC) that are
encoded by arsC of the ars operon.
These arsenate reductases are categorically and
structurally different than arsenate reductases important in mediating the dissimilatory
reduction of arsenate in anaerobic respiration (see below), and act to reduce arsenate to
arsenite in the cytoplasm, which is then extruded from the cell via an arsenite efflux pump
(ArsB). The arrangement of the ars operon varies among microorganisms, but can contain
up to five genes, which encode for various protein function (Table 1.1). Generally, the
operon begins with arsR, which encodes for a transcriptional regulator, followed by arsB, an
arsenite efflux pump encoder, and concludes with the arsenate reductase gene, arsC (Silver et
11
al., 2002). All arsR regulator genes known to date appear closely related and can be induced
by arsenate, arsenite, antimonite, and bismuth. Several families of arsB genes exist though
all appear functionally similar. They encode a membrane efflux pump which extrudes both
arsenite and antimonite from the cell and is driven by membrane potential. Two main
families of arsC genes have been characterized, though both encode for a small cytoplasmic
protein that reduces arsenate to arsenite. The primary difference between the families lies in
their electron source. One family, exemplified by the well studied E. coli plasmid R773, uses
both reduced glutathione and glutaredoxin as electron donors while the other family, typified
by the S. aureus plasmid pI258, uses thioredoxin as an electron donor (Silver et al., 2002;
Mukhopadhyay et al., 2002). In both cases, ATP is required for reducing arsenate via this
mechanism.
Table 1.1. Various primary gene products of known ars operons (Mukhopadhyay et al.,
2002; Silver et al., 2002; Oremland and Stoltz, 2003).
Protein
Function
ArsA
soluble ATPase subunit of the AsIII efflux pump; interacts with ArsB to convert
to an ATP driven efflux pump
ArsB
membrane AsIII efflux pump; removes AsIII from cytoplasm
ArsC
arsenate reductase; mediates cytoplasmic reduction of AsV to AsIII
ArsD
small secondary transcriptional regulator; proposed to regulate arsB expression
ArsH
undetermined
ArsR
transcriptional regulator
In addition to these essential ars genes, two others (arsD and arsA) have been found most
often inserted between the arsR and arsB genes.
The arsD gene encodes for a small
12
secondary transcriptional regulator protein and has not been studied extensively. In fact,
only six arsD genes have been discovered, though it is suggested they all function to repress
the overexpression of ArsB, which could be toxic to the cell in large quantities (Silver et al.,
2002). Interestingly, all arsD genes have been found in combination with arsA, which
encodes a protein that complexes with the ArsB efflux pump and converts it to an ATPhydrolysis driven ArsAB complex pump for more efficient arsenite extrusion from the cell
(Mukhopadhyay et al., 2002; Silver et al., 2002). The presence of arsAD appears to be
exclusive to plasmids of gram-negative bacteria. Further gene additions to the ars operon of
specific organisms have also been documented, though the function of these genes are yet
undetermined.
For instance, arsH is part of both the plasmid-encoded Yersinia and
chromosome-encoded Acidothiobacillus ferrooxidans ars operon and orf2 is a component of
the Bacillus subtilis chromosomal operon (Mukhopadhyay et al., 2002; Silver et al., 2002).
While the three-gene arsRBC and five-gene arsRDABC arrangements are the bestcharacterized ars operons, variation is common, including shuffled gene arrangement and
even separate transcription. Furthermore, in Mycobacterium tuberculosis, the arsB and arsC
genes are fused and the existence of two separate sequence-unrelated arsC genes has been
discovered in Pseudomonas aeruginosa (Mukhopadhyay et al., 2002). The gene-structure of
the ars operon may also differ within a single microorganism based on the loci of the operon.
For example, the E. coli chromosomal ars operon includes arsB, arsC and arsR genes while
its plasmid, R733, contains these three plus arsD and arsR (Oremland and Stolz, 2003).
In
short, while the ars operon is diverse in structure, its function across disparate phyla is
largely conserved.
13
The other predominant mechanism responsible for the reduction of arsenate to arsenite is
dissimilatory reduction. This mechanism involves a membrane-bound arsenate reductase
that uses arsenate as a terminal electron acceptor coupled to an electron donor to support
anaerobic respiration (Saltikov and Newman, 2003). It appears that all known arsenaterespirers are facultative arsenate-reducers, capable of using one or more alternate electron
acceptors, with the most common alternative being nitrate (Stolz and Oremland, 1999; Macy
and Santini, 2002). The vast majority of arsenate-respirers have been isolated using lactate
as the electron donor (Macy and Santini, 2002; Oremland and Stolz, 2003). Dissimilatory
reduction is mediated by a two-gene (arrA and arrB) cluster (Saltikov and Newman, 2003),
and the structure of the respiratory arsenate reductase (Arr) appears to be highly conserved
across disparate phyla, as suggested by the similarity of three characterized reductases in
Chrysiogenes arsenatis (Macy and Santini, 2002), Bacillus selenitireducens (Afkar et al.,
2003) and Shewanella sp. str. ANA-3 (Saltikov and Newman, 2003). Dissimilatory arsenate
reductases appear to be membrane-associated and consist of a large (ArrA) molybdenum
containing subunit and a small (ArrB) subunit composed of several iron-sulfur clusters. The
large Mo-pterin sub-unit (ArrA) is also a member of the DMSO reductase family and is
related to the Mo-pterin sub-units of aerobic arsenite oxidases (AsoA/AroA/AoxB) (Saltikov
and Newman, 2003; Inskeep et al., 2007). The mechanism by which this reductase reduces
arsenate while harvesting energy is still not clearly understood nor is the physiological
function of the two reductase subunits clear, though both were shown to be required for
arsenate respiration (Saltikov and Newman, 2003).
14
Microbial arsenate reduction via detoxification may play a greater role than dissimilatory
reduction in transforming arsenate to arsenite in phylogenetically diverse soil environments
(Inskeep et al., 2002).
The reasoning behind this hypothesis is that soil arsenate
concentrations are generally insufficient to support significant growth of microbial
populations utilizing arsenate as the dominant electron acceptor. Moreover, the dissimilatory
arsenate reducing organisms isolated to date are strict anaerobes or microaerophiles
(Oremland et al., 1994; Macy et al., 1996; Blum et al., 1998); consequently, this mechanism
is only expected to be important when levels of oxygen are depleted. Given that arsC genes
appear
widely
distributed
throughout
the
prokaryotic
domains,
numerous
soil
microorganisms are likely capable of arsenate reduction via this mechanism. Indeed, the
reduction of arsenate to arsenite has often been documented in highly oxic environments and
in fully aerobic pure cultures (Jones et al., 2000; Macur et al., 2001) where dissimilatory
reduction is not occurring.
Multiplicity and Ubiquity of Arsenic Transforming Mechanisms
Several microorganisms have been shown to be capable of transforming arsenic via
several of the aforementioned possible mechanisms depending on the environmental
conditions. For instance, Thermus sp. str. HR 13 has the ability to oxidize arsenite via
presumed detoxification in aerobic conditions, while generating energy by reducing arsenate
as its electron acceptor in anaerobic environments (Oremland and Stolz, 2003). Another
study (Macur et al., 2004) identified two Agrobacterium tumefaciens-like isolates with an
identical 16S rDNA sequence (across 1400 bp); however, under aerobic conditions one
isolate exhibited arsenite oxidation (str. 5A) and the other isolate was capable only of
15
arsenate reduction under aerobic conditions (str. 5B). Interestingly, a putative arsC gene was
identified in both strains; however, isolate 5A was not capable of arsenate reduction (Macur
et al., 2004). The oxidizing strain was also shown to possess an arsenite oxidase operon
(aoxSRAB); transposon mutagenesis not only resulted in the loss of arsenite-oxidizing
capability, but also revealed that isolate 5A now had the ability to reduce arsenate (Kashyap
et al., 2006).
Therefore, it is hypothesized that A. tumefaciens str. 5A is capable of
simultaneous arsenite oxidation and arsenate reduction, but the oxidation rate exceeds that of
reduction yielding a net arsenite oxidizing phenotype (Kashyap et al., 2006).
Previous research further supports that microorganisms with the ability to oxidize
arsenite and/or reduce arsenate coexist within localized areas and are ubiquitous across
myriad soil environments (Jackson and Dugas, 2003; Macur et al., 2004; Inskeep et al.,
2007). Arsenic resistance mechanisms similar to those found in bacteria also appear to be
present in yeast, plants and animals (Silver et al., 2002). Currently, it is unknown what role
detoxification plays versus energy generation in controlling the speciation of As in natural
environments, or what soil conditions may shift the net microbial transformation towards
AsIII or AsV. Certainly, anaerobiosis will generally result in the production of arsenite, and
microbial oxidation of arsenite to arsenate is likely in aerobic environments (Oremland and
Stoltz, 2003). Yet, there has not been sufficient progress in assessing relative contributions of
detoxification versus energy generation pathways in controlling arsenic speciation in natural
systems.
The methylation of arsenic is also an important component of the global As cycle. Like
ars operons, the genes responsible for methylation, arsM, are often located downstream of an
16
arsR transcriptional regulator gene, suggesting the evolutionary purpose of arsM genes was
arsenic resistance. This has been disputed as the gene’s protein product, ArsM, catalyzes the
transfer of methyl groups from S-adenosylmethionine to AsIII to form monomethylarsenite
(MMAIII), dimethylarsenite (DMAIII) and trimethylarsine (TMAIII), which are considerably
more toxic than inorganic As forms. However, it appears that MMAIII and DMAIII are nonaccumulating “transient intermediates” (Qin et al., 2006), and that TMAIII is the end product
of methylation, which volatilizes to remove arsenic from the cell (i.e. As resistance). Over
140 methylase (ArsM) homologues have been identified to date in bacteria and archaea (Qin
et al., 2006).
The global arsenic cycle is not only defined by the aforementioned microbial
transformations (Figure 1.5), but also by abiotic processes such as mineral precipitation and
dissolution (Inskeep et al., 2002; Mukhopadhyay et al., 2002). However, certain sets of soil
conditions may result in microbial processes that favor arsenate, the less bioavailable and
mobile form. Information on factors controlling microbial transformations of arsenic in soilwater systems will be useful in meeting compliance with current arsenic health standards,
and for achieving bioremediation goals for contaminated lands, such as abandoned mines.
17
Figure 1.5. Microbiological components of the global arsenic cycle. (Reproduced
from Cervantes et al., 1994.)
Soil Arsenic Contamination
Arsenic is a leading contaminant in many soil environments worldwide with background
soil levels typically varying from < 10 to 40 mg kg-1 (Belluck et al., 2003). Specific rock
types naturally vary in their average As content.
For example, the mean arsenic
concentration in shale, igneous rock and sandstone is 13, 1.8, and 1 mg kg-1, respectively,
while some sedimentary deposits, such as coal, have been found to contain 2900 mg As kg-1
(Tamaki and Frankenberger, 1992). The origin of arsenic in a specific soil is not limited to
the inherent rock material, but can also emanate from adjacent water sources. Specifically,
18
As-rich geothermal water originating from Yellowstone National Park discharges to the
Madison River (Madison County, Montana) yielding river water with an As concentration
range of 1-4 µM. High concentrations of As in the Upper Madison River Basin may impact
biota throughout the watershed. For example, it has been shown that elk herds concentrating
in the upper Madison-Firehole River geothermal basins have elevated levels of As in tissues
compared to elk residing in sites with low background As concentrations (Kocar et al., 2003).
Contamination problems from natural arsenic sources can be further exacerbated by
anthropogenic activities. For example, As-rich Madison River water is used as a primary
source of irrigation for surrounding lands, thus widening the distribution of arsenic.
Regional water and soil quality issues have developed as a result of these irrigation practices
and several residential wells in the Madison River aquifer have revealed concentrations of As
5 to 10 times the current drinking water standard (10 µg L-1); the groundwater concentrations
in this region however, may be more affected by long-term natural processes originating
ultimately from the geothermal sources in Yellowstone (Jones et al., 1999).
Another regionally important example of anthropogenic contamination is the aerial
deposition of trace elements from smelting operations, including Cu, Zn, As, Pb and others.
For example, smelters in Anaconda, Montana dealt in the milling, smelting and refining of
primarily copper ore and were in operation from 1884 to 1980 (nearly one hundred years).
One Anaconda smelting operation, added to the National Priorities List (NPL) in 1983, has
left a legacy of soil and water Cu, As and Zn contamination that remains a significant
environmental health issue decades after operations have ceased.
Soil and house dust
samples from residences in nearby Anaconda have been shown to contain 410 mg kg-1 and
19
170 mg kg-1 As, respectively (Freeman et al., 1995). The extent of aerial deposition was
farther reaching than the adjacent town of Anaconda, and the now vacated smelter is
surrounded by approximately 6,000 acres affected by solid mine waste, 13,000 acres
contaminated by aerial emissions, 4,800 acres of alluvial ground water with elevated metal
concentrations, and 28,600 acres of bedrock ground water that exceeds the state standards for
arsenic (Jones et al., 1997; US EPA, 1998). Similar situations can be described throughout
the northwestern U.S. including former smelting turned Superfund sites in Kellogg and
Smelterville, ID, East Helena, MT and Sandy, UT to name a few (US EPA 2006c). Other
noteworthy examples of large scale environmental As contamination from mining can be
found across the U.S. and abroad.
The prevalence and cost effectiveness of smelting by-products containing high arsenic
fostered its use as a pesticide. The application of arsenical pesticides is another primary
source of anthropogenic arsenic contamination. Prominent commercial arsenical pesticide
usage began in the U.S. with Paris green (copper arsenate) in the late 1860s, followed by
London purple (a calcium arsenate, arsenite, and organic matter mixture) by 1872, lead
arsenate in 1892 and calcium arsenate in 1906. The next major commercial addition to the
arsenical suite came in the 1940s with the introduction of synthetic organoarsenicals such as
monosodium methanoarsonate (MSMA), dimethylarsenic acid (DMSA) and arsonic acid.
Though these compounds are less toxic than arsenate salts, they eventually are transformed
into the more toxic, inorganic forms. As such, their immense usage in the U.S. through the
1980s has contributed to elevated levels of arsenic in numerous soils (Nriagu, 2002).
20
The various sources of arsenic contamination have resulted in many sites where arsenic
concentrations exceed 100 mg kg-1. In fact, Freeman et al. (1995) estimates that “in the
United States alone, 100,000 to 1,000,000 hectares of current and former agricultural land
contain soil As concentrations of 200 mg kg-1or more while tens of millions of hectares
contain arsenic residues in the range of 20 to 30 mg kg-1.”
In some instances soil
concentrations exceed 1000 mg kg-1. For example, arsenical use has resulted in soil arsenic
levels upwards of 2,550 mg kg-1 as measured in an orchard soil in Washington State and in
Tacoma, Washington arsenic levels of approximately 3000 mg kg-1were reported in soils of
property adjacent to a post-operational smelter (Belluck et al., 2003).
Generally accepted methods of remediating extremely disturbed lands can be complicated
by the environmental chemistry and microbiology of arsenic. Specifically, the addition of
liming agents, generally in the form of CaCO3, Ca(OH)2 or CaO, or phosphate is common in
acidic mine tailing environments to increase soil pH, which is conducive to both revegetation
and immobilization of trace metals such as Cu, Zn, Pb and Cd. However, the pH dependence
of arsenate sorption reactions is reversed relative to metal ions, and increases in pH above 8
can result in an increase in the mobility of arsenic (Jones et al., 1997; Macur et al., 2001).
Therefore, it is important to understand how major environmental parameters such as pH or
phosphate concentration influence the behavior and microbial transformations of arsenic.
Summary and Project Goals
The extent of arsenic contamination in landscapes across the globe provides significant
incentive for understanding factors controlling the fate and transport of this toxic trace
21
element.
The need to remediate widespread arsenic soil contamination resulting from
various anthropogenic sources, treatment of arsenic in the groundwater of countries such as
Bangladesh and compliance with the lowered U.S. arsenic drinking water standard represent
specific examples where a thorough understanding of these factors will contribute to longterm solutions. As such, the overall goal of my project was to elucidate interactions of
arsenic and indigenous soil microbial populations and to determine factors important in the
microbial oxidation or reduction of arsenic in soil environments. Two detailed objectives
were formulated to support this goal along with a respective hypothesis:
(i)
Evaluate the influence of arsenite and arsenate concentrations on microbial
community composition under unsaturated (aerobic) conditions and link As redox
transformations with specific microbial populations.
H1. It is hypothesized that low concentrations of As contamination will
not result in large reductions in microbial diversity in soils given the
ubiquity of As detoxification genes. However, high concentrations of
As will likely select for organisms that either are capable of As
detoxification or that utilize As in energy conservation.
(ii)
Determine the affect of phosphate concentration on microbial arsenite oxidation
and microbial arsenate reduction using known As transforming organisms.
H2. It is hypothesized that high concentrations of phosphate in the soil
environment may inhibit a cell’s ability to uptake arsenate (via
phosphate transporters), thus preventing its reduction to arsenite through
the ars operon.
22
The experiments and results obtained to address each objective are detailed in the
following chapters: Chapter 2: Effects of Arsenic Pressure on Microbial Diversity and AsRedox Capability; Chapter 3: Inhibition of Microbial Arsenate Reduction by Phosphate. The
final chapter is a summary of the thesis project, with additional comments on broader
implications.
23
2. EFFECTS OF ARSENIC PRESSURE ON MICROBIAL
DIVERSITY AND AS-REDOX CAPABILITY
Introduction
The fate and toxicity of As is largely dependent on its predominant valence state and
chemical form. In soil environments two arsenic oxidation states dominate: AsIII (arsenite, as
H3AsO30) and AsV (arsenate, as H2AsO4- and HAsO4-2) (Oremland and Stolz, 2003).
Pentavalent arsenic is generally more strongly sorbed to common soil components, thus
considered less mobile and bioavailable than the more reduced, trivalent form (Pierce and
Moore, 1982; Xu et al., 1991). Given its neutral charge in soil environments, arsenite is able
to enter cells via membrane aqua-glyceroporins (Cullen and Reimer, 1989; Winegardener,
1995; Rosen, 2002) and is toxic to cells because of its affinity to protein sulfhydryl groups
(Oremland and Stolz, 2003). Conversely, arsenate is typically negatively charged in soils,
therefore unable to enter non-specific membrane porins (Cullen and Reimer, 1989); however,
as a chemical analog of phosphate, AsV is able to enter cells via membrane phosphate
transporters (Mukhopadhyay et al., 2002; Oremland and Stolz, 2003). As a consequence of
its chemical and biological similarities to phosphate, arsenate toxicity arises when substituted
for phosphate in cell metabolic processes such as oxidative phosphorylation (Mukhopadhyay
et al., 2002; Oremland and Stolz, 2003).
The relative abundance of arsenite and arsenate in soil environments is influenced by
geochemical conditions as well as a myriad of possible microbial transformations including
methylation, detoxification or energy-yielding oxidation-reduction (Inskeep et al., 2002).
Cultured microorganisms have been shown to possess arsenic transforming mechanisms and
24
several have multiple As regulatory pathways (Oremland and Stolz, 2003; Macur et al., 2004;
Kashyap et al., 2006). Previous research further supports that microorganisms with the
ability to oxidize arsenite and or reduce arsenate coexist within localized areas and are
ubiquitous across myriad soil environments (Jackson and Dugas, 2003; Macur et al., 2004;
Inskeep et al., 2007). While arsenate is often the predominant valence state in oxidized
environments (Oremland and Stolz, 2003), microbial reduction to arsenite in aerobic or
anaerobic environments is an important factor increasing the bioavailability of arsenic.
The reductive dissolution of iron oxides or sulfides and resultant release of sorbed arsenic
has been widely accepted as one of the primary mechanisms responsible for high arsenic
concentrations in South Asian groundwater aquifers (Nickson et al., 1998; Fazal et al., 2001).
The dissolution of iron oxides containing sorbed arsenic has been implicated in other cases of
arsenic release in soil-water environments including the high total arsenic (>500 mg kg-1)
found in sediments of Coeur d’Alene Lake (Idaho) (Harrington et al., 1998) and the elevated
arsenic (up to 1100 mg kg-1) found in the hyporheic zone of Silver Bow Creek (Montana). In
the Silver Bow-Clark Fork River system, over 100 million metric tons of As-rich mining
waste were discharged throughout the basin, accumulating arsenic rich Fe-oxide sediments
(Smith et al., 1998; Nagorski and Moore, 1999).
In addition to the potential release of As via Fe-oxide dissolution, a recent study on the
cause of the South Asian arsenic crisis (Polizzotto et al., 2005) put greater emphasis on
sulfide dissolution as a predominant control on arsenic release. Sediment core sampling in
the highly impacted central Bangladesh district revealed that arsenic-sulfides constituted the
largest solid phase arsenic fraction while reactive iron oxides were, in fact, largely depleted
25
in arsenic. The study further suggested that seasonal wetting and drying periods in nearsurface sediments induced cyclic redox processes which increase sulfide weathering rates
and thus arsenic release. Aqueous arsenic is then thought to be transported to well depth
(~30-50 m) by lateral groundwater flow and downward migration due to recharge (Polizzotto
et al., 2005).
These cases illustrate the complexity of geochemical and hydrodynamic
conditions that combine to control arsenic mobility.
Prior work has been successful in isolating As transforming organisms that contribute to
As geochemical cycling in soils and natural water systems. Salmassi et al. (2006) recently
isolated Hydrogenophaga-like organisms from Hot Creek, CA sediments that were shown to
be important arsenite-oxidizing organisms in this system, and it was recently shown that
three arsenite-oxidizing strains of Hydrogenophaga contain an aroA-like gene (Inskeep et al.,
2007). Prior work in irrigated soils of the Madison River Valley demonstrated that both
arsenic oxidizers and reducers were present in the same soil system (Macur et al., 2004).
Three arsenite-oxidizing organisms were isolated from this study (Agrobacterium
tumefaciens str. 5A, Variovorax paradoxous str. RM1, and Pseudomonas fluorescens str. 3)
and the arsenite oxidase of the Agrobacterium tumefaciens str. 5A was further shown to be
important in the detoxification of arsenite (Kashyap et al., 2006). Earlier work by Macur et
al. (2001) identified aerobic arsenate reduction as a net arsenic transforming process in a
mine tailing soil.
In summary, past studies clearly indicate that arsenic transforming
microorganisms are important members of microbial communities present in soil-water
systems. However, it is not clear what generalizations can be made at this point in time
26
regarding the organisms and factors controlling As oxidation-reduction reactions in natural
environments, especially in soils where the microbial diversity is extremely high.
More knowledge regarding the distribution of organisms with regulatory pathways
contributing to arsenic redox transformations will be beneficial for understanding biotic
processes contributing to the fate and transport of As in soils. Although previous work in As
contaminated soils has shown that As transforming organisms are common, less effort has
focused on the microbial community response to As perturbation in previously
uncontaminated systems. Consequently, the objectives of the current work were to (i)
evaluate the influence of arsenite and arsenate concentration on the resident microbial
community composition of a previously non-contaminated oxic soil, and (ii) identify the
organisms and mechanisms responsible for As transformations occurring under unsaturated
(aerobic) soil conditions. It was hypothesized that low concentrations of As contamination
would not result in large reductions in microbial diversity in soils given the ubiquity of As
detoxification genes; however, high concentrations of arsenic would select for organisms that
either are very efficient in detoxifying As or that utilize arsenic in energy conservation.
Materials and Methods
Soil Collection and Chemical Characterization
The top 20 cm of a previously non-contaminated agricultural soil (Amsterdam soil series,
Post Farm, Bozeman, MT) was collected, sieved (2 mm), physically and chemically
characterized (Table 2.1), and stored field moist at 4oC until use.
27
Column Experiments
Autoclaved polycarbonate columns (10.4 cm long, 3.48 cm diameter) were packed with a
mixture of 2.4% Amsterdam soil and 97.6% sterilized quartz sand (50-70 mesh, Sigma
Chemical, St. Louis, MO) for a total mass of ~153 g (bulk density ~1.52 g cm-3). The sand
used in all columns, as well as the soil used in two duplicate sterile control columns were
sterilized by autoclaving a 5 cm layer for 90 minutes, repeated for three consecutive days.
Table 2.1. Physical and chemical characteristics of Amsterdam soil used in column
experiments.
NO3-N
Texture Sand Silt Clay pH
K
EC
% % %
mg/kg mmhos/cm mg/kg
Silty
clay
loam
15
49
36
6.7
474
0.19
36.7
OM
%
2.15
Olsen P % H2O
mg/kg 1/3 bar
44.4
25.6
Column treatments were continuously supplied with soil solution media (SSM) for
approximately 57 days at room temperature. This media was used to reduce the bias inherent
with glucose-based media, and to provide the resident microbial community with carbon and
vitamin sources that would be common in the actual soil environment. The media was
prepared by first thoroughly mixing sieved Amsterdam soil and deionized water (1:5 ratio)
for approximately 6 hours at 100 oscillations min-1.
Aliquots of the suspension were
centrifuged at 9000 rpm for 40 minutes and the supernatant was filtered (0.2 µm) with a
Gelman filter apparatus attached to a vacuum pump.
Filtered media was chemically
characterized (Table 2.2), then stored at -20oC until use. Inductively coupled plasma (ICP)
spectrometry was employed to determine Ca, K, Fe, Mg, Na, Al, and P concentrations, while
F, Cl, and SO4 concentrations were determined using ion chromatography (IC).
28
Concentrations of NH4 and NO3 were determined using a flow injection analyzer (Lachat
Corp., Loveland, CO) and dissolved organic carbon (DOC) was determined with a
Dohrmann carbon analyzer. The final solutions used for column experiments contained SSM
(diluted 1:4 with triple-deionized water), and were autoclaved prior to spiking the solutions
with either arsenite (as NaAsO2) or arsenate (as Na2HAsO4) to obtain three influent As
concentrations (2 mg L-1, 20 mg L-1 and 200 mg L-1). Diluted, As-spiked SSM was delivered
to two duplicate sterile controls and two duplicate non-sterile “treatment” columns via a
continuous-flow pump set at 1.3 mL h-1 (pore water velocity = 0.64 cm h-1) in each of the
three concentration experiments.
Two duplicate non-sterile control columns were also
included which received dilute SSM only (no As). The first experiment was conducted for
57 days; however, subsequent experiments were shortened to approximately 25 days after
observations showed that arsenic transformations (i.e., arsenite oxidation or arsenate
reduction) were complete by approximately 28 days.
Table 2.2. Chemical composition of soil solution media (SSM) used in column
experiments. A 1:4 dilution of this extract into water was prepared prior to use as influent
in column studies.
All dissolved constituent concentrations reported in µM.
Concentrations of dissolved organic carbon (DOC), Ca, Mg and NO3 are listed as the
average of three separate soil:water extractions.
pH
6.7
DOC
1040
Ca
318
K
136
Fe
4
Mg
145
Na
43
Al
10
NH4
NO3
P
F
Cl
SO4
As
5
462
38
11
33
243
<0.67
The abiotic oxidation of arsenite was minimized by continuously bubbling the arsenite
input solution with N2 (g) and by separating the SSM and arsenite influent components until
29
just prior to addition to columns. Unsaturated flow conditions (volumetric water content, θv
= ~0.22 cm3 cm-3, ~50.9% of saturation) were maintained in the columns using a second
peristaltic pump set at approximately 100 mL h-1 to draw 0.22 µm filter-sterilized air through
the column via a separate inlet from that which delivered the liquid media. Flow rates were
confirmed periodically by measuring column effluent volume as a function of time.
Two pore-water samples per column were obtained approximately every third day
throughout the experiment by drawing 1.5 mL of effluent from two separate column samplers
(Figure 2.1) constructed using a 1-bar ceramic lysimeter (Soil Moisture, Goleta, CA)
connected with a sterile needle and peristaltic tubing to a sterile Vacutainer vile (BD,
Franklin Lakes, NJ). Two 1:10 dilutions were prepared from each port sample by adding 0.5
mL filtered (0.22 µm) sample to 4.5 mL deionized water in a 15-mL polyethylene bottle; a
total of four samples were prepared per column per sampling time. One sample per port was
pre-treated with sodium-borohydride to remove arsenite using the method modified from
Masscheleyn et al. (1991): 1 mL of 2 M TRIS buffer (pH 6.5) was added to 5-mL filtered
sample, then the buffered solution was sparged with N2 (g) while 1 mL of 3% NaBH4/0.1%
NaOH was added (in 0.2 mL aliquots) over a 3 min. period, then sparged for an additional 7
min. Samples were then acidified with HCl to a final concentration of 1% (v/v) and stored at
4oC. Before analysis with hydride generation atomic absorption spectrometry (HG-AAS;
Varian VGA 77; Perkin Elmer 3100), samples were further diluted to within the instrument’s
2 – 30 µg L-1 linear detection range (Jones et al., 2000). For treatments receiving 2 mg L-1
(26.7 µM) and 20 mg L-1 (267 µM) arsenite or arsenate, the arsenic speciation results were
essentially identical at the two column sampling positions, therefore only data from the top
30
sampling position is presented here; sampling at the lower column position was excluded in
the 200 mg As L-1 (2.7 mM) experiment.
Two additional treatment columns were included for experiments performed at 2 and 20
mg L-1. These columns were set-up (Figure 2.2) and sampled as described previously, but
were sacrificed at various time points to monitor changes in microbial community
composition potentially occurring during the ~30 d experiments.
(Analytical data for
columns sacrificed for molecular sampling is included in Appendix A, Figures A.1,2.) At
termination, the treatment columns were disassembled, and two soil samples per port were
obtained; one soil sample was stored at -80oC for molecular analysis, and the other was kept
at 4oC for cultivation of As-transforming organisms.
3 cm
8 cm
Figure 2.1. Polycarbonate column apparatus used in As transformation experiments
showing placement of two ceramic lysimeters (sampling ports).
31
Figure 2.2. Arrangement of column experiment.
Cultivation and Characterization of Pure-culture Isolates
Microorganisms were cultivated from the column soil samples by adding 1 g
homogenized soil to 9 mL of 10 mM NaCl. The resulting slurry was serially diluted to 10-7
and 100 mL of the three most dilute solutions were plated using a spread-plate technique on
SSM media (prepared with autoclaved, concentrated SSM media containing Difco granulated
agar). Colonies were picked from the SSM plates via a sterilized loop and replated on
nutrient-rich R2A media plates. Isolates were subsequently replated a third time on R2A
media to ensure their purification, after which, their colony morphologies were recorded
(Appendix A, Tables A.1-4).
DNA templates of the isolates were obtained by loop inoculating a single pure colony
into a 0.2 mL tube containing 50 µL DNA-free water. Two µL of this DNA template were
used in polymerase chain reactions (PCR) to amplify a 1384 bp region of the 16S rRNA gene
using the Bacteria-specific Bac-8F and universal 1392R primers and the following optimized
32
thermocycler (Techne model FGENO5TP, Burlington, NJ) protocol: initial denaturing at
95oC for 8 min; 32 cycles of 95o for 1 min, 55o for 1 min and 72o for 2 min; and final
extension at 72o for 10 min.
PCR products were purified using the QIAquick PCR
Purification Kit (Qiagen Inc., Valencia, CA) and quantified by electrophoresing 2 µL product
and 2 µL Low DNA Mass Ladder (Gibco-BRL) on a 1.2% agarose gel containing ethidium
bromide.
Purified products were sequenced by TGen (Phoenix, AZ) and Ohio State
University (Columbus, OH).
Sequences were edited and aligned with Sequencer 4.2
software (Gene Codes Corporation, Ann Arbor, MI) and compared to known sequences in
the GenBank database using BLAST (NCBI, Bethesda, MD).
Isolates were tested for their ability to oxidize or reduce As by incubating in two
duplicate 50 mL Falcon tubes containing 15 mL of autoclaved media made from 25%
concentrated SSM solution (described above) and 75% Soil Solution Equivalent (SSE)
media, modified as described in the Materials and Methods section of Chapter 3 to include
an additional 3 mg L-1 yeast extract. Prior to isolate inoculation, media was spiked with
equal amounts (2.9 mg L-1) of both arsenite and arsenate to a final concentration of 5.8 (±0.5)
mg As L-1 (77 µM). Vials were incubated at 30oC while agitating at 120 rpm and were
aseptically aerated periodically. To confirm that observed transformations were not abiotic,
duplicate sterile control vials containing only spiked media were also included.
After
approximately 8 days, 3.5 mL of the suspended cell solution was extracted from each vial for
optical density determination via A500 spectrometry.
To confirm that vials were not
contaminated, 10 µL of suspended cells were extracted from each vial at the experiment’s
conclusion, plated on R2A media and evaluated for purity and consistency with colony
33
morphology prior to inoculation. Two additional 5 mL extracts were filtered (0.22 µm) and
added to separate 15-mL polyethylene bottles; one sample was treated with sodiumborohydride to remove arsenite (described above). The total soluble As and arsenate only
(borohydride treated) samples were 2 and 3% acidified with 12N HCl, respectively, and
stored at 4oC until analysis with inductively coupled plasma (ICP) spectrometry. Isolates
were assigned an As redox phenotype based on a comparison of the arsenate remaining after
8 days in treated versus control vessels. Organisms were considered ‘arsenate reducing’ if the
amount of arsenate remaining was less than ~2.2 mg L-1, or ‘arsenite oxidizing’ if the
arsenate remaining was greater than ~3.6 mg L-1. Consequently, isolates exhibiting arsenate
values between ~2.2 and 3.6 mg L-1 were assumed to have no As-transforming capability
under the growth conditions tested (see Appendix A, Tables A.5–8 for data).
Molecular Analysis of Column Soils and Isolate DNA Extracts
DNA was extracted from column porous media samples using the FastDNA SPIN Kit for
Soil (Bio 101, Vista, CA). Both these column sample DNA extracts and the pure culture
templates described previously were used to PCR-amplify a 322 bp region of the 16S rRNA
gene using the Bacteria-specific 1070F primer and the universal 1392R primer containing a
40 bp GC clamp (Integrated DNA Technologies, Coralville, IA). Optimized thermocycler
protocol for the PCR reaction included initial denaturing at 94oC for 6 min; 32 cycles of 94o
for 45 s, 55o for 45 s and 72o for 55 s; and a final extension of 72o for 7 min. Two µL PCR
product and 2 µL Low DNA Mass Ladder (Gibco-BRL) were electrophoresed on a 1.2%
agarose gel containing ethidium bromide for quantification. The PCR products were loaded
onto a denaturing gradient gel electrophoresis (DGGE) gel consisting of 8% acrylamide, 1x
34
TAE (40 mM Tris, 20 mM acetic acid, and 2 mM EDTA at pH 8.5), and a 40–70%
urea/formamide gradient that increased from gel top to bottom.
PCR products were
separated via DGGE by inserting gels into the DCode Universal Detection System (Bio-Rad,
Hercules, CA) and running at 60 V for 17 hours at 60oC in 1x TAE buffer. DGGE gels were
stained by agitating at low speed for 40 min in a 200 mL 1x TAE buffer/20 µl SYBR green
solution, then rinsed with triple-DI water before they were visualized and photographed
using UV transillumination. Gels were poured in multiple arrangements to allow comparison
of microbial signatures across the three As concentrations, among duplicate columns and
between pure isolates and the column soil sample from which they were extracted (Appendix
A, Figures A.3-14).
Visually dominant DGGE bands from selected environmental lanes, including all which
showed co-migration with a pure isolate PCR product, were stabbed with a sterile pipette tip,
inserted into a 0.2 mL tube containing 10 µL DNA-free water and used as template for PCR
amplification using 1070 forward and 1392 reverse (without the GC clamp) primers as
described above. Amplification and DGGE analysis were repeated until a pure band was
obtained. Pure band PCR products were purified, quantified and sequenced (TGen, Phoenix,
AZ).
Resultant sequences were edited and aligned with Sequencer 4.2 software and
compared to known sequences in the GenBank database using BLAST.
35
Amplification of Arsenite Oxidase Gene Fragments
Degenerate primers were used to PCR amplify approximately 500 bp within the Mopterin (aroA/asoA/aoxB) subunit of arsenite oxidases detected in the DNA extracts from soil
columns and isolates from each As concentration experiment.
Forward (5'-
GTSGGBTGYGGMTAYCABGYCTA-3') and reverse (5'-TTGTASGCBGGNCGRTTR
TGRAT-3') primers bind at nucleotide positions 85-107 and 592-614, respectively, within the
aroA-like gene of Rhizobium sp. str. NT26 (Santini and vanden Hoven, 2004). The PCR mix
contained 1 μM of each primer. Optimized PCR conditions were: 95oC for 4 min followed
by 9 cycles of 95oC for 45 s, 50oC (decreased by 0.5oC after each cycle) for 45 s, 72oC for 50
s, followed by 25 cycles of 95oC for 45 s, 46oC for 45 s, and 72oC for 50 s, and a final
extension of 72oC for 5 min. Purified PCR products were cloned into the pGEM-T Vector
System (Promega, Madison, WI) and the clones were sequenced (TGen, Phoenix, AZ) using
T7P and SP6 primers that targeted vector sites.
Results and Discussion
Column Experiments
Columns amended with 2 mg L-1 (26.7 µM) arsenite exhibited a steady increase in the
amount of arsenate measured in pore waters from column sampling ports, culminating in near
complete (~97%) oxidation to arsenate (Figure 2.3B).
Although arsenate is the
thermodynamically favored As species under these (aerobic) conditions, arsenite-treated
sterile controls showed little (less than 13%) oxidation to arsenate (Figure 2.3A). These data
36
confirm that the abiotic oxidation rate is relatively slow and that biota were responsible for
the observed oxidation of arsenite in the non-sterile treatments.
Neither sterile controls nor treatment columns amended with 2 mg L-1 arsenate showed
appreciable reduction to arsenite; arsenate remained an average of 93.4% (standard deviation,
sd, = 11.3) of the total soluble As in sterile controls and 92.9% (sd = 7.1) in treatment
columns (Figure 2.3C,D). Over the course of the 2 mg L-1 As experiment, total As values
fluctuated between 1.4 and 2.6 mg L-1 (average concentration 2.0 mg L-1, sd = 0.31).
A) Arsenite Sterile Controls
B) Arsenite Treatment Columns
C) Arsenate Sterile Controls
D) Arsenate Treatment Columns
Figure 2.3. Percent of total soluble arsenic present as arsenite (gray) or arsenate (black) as
a function of time in columns amended with 2 mg L-1 (26.7 µM) of either arsenite or
arsenate. Data presented represents the average of two duplicate columns.
37
Similar As transformations were observed in columns amended with 20 mg L-1 (267 µM)
arsenite or arsenate. Within 9 to 12 days, nearly 60% of the arsenite was oxidized and by 24
days approximately 97% of the arsenite had been oxidized to arsenate (Figure 2.4B).
Conversely, sterile controls receiving 20 mg L-1 arsenite showed no appreciable oxidation to
arsenate (Figure 2.4A).
Meanwhile, arsenate-amended sterile controls and treatment
columns remained predominately arsenate over the experiment duration with an average 96 97% (sd = 3) of total soluble As comprised by arsenate (Figure 2.4C,D). Total arsenic in the
20 mg L-1 experiments fluctuated between 14.9 and 30.8 mg L-1 with an average As
concentration of 20.4 mg L-1 (sd = 3.95).
The upper limit of As tolerance and the maximum concentrations of As which can be
transformed by various organisms is also useful for broadening our understanding of
microbial As interactions.
Consequently, although rarely observed in uncontaminated
systems, an additional experiment was performed at 200 mg As L-1 (2.7 mM). In contrast to
results obtained at 2 and 20 mg As L-1, no appreciable redox transformations were observed
at 200 mg L-1. Arsenite remained the predominant aqueous species of As in arseniteamended sterile controls and treatment columns throughout the experiment (Figure 2.5A,B).
Similarly, arsenate was the predominant solution species in arsenate-amended sterile controls
and treatment columns, remaining >99.6% (sd = 0.57) and 99.7% (sd = 0.31), respectively, as
arsenate (Figure 2.5C,D). Total soluble As values in the 200 mg L-1 experiments fluctuated
between 173.5 and 228.7 mg L-1 with an average concentration of 207.4 mg L-1 (sd = 10.90).
Two duplicate non-sterile control columns treated with only SSM (no arsenic) for
approximately 26 days were included in the experimental suite to monitor microbial
38
composition response to media in the absence of As-pressure. As expected, concentrations
of soluble As were significantly lower than treatment columns (average = 0.07 mg L-1, sd =
0.11).
A) Arsenite Sterile Controls
B) Arsenite Treatment Columns
C) Arsenate Sterile Controls
D) Arsenate Treatment Columns
Figure 2.4. Percent of total soluble arsenic present as arsenite (gray) or arsenate (black) as
a function of time in columns amended with 20 mg L-1 (267 µM) of either arsenite or
arsenate. Data presented represents the average of two duplicate columns.
39
A) Arsenite Sterile Controls
B) Arsenite Treatment Columns
C) Arsenate Sterile Controls
D) Arsenate Treatment Columns
Figure 2.5. Percent of total soluble arsenic present as arsenite (gray) or arsenate (black) as
a function of time in columns amended with 200 mg L-1 (2.7 mM) of either arsenite or
arsenate. Data presented represents the average of two duplicate columns.
Isolate Cultivation
A total of 62 bacterial isolates (confirmed using long-fragment [~1380 bp] 16S rRNA
gene sequence analysis) were obtained from column samples taken at the conclusion of each
experiment. No significant trends were observed in the absolute number of unique isolates
retrieved in treatments ranging from 2 to 200 mg As L-1. Specifically, 24 isolates from 14
40
genera were obtained from the 2 mg As L-1 treatments, 14 isolates from 12 genera were
obtained from the 20 mg As L-1 treatments and 17 isolates from 10 genera were obtained
from the 200 mg As L-1 treatments (Tables 2.3A-C). Interestingly, only 7 isolates from 4
genera were retrieved from non-sterile control columns receiving no As (Table 2.3D). The
most common phylogenetic groups represented in this study include members of the
Actinobacteria, and the α- and β- Proteobacteria. Members of the Actinobacteria were the
only organisms isolated in column treatments receiving no As pressure (Table 2.3D). The
importance of these bacterial genera in soil environments has been documented in several
prior studies on soil microbial diversity (Ulrich and Wirth, 1999; Smit et al., 2001; Torsvik
and Ovreas, 2002; Salmassi et al., 2006). Streptomyces, Rhodococcus, Nocardioides and
Arthrobacter are all common genera in soil environments, and they are often represented in
culture libraries (Ulrich and Wirth, 1999; Smalla et al., 2001; Smit et al., 2001).
Although the microbial diversity in terms of number of genera cultivated showed
essentially no decrease with increasing As-pressure, the most commonly encountered genera
appeared to shift from Streptomyces in the 2 mg As L-1 treatments to Variovorax in the 200
mg As L-1 treatments. The cultivation of numerous Variovorax spp. at the highest As
concentration suggests this genera may be more efficient in detoxifying As or may utilize As
in energy generation, which is consistent with previous literature that has identified
Variovorax-like populations in As-rich environments (Ellis et al., 2003; Macur et al., 2004;
Battaglia-Brunet et al., 2006). There was a significant shift from primarily Actinobacteria
obtained from columns treated with a lower [As] 0 – 20 mg L-1) (Tables 2.3A,B,D) to βProteobacteria in columns treated with a more extreme [As] (200 mg L-1) (Table 2.3C).
41
Table 2.3A. Pure cultures of bacteria isolated from columns receiving 2 mg L-1 (26.7 µM)
of either arsenite or arsenate. Isolates are identified based on strain number, closest
cultivated relative in GenBank, major phylogenetic division, and experimentally
determined As phenotype (i.e. arsenite oxidizer or arsenate reducer). All isolates were
obtained from 10-4 serial dilutions.
Isolate
Strain
Name
1
Closest GenBank Neighbor (% sim.)
Phylogenic Group
Isolates cultivated from As III treatment columns.
DM1A Leptothrix sp. str. S1.1 (98.0)
DM1AA Streptomyces sp. str. FXJ14 (99.9)
DM1B Streptomyces argillaceus (99.8)
DM1BB Rhodococcus sp. str. 17 (99.9)
DM1E Janthinobacterium lividum (99.8)
DM1M Streptomyces luteogriseus str. ISP 5483 (99.6)
DM1N Acidovorax delafieldii (98.8)
DM1O Streptomyces peruviensis str. DSM 40592 (98.5)
DM1Q Caulobacter henricii str. ATCC 15253 (99.6) 4
DM1R Saccharothrix texasensis str. NRRL B-16107T (99.1)
DM1S Rhodococcus marinonanscens str. ATCC 35653T (98.0)
DM1T Arthrobacter chlorophenolicus (98.8)
Betaproteobacteria
Actinobacteria
Actinobacteria
Actinobacteria
Betaproteobacteria
Actinobacteria
Betaproteobacteria
Actinobacteria
Alphaproteobacteria
Actinobacteria
Actinobacteria
Actinobacteria
Isolates cultivated from As V treatment columns.
DM1U Janthinobacterium sp. str. IC161 (98.6)
DM1V Flavobacterium sp. str. TB4-10-II (99.0)
DM1W Janthinobacterium agaricidamnosum str. SAFR-022 (97.5)
DM1X Sinorhizobium sp. str. TB2-T-5 (95.8)
DM1Y Arthrobacter aurescens str. TC1 (99.9)
DM1F Nocardioides sp. str. RS3-1 (98.3)
DM1G Mesorhizobium sp. str. 98_RREM2003 (99.8)
DM1H Mesorhizobium sp. str. HB5A4 (99.9)
DM1I Streptomyces mirabilis str. ATCC27447 (98.5)
Betaproteobacteria
Flavobacteria
Betaproteobacteria
Alphaproteobacteria
Actinobacteria
Actinobacteria
Alphaproteobacteria
Alphaproteobacteria
Actinobacteria
Isolates cultivated from As III sterile control columns.
DM1J Microbacterium ginsengisoli (99.9)
DM1K Caulobacter henricii str. ATCC 15253 (99.4) 4
Actinobacteria
Alphaproteobacteria
DM1L
Betaproteobacteria
Zoogloea ramigera str. ATCC 25935 (99.2)
Isolate did not grow well in liquid media.
Unable to replate isolate from glycerol stock.
3
Isolate showed neither oxidation or reduction capability.
4
Caulobacter henricii -like strs. DM1Q and str. DM1K are 99.4% similar.
2
As
Phenotype
2
reducer
reducer
reducer
reducer
reducer
reducer
reducer
3
reducer
reducer
reducer
2
reducer
reducer
3
reducer
reducer
3
oxidizer
reducer
1
3
reducer
42
Table 2.3B. Pure cultures of bacteria isolated from columns receiving 20 mg L-1 (267 µM)
of either arsenite or arsenate. Isolates are identified based on strain number, closest
cultivated relative in GenBank, major phylogenetic division, and experimentally
determined As phenotype (i.e. arsenite oxidizer or arsenate reducer). All isolates obtained
from 10-3 serial dilutions.
Isolate
Strain
Name
Closest GenBank Neighbor (% sim.)
Phylogenic Group
As
Phenotype
Isolates cultivated from AsIII treatment columns.
DM2A
DM2B
DM2C
DM2D
DM2T
DM2U
Rhodococcus sp. str. I7 (99.7)
Arthrobacter sp. str. c311 (99.5)
Methylobacterium sp. str. G296-5 (98.3) 4
Afipia massiliensis (95.5)
Hongia koreensis (99.1)
Bradyrhizobium sp. str. Cmey 1 (97.4)
Actinobacteria
Actinobacteria
Alphaproteobacteria
Alphaproteobacteria
Actinobacteria
Alphaproteobacteria
reducer
reducer
reducer
Actinobacteria
Actinobacteria
Actinobacteria
Firmicutes
Actinobacteria
reducer
reducer
reducer
reducer
Alphaproteobacteria
Actinobacteria
reducer
reducer
Alphaproteobacteria
3
1
reducer
3
Isolates cultivated from AsV treatment columns.
DM2N
DM2O
DM2P
DM2Q
DM2R
Streptomyces griseus (99.7)
Actinomadura citrea str. DSM 43461T (99.7)
Nocardioides kribbensis str. KSL-6 (98.4)
Paenibacillus alginolyticus (97.5)
Streptosporangium roseum str. DSM44111 (99.1)
2
Isolates cultivated from AsIII sterile control columns.
DM2E
DM2F
Methylobacterium radiotolerans str. P (99.5) 4
Mycobacterium sacrum str. BN 3151 (97.6)
Isolates cultivated from AsV sterile control columns.
DM2S
Methylobacterium fujisawaense str. DSM 5686 (99.8) 4
1
Isolate did not grow well in liquid media.
Unable to replate isolate from glycerol stock.
3
Isolate showed neither oxidation or reduction capability.
2
4
Methylobacterium -like str. DM2C is 96.0% similar to str. DM2E and 95.6% similar to str. DM2S. And
Methylobacterium -like str. DM2E is 99.5% similar to str. DM2S.
43
Table 2.3C. Pure cultures of bacteria isolated from columns receiving 200 mg L-1 (2.7
mM) of either arsenite or arsenate. Isolates are identified based on strain number, closest
cultivated relative in GenBank, major phylogenetic division, and experimentally
determined As phenotype (i.e. arsenite oxidizer or arsenate reducer).
Isolate
Strain
Closest GenBank Neighbor (% sim.)
Name
Isolates cultivated from AsIII treatment columns.
DM3A9
DM3B9
DM3C9
DM3D9
DM3E9
DM3F9
Rhodoferax ferrireducens str. DSM 15236 (97.7) 6
Pseudomonas frederiksbergensis str. OUCZ24 (99.6)
Variovorax sp. str. K6 (99.9) 4
Arthrobacter nitroguajacolicus str. Rue61a (99.8) 5
Variovorax sp. str. K6 (99.4) 4
Variovorax paradoxus (100) 4
Phylogenic Group
Betaproteobacteria
Gammaproteobacteria
Betaproteobacteria
Actinobacteria
Betaproteobacteria
Betaproteobacteria
Isolates cultivated from AsV treatment columns.
Betaproteobacteria
DM3G7 Variovorax paradoxus (99.4) 4
7
DM3H Janthinobacterium agaricidamnosum str. SAFR-022 (99.4) Betaproteobacteria
DM3I8 Variovorax sp. str. WDL1 (97.6) 4
Betaproteobacteria
DM3J8 Streptomyces griseus (99.3)
Actinobacteria
DM3K8 Phyllobacterium trifolii (99.8)
Alphaproteobacteria
DM3L8 Afipia genospecies 9 strain G8993 (99.5)
Alphaproteobacteria
As
Phenotype
reducer
reducer
3
reducer
3
reducer
reducer
3
reducer
2
reducer
1
Isolates cultivated from AsV sterile control columns.
DM3M9
DM3N8
DM3O7
DM3P9
DM3R9
Herbaspirillum seropedicae str. X8 (97.6)
Sinobacter albidoflavus str. 45 (99.1)
Arthrobacter nitroguajacolicus str. Rue61a (99.8) 5
Rhodoferax ferrireducens (97.7) 6
Variovorax sp. str. K6 (99.7) 4
Betaproteobacteria
Betaproteobacteria
Actinobacteria
Betaproteobacteria
Betaproteobacteria
1
reducer
reducer
1
reducer
1
Isolate did not grow well in liquid media.
Unable to replate isolate from glycerol stock.
3
Isolate showed neither oxidation or reduction capability.
2
4
Variovorax -like str. DM3R is 99.7% similar to str. DM3C, 99.3% similar to str. DM3E, 98.4% similar
to str. DM3F, 98.9% similar to str. DM3G, and 96.84% smilar to str. DM3I.
5
Arthrobacter nitroguajacolicus-like strs. DM3O and DM3D are 100% similar.
Rhodoferax ferrireducens-like strs. DM3P and DM3A are 99.9% similar.
7
Cultured from 10-2 serial dilution plate.
8
Cultured from 10-3 serial dilution plate.
9
Cultured from 10-4 serial dilution plate.
6
44
Table 2.3D. Pure cultures of bacteria isolated from columns withstanding no As pressure.
Isolates are identified based on strain number, closest cultivated relative in GenBank,
major phylogenetic division, and experimentally determined As phenotype (i.e. arsenite
oxidizer or arsenate reducer). All isolates obtained from 10-3 serial dilutions.
Isolate
Strain
Name
DM2G
DM2H
DM2I
DM2J
DM2K
DM2L
DM2M
Closest GenBank Neighbor (% sim.)
Streptomyces novaecaesareae str. NBRC 13368 (99.2)
Streptomyces turgidiscabies (99.6)
Rhodococcus erythropolis str. HS11 (97.0)
Streptomyces novaecaesareae str. NBRC 13368 (99.2)
Arthrobacter sp. str. 110 (99.4)
Nonomuraea sp. str. TFS 1165 (98.8)
Arthrobacter aurescens str. TC1 (100)
Phylogenic Group
Actinobacteria
Actinobacteria
Actinobacteria
Actinobacteria
Actinobacteria
Actinobacteria
Actinobacteria
As
Phenotype
reducer
reducer
reducer
reducer
reducer
reducer
reducer
Despite employing standard autoclaving protocol (described above) to sterilize soil
before use in the column studies, a number of isolates were cultivated from the sterile control
columns in each experiment.
These findings suggest that the columns were either
contaminated by an extraneous source during the course of the experimentation, or that
specific organisms within the soil inoculum survived autoclaving. However, organisms
belonging to the same genera were cultivated both from sterile control and treatment
columns, and were highly similar in their 16S rRNA sequence identity (Tables 2.3A-C).
Therefore, it is probable these isolates also shared the same environmental origin, the soil
inoculum.
These strains represent populations that are likely particularly resilient,
indigenous members of the Amsterdam soil microbial community and were able to survive
autoclave suppression and restore growth over the one-to-two month experiment.
This
phenomenon has been documented in the literature previously where one study suggested
that high concentrations of dissolved organic carbon can result after autoclaving soil,
45
providing surviving bacterial spores a nutrient rich environment following autoclaving
(Tuominen et al. 1994).
Although arsenite oxidation was the net As transformation process observed in soil
column experiments, 43 of the 62 isolates obtained from the column treatments exhibited an
arsenate-reducing phenotype (Tables 2.3A-D). In fact, the Mesorhizobium-like isolate (str.
DM1) cultivated from a 2 mg As L-1 arsenate-treated column was the sole arsenite-oxidizing
isolate cultivated from all experiments. The reasoning for this anomaly is unclear. Perhaps
the net activity of the seemingly smaller population of arsenite oxidizers is greater than that
of the arsenate-reducers. It is also possible that the cultivation approach utilized in this study
favored the growth of arsenate-reducing microorganisms; although, it is not clear what
specific attributes of the media would favor arsenate reducers. It is further possible that the
relevant and numerous arsenite-oxidizing populations present in the columns were not
cultivatable using the current approach. Finally, it is possible that a subset of the isolates
characterized could be shown to oxidize arsenite under different conditions, as it has been
documented that some organisms have the potential to oxidize and or reduce As under
different growth conditions (Oremland and Stolz, 2003). Nevertheless, it is certainly the case
that arsenate reduction is a common attribute among soil bacteria.
Molecular Analysis
Porous media samples from column experiments were used to obtain a molecular
fingerprint of the microbial community using DNA extraction, PCR-amplification and
separation of short-fragment 16S sequences (~300 bp) via denatured gradient gel
electrophoresis (DGGE). DGGE banding patterns showed no evidence for a decrease in
46
microbial diversity with increasing As concentration. In fact, the significant number of
different microbial bands within each sample, evidenced by “smearing” of the signatures on
the DGGE gel, precluded sequence analysis of all possible bands (Appendix A, Figures A.38). Banding patterns of columns sampled at the conclusion of each experiment demonstrated
up to 15 distinct bands in a particular column treatment and consistency in the signatures of
arsenite- versus arsenate-amended treatment columns at each respective As concentration
ranged from ~20 to 80% (Appendix A, Figures A.3,5,7,8). Banding patterns were generally
consistent (~80 - 90%) across the same column treatment regardless of the time sampled, that
is, columns “sacrificed” and sampled prior to experiment’s conclusion exhibited similar
banding patterns to columns sampled at the end of the experiment (Appendix A, Figures
A.4,6).
A total of 18 dominant DGGE bands were purified and sequenced in an attempt to
elucidate potential dominant microbial populations within selected columns across the
experimental arsenic concentration gradient. Eight of the isolate DGGE bands appeared to
co-migrate with one of the environmental bands when run simultaneously on DGGE (Table
2.4). However, the 16S sequence of only one of these isolates, Herbaspirillum-like DM3M
(as assigned by closest GenBank neighbor), agreed with the BLAST-named sequence of the
co-migrating dominant environmental band (Figure 2.6). A second isolate, Sinobacter-like
DM3N, also co-migrated with this environmental band (Figure 2.6); a direct comparison of
the 16S sequences of the co-migrating BLAST-named Herbaspirillum-like environmental
band and Sinobacter-like DM3N isolate confirmed they are also closely related (99.6%)
phylogenetically.
Variovorax limosa str. EMB320 (86.8)
Herbaspirillum sp. str. SE1 (99.6)
Chelatococcus sp. str. A (99.7)
200 mg AsV L-1
V -1
200 mg As L
200 mg AsV L-1
200-C7-1
200-C6-1
200-C6-2
Methylobacterium extorquens (97.4)
Massilia sp. str.VA23069_03 (98.7)
-1
III
200 mg As L
200 mg AsIII L-1
V -1
200 mg As L
200-C3-3
200-C3-4
200-C3-5
200-C5-2
-1
III
200-C5-3
200-C5-4
200 mg As L
200 mg As L
-1
V
200 mg As L
200 mg As L
Bosea thiooxidans str. TJ1 (94.6)
Duganella violaceinigra str. YIM 31327 (98.6)
Achromobacter xylosoxidans str. 27 (93.1)
Sphingomonas sp. str. D31C2 (98.8)
Variovorax sp. str. Yged136 (99.6)
-1
III
200-C3-2
Variovorax sp. str. Yged136 (99.6)
-1
III
200-C3-1
200 mg As L
-1
III
Pseudomonas sp. str. BCL-63 (90.2)
20 mg As L
20-C12-1
Herbaspirillum sp. str. B601 (99.1)
-1
2 mg As L
V
2 mg AsIII L-1
2-C2-1
2-C2-2
Methylobacillus sp. JER103 (94.1)
III -1
2 mg As L
2-C3-5
-1
Leptothrix sp. str. SAYI-C (94.8)
2 mg AsIII L-1
2-C3-3
III
Phormidium sp. str. MBIC10025 (96.3)
2 mg As L
Proteobacterium Sva0812b (85.2)
Bacterium clone ITOD25 (95.9)
Closest GenBank Neighbor (% sim.)
2-C3-2
-1
2 mg AsIII L-1
2-C3-1
III
As Treatment of
Source Column
Band ID
Variovorax -like str. DM3R
Variovorax -like str. DM3F
Variovorax -like str. DM3C
Variovorax -like str. DM3E
Herbaspirillum -like str. DM3M
Sinobacter -like str. DM3N
Rhodococcus -like str. DM1BB
Rhodococcus -like str. DM1S
Co-migrating Isolates
Table 2.4. Potentially dominant microbial populations within column treatments based on 16S rRNA DGGE
banding patterns and co-migrating cultivars.
47
Herbaspirillum sp. str. SE1
(99.6)
Sinobacter-like Isolate DM3N
Herbaspirillum-like Isolate DM3M
Band Sequence Closest GenBank
Neighbor
(% sim)
AsV-treated Column 200-6
48
sequences
99.6%
similar
sequences
98.8%
similar
Chelatococcus sp. str. A
(99.7)
Figure 2.6. Representative co-migration using DGGE of short fragment 16S rDNA
sequences from two isolates compared to environmental DGGE bands from the 200 mg L-1
(2.7 mM) arsenate-treated column from which they were isolated.
Arsenite Oxidase Gene Amplification
PCR amplification using aroA-like specific primers resulted in the detection of three
novel aroA-like gene sequences from soil columns treated with 2 mg arsenite L-1 (GenBank
accession numbers: DQ380572, DQ380571 and DQ380573). No aroA-like PCR product was
detected in any of the other AsIII or AsV treated columns, or the untreated Amsterdam soil,
which had no history of As application. The fact that aroA-like was only detected in the 2
49
mg AsIII L-1 treatment and not in the 20 mg AsIII L-1 treatment was perplexing, since arsenite
oxidation was the dominant redox process in both treatments. Possible explanations for this
result are that the primer sequences lacked sufficient homology to anneal to the aroA-like
gene sequences present in the 20 mg L-1 treatment or that the DNA extraction or PCR
conditions are not yet optimized for these types of samples. Negative results with functional
gene primers applied to environmental samples are not uncommon and are difficult to
interpret. The results obtained with these soils suggest that aroA-like sequences are present
(2 mg As L-1) and may play a role in arsenite oxidation, but the inability to detect aroA-like
in the 20 mg As L-1 treatments leaves a question regarding the efficacy of the technique and
or other mechanisms of arsenite oxidation.
PCR amplification using genomic DNA from 62 isolates representing α−, β−, and γProteobacteria, Actinobacteria, Firmicutes and Flavobacteria, yielded only one aroA-like
sequence. Detection of an aroA-like gene in Mesorhizobium str. DM1 correlated with the
observation that it was the only isolate capable of oxidizing arsenite under the experimental
conditions employed in the current study. It is interesting that this aroA-like sequence was
not detected in the 2 mg L-1 AsV-treated soil column from which the Mesorhizobium-like
isolate was cultivated. This was possibly due to inhibition of PCR by the relatively high
concentrations of dissolved humic substances in the Amsterdam soil. It is also possible that
only very low numbers of this population were present in this soil column.
Phylogenetic analysis of the aroA-like genes detected in soil columns revealed that they
clustered with aroA-like genes found in Variovorax, Acinetobacter and Hydrogenophaga
strains (Figure 2.7). Conversely, the aroA-like gene detected in Mesorhizobium sp. str. DM1
50
groups with the oxidase genes found in Agrobacterium and Rhizobium strains, all
phylogenetically closely related organisms.
Achromobacter sp. str. NT10 (DQ412673)
‘Alcaligenes faecalis’ (AAQ19838)
H. Arsenicoxydans str. ULPAs1 (AAN05581)
73
Variovorax sp. str. RM1 (DQ380569)
66
Acinetobacter sp. str. WA19 (DQ412677)
73
Rhodoferax ferrireducens (ZP_00691323)
Hydrogenophaga sp. str. NT14 (DQ412672)
100
Amst soil column aroA (DQ380572)
Thiomonas sp. VB-2002 (CAD53341)
Agrobacterium sp. str. Ben5 (DQ412675)
100
98
74
Sinorhizobium sp. str. NT4 (DQ412674)
Ag. tumefaciens str. 5A (ABB51928)
Rhizobium sp. str. NT26 (AARO5656)
Mesorhizobium sp. str. DM1 (DQ380570)
73
94
100
58
Rosevarius sp. 217 (ZP_01034989)
Nitrobacter hamburgensis (ZP_00627780)
Sargasso Sea metagenome (EAI76964)
96
Thermus thermophilus HB8 (BAD71923)
Chlorobium phaeobacteroides (ZP_00530522)
Sulfolobus tokodaii str. 7 (NP_378391)
0.05 changes
Figure 2.7. Phylogenetic tree of selected, deduced prokaryotic amino acid sequences of
the large subunit of the aerobic bacterial arsenite oxidase (AroA-like). Tree shows relative
positioning of aroA-like sequences obtained from a pure culture isolate, Mesorhizobium
sp. str. DM1, and an arsenite-treated soil column sample (Amst soil column). Bootstrap
values (per 100 trials) of major branch points are shown. Tree = neighbor-joining method;
bar = 0.05 substitutions/sequence position; tree rooted with Fdh from Methanocaldococcus
jannaschii NP_248356 (not shown); accession numbers for the sequences are shown in
parentheses.
Conclusions and Implications
The primary goal of this study was to determine the influence of As concentration on the
microbial community composition of a previously non-contaminated oxic soil. Cultivation
and short-fragment DNA analysis via DGGE provided no evidence for a decrease in
microbial diversity as a function of increasing As concentration, even in the presence of
51
extreme (200 mg L-1) As contamination. Greater microbial selection and less diversity were
expected at the highest level of As perturbation. The fact that numerous organisms appeared
to be stimulated in this treatment suggests that detoxification mechanisms allow them to
withstand As pressure. However, while concentrations of 200 mg As L-1 did not greatly
decrease microbial diversity, a shift in phylogeny was noted, and the concentrations were too
high for organisms within this community to maintain arsenite oxidation (the net redox
capability of the resident microbial population at 2 and 20 mg As L-1). Contaminated soil
environments containing 200 mg As kg-1 or greater are not uncommon (Freeman et al., 1995;
Belluck et al., 2003), consequently, future studies should continue to focus on cultivating or
stimulating microorganisms in situ that are capable of oxidizing arsenite to the less mobile
arsenic form, thereby assisting in bioremediation goals.
The second experimental goal was to elucidate the organisms and mechanisms
responsible for observed As redox transformations. Sixty-two pure culture isolates obtained
across respective column treatments are potential candidates for contributing to As oxidationreduction. The primary genera cultivated shifted from members of the Actinobacteria (e.g.
Streptomyces and Arthrobacter spp.) to β-proteobacteria (e.g. Variovorax spp.) with
increasing As concentration. However, 43 of the isolates exhibited an arsenate-reducing
phenotype while only one isolate (Mesorhizobium-like str. DM1) appeared capable of
arsenite oxidation. Although aerobic arsenate-reducing organisms appear easy to cultivate,
arsenite-oxidizing organisms are either considerably less frequent or are more difficult to
cultivate. Results from the isolation and characterization of As transforming microorganisms
from the column treatments do not explain the net biotic oxidation that was observed at 2 and
52
20 mg As L-1, where we would have expected greater success in cultivating arseniteoxidizing organisms.
Molecular analysis of short-fragment 16S rRNA genes obtained from column porous
media samples was used to identify potentially important As transforming organisms at
different As treatment concentrations. Eighteen different DGGE bands were sequenced
based on their intensity and apparent co-migration compared to DNA fragments of pure
culture isolates. There was little consistency between the dominant genera identified by
cultivation versus molecular methods, and only one genus, Herbaspirillum, was suggested as
a dominant population by both molecular and isolation techniques. However, this isolate was
identified as an arsenate-reducing organism in the phenotypic screening, thus its contribution
to the net oxidation of As observed in the column experiments is not clear. While this study
has contributed information regarding the ubiquity of aerobic arsenate-reducing
microorganisms, the role of specific biota in arsenite oxidation remains elusive. Additional
work with a greater number of previously uncontaminated soils would be helpful for
identifying consistent trends in microbial response to As perturbation, and improvement in
molecular tools and methods will be necessary for quantifying arsenite oxidase genes from
environmental samples.
53
3. INHIBITION OF MICROBIAL ARSENATE REDUCTION BY PHOSPHATE
Introduction
Arsenic (As) is the twentieth most abundant element in the earth’s crust and is ubiquitous
in soils (average concentration: ~ 2-3 mg kg-1) across the globe (Francesconi and Kuehnelt,
2002). Thus, most soil microorganisms are exposed to some level of arsenic, and in certain
circumstances elevated levels of As may enrich for organisms capable of utilizing arsenic in
energy conservation. The source of As in soils and natural water systems is ultimately
volcanic (Cullen and Reimer, 1989), it’s distribution generally correlating with the
distribution of sulfides in hydrothermal, geothermal and heavily mineralized, sulfidic veins.
Sedimentary (alluvial) sources of arsenic are also significant in some geographic locations
such as Bangladesh and India where naturally occurring As associated with pyritic sediments
has contaminated drinking water supplies for millions of rural residents (Nickson et al., 1998;
Fazal et al., 2001; Nordstrom, 2002; Polizzotto et al., 2005). Anthropogenic sources of
arsenic include mining and smelting of pyritic-ores and the application of As-based
pesticides; these inputs can result in elevated levels of arsenic, in many cases resulting in soil
concentrations exceeding 100 mg As kg-1 (Belluck et al., 2003).
The distribution of As in soils and natural waters has serious environmental health
implications; symptoms of chronic exposure to arsenic range from nausea, skin discoloration
and lesions to various organ cancers and may result in death (US EPA, 2006a). Fatalities
linked to As poisoning have been recorded since antiquity; however, in the last two decades
arsenic has gained heightened global attention following its discovery in high concentrations
54
in the drinking water of many Asian and North and South American countries (Fazal et al.,
2001; Nordstrom, 2002). The most severely impacted country is Bangladesh where an
estimated 21 to 40 million citizens have been exposed to concentrations above the nation’s
current As standard (50 µg L-1) (Fazal et al., 2001). In part due to calls from the World
Health Organization (WHO) and the U.S. EPA, the U.S. recently (February, 2006) adopted a
new drinking water standard for As equal to 10 µg L-1, commonly accepted across the
developed countries as a reasonable exposure level based on chronic risks (Smith et al.,
2002).
The environmental fate and toxicity of As is highly dependent on its predominant valence
state and chemical form. Arsenic is known to occur naturally in four oxidation states: As-III
(arsine), As0 (elemental As), AsIII (arsenite), and AsV (arsenate), where the latter two are most
prevalent in soil environments (Oremland and Stolz, 2003).
Pentavalent arsenic is
thermodynamically favored in oxic environments as the oxyanion arsenate, which generally
sorbs more strongly and to a wider variety of minerals than trivalent arsenic (Pierce and
Moore, 1982; Xu et al., 1991). Hence, arsenate is generally considered to be less mobile and
less bioavailable than arsenite. These two forms of arsenic also vary in their molecular
properties, mode of cell entry and biological toxicity. The predominant arsenite species in
soils and natural waters is usually H3AsO30, with an exception that sulfidic environments can
yield significant levels of AsIII-sulfide complexes (Rochette et al., 2000). The weak acid
H3AsO30 has a pKa value of 9.2; consequently, across most environmental systems, arsenite
will exist as an uncharged species (Cullen and Reimer, 1989). The neutrally charged arsenite
species can enter cells via aqua-glyceroporins (large pores in the cell membrane) that allow
55
passage of water and uncharged solutes (Rosen, 2002). Once internalized, arsenite toxicity
occurs as it binds to the sulfhydryl groups of proteins and impairs their function (Oremland
and Stolz, 2003).
The predominant forms of arsenate in soil solutions and natural waters will generally be
either the H2AsO4- (pKa = 7.0) or HAsO42- (pKa = 11.5) species, which due to their negative
charge, are unable to enter nonspecific membrane porins (Cullen and Reimer, 1989).
However, due to structural similarities between arsenate and phosphate, arsenate species can
enter cells via membrane-associated phosphate transporters (Mukhopadhyay et al., 2002;
Oremland and Stolz, 2003). The mechanism of arsenate toxicity is quite different than for
arsenite, and is again due to the chemical similarities between arsenate and phosphate.
Specifically, arsenate is detrimental to basic cell function when it is substituted for phosphate
in cell metabolic processes such as oxidative phosphorylation (Mukhopadhyay et al., 2002;
Oremland and Stolz, 2003).
The relative abundance of AsIII and AsV in soil environments is influenced by
geochemical conditions and microbial transformations including detoxifying or energyyielding redox pathways (Inskeep et al., 2002). Most microorganisms in culture have been
shown to possess at least one type of arsenic transforming mechanism. While arsenate is
often the predominant valence state in oxidized environments (Oremland and Stolz, 2003),
microbial reduction to arsenite in both aerobic and anaerobic systems is an important factor
increasing the mobility and potential bioavailability of As (Macur et al., 2001; Harvey et al.,
2002).
56
Commonly practiced methods of remediating damaged lands, such as the addition of lime
to increase soil pH, can also exacerbate arsenic bioavailability. Past work has shown that
increases in pH values above 8 increase arsenic mobility (Darland and Inskeep, 1997; Jones
et al., 1997; Macur et al., 2001).
Therefore, careful liming and pH management is
recommended to minimize solubilization of sorbed arsenic (Jones et al., 1997; Heeraman et
al., 2001). In theory, the addition of phosphate can help prevent microbial reduction of
arsenate to the more mobile and toxic arsenite form, thus phosphate application is also a
potential component of As bioremediation strategies. However, there have been indications
that the possible effects of soil phosphate additions on arsenic mobility may complicate its
use. Literature supports that increased phosphate concentrations lead to an increase in
arsenic desorption and thus mobility, consistent with the fact that phosphate strongly
competes with arsenate for soil adsorption sites (Darland and Inskeep, 1997; Peryea and
Kammereck, 1997; Alam et al., 2001). However, there is also evidence that As desorption
rates decrease in “aged” systems where the presence of arsenic precedes the introduction of
phosphate giving arsenic sufficient time to react to soil binding sites (Darland and Inskeep,
1997). Many As-contaminated sites are “aged,” where their arsenic exposure dates decades
or even centuries (Freeman et al., 1995; Harrington et al., 1998; Nagorski and Moore, 1999;
Belluck et al., 2003), the potential use of phosphate in arsenic bioremediation should not be
ruled out.
As mentioned previously, phosphate and arsenate are in many ways chemically and
biologically analogous and this relationship is what allows phosphate to inhibit microbial
AsV reduction. Phosphate has long-since been shown to compete with the cell uptake of
57
arsenate in numerous biological species (Rothstein and Donovan, 1963; Harold and Baarda,
1966; Willsky and Malamy, 1980; Thiel, 1988); therefore, high concentrations of phosphate
may completely inhibit microbial uptake of arsenate, thus preventing its reduction to AsIII via
ArsC (arsenate reductase) of the ars operon.
To date, the ars operon is the best understood arsenic regulatory mechanism in
microorganisms and appears to be widely distributed phylogenetically. For example, a recent
gene search of The Institute for Genomic Research-Comprehensive Microbial Resource
(TIGR-CMR) database shows that an arsenate reductase gene has been at least putatively
detected either chromosomally or in a plasmid in over 100 microbial genera, including three
of particular interest for this study: Agrobacterium, Arthrobacter and Bacillus
(http://cmr.tigr.org). Previous research further supports that microorganisms with the ability
to reduce arsenate are ubiquitous across myriad soil environments (Macur et al., 2001;
Jackson and Dugas, 2003; Macur et al., 2004). Although there are interesting variations in
both the structure of the ars operon and the mechanisms by which particular ars genes are
regulated, the primary function of this operon (reduction of arsenate to arsenite via ArsC and
exclusion of arsenite from the cell via ArsA,B efflux pumps) is largely conserved across
different phyla (Mukhopadhyay et al., 2002; Silver et al., 2002).
Previous literature
(Gladysheva et al., 1994; Ji et al., 1994; Mukhopadhyay et al., 2000; Zhou et al., 2004)
supports that the ability of phosphate to inhibit arsenate reductase activity varies by species.
For example, phosphate did not inhibit the activity of LmACR2, the arsenate reductase in
Leishmania major (Zhou et al., 2004), the Saccharomyces cerevisiae arsenate reductase,
Acr2p (Mukhopadhyay et al., 2000) or the arsenate reductase located in plasmid pI258 of
58
Staphylococcus aureus (Ji et al., 1994); however, it was shown to be an inhibitor of arsenate
reduction by the arsenate reductase located in the Escherichia coli plasmid R773
(Gladysheva et al., 1994). How phosphate may affect the uptake of arsenate or ArsC activity
in countless other organisms is yet to be determined.
Consequently, the goal of the current study was to evaluate the effects of phosphate on
the microbial transformation of arsenic in five As-transforming bacteria. Specific objectives
were to (i) evaluate the oxidation of arsenite by several known arsenite oxidizing organisms
as a function of P:As ratios, (ii) determine the effects of phosphate on arsenate reduction and
cell growth in organisms known to possess arsC genes (e.g. known arsenate reducers), and
(iii) evaluate whether high phosphate:arsenate has any indirect role on the expression of arsC
in a specific arsenate-reducing microorganism, Agrobacterium tumefaciens str. 5B. It was
hypothesized that high ratios of phosphate relative to arsenate will competitively reduce AsV
uptake by cells via phosphate transporters. As a result, it was further hypothesized that (i)
microbial arsenate reduction would be inhibited, (ii) the upper concentration threshold of As
exposure before a reduction in cell growth is observed would be extended, and (iii) arsC
expression would be reduced. Conversely, AsIII oxidation was not expected to be affected by
phosphate given that (i) no competition for uptake between arsenite and phosphate has been
suggested and (ii) phosphate is transported into the cell’s cytoplasm while arsenite oxidases
are believed to exist in the periplasm.
59
Materials and Methods
Isolate Selection and Preparation
Five As-transforming microorganisms previously isolated from soils with long-term As
contamination were selected for use in liquid culture experiments.
Specifically, the
microorganisms included were two known arsenite-oxidizing organisms – Variovorax
paradoxus-like (99.3% similarity) str. RM1 and Agrobacterium tumefaciens-like (99.9%
similarity) str. 5A – and three known arsenate-reducing organisms – Agrobacterium
tumefaciens-like (99.9% similarity) str. 5B, Bacillus sp.-like (99.8% similarity) str. S18, and
Arthrobacter sp.-like (99.5% similarity) str. S6. Three of these organisms (Variovorax
paradoxus-like str. RM1 and Agrobacterium tumefaciens-like strs. 5A and 5B) were isolated
previously (Macur et al., 2004) from aerobic column experiments conducted using
agricultural soil with prior exposure to As-rich irrigation water from the Madison River
(Gallatin County, MT).
The remaining two organisms (Bacillus sp.-like str. S18 and
Arthrobacter sp.-like str. S6) were isolated from soil samples impacted by aerial As
contamination from several copper smelters near Anaconda, MT (N 46.10313o W 112.87296o
(see Appendix C, Near Substation section for further details).
Prior to their use as inoculum for As transformation experiments, these microorganisms
(which will hereafter be identified simply by genus and strain designation) were grown in
individual autoclaved centrifuge vials containing 200 mL of synthetic soil solution media
(SSE), modified from Macur et al. (2004) to contain NH4NO3 (1.25 mM), MgCl2 (1 mM),
KH2PO4 (0.05 mM), KOH (0.25 mM), FeCl2-4H2O/disodium EDTA (0.02 mM), CaSO4 (2
mM), glucose (5 mM) yeast extract (2 mgL-1), MOPS buffer (5 mM) and 1 mL L-1 trace
60
metals solution modified from Newman et al. (1997) to exclude FeCl2. Once stationary
growth was achieved, the culture was centrifuged at 5000 rpm for 40 – 60 min. Pelleted cells
were sterilely aspirated to remove supernatant then re-suspended in approximately 30 mL of
sterile SSE media containing no KH2PO4 and an original optical density was determined at
A500 (Hitachi U-2000).
Liquid Culture Experiments
Duplicate autoclaved glass 118 mL culture vials were filled with sterile SSE media
(without KH2PO4) and spiked with either arsenite (as NaAsO2) or arsenate (as Na2HAsO4)
plus phosphate (as KH2PO4) in one of the P:As ratio combinations listed in Table 3.1 (to a
known total volume of 80 - 85 mL). Filled vials were inoculated with stationary cells to
obtain a uniform initial optical density. Cultures and duplicate sterile controls (containing
only SSE and either AsIII or AsV) were incubated (30oC) for approximately a 48-hour period
on a shaker (120 rpm). Two 5-mL solution samples were extracted and filtered (0.22 µm) at
0, 8, 24 and ~48 hours; one sample was oxidized to contain only AsV via the sodiumborohydride pre-treatment method described in the Materials and Methods section of
Chapter 2. The original, “AsTotal,” sample and the borohydride-treated sample, containing
only AsV, were 2 and 3% acidified, respectively, with 12M HCl and stored at 4oC until
analyzed for As using hydride generation-atomic absorption spectrometry (HG-AAS;
Hydride system: Varian VGA 77; AAS: Perkin Elmer 3100) or inductively coupled plasma
spectrometry (ICP). The AsIII concentration was calculated from the difference of these
values. This method has been thoroughly evaluated in a past study (Jones et al., 2000) and
“check standards” prepared using fresh arsenite and arsenate stock solutions were included in
61
the analysis suite to further validate sample data. At each sampling point, samples were also
obtained for optical density measurement (3.5 mL) at 500 nm (OD500) and for subsequent
analysis of mRNA transcripts (3 mL were immediately frozen at -80oC).
Table 3.1. Concentrations of phosphate and arsenic, and corresponding P:As ratios used in
experiments to examine effects of phosphate on either the oxidation of arsenite or the
reduction of arsenate. A P:As ratio of 0.5 was used to confirm organism As phenotype.
Increases in phosphate were examined at different absolute concentrations of As and P to
achieve similar ratios of either 5 or 10.
P (µM)
As (µM)
P:As
50
500
1000
50
50
100
100
100
10
5
0.5
5
10
5
10
Isolate Confirmation through Full-Length Sequencing
Periodically during experimentation, 10 µL of suspended cells were extracted from the
culture vials and diluted with 90 µL DNase-free water. This DNA template was used in
polymerase chain reactions (PCR) which amplified a 1384 bp region of the 16S rRNA gene
using the Bacteria-specific Bac-8 forward (5’-AGAGTTTGATCCTGGCTCAG-3’) and
1392 reverse (5’-ACGGGCGGTGTGTA-3’) primers. The optimized thermocycler protocol
included 6 min initial denaturing at 94oC, 25 cycles including 94oC for 45 s, 55oC for 45 s
and 72oC for 55 s followed by a 7 min final extension at 72oC. The products were purified,
quantified and sequenced by TGen (Pheonix, AZ). Sequences were edited and aligned with
Sequencer 4.2 software (Gene Codes Corporation, Ann Arbor, MI) and compared to known
62
sequences in the GenBank database using BLAST (NCBI, Bethesda, MD) to confirm
inoculum purity.
Amplification of arsC mRNA transcripts
The effect of phosphate and arsenic concentration on the expression of arsC during
growth of Agrobacterium str. 5B was evaluated by first extracting RNA from cell
suspensions using the FastRNA Pro Blue Kit (Q-Biogene). A 346 bp fragment of arsC
mRNA
was
PCR-amplified
using
the
Atume-arsC
forward
(5’-ACCCTCGCACTCATTGAGC-3’) and reverse (5’-ACCTGCTCGCCGTCTTCT-3’)
primers; the design was based on the known arsC sequences in the Agrobacterium strains 5A
and 5B (Macur et al., 2004; accession # AY286230 and AY286231).
The PCR mix
contained 1 μM of each primer. The initial generation of cDNA using the Access RT-PCR
System (Promega Corp.) was followed by a PCR protocol of 95oC for 2 min, 40 cycles of
95oC for 45 s, 50oC for 45 s, 72oC for 50 s, and final extension of 72oC for 5 min. To ensure
that the correct target sequences were amplified, purified PCR products were cloned into the
pGEM-T Vector System (Promega, Madison, WI) and the clones were sequenced.
Results and Discussion
Effects of Phosphate on Microbial Oxidation of Arsenite
The 0.5 (50 µM P and 100 µM As) ratio treatment served as a baseline to verify that the
isolates used in the study were expressing their usual ‘As-phenotype’. As expected, the two
oxidizing strains (Variovorax str. RM1 and Agrobacterium str. 5A) converted arsenite to
arsenate within ~ 48 hours after inoculation (Figure 3.1). Sterile control trials showed
63
average abiotic oxidation of 1.25% and neither arsenite-oxidizer showed considerable
arsenate reducing capability when grown in media containing only arsenate, regardless of
phosphate concentration applied. The amount of arsenite relative to total As decreased from
100% to less than 10% by the ~48-hour time point in all ten experiments involving different
ratios of phosphate to arsenite involving either of the arsenite-oxidizing strains. Significant
differences (α =0.05; univariate ANOVA; see Appendix B, Tables B.1-5 for all ANOVA
data) in the amount of arsenite oxidized were minimal at the conclusion of each P:arsenite
ratio experiment (Figure 3.1B). Despite the statistical difference, the data indicate that the
two arsenite-oxidizing organisms still achieved near complete arsenite oxidation (>90%)
regardless of the P:arsenite ratio treatment applied, suggesting that the presence of phosphate
has no apparent effect on the oxidation process (Figure 3.1).
Effects of Phosphate on Microbial Reduction of Arsenate
All three AsV-reducing isolates (Agrobacterium str. 5B, Arthrobacter str. S6, and
Bacillus str. S18) exhibited efficient reduction of arsenate to arsenite within 48 hours in the
presence of 100 µM arsenate and 50 µM phosphate (P:As ratio = 0.5) (Figure 3.2). The
Agrobacterium str. 5B and Arthrobacter str. S6 showed near complete reduction (>98%) of
arsenate within 48 hours, while the Bacillus str. S18 isolate showed approximately 74%
arsenate reduction by 48 hours. The Bacillus str. S18 is a less vigorous arsenate-reducing
strain when compared to the Agrobacterium and Arthrobacter strains used in this study.
Sterile controls sampled over the same time frame exhibited less than 4% reduction to
arsenite.
64
A) Agrobacterium tumefaciens str. 5A
a
B) Variovorax sp. str. RM1
P (µM) As (µM)
50
50
500
50
1000
100
5
100
10
100
P:As
0.5
10
5
5
10
a
b
Figure 3.1. Percent of arsenite relative to total As plotted as a function of time in
experiments containing arsenite-oxidizing strains (A) Agrobacterium tumefaciens str. 5A,
and (B) Variovorax sp. str. RM1, each subjected to five phosphate:arsenite ratios. The
amount of arsenite oxidized by 48 hrs is essentially independent of the P:AsIII ratio. Lower
case letters denote significant differences at α=0.05, and error bars indicate standard
deviation of duplicate cultures; when apparently absent, error bars are contained within
symbol.
65
A) Agrobacterium tumefaciens str. 5B
a
P (µM) As (µM)
50
50
500
50
1000
100
5
100
10
100
P:As
0.5
10
5
5
10
b
B) Arthrobacter sp. str. S6
a
b
c
C) Bacillus sp. str. S18
a
b
c
d
Figure 3.2. Percent of arsenate relative to total As plotted as a function of time in
experiments containing arsenate–reducing strains (A) Agrobacterium tumefaciens str. 5B,
(B) Arthrobacter sp. str. S6 and (C) Bacillus sp. str. S18, each subjected to five
phosphate:arsenate ratios. The amount of arsenate oxidized by 48 hrs is dependent on the
P:AsV ratio. Lower case letters denote significant differences at α=0.05, and error bars
indicate standard deviation of duplicate cultures; when apparently absent, error bars are
contained within symbol.
66
At 100 µM arsenate, reduction was nearly completely inhibited (Figure 3.2) when
phosphate concentrations were increased to 500 and 1000 µM (P:As ratios of 5 and 10).
These effects were observed for all three arsenate-reducing isolates, suggesting that high
phosphate to arsenate ratios in natural systems may decrease the likelihood that arsenate will
become reduced to arsenite via microbial detoxification processes. However, results from
experiments conducted at identical P:As ratios (5 and 10), but with relatively low arsenate
concentrations (5 and 10 µM) and a constant P concentration (50 μM), showed no significant
differences (α=0.05) in arsenate reduction compared to baseline (50 µM P and 100 µM AsV)
results, where nearly 100 percent of the arsenate was reduced (Figure 3.2). Consequently,
high P:As ratios alone are not sufficient to inhibit microbial reduction of arsenate, and the
absolute concentration of phosphate is an important factor in addition to the overall P:As
ratio. Finally, none of the three arsenate-reducers showed considerable AsIII oxidation when
grown in media containing only arsenite, independent of the phosphate concentration
applied. (Growth curves and arsenic totals remained uniform, respectively, for each species
across the five ratio trials; see Appendix B for plots of these relationships and additional data,
Figures B.3-15.)
Effects of Phosphate on Cell Growth in the Presence of High Arsenic
Isolates were subjected to additional experiments in the presence of either 1000 µM
arsenite or arsenate (for reducing or oxidizing isolates, respectively) plus 50 µM phosphate
(P:As = 0.05) to query the organisms ability to transform significant levels of As (Figures 3.3
and 3.4). All isolates except the Bacillus str. S18 continued to exhibit their expected Asphenotype at this elevated As concentration. However, a lag or slower AsIII oxidation
67
(Figure 3.3) or AsV reduction (Figure 3.4) is noted when plotted as a percent of AsTotal. The
actual maximum rates of AsV-reduction in the 50 µM P:1000 µM As experiments for the
Agrobacterium, Arthrobacter and Bacillus strains were 4.46 x 10-11, 1.64 x 10-11, and 1.75 x
10-9 millimoles AsV reduced hour-1 cell-1, respectively, and each was higher than the
maximum rate of reduction calculated at 100 µM arsenate and an identical phosphate
concentration of 50 µM for each organism.
A 10 - 82% decrease in cell growth was observed in experiments containing 1000 µM As
(Figure 3.5). It is noteworthy that all isolates with the exception of Bacillus str. S18 were
more vigorous than expected despite this extreme As pressure.
The growth of the
Arthrobacter strain appeared to be affected the least by the high As pressure (Figure 3.5B),
although its arsenate-reducing capability at 1000 µM As appeared delayed compared to the
100 µM As treatment (Figure 3.4B). To determine if increased phosphate concentration
would ameliorate high As-induced reduction in cell growth, the AsV-reducing bacteria were
treated with 1000 µM As and 1000 µM P. (Arsenite-oxidizing isolates were not subjected to
this treatment as phosphate previously showed no effect on studies utilizing arsenite.) The
high phosphate treatment appeared to have the greatest affect on the growth of Bacillus str.
S18. This organism is clearly sensitive to arsenate at concentrations of 1000 µM; however,
the addition of 1000 µM phosphate is sufficient to alleviate this toxicity (Figure 3.5C).
Increased phosphate concentrations of 1000 µM also successfully inhibited arsenate
reduction for all three arsenate-reducing isolates, similar to results observed in the baseline,
0.5 P:As ratio, experiment (Figure 3.4).
68
A) Agrobacterium tumefaciens str. 5A
P (µM) As (µM)
50
50
P:As
1000
0.05
100
0.5
B) Variovorax sp. str. RM1
Figure 3.3. Oxidation of arsenite by arsenite-oxidizing strains (A) Agrobacterium
tumefaciens str. 5A and (B) Variovorax sp. str. RM1 in the presence of 1000 µM As. The
0.5 ratio study (50 µM P:100 µM As) serves as the organism’s baseline oxidation trend.
Error bars indicate standard deviation of duplicate cultures; when apparently absent, error
bars are contained within symbol.
69
A) Agrobacterium tumefaciens str. 5B
P (µM) As (µM)
P:As
50
1000
50
0.05
1.0
1000
1000
100
0.5
B) Arthrobacter sp. str. S6
C) Bacillus sp. str. S18
Figure 3.4. Reduction of arsenate by arsenate-reducing isolates (A) Agrobacterium
tumefaciens str. 5B, (B) Arthrobacter sp. str. S6 and (C) Bacillus sp. str. S18 at high levels
of arsenate (1000 µM) and the effects of elevated phosphate (1000 µM) on arsenate
reduction. The 0.5 ratio study (50 µM P:100 µM As) serves as the organism’s baseline
reduction trend. Error bars indicate standard deviation of duplicate cultures; when
apparently absent, error bars are contained within symbol.
70
A) Agrobacterium tumefaciens str. 5B
B) Arthrobacter sp. str. S6
C) Bacillus sp. str. S18
D) Agrobacterium tumefaciens str. 5A
E) Variovorax sp. str. RM1
Figure 3.5. Effect of 1000 µM As on cell vitality and ameliorating effects of elevated
phosphate (1000 µM). Arsenate-reducing isolates were subjected to treatments at 50
µM P:100 µM As (
), 50 µM P:1000 µM As (
), and 1000 µM P:1000 µM
As (
). Arsenite-oxidizing isolates were not treated with the high phosphate
treatment. Error bars indicate standard deviation of duplicate cultures; when
apparently absent, error bars are contained within symbol.
71
Possible Mechanisms Controlling Phosphate Inhibition of Arsenate Reduction
The inhibition of arsenate reduction by phosphate could be the result of several possible
factors. One of the hypotheses tested was that increased phosphate:arsenate causes arsenatereducing isolates to down-regulate the production of ArsC at, or prior to, transcription.
However, analysis of arsC mRNA using gene-specific PCR primers revealed that high P:AsV
failed to inhibit transcription of arsC in Agrobacterium str. 5B, even when a high absolute
phosphate concentration (1000 µM) was used (Figure 3.6). A similar result was reported in a
previous study (Saltikov et al., 2005), where three different P:AsV ratios (0.02, 0.12 and 1)
failed to reduce arsC expression in Shewanella sp. str. ANA-3.
- Reverse
Transcriptase
+ Reverse
Transcriptase
Time (hrs) 0
Replicate
8
a
24
b a
b
0
8
a
24
b a
b
100 µM As:50 µM PO4
arsC
mRNA
100 µM As:1000 µM PO4
arsC
mRNA
Figure 3.6. Expression of the arsenate reductase gene, arsC, in Agrobacterium
tumefaciens str. 5B as a function of time (0 – 24 hours) and phosphate concentration (50
and 1000 µM).
72
In the current study, expression of the arsenate reductase gene was not observed at T0,
suggesting that either arsC is not a constitutively expressed gene in this organism or that the
copy number was below detection for this protocol. However, expression was observed at
the 8 and 24-hour time points for both the 50 µM P:100 µM As and 1000 µM P:100 µM As
ratio experiments, despite the fact that essentially no AsV reduction was observed in the latter
experiment. These results suggest that transcription of the isolate’s arsenate reductase is not
inhibited by high phosphate to arsenate ratio or P concentration and that the decreased
reduction of AsV observed in the high phosphate:arsenate experiments (ratios of 5 and 10) is
the result of some other mechanism, such as competitive inhibition at the uptake or
enzymatic level.
It is possible that arsenate reduction may have been inhibited as a result of enzymatic
substrate (arsenate) exclusion at the cell transport level. Literature supports that to conserve
time and energy many organisms possess at least two genetically differentiated systems of
phosphate transport. Both high- and low-affinity phosphate transport systems have been
identified in Bacillus cereus (Rosenberg et al., 1969), Escherichia coli (Rosenberg et al.,
1977; Willsky and Malamy, 1980), Acinetobacter johnsonii (van Veen et al., 1993),
Saccharomyces cerevisiae (Tamai et al., 1985; Martinez et al., 1998), Pseudomonas
aeruginosa (Lacoste et al., 1981), Rhizobium meliloti (Voegele et al., 1997) and Rhizobium
tropici (Botero et al., 2000). Though differences exist in the gene structure, specific affinity
(KM) and ion specificity of phosphate transport systems, many characteristics appear
universal within each affinity classification.
73
The low-affinity system appears less efficient at scavenging phosphate at very low
external concentrations as indicated by KM values typically one to two orders of magnitude
higher than the same organism’s high-affinity system (Willsky and Malamy, 1980; Lacoste,
1981; Tamai et al., 1985; van Veen et al., 1993; Voegele et al., 1997; Botero et. al, 2000).
Additionally, low-affinity systems are used for a variety of substrates therefore they are
typically constitutively expressed; this can be afforded as these systems are typically driven
by chemiosmotic gradients (van Veen, 1997). Conversely, high-affinity systems are more
efficient in the binding and thus transport of phosphate ions and are likely less efficient at
arsenate transport (Willsky and Malamy, 1980; Rosen, 2002). In fact, van Veen et al. (1997)
estimated that the E. coli high-affinity system (Pst) has a 100-fold greater affinity for
phosphate than arsenate. Often driven by energy-expensive ATP hydrolysis, high-affinity
systems are generally induced only in stress conditions when phosphate concentrations are
limited.
The external phosphate concentration necessary to induce high-affinity transport varies
with the organism, but is loosely defined in the literature as being “low (µM) concentrations”
(Martinez et al., 1998) or “below the millimolar range” (Harris et al., 2001). Additional
literature based on specific species suggests a more concrete quantification of the phosphate
concentration necessary to derepress high-affinity transport is between 1 and 100 µM.
Specifically, alkaline phosphatase activity, a good indicator that the high-affinity system has
been derepressed, was shown to greatly increase in Rhizobium meliloti when external
phosphate concentration decreased to approximately 10 µM (Al-Niemi et al., 1997).
Interestingly, alkaline phosphatase activity in Rhizobium tropici does not induce until
74
concentrations reach 1 µM (Botero et al., 2000). Derepression of the high-affinity phosphate
transport system in Saccharomyces cerevisiae is reported to occur when external phosphate
concentrations are below 100 µM, with highest transport activity when concentrations near
30-40 µM (Mouillon and Persson, 2005). At the other extreme, Rosenberg et al. (1977)
showed that the E. coli high-affinity system was repressible when cells were grown in media
containing phosphate concentrations of 1 mM or higher. The inhibition of arsenate reduction
was observed in the current study when high concentrations of phosphate (>500 µM) were
utilized. Consequently, based on previous literature, it is likely that the isolates’ low-affinity
systems were responsible for primary phosphate transport activity. Since the high affinity
transport system was likely not employed, arsenic should not have been systematically
excluded from cell uptake. However, it is conceivable that low affinity P transport systems
exhibit some competition between phosphate and arsenate, where very high phosphate
concentrations may reduce arsenate uptake simply due to competitive exclusion.
Alternatively, once inside the cell, phosphate may have acted as a competitive inhibitor,
binding to the active site of ArsC, thus preventing arsenate from binding and subsequently
being reduced to arsenite. As mentioned previously, this mechanism has been documented in
E. coli (Gladysheva et al., 1994); however, phosphate was shown to only weakly inhibit
ArsC activity in this organism (inhibition constant, Ki = 30 mM).
The inhibitory effect of phosphate in the P:AsV experiments involving Agrobacterium
tumefaciens str. 5B, Arthrobacter sp. str. S6 and Bacillus sp. str. S18 was modeled using
Michaelis-Menten enzyme kinetic expressions. The maximum velocity of the reduction
reaction (VMAX) and the Michaelis constant (KM) for each isolate were estimated based on
75
Equation 3.1 and optimized using MS Excel Solver software, using the respective maximum
arsenate-reduction velocity (υ, in mmol AsV hr-1 cell-1) and corresponding arsenate
concentration (S1, in mM) for the four P:AsV experiments that included an initial phosphate
concentration of 50 µM (Table 3.1).
(See Appendix B, Figures B.16, 18, and 20 for
calculations.)
v =
VMAX
K
1+ M
[ S1 ]
(3.1)
The predicted VMAX and KM values were then used to estimate and optimize the phosphate
inhibition constant (Ki, in mM) for each isolate via Equation 3.2 using the respective
maximum arsenate-reduction velocity (υ, in mmol AsV hr-1 cell-1) in addition to the
corresponding phosphate concentration (I, in mM) and arsenate concentration (S2, in mM) for
the three P:As experiments that included an initial arsenate concentration of 100 µM (Table
3.1). (See Appendix B, Figures B.17, 19, and 21 for calculations.)
v =
KM
VMAX ⋅ [ S 2 ]
⎛
[I ] ⎞
⎜⎜1 +
⎟ + [S 2 ]
K i ⎟⎠
⎝
(3.2)
The results of this modeling effort (Table 3.2) confirm that phosphate is a strong
competitive inhibitor of arsenate reduction in Agrobacterium str. 5B (Ki = 0.024 mM),
Arthrobacter str. S6 (Ki = 0.090 mM), and Bacillus str. S18 (Ki = 0.105 mM). Although this
data supports the idea that phosphate acts as a competitive inhibitor of arsenate reduction in
76
these arsenate-reducing isolates, it does not specify whether inhibition occurs during
phosphate transport or arsenate reduction. Therefore, the inhibition of arsenate reduction that
was observed in this study may be due to reduced arsenate uptake into the cells and or the
reduced binding of arsenate to the ArsC active sites, both seemingly as a result of
competition from the increased concentration of phosphate ions present at P concentrations
exceeding 500 µM.
Table 3.2. Calculated kinetic values for three known arsenate-reducing isolates
(Agrobacterium str. 5B, Arthrobacter str. S6 and Bacillus str. S18) based on MichaelisMenten modeling. Calculated parameters included: maximum reduction velocity (VMAX),
Michaelis constant (KM), and inhibition constant (Ki).
Isolate
Agrobacterium str. 5B
Arthrobacter str. S6
Bacillus str. S18
VMAX
-1
-1
(mmol hr cell )
KM
R
(mM)
2
2
Ki
R
(%)
(mM)
(%)
5.0 x 10
-11
0.055
99.9
0.024
97.1
1.9 x 10
-11
0.070
99.0
0.090
88.7
-9
0.075
98.8
0.105
88.7
1.75 x 10
Conclusions and Implications
High phosphate concentrations were shown to inhibit the microbial reduction of arsenate
to the more toxic arsenite form both empirically and through Michaelis-Menten enzyme
kinetic modeling. Conversely, P:As ratios had no affect on the microbial oxidation of
arsenite by known arsenite-oxidizing organisms. The absolute concentration of phosphate is
important as well as the ratio of phosphate to arsenate. It appears that for the three arsenatereducing isolates, a phosphate concentration threshold must be exceeded to effectively inhibit
reduction; based on the current study, this threshold is between 50 and 500 µM. The
inhibition of arsenate reduction observed in this study was not a result of decreased arsenate
77
reductase expression. Furthermore, systematic exclusion of arsenate by the cell’s transport
system did not appear responsible for limiting arsenate reduction. Rather, less arsenate was
microbially reduced to arsenite via the ars operon because the elevated amount of phosphate
ions present in the media (i) allowed less arsenate to be taken into the cells due to
competition for phosphate membrane transporters, and or (ii) prevented arsenate from
binding to the ArsC active sites.
All five isolates included in this study showed lessened growth or As-transforming
capability when grown in high As (1000 µM) conditions, though the Arthrobacter sp. str. S6
appeared most tenacious. The addition of a high phosphate concentration (1000 µM) both
increased cell growth and inhibited reduction in arsenate-reducing isolates, likely by limiting
cell exposure to arsenate through competition for cell uptake.
A better understanding of the factors which control formation of the more toxic and
bioavailable arsenite form in soils and natural waters has implications for reducing chronic
human As poisoning, increasing compliance with the new U.S. drinking water regulation for
arsenic and for bioremediation of As-contaminated lands. The findings of this study add
justification to the theory that phosphate concentration in soils or natural waters may be an
important parameter indirectly limiting the formation of arsenite via microbial processes.
Hence, phosphate addition may have potential use in As-bioremediation; however, additional
column studies and field experiments which consider the effects of increased phosphate on
arsenic desorption are necessary before this method is put into practice. Tassi et al. (2004)
reported increased success of lupine As-phytoextraction when this As-accumulating crop was
grown in previously As-contaminated soil amended with multiple biammonium phosphate
78
applications. Hence, the use of phosphate in arsenic bioremediation may be more effective if
coupled with the phytoextraction of desorbed arsenic by As-hyperaccumulating plants.
79
4. SUMMARY AND CONCLUSIONS
Summary of Problem
Arsenic is a naturally ubiquitous element in the Earth’s crust; however, anthropogenic
actions have led to an increase in concentrations of arsenic in soil and water systems. In
some instances, concentrations are up to three orders of magnitude greater than values
accepted as background levels (Francesconi and Kuehnelt, 2002; Nriagu, 2002; Belluck et
al., 2003).
While century-old former mining and smelting operations have contributed
significantly to the global arsenic dilemma (Freeman et al., 1995; Jones et al., 1997; US EPA,
1998; US EPA 2006c), arsenic soil and water contamination should not be perceived as only
a past transgression that we are presently remediating.
For example, since the 1970s
chromated copper arsenic (CCA) has been the primary wood preservative used in the U.S.
(US EPA, 2006b). Although it has been phased out for residential use since 2004 (US EPA,
2006b), CCA-treated lumber is still quite prevalent in outdoor wooden structures, which can
readily release arsenic by leaching or decomposing when discarded (Nriagu, 2002; Belluck et
al., 2003). Furthermore, 3-nitro-4-hydroxybenzene arsonic acid (roxarsone) is presently used
extensively in the U.S. as a chicken feed additive to stimulate growth and prevent disease in
commercial broiler operations (Schaefer, 2007).
However, Stolz et al. (2007) recently
reported that Clostridium species prevalent in chicken excrement can transform this
organoarsenical to the more toxic inorganic arsenic form in less than 10 days. Roxarsone is
not readily retained by the broilers but rather excreted; since an estimated 90% of
commercial chicken litter is applied to land (Stolz et al., 2007), roxarsone usage is a very real
example of present day land and water As contamination.
80
Summary of Objectives and Conclusions
Microbial interactions can increase the bioavailability of arsenic in soil and water
environments by inducing the reduction of arsenate to the more mobile arsenite form
(Oremland and Stolz, 2003). Biogeochemical processes catalyzed by microbiota have even
been implicated as important reactions contributing to the massive arsenic epidemic in
Bangladesh (Harvey et al., 2002). Therefore, a better understanding of the factors that cause
microbial communities to exhibit net arsenite oxidation or net arsenate reduction is
imperative for preventing further pollution, limiting human exposure, and ameliorating
present contamination. In this thesis, experiments were conducted to address arsenic redox
transformations as a function of arsenic concentration in column experiments (Chapter 2),
and to investigate the affect of phosphate on arsenate oxidation-reduction mediated by known
As-transforming microorganisms (Chapter 3). Specific objectives of these studies were to:
(i)
Determine the effects of arsenite or arsenate concentration on microbial diversity
and ability to transform arsenic by soil populations within an aerobic column
transport system.
(ii)
Isolate microorganisms from column transport study and elucidate their dominant
arsenic redox phenotype.
(iii)
Evaluate the effects of phosphate concentration and P:As ratio on biological
responses to arsenate and arsenite.
The primary conclusions of experiments designed to address these objectives are as
follows:
81
(i)
Arsenite oxidation was the net microbial transformation process observed in
aerobic column experiments using previously uncontaminated soil at arsenite or
arsenate concentrations of 2 and 20 mg L-1. Conversely, values of 200 mg L-1
inhibited microbial arsenite oxidation. However, microbial diversity was not
significantly reduced as a function of increasing As concentration and numerous
isolates were obtained from columns treated with high arsenic (200 mg L-1).
(ii)
Sixty-two isolates were cultivated from porous media samples obtained at the
termination of column experiments including representatives from α−, β−, and γProteobacteria, Actinobacteria, Firmicutes and Flavobacteria. Forty-three of these
isolates were shown to be capable of arsenate reduction, while only one isolate (a
Mesorhizobium sp.-like strain) was capable of arsenite oxidation.
(iii)
The addition of phosphate inhibited the microbial reduction of AsV in three
arsenate-reducing organisms. Further, high phosphate ameliorated As-induced
cell growth reduction in the presence of high (1 mM) arsenic pressure; however,
the absolute concentration of phosphate, in addition to a high phosphate to
arsenate ratio, was necessary to inhibit microbial arsenate reduction. While high
phosphate:arsenate ratios effectively shut down microbial arsenate reduction, the
expression of the arsenate reductase gene (arsC) was not inhibited under these
conditions in the arsenate-reducing isolate, Agrobacterium tumefaciens str. 5B.
Consequently, phosphate likely inhibits arsenate reduction by impeding arsenate
uptake by the cell via phosphate transport systems or by competitively binding to
the active site of the ArsC.
82
Implications of the Current Work
Soils are among the most complex and difficult systems to analyze accurately, in part due
to the tight coupling of physical, chemical and biological processes. For example, it is
currently very difficult to quantify the immense diversity and abundance of microflora in
soil, frequently exceeding one billion individuals g-1 (Sylvia et al., 1999), much less
understand whether several populations may dominate microbial community activity at any
given time. Many molecular techniques may be complicated by biases such as the presence
of humic or fulvic acids known to inhibit PCR reactions, depending on the primer sets and
the templates. Results obtained in the current work using a previously uncontaminated soil
raise numerous questions regarding the impacts of As on soil microbial communities and
provide another example of the difficulty involved with linking microbial processes with
specific individual species and their physiologies in culture. For example, although the
results clearly documented that arsenite oxidation was the net As transformation mediated by
the microbial community, only 1 of the 62 isolates cultivated from this soil was shown to be
capable of oxidizing arsenite while 43 were shown to reduce arsenate.
There are many possible reasons for this apparent inconsistency. Arsenite-oxidizing
organisms could be less prevalent in soil environments than arsenate reducers. However, a
recent study (Inskeep et al., 2007) identified over 170 phylogenetically diverse arsenite
oxidase gene sequences from geographically widespread soil environments, including 3
sequences from the Amsterdam soil extracted from the aforementioned column studies
(Chapter 2) and 12 from the Gardner Ditch and Substation soils detailed in Appendix C.
83
Prior to the Inskeep et al. (2007) study, a mere 13 microbial genera had been shown to
possess an arsenite oxidase-like gene sequence, while arsenate reductase genes (arsC) had
been confirmed or putatively identified in over 100 genera (TIGR-CMR; http://cmr.tigr.org).
Since fewer arsenite oxidase sequences are available for the construction of degenerate
primers compared to arsenate reductases, it is conceivable that current amplification
techniques are biased against the identification of oxidase gene sequences of common soil
organisms.
Moreover, arsenite oxidases are not highly conserved across phylogenetically
diverse organisms (Inskeep et al., 2007), which further complicates effective primer design.
Another important consideration is that the currently known bacterial arsenite oxidases
(AroA, AsoA, AoxB) may only represent one type of enzymes capable of mediating this
redox process.
The precedent for this hypothesis originates in part from the fact that
microorganisms clearly exhibit diverse strategies for mediating As transformation and that
some known arsenite oxidizing organisms do not appear to have aroA-like genes (Inskeep et
al., 2007). Finally, it is critical to remember that cultivation protocols certainly have their
biases and it is possible that the cultivation techniques employed in the current study favored
the isolation of arsenate reducing microorganisms over arsenite oxidizers. Together, these
potential biases and unknowns pose a significant limitation to fully understanding microbial
contributions to the global arsenic cycle.
Many environmental factors beyond the scope of this project also contribute to arsenic
cycling, such as soil water content. Water saturated soils lead to low oxygen diffusion rates,
which would exert a greater selection pressure towards dissimilatory arsenate reducing
microorganisms (Stolz and Oremland, 1999).
Microbial transformations under anoxic
84
conditions were not addressed in the current study; nonetheless, microorganisms with the
capability to respire on arsenate may become active in microaerobic to anaerobic
environments and play an important role in the chemical speciation of As. Future studies to
determine if these organisms contribute to arsenate reduction in anaerobic conditions are
warranted.
The complexity of microbial interactions with arsenic is further complicated by the
competitive relationship between arsenate and phosphate, and the fact that the P:As ratio may
also influence the net microbial transformation of As in soils and natural waters. In the
current work (Chapter 3), the presence of high phosphate was shown to inhibit microbial
reduction of arsenate by aerobic arsenate reducing organisms, thus high P:As may minimize
the formation of arsenite, a more mobile and toxic arsenic form.
However, the same
phenomenon responsible for phosphate inhibition of arsenate reduction (i.e. the chemical
similarity between phosphate and arsenate) can also result in increased mobilization of
arsenic (Darland and Inskeep, 1997; Peryea and Kammereck, 1997; Alam et al., 2001). As a
chemical analog, phosphate often competes with arsenate for soil sorption sites, thus leading
to an increase in soluble arsenic concentrations. Consequently, the potential for phosphate to
lessen or increase arsenic mobility in soils will depend in part on the balance between
competition for sorption sites and ability to prevent reduction to the more mobile arsenite
form. From a practical viewpoint, how do we prevent further human or land exposure to
arsenic and economically comply with the current arsenic statutes?
Further research
regarding abiotic and biotic controls on arsenic cycling is, in part, the answer. In the
meantime, more emphasis needs to be placed on minimizing the amount of As introduced to
85
the environment, and based on recent research, it appears that the U.S. EPA and the poultry
industry should start by ensuring that arsenic is kept out of chicken feed!
86
LITERATURE CITED
Afkar, E., J. Lisak, C. Saltikov, P. Basu, R.S. Oremland, and J.F. Stolz. 2003. The
respiratory arsenate reductase from Bacillus selenitireducens strain MLS10. FEMS
Microbiol. Letters 226: 107-112.
Alam, M.G.M., S. Tokunaga, and T. Maekawa. 2001. Extraction of arsenic in a synthetic
arsenic-contaminated soil using phosphate. Chemosphere 43: 1035-1041.
Allègre, C.J., T. Staudacher, and P. Sarda. 1986. Rare gas systematics: formation of the
atmosphere, evolution and structure of the Earth’s mantle. Earth and Planetary Science
Letters 81: 127-150.
Al-Niemi, T.S., M.L. Summers, J.G. Elkins, M.L. Kahn, and T.R. McDermott. 1997.
Regulation of the phosphate stress response in Rhizobium meliloti by PhoB. Appl. and
Environ. Microbiol. 63: 4978-4981.
Anderson, G.L., J. Williams, and R. Hille. 1992. The purification and characterization of
arsenite oxidase from Alacaligenes faecalis, a molybdenum-containing hydroxylase. J.
Biol. Chem. 267: 23674-23682.
Battaglia-Brunet, F., Y. Itard, F. Garrido, F. Delorme, C. Crouzet, C. Greffiè, and C. Joulian.
2006. A simple biogeochemical processes removing arsenic arsenic from a mine
drainage water. Geomicrobiology J. 23: 201-211.
Belluck, D.A., S.L. Benjamin, P. Baveye, J. Sampson and B. Johnson. 2003. Widespread
arsenic contamination of soils in residential areas and public spaces: An emerging
regulatory or medical crisis? Int. J. Toxicol. 22: 109-128.
Blum, J.S., A.B. Bindi, J. Buzzelli, J.F. Stolz, and R.S. Oremland. 1998. Bacillus
arsenicoselenatis, sp. nov., and Bacillus selenitireducens, sp. nov.: two haloalkaliphiles
from Mono Lake, California that respire oxyanions of selenium and arsenic. Arch.
Microbiol. 171: 19-30.
Botero, L.M., T.S. Al-Niemi, and T.R. McDermott. 2000. Characterization of two inducible
phosphate transport systems in Rhizobium tropici. Appl. and Environ. Microbiol. 66: 1522.
Cervantes, C., G. Ji, J.L. Ramirez, and S. Silver. 1994. Resistance to arsenic compounds in
microorganisms. FEMS Microbiol. Rev. 15: 355-367.
87
Chowdhury, U.K., B.K. Biswas, T.R. Chowdhury, G. Samanta, B.K. Mandal, G.C. Basu,
C.R. Chanda, D. Lodh, K.C. Saha, S.K. Mukherjee, S. Roy, S. Kabir, Q. Quamruzzaman,
and D. Chakraborti. 2000. Groundwater arsenic contamination in Bangladesh and West
Bengal, India. Environ. Health Perspect. 108: 393-397.
Cullen, W.R. and K.J. Reimer. 1989. Arsenic speciation in the environment. Chemical
Reviews 89: 713-764.
Darland, J. E. and W. P. Inskeep. 1997. Effects of pH and phosphate competition on the
transport of arsenate. J. Environ. Qual. 26: 1133-1139.
Ehrlich, H.L. 2002. Bacterial oxidation of As(III) compounds. In W.T. Frankenberger (ed.)
Environmental chemistry of arsenic. New York: Marcel Dekker, pp 313-327.
Ellis, P.J., T. Conrads, and R. Hille. 2001. Crystal structure of the 100 kDa arsenite oxidase
from Alcaligenes faecalis in two crystal forms at 1.64 Å and 2.03 Å. Structure 9: 125132.
Ellis, R.J., P. Morgan, A.J. Weightman, and J.C. Fry. 2003. Cultivation-dependent and –
independent approaches for determining bacterial diversity in heavy-metal-contaminated
soil. Appl. and Environ. Microbiol. 69: 3223-3230.
Fazal, M.A., T. Kawachi, and E. Ichion. 2001. Extent and severity of groundwater arsenic
contamination in Bangladesh. Water Int. 26: 370-379.
Francesconi, K.A. and D. Kuehnelt. 2002. Arsenic compounds in the environment. In W.T.
Frankenberger (ed.) Environmental chemistry of arsenic. New York: Marcel Dekker, pp
51-94.
Freeman, G.B., R.A. Schoof, M.V. Ruby, A.O. Davis, J.A. Dill, S.C. Liao, C.A. Lapin, and
P.D. Bergstrom. 1995. Bioavailability of arsenic in soil and house dust impacted by
smelter activities following oral administration in cynomolgus monkeys. Fundam. Appl.
Toxicol. 28: 215-222.
Gladysheva, T.B., K.L. Oden, and B.P. Rosen. 1994. Properties of the arsenate reductase of
plasmid R773. Biochemistry 33: 7288-7293.
Green, HH. 1918. Description of a bacterium which oxidizes arsenite to arsenate, and of one
which reduces arsenate to arsenite, isolated from a cattle-dipping tank. S. Afr. J. Sci. 14:
465-467.
Harold, F.M. and J.R. Baarda. 1966. Interaction of arsenate with phosphate-transport
systems in wild-type and mutant Streptococcus faecalis. J. Bacteriol. 91: 2257-2262.
88
Harrington, J.M., S.E. Fendorf., and R.F. Rosenzweig. 1998. Biotic generation of
arsenic(III) in metal(loid)-contaminated freshwater lake sediments. Environ. Sci.
Technol. 32: 2425-2430.
Harris, R.M, D.C. Webb, S.M. Howitt, and G.B. Cox. 2001. Characterization of PitA and
PitB from Escherichia coli. J. Bacteriol. 183: 5008-5014.
Harvey C.F., C.H. Swartz, A.B.M. Badruzzaman, N. Keon-Blute, W. Yu, M.A. Ali, J. Jay, R.
Beckie, V. Niedan, D. Brabander, P.M. Oates, K.N. Ashfaque, S. Islam, H.F. Hemond,
M.F. Ahmed. 2002. Arsenic mobility and groundwater extraction in Bangladesh.
Heeraman, D.A., V.P. Claassen and R.J. Zasoski. 2001. Interaction of lime, organic matter
and fertilizer on growth and uptake of arsenic and mercury by Zorro fescue (Vulpia
myuros L.) Plant and Soil 234: 215-231.
Hoque, B.A., A.A. Mahmood, M. Quadiruzzaman, F. Khan, S.A. Ahmed, S. Shafique, M.
Rahman, G. Morshed, T. Chowdhury, M.M Rahman, F.H. Khan, M. Shahjahan, M.
Begum and M.M. Hoque. 2000. Recommendations for water supply in arsenic
mitigation: a case study from Bangladesh. Public Health 114: 488-494.
Inskeep, W.P., T.R. McDermott and S. Fendorf. 2002. Arsenic (V)/(III) cycling in soils and
natural waters: Chemical and microbiological processes. In W.T. Frankenberger (ed.)
Environmental chemistry of arsenic. New York: Marcel Dekker, pp 183-215.
Inskeep, W.P., G.G. Ackerman, W.P. Taylor, M. Kozubal, S. Korf, and R.E. Macur. 2005.
On the energetics of chemolithotrophy in nonequilibrium systems: case studies of
geothermal springs in Yellowstone National Park. Geobiology 3: 297-317.
Inskeep, W.P., R.E. Macur, N. Hamamura, T.P. Warelow, S.A. Ward, and J.M. Santini.
2007. Detection, diversity and expression of aerobic bacterial arsenite oxidase genes.
Environ. Microbiol. 9: 934–943.
Jackson, C.R., and S.L. Dugas. 2003. Phylogenetic analysis of bacterial and archaeal arsC
sequences suggests an ancient, common origin for arsenate reductase.
BMC
Evolutionary Biology 3:18.
Ji, G., E.A.E. Garber, L.G. Armes, C.M. Chen, J.A. Fuchs, and S. Silver. 1994. Arsenate
reductase of Staphylococcus aureus plasmid pI258. Biochemistry 33: 7294-7299.
Jones, C.A., W.P. Inskeep, and D.R. Neuman. 1997. Arsenic transport in contaminated
mine tailings following liming. J. Environ. Qual. 26: 433-439.
89
Jones, C.A., W.P. Inskeep, J.W. Bauder, and K.E. Keith. 1999. Arsenic solubility and
attenuation in soils of the Madison River basin, Montana: impacts of long-term irrigation.
J. Environ. Qual. 28: 1314-1320.
Jones, C.A., H.W. Langner, K. Anderson, T.R. McDermott, and W.P. Inskeep. 2000. Rates
of microbially mediated arsenate reduction and solubilization. Soil Sci. Soc. Am. J. 64:
600-608.
Kashyap, D.R., L.M. Botero, W.L. Franck, D.J. Hassett, and T.R. McDermott. 2006.
Complex regulation of arsenite oxidation in Agrobacterium tumefaciens. J. Bacteriol.
188: 1081-1088.
Kasting, J.F. and D. Catling. 2003. Evolution of a habitable planet. Annu. Rev. of Astron.
Astrophys. 41: 429-463.
Kocar, B.D., R.A. Garrott, and W.P. Inskeep. 2003. Elk exposure to arsenic in geothermal
watersheds of Yellowstone National Park, USA. Environ. Toxicol. and Chem. 23: 982989.
Lacoste, A.M., A. Cassaigne, and E. Neuzil. 1981. Transport of inorganic phosphate in
Pseudomonas aeruginosa. Current Microbiol. 6: 115-120.
Macur, R.E., J.T. Wheeler, T.R. McDermott, and W.P. Inskeep. 2001. Microbial
populations associated with the reduction and enhanced mobilization of arsenic in mine
tailings. Environ. Sci. Technol. 35: 3676-3682.
Macur, R.E., C.R. Jackson, L.M. Botero, T.R. McDermott, and W.P. Inskeep. 2004.
Bacterial populations associated with the oxidation and reduction of arsenic in an
unsaturated soil. Environ. Sci. Technol. 38: 104-111.
Macy, J.M., K. Nunan, K.D. Hagen, D.R. Dixon, P.J. Harbour, M. Cahill, and L.I. Sly. 1996.
Chrysiogenes arsenatis gen. nov., sp. nov., a new arsenate-respiring bacterium isolated
from gold mine wastewater. Int. J. Syst. Bacteriol. 46: 1153-1157.
Macy, J.M. and J.M. Santini. 2002. Unique modes of arsenate respiration by Chrysiogenes
arsenatis and Desulfomicrobium sp. str. Ben-RB. In W.T. Frankenberger (ed.)
Environmental chemistry of arsenic. New York: Marcel Dekker, pp 297-312.
Martinez, P., R. Zvyagilskaya, P. Allard, and B.L. Persson. 1998. Physiological regulation
of the derepressible phosphate transporter in Saccharomyces cerevisiae. J. Bacteriol.
180: 2253-2256.
Masscheleyn, P.H., R.D. Delaune, and W.H. Patrick, Jr.
1991.
A hydride
generation/separation technique for arsenic speciation. J. Environ. Qual. 20: 96-100.
90
Mouillon, J.M. and B.L. Persson. 2005. Inhibition of the protein kinase A alters the
degradation of the high-affinity phosphate transporter Pho84 in Saccharomyces
cerevisiae. Curr. Genet. 48: 226-234.
Mukhopadhyay, R., J. Shi, and B.P. Rosen. 2000. Purification and characterization of Acr2p,
the Saccharomyces cerevisiae arsenate reductase. J. Biol. Chem. 275: 21149-21157.
Mukhopadhyay, R., B.P. Rosen, L.T. Phung, and S. Silver. 2002. Microbial arsenic: from
geocycles to genes and enzymes. FEMS Microbiol. Rev. 26: 311-325.
Muller D., D. Lièvremont, D.D. Simeonova, J.C. Hubert and M.C. Lett. 2003. Arsenite
oxidase aox genes from a metal-resistant β–proteobacterium. J. Bacteriol. 185: 135-141.
Nagorski, S.A. and J.N. Moore. 1999. Arsenic mobilizationin the hyporheic zone of a
contaminated stream. Water Resources Research 35: 3441-3450.
Nakamuro, K. and Y. Sayato. 1981. Comparative studies of chromosomal aberration
induced by trivalent and pentavalent arsenic. Mutation Research 88: 73-80.
National Research Council (NRC). 2005. Mineral tolerance of animals. Washington, D.C.:
The National Academies Press. pp. 31-41.
Newman, D.K, T.J. Beveridge, and F.M. Morel. 1997. Precipitation of arsenic trisulfide by
Desulfotomaculum auripigmentum. Appl. and Environ. Microbiol. 63: 2022-2028.
Nickson, R., J. McArthur, W. Burgess, K.M. Ahmed, P. Ravenscroft, and M. Rahmanñ.
1998. Arsenic poisoning of Bangladesh groundwater. Nature 395: 338.
Nordstrom, D.K. 2002. Worldwide occurrences of arsenic in ground water. Science 296:
2143-2145.
Novick, R.P. and C. Roth. 1968. Plasmid-linked resistance to inorganic salts in
Staphylococcus aureus. J. Bacteriol. 95: 1335-1342.
Nriagu, J.O. 2002. Arsenic poisoning through the ages. In W.T. Frankenberger (ed.)
Environmental chemistry of arsenic. New York: Marcel Dekker, pp 1-26.
Oremland, R.S., J.S. Blum, C.W. Culbertson, P.T. Visscher, L.G. Miller, P. Dowdle and F.E.
Strohmaier. 1994. Isolation, growth, and metabolism of an obligately anaerobic,
selenate-respiring bacterium, strain SES-3. Appl. and Environ. Microbiol. 60: 30113019.
91
Oremland, R.S., S.E. Hoeft, J.M. Santini, N. Bano, R.A. Hollibaugh, and J.T. Hollibaugh.
2002. Anaerobic oxidation of arsenite in Mono Lake water and by a facultative, arseniteoxidizing chemoautotroph, strain MLHE-1. Appl. and Environ. Microbiol. 68: 47954802.
Oremland, R.S., and J.F. Stolz. 2003. The ecology of arsenic. Science 300: 939-943.
Peryea, F. J. and R. Kammereck. 1997. Phosphate-enhanced movement of arsenic out of
lead arsenate-contaminated topsoil and through uncontaminated subsoil. Water, Air, and
Soil Pollution 93: 243-254.
Pierce, M.L. and C. B. Moore. 1982. Adsorption of arsenite and arsenate on amorphous iron
hydroxide. Water Res. 16: 1247-1253.
Polizzotto, M.L., C.F. Harvey, S.R. Sutton, and S. Fendorf. 2005. Processes conducive to
the release and transport of arsenic into aquifers of Bangladesh. PNAS 102: 1881918823.
Qin, J., B.P. Rosen, Y. Zhang, G. Wang, S. Franke, and C. Rensing. 2006. Arsenic
detoxification and evolution of trimethylarsine gas by a microbial arsenite Sadenosylmethionine methyltransferase. PNAS 103: 2075-2080.
Rochette, E.A., B.C. Bostick, G. Li, and S. Fendorf. 2000. Kinetics of arsenate reduction by
dissolved sulfide. Environ. Sci. Technol. 34: 4714-4720.
Rosen, B.P. 2002. Biochemistry of arsenic detoxification. FEBS Letters 529: 86-92.
Rosenberg, H., N. Medveczky and J.M. La Nauze. 1969. Phosphate transport in Bacillus
cereus. Biochemica et Biophysica Acta 193: 159-167.
Rosenberg, H., R.G. Gerdes and K. Chegwidden. 1977. Two systems for the uptake of
phosphate in Escherichia coli. J. Bacteriol. 131: 505-511.
Rothstein, A. and K. Donovan. 1963. Interactions of arsenate with the phosphatetransporting system of yeast. J. Gen. Physiol. 46: 1075-1085.
Ryker, S.J. 2001. Mapping arsenic in groundwater. Geotimes 46: 34-36.
Sadiq, M. 1997. Arsenic chemistry in soils: An overview of thermodynamic predictions and
field observations. Water, Air and Soil Pollution 93: 117-136.
Salmassi, T.M., J.J. Walker, D.K. Newman, J.R. Leadbetter, N.R. Pace, and J.G. Hering.
2006. Community and cultivation analysis of arsenite oxidizing biofilms at Hot Creek.
Environ. Microbiol. 8: 50-59.
92
Saltikov, C. and D. Newman. 2003. Genetic identification of a respiratory arsenate
reductase. PNAS 100: 10983-10988.
Saltikov, C.W., R.A. Wildman, and D.K. Newman. 2005. Expression dynamics of arsenic
respiration and detoxification in Shewanella sp. strain ANA-3. J. Bacteriol. 187: 73907396.
Smalla, K., G. Wieland, A. Buchner, A. Zock, J. Parzy, S. Kaiser, N. Roskot, H. Heuer, and
G. Berg. 2001. Bulk and rhizosphere soil bacterial communities studied by denaturing
gradient gel electrophoresis: Plant-dependent enrichment and seasonal shifts revealed.
Appl. and Environ. Microbiol. 67: 4742-4751.
Santini, J.M., L.I. Sly, R.D. Schnagl, and J.M. Macy. 2000. A new chemolithoautotrophic
arsenite-oxidizing bacterium isolated from a gold mine: Phylogenetic, physiological, and
preliminary biochemical studies. Appl. and Environ. Microbiol. 66: 92-97.
Santini, J.M. and R.N. vanden Hoven. 2004. Molybdenum-containing arsenite oxidase of
the chemolithoautotrophic arsenite oxidizer NT-26. J. Bacteriol. 186: 1614-1619.
Schaefer, A. 2007. Chicken litter and arsenic. Environ. Sci. Technol. 41: 668-669.
Silver, S., L.T. Phung, and B.P. Rosen. 2002. Arsenic metabolism: Resistance, reduction
and oxidation. In W.T. Frankenberger (ed.) Environmental chemistry of arsenic. New
York: Marcel Dekker, pp 247-272.
Smit, E., P. Leeflang, S. Gommans, J. van den Broek, S. van Mil, and K. Wernars. 2001.
Diversity and seasonal fluctuations of the dominant members of the bacterial soil
community in a wheat field as determined by cultivation and molecular methods. Appl.
and Environ. Microbiol. 67: 2284-2291.
Smith, A.H., P.A. Lopipero, M.N. Bates, and C.M. Steinmaus. 2002. Arsenic epidemiology
and drinking water standards. Science 296: 2145-2146.
Smith, J.D., J.H. Lambing, D.A. Nimick, C. Parrett, M. Ramey, and W. Schafer. 1998.
Geomorphology, flood-plain tailings, and metal transport in the Upper Clark Fork River
Valley, Montana. Water-Resources Investigation Report 98-4170, U.S. Geological
Survey.
Stoiber, R.E. 1995. Volcanic gases from subaerial volcanoes on Earth. In T.J. Ahrens (ed.)
Global Earth Physics: A Handbook of Physical Constants. Washington D.C.: AGU, pp
308-319.
Stolz, J.F. and R.S. Oremland. 1999. Bacterial respiration of arsenic and selenium. FEMS
Microbiol. Rev. 23: 615-627.
93
Stolz, J.F., E. Perera, B. Kilonzo, B. Kail, B. Crable, E. Fisher, M. Ranganathan, L. Wormer,
and P. Basu. 2007. Biotransformation of 3-nitro-4-hydroxybenzene arsonic acid
(roxarsone) and release of inorganic arsenic by Clostridium species. Environ. Sci.
Technol. 41: 818-823.
Sylvia, D.M., J.J. Fuhrmann, P.G. Hartel, and D.A. Zuberer. 1999. Principles and
applications of soil microbiology. New Jersey: Prentice Hall, Inc., p 7.
Tamai, Y., A. Toh-e, and Y. Oshima. 1985. Regulation of inorganic phosphate transport
systems in Saccharomyces cerevisiae. J. Bacteriol. 164: 964-968.
Tamaki, S. and W.T. Frankenberger. 1992. Environmental biochemistry of arsenic. Rev. of
Environ. Contam. and Toxicol. 124: 79-110.
Tassi, E., F. Pedron, M. Barbafieri, and G. Petruzzelli. 2004.
phytoextraction in As-contaminated soil. Eng. Life Sci. 4: 341-346.
Phosphate-assisted
Thiel, T. 1988. Phosphate transport and arsenate resistance in the cyanobacterium Anabaena
variabilis. J. Bacteriol. 170: 1143-1147.
Torsvik, V. and L. Ovreas. 2002. Microbial diversity and function in soil: from genes to
ecosystems. Current Opinion in Microbiol. 5: 240-245.
Tuominen, L., T. Kairesalo, and H. Hartikainen. 1994. Comparison of methods for
inhibiting bacterial activity in sediment. Appl. and Environ. Microbiol. 60: 3454-3457.
Ulrich, A. and S. Wirth. 1999. Phylogenetic diversity and population densities of culturable
cellulolytic soil bacteria across an agricultural encatchment. Microb. Ecol. 37: 238-247.
United States Environmental Protection Agency (US EPA). 1998. EPA Superfund Record of
Decision: Anaconda Co. Smelter. Publication Number EPA/ROD/R08-98/096.
United States Environmental Protection Agency (US EPA). 2006a. Arsenic in drinking
water. Available at: http://www.epa.gov/safewater/arsenic/.
United States Environmental Protection Agency (US EPA). 2006b. Chromated copper
arsenic (CCA). Available at: http://www.epa.gov/oppad001/reregistration/cca/.
United States Environmental Protection Agency (US EPA). 2006c. Locate NPL sites.
Available at: http://www.epa.gov/superfund/sites/npl/locate.htm.
van Veen, H.W., T. Abee, G.J. Kortstee, W.N. Konings, and A.J. Zehnder. 1993.
Characterization of two phosphate transport systems in Acinetobacter johnsonii 210A. J.
Bacteriol. 175: 200-206.
94
van Veen, H.W. 1997. Phosphate transport in prokaryotes: molecules, mediators, and
mechanisms. Antonie van Leeuwenhoek 72: 299-315.
Voegele, R.T., S. Bardin, and T.M. Finan. 1997. Characterization of the Rhizobium
(Sinorhizobium) meliloti high- and low-affinity phosphate uptake systems. J. Bacteriol.
179: 7226-7232.
Willsky, G.R. and M.H. Malamy. 1980. Characterization of two genetically separable
inorganic phosphate transport systems in Escherichia coli. J. Bacteriol. 144: 356-365.
Winegardner, D.L. 1995. An introduction to soils for environmental professionals. CRC
Press, Boca Raton, FL, p 180.
World Bank. 1999. The Bangladesh arsenic mitigation water supply project: Addressing a
massive public health crisis. A project brief of the World Bank/UNDP Water and
Sanitation Program.
Xu, H., B. Allard and A. Grimvall. 1991. Effects of acidification and natural organic
materials on the mobility of arsenic in the environment. Water, Air, and Soil Pollution
57-58: 269-278.
Zhou, Y., N. Messier, M. Ouellette, B.P. Rosen, and R. Mukhopadhyay. 2004. Leishmania
major LmACR2 is a pentavalent antimony reductase that confers sensitivity to the drug
pentostam. J. Biol. Chem. 279: 37445-37451.
95
APPENDICES
96
APPENDIX A:
SUPPLEMENTARY MATERIAL FOR CHAPTER 2
97
A) Arsenite 20-Day Treatment Column
B) Arsenite 28-Day Treatment Column
C) Arsenate 20-Day Treatment Column
D) Arsenate 28-Day Treatment Column
Figure A.1. Percent of total soluble arsenic present as arsenite (gray) or arsenate
(black) as a function of time in columns amended with 2 mg L-1 (26.7 µM) of either
arsenite or arsenate. Columns were sacrificed prior to experiment’s conclusion to
monitor changes in microbial community composition over time and were later
excluded from the experimental design as they exhibited no significant differences in
microbial or As redox profiles as treatment columns completing the entire experiment
duration.
98
A) Arsenite 5-Day Treatment Column
B) Arsenite 16-Day Treatment Column
C) Arsenate 5-Day Treatment Column
D) Arsenate 16-Day Treatment Column
Figure A.2. Percent of total soluble arsenic present as arsenite (gray) or arsenate
(black) as a function of time in columns amended with 20 mg L-1 (267 µM) of either
arsenite or arsenate. Columns were sacrificed prior to experiment’s conclusion to
monitor changes in microbial community composition over time and were later
excluded from the experimental design as they exhibited no significant differences in
microbial or As redox profiles as treatment columns completing the entire experiment
duration.
99
Table A.1. Colony morphology of isolates cultivated from 2 mg As L-1 (26.7 µM)
Amsterdam column soil samples. (Plated on R2A media.)
Isolate Strain
Name
DM1A
DM1AA
DM1B
DM1BB
Description
yellowish-cream, opaque, convex, circular colony with translucent outer ring
cream, opaque, dull, tiny, punctiform colony with smooth margin
orangish, opaque, dull, circular colony with white fuzzy umbonate top and smooth
margin
pinkish, opaque, shiny, tiny, punctiform colony with smooth margin
DM1E
cream, opaque, shiny, irregular colony diffusing to clear, translucent smooth margin
DM1F
creamish-clear, translucent, shiny, small, irregular colony with smooth margin
DM1G
cream, opaque, shiny, convex, circular colony with diffuse translucent margin; tends
to smear
DM1
creamish-clear, translucent, shiny, tiny, punctiform colony; tends to smear
DM1I
grey, opaque, dull, punctiform colony with white fuzzy top and undulate margin
DM1J
yellowish, translucent, shiny, irregular colony
DM1K
bright yellow, shiny, raised, irregular colony
DM1L
yellowish-cream, opaque, shiny, large, irregular colony with scalloped margin
DM1M
cream, opaque, dull, circular colony with white fuzzy umbonate top and smooth
margin
DM1N
cream, shiny, punctiform colony with undulate margin
DM1O
cream, opaque, dull, circular colony with white fuzzy slightly umbonate top and
smooth margin
DM1Q
orangish, translucent, shiny, convex, circular colony with smooth margin; tends to
smear
DM1R
cream, opaque, shiny, irregular colony with lobate margin; tends to smear
DM1S
peachish-cream, opaque, shiny, punctiform colony with smooth margin
DM1T
cream, opaque, shiny, irregular colony with smooth margin; tends to smear
DM1U
creamish-clear, translucent, concentric, circular colony with undulate margin
DM1V
peach, translucent, shiny, punctiform colony with undulate margin
DM1W
yellowish, shiny, irregular colony with translucent undulate margin; tends to smear
DM1X
cream, translucent, shiny, irregular colony with translucent smooth margin
DM1Y
cream, opaque, shiny, medium puntiform colony with smooth margin
100
Table A.2. Colony morphology of isolates cultivated from 20 mg As L-1 (267 µM)
Amsterdam column soil samples. (Plated on R2A media.)
Isolate Strain
Name
Description
DM2A
peach, opaque, punctiform, convex, dull colony with smooth margin
DM2B
cream, translucent, irregular, shiny colony with undulate margin; tends to smear
DM2C
pink, tiny, translucent, punctiform, shiny colony with smooth margin
DM2D
yellowish, tiny, shiny, punctiform colony
DM2E
pink, tiny, translucent, punctiform, shiny colony with smooth margin
DM2F
peachish, translucent, punctiform, shiny, convex colony with smooth margin
DM2N
tan, opaque, dull, convex papillate, circular colony with undulate margin and fuzzy
white top
DM2O
cream, opaque, shiny, circular colony with concentrated cream center
DM2P
yellowish, opaque, shiny, irregular colony with undulate margin; tends to smear
DM2Q
cream, opaque, shiny, irregular colony with smooth translucent margin
DM2R
cream, opaque, concentric, raised with concave beveled edge, circular colony with
thick white outer ring
DM2S
red, translucent, tiny, shiny, punctiform colony with smooth margin
DM2T
tan, dull, circular colony with fuzzy white umbonate top and undulate white margin
DM2U
cream, opaque, shiny, circular colony with smooth translucent margin
101
Table A.3. Colony morphology of isolates cultivated from 200 mg As L-1 (2.7 mM)
Amsterdam column soil samples. (Plated on R2A media.)
Isolate Strain
Name
Description
DM3A
peachish, opaque, moderately shiny, irregular colony with undulate margin; tends to
smear
DM3B
yellowish, opaque, shiny, irregular colony with translucent cream lobate margin
DM3C
yellowish-green, translucent, shiny, irregular colony with transparent undulate margin
DM3D
yellowish, opaque, shiny, convex, circular colony with opaque cream smooth margin
DM3E
yellow, translucent, shiny, convex, irregular colony with smooth margin
DM3F
yellowish-green, transparent, shiny, raised with concave beveled edge, irregular
colony with clear transparent, diffusing ciliate margin
DM3G
yellow, opaque, shiny, convex, irregular colony with smooth margin
DM3H
yellow, translucent, shiny, raised with concave beveled edge, irregular colony with
clear undulate margin
DM3I
yellow, translucent, shiny, punctiform colony with smooth margin
DM3J
brownish-cream, opaque, dull, circular colony with white fuzzy top
DM3K
peachish, opaque, shiny, irregular colony with cream translucent, filamentous outer
ring and smooth margin
DM3L
cream, translucent, shiny, punctiform colony with smooth margin; tends to smear
DM3M
cream, translucent, shiny, punctiform colony with smooth margin
DM3N
yellow, translucent, shiny, convex papillate, irregular colony with scalloped margin
DM3O
yellowish, translucent, shiny, irregular colony with cream smooth margin
DM3P
cream, translucent, shiny, umbonate, circular colony with smooth margin
DM3R
yellowish-green, transparent, shiny, raised with concave beveled edge, irregular
colony with clear transparent, diffusing ciliate margin
102
Table A.4. Colony morphology of isolates cultivated from non-sterile control (no-arsenic
pressure) Amsterdam column soil samples. (Plated on R2A media.)
Isolate Strain
Name
Description
DM2G
white, opaque, dull, circular, umbonate colony with smooth margin; tends to harden
DM2H
brown, opaque, dull, puntiform colony with smooth margin and fuzzy white top
DM2I
peachish, opaque, shiny, irregular colony with smooth margin; tends to smear
DM2J
pinkish, opaque, shiny, circular, moderately umbonate colony with smooth margin
DM2K
cream, opaque, shiny, convex, circular colony with smooth, translucent margin
DM2L
pinkish, opaque, shiny, circular, moderately umbonate colony with smooth margin
DM2M
yellowish, opaque, shiny, irregular colony with undulate, translucent margin
103
Table A.5. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding 2 mg L-1 (26.7 µM) As pressure. Isolates treated with 2.9 mg
L-1 (38.7 µM) of both AsIII and AsV and grown in duplicate for approximately 8 days in
SSE/SSM media.
Isolate Strain
Name
Measured [As]
(mg/L)
Dilution
Factor
Samples speciated to include As V only.
DM1AA
2.0
1.24
DM1AA
1.8
1.24
DM1AA2
0.3
1.40
DM1AA2
0.3
1.40
DM1B
0.9
1.24
DM1B
0.4
1.24
DM1BB
0.2
1.24
DM1BB
0.2
1.24
DM1C
0.4
1.24
DM1C
0.3
1.24
DM1D
0.2
1.24
DM1D
0.2
1.24
DM1E
0.2
1.24
DM1E
0.2
1.24
DM1F
0.2
1.24
DM1F
0.2
1.24
DM1G
2.2
1.40
DM1G
2.1
1.40
DM1
3.8
1.24
DM1
3.7
1.24
DM1I
0.3
1.24
DM1I
0.3
1.24
DM1J
2.2
1.24
DM1J
2.1
1.24
DM1J2
2.2
1.40
DM1J2
2.1
1.40
DM1K
2.1
1.24
DM1K
2.1
1.24
DM1L
0.2
1.24
DM1L
0.2
1.24
DM1M
0.2
1.24
DM1M
0.3
1.24
Final
[As]
(mg/L)
AsV in Isolate Media:AsV in
Sterile Control (SC) Media
(%)
Observed As
Transformation1
OD
(A 500)
2.43
2.20
0.38
0.40
1.07
0.43
0.22
0.21
0.49
0.42
0.28
0.28
0.29
0.27
0.29
0.25
3.14
2.93
4.74
4.57
0.34
0.33
2.76
2.61
3.03
2.98
2.55
2.60
0.25
0.26
0.30
0.32
90
82
10
11
40
16
8
8
18
16
10
11
11
10
11
9
111
104
176
170
12
12
103
97
107
105
95
97
9
10
11
12
none
none
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
none
none
oxidation
oxidation
reduction
reduction
none
none
none
none
none
none
reduction
reduction
reduction
reduction
0.058
0.214
0.251
0.004
0.752
0.178
0.095
0.415
0.844
0.012
0.016
0.61
0.056
0.087
0.037
0.028
0.483
0.591
0.078
1
The observed As transformation of an isolate was assigned based on the ratio of AsV in isolate media:AsV in sterile
control media. The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% =
oxidation and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical
density (OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
104
Table A.5 con’t. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding 2 mg L-1 (26.7 µM) As pressure. Isolates treated with 2.9 mg
L-1 (38.7 µM) of both AsIII and AsV and grown in duplicate for approximately 8 days in
SSE/SSM media.
Final
AsV in Isolate Media:AsV in
Observed As
OD
[As]
Sterile Control (SC) Media
(A 500)
Transformation1
(mg/L)
(%)
DM1N
1.5
1.40
2.11
75
reduction
0.252
DM1N
1.2
1.40
1.69
60
reduction
0.265
DM1O
1.6
1.24
1.98
74
reduction
0.046
DM1O
1.4
1.24
1.72
64
reduction
DM1Q
1.8
1.24
2.25
84
none
0.16
DM1Q
2.0
1.24
2.43
90
none
DM1R
2.0
1.24
2.51
93
none
0.054
DM1R
1.9
1.24
2.31
86
none
2
DM1R
0.3
1.40
0.44
16
reduction
0.773
DM1R2
0.3
1.40
0.47
17
reduction
DM1S
0.2
1.24
0.19
7
reduction
1.656
DM1S
0.2
1.24
0.22
8
reduction
DM1T
0.2
1.24
0.26
10
reduction
0.739
DM1T
0.2
1.24
0.26
10
reduction
DM1V
0.2
1.24
0.26
10
reduction
0.343
DM1V
0.2
1.24
0.28
10
reduction
DM1W
0.3
1.40
0.38
14
reduction
1.169
DM1W
0.3
1.40
0.36
13
reduction
1.131
DM1X
2.2
1.24
2.73
101
none
0.371
DM1X
2.2
1.24
2.73
101
none
DM1Y
0.2
1.24
0.28
11
reduction
1.245
DM1Y
0.3
1.24
0.36
13
reduction
DM1Z
2.1
1.24
2.60
97
none
0.052
DM1Z
2.1
1.24
2.60
97
none
DM1Z2
0.3
1.40
0.39
15
reduction
0.161
2
DM1Z
0.3
1.40
0.45
17
reduction
0.39
SC3 Trial 1
2.2
1.24
2.69
SC3 Trial 2
2.7
1.40
3.75
SC3 Trial 3
2.0
1.40
2.83
1
The observed As transformation of an isolate was assigned based on the ratio of AsV in isolate media:AsV in sterile
control media. The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% =
oxidation and 75 - 125% = none.
Isolate Strain
Name
Measured [As]
(mg/L)
Dilution
Factor
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical
density (OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
105
Table A.5 con’t. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding 2 mg L-1 (26.7 µM) As pressure. Isolates treated with 2.9 mg
L-1 (38.7 µM) of both AsIII and AsV and grown in duplicate for approximately 8 days in
SSE/SSM media.
Isolate Strain
Name
Measured [As]
(mg/L)
Dilution
Factor
Final
[As]
(mg/L)
Total arsenic samples; run to ensure no loss of arsenic
via methylation, etc.
DM1AA
4.7
1.11
5.26
DM1AA2
6.0
1.00
5.99
DM1B
4.8
1.11
5.37
DM1BB
4.8
1.11
5.31
DM1C
4.8
1.11
5.27
DM1D
4.8
1.11
5.28
DM1E
4.6
1.11
5.11
DM1F
4.6
1.11
5.16
DM1G
5.6
1.00
5.56
DM1
5.0
1.11
5.60
DM1I
4.8
1.11
5.28
DM1J
4.8
1.11
5.29
DM1J2
5.5
1.00
5.52
DM1K
4.8
1.11
5.33
DM1L
4.9
1.11
5.48
DM1M
4.8
1.11
5.37
DM1N
5.4
1.00
5.39
DM1O
4.8
1.11
5.34
DM1Q
4.8
1.11
5.36
DM1R2
5.4
1.00
5.37
DM1S
4.9
1.11
5.43
DM1T
4.8
1.11
5.35
DM1V
4.9
1.11
5.44
DM1W
5.4
1.00
5.43
DM1X
4.8
1.11
5.33
DM1Y
4.7
1.11
5.26
DM1Z
4.8
1.11
5.30
DM1Z2
6.0
1.00
6.00
SC3 Trial 1
5.4
1.00
5.40
3
6.3
1.00
6.26
3
5.3
1.00
5.33
SC Trial 2
SC Trial 3
1
The observed As transformation of an isolate was assigned based on the ratio of AsV in isolate media:AsV in sterile
control media. The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% =
oxidation and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical
density (OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
106
Table A.6. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding 20 mg L-1 (267 µM) As pressure. Isolates treated with 2.9
mg L-1 (38.7 µM) of both AsIII and AsV and grown in duplicate for approximately 8 days in
SSE/SSM media.
Isolate Strain
Name
Measured
[As]
(mg/L)
Dilution
Factor
Final
[As]
(mg/L)
AsV in Isolate Media:AsV in
Sterile Control (SC) Media
(%)
Observed As
Transformation1
OD
(A 500)
0.36
0.43
0.39
0.38
0.42
0.45
2.98
3.00
3.21
3.37
1.85
0.62
0.38
0.46
0.40
0.39
0.43
0.44
1.34
2.14
0.47
0.40
2.68
2.71
0.53
0.60
2.94
2.96
3.75
2.83
3.26
10
11
14
13
11
12
105
106
98
103
49
17
10
12
11
10
15
16
36
57
13
11
95
96
14
16
104
105
reduction
reduction
reduction
reduction
reduction
reduction
none
none
none
none
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
none
none
reduction
reduction
none
none
0.987
0.991
1.013
1.077
0.442
0.639
0.038
0.027
0.062
0.08
0.159
0.735
0.919
0.918
0.48
0.404
1.268
1.118
0.208
0.203
0.201
0.14
0.095
0.102
0.149
0.086
0.045
0.051
Samples speciated to include As V only.
DM2A
DM2A
DM2B
DM2B
DM2C
DM2C
DM2D
DM2D
DM2D2
DM2D2
DM2E
DM2E
DM2F
DM2F
DM2N
DM2N
DM2P
DM2P
DM2Q
DM2Q
DM2R
DM2R
DM2S
DM2S
DM2T
DM2T
DM2U
DM2U
SC3 Trial 1
SC3 Trial 2
SC3 Trial 3
0.3
0.3
0.3
0.3
0.3
0.3
2.1
2.1
2.3
2.4
1.3
0.4
0.3
0.3
0.3
0.3
0.3
0.3
1.0
1.5
0.3
0.3
1.9
1.9
0.4
0.4
2.1
2.1
2.7
2.0
2.3
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1
The observed As transformation of an isolate was assigned based on the ratio of AsV in isolate media:AsV in sterile control media.
The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% = oxidation and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical density
(OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
107
Table A.6 con’t. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding 20 mg L-1 (267 µM) As pressure. Isolates treated with 2.9
mg L-1 (38.7 µM) of both AsIII and AsV and grown in duplicate for approximately 8 days in
SSE/SSM media.
Measured
Final
Dilution
[As]
[As]
Factor
(mg/L)
(mg/L)
Total arsenic samples; run to ensure no loss of arsenic via
methylation, etc.
Isolate Strain
Name
DM2A
DM2A
DM2B
DM2B
DM2C
DM2C
DM2D
DM2D
DM2E
DM2E
DM2F
DM2F
DM2N
DM2P
DM2P
DM2Q
DM2Q
DM2R
DM2S
DM2S
DM2T
DM2U
DM2U
3
SC Trial 1
3
SC Trial 2
SC Trial 3
6.2
6.2
5.5
5.5
6.3
6.0
5.4
5.4
6.2
5.6
5.8
6.3
6.1
5.4
5.3
6.2
6.1
6.2
5.3
5.3
6.2
5.3
5.3
6.3
5.3
6.0
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
6.20
6.23
5.45
5.46
6.28
6.01
5.39
5.42
6.24
5.59
5.85
6.25
6.14
5.36
5.31
6.20
6.15
6.19
5.32
5.32
6.16
5.32
5.34
6.26
5.33
6.04
1
The observed As transformation of an isolate was assigned based on the ratio of AsV in isolate media:AsV in sterile control media.
The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% = oxidation and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical density
(OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
108
Table A.7. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding 200 mg L-1 (2.7 mM) As pressure. Isolates treated with 2.9
mg L-1 (38.7 µM) of both AsIII and AsV and grown in duplicate for approximately 8 days in
SSE/SSM media.
Isolate Strain
Name
Measured
[As]
(mg/L)
Dilution
Factor
V
Final
[As]
(mg/L)
V
V
As in Isolate Media:As in
Sterile Control (SC) Media
(%)
Observed As
1
Transformation
OD
(A 500)
Samples speciated to include As only.
DM3A
0.822
1.40
1.15
35
reduction
1.033
DM3A
0.53
1.40
0.74
23
reduction
1.057
DM3B
0.4
1.40
0.62
17
reduction
0.369
DM3B
0.3
1.40
0.49
13
reduction
0.269
DM3C
2.4
1.40
3.37
119
none
0.08
DM3C
2.5
1.40
3.49
123
none
0.105
DM3D
0.4
1.40
0.53
14
reduction
0.847
DM3D
0.4
1.40
0.51
14
reduction
DM3E
2.2
1.40
3.09
83
none
0.155
DM3E
2.5
1.40
3.50
94
none
0.13
DM3F
0.3
1.40
0.42
11
reduction
0.076
DM3F
0.3
1.40
0.36
10
reduction
0.067
DM3G
0.3
1.40
0.38
10
reduction
0.351
DM3G
0.3
1.40
0.38
10
reduction
0.325
DM3H
2.5
1.40
3.50
94
none
0.122
DM3H
2.5
1.40
3.52
95
none
0.288
DM3I
0.3
1.40
0.39
14
reduction
1.384
DM3I
0.3
1.40
0.39
14
reduction
1.381
DM3K
1.5
1.40
2.07
73
reduction
0.117
DM3K
1.5
1.40
2.07
73
reduction
0.126
DM3L
2.3
1.40
3.27
116
none
0.054
DM3L
2.4
1.40
3.34
118
none
0.082
DM3M
2.0
1.40
2.86
101
none
0.058
DM3M
2.0
1.40
2.82
100
none
0.065
DM3N
0.3
1.40
0.37
10
reduction
0.488
DM3N
0.3
1.40
0.44
12
reduction
0.336
DM3O
0.3
1.40
0.41
11
reduction
0.784
DM3O
0.3
1.40
0.36
10
reduction
0.864
DM3P
2.4
1.40
3.30
88
none
0.066
DM3P
2.4
1.40
3.39
91
none
0.064
2
DM3P
2.0
1.40
2.78
98
none
0.043
2
DM3P
2.0
1.40
2.84
100
none
0.04
1
V
V
The observed As transformation of an isolate was assigned based on the ratio of As in isolate media:As in sterile
control media. The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% = oxidation
and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical
density (OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
109
Table A.7 con’t. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding 200 mg L-1 (2.7 mM) As pressure. Isolates treated with 2.9
mg L-1 (38.7 µM) of both AsIII and AsV and grown in duplicate for approximately 8 days in
SSE/SSM media.
Isolate Strain
Name
Measured
[As]
(mg/L)
Dilution
Factor
Final
[As]
(mg/L)
V
V
As in Isolate Media:As in
Sterile Control (SC) Media
(%)
Observed As
1
Transformation
OD
(A 500)
DM3R
0.3
1.40
0.43
12
reduction
0.591
DM3R
0.3
1.40
0.48
13
reduction
1.444
3
SC Trial 1
2.7
1.40
3.73
3
SC Trial 2
2.0
1.40
2.83
3
SC Trial 3
2.3
1.40
3.29
Total arsenic samples; run to ensure no loss of arsenic via methylation, etc.
DM3A
6.128
1.00
6.13
DM3B
5.7
1.00
5.71
DM3B
5.6
1.00
5.63
DM3C
5.3
1.00
5.32
DM3C
5.3
1.00
5.34
DM3D
5.9
1.00
5.88
DM3D
5.9
1.00
5.92
DM3E
6.0
1.00
5.98
DM3E
6.1
1.00
6.11
DM3F
5.8
1.00
5.82
DM3F
5.9
1.00
5.90
DM3G
5.9
1.00
5.91
DM3G
6.0
1.00
6.04
DM3H
6.0
1.00
5.95
DM3H
6.0
1.00
6.04
DM3I
5.3
1.00
5.26
DM3I
5.3
1.00
5.28
DM3K
5.5
1.00
5.50
DM3K
5.5
1.00
5.46
DM3L
5.4
1.00
5.36
DM3L
5.3
1.00
5.31
DM3M
5.4
1.00
5.36
DM3M
5.3
1.00
5.35
DM3N
5.8
1.00
5.84
DM3N
5.8
1.00
5.83
DM3O
5.8
1.00
5.84
DM3O
5.8
1.00
5.83
1
V
V
The observed As transformation of an isolate was assigned based on the ratio of As in isolate media:As in sterile
control media. The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% = oxidation
and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical
density (OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
110
Table A.7 con’t. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding 200 mg L-1 (2.7 mM) As pressure. Isolates treated with 2.9
mg L-1 (38.7 µM) of both AsIII and AsV and grown in duplicate for approximately 8 days in
SSE/SSM media.
Final
Measured
Dilution
[As]
[As]
Factor
(mg/L)
(mg/L)
DM3P
6.0
1.00
5.97
DM3P
5.9
1.00
5.94
2
DM3P
5.3
1.00
5.33
2
DM3P
5.3
1.00
5.33
DM3R
5.9
1.00
5.89
DM3R
5.9
1.00
5.90
3
SC Trial 1
6.0
1.00
6.01
3
SC Trial 2
5.3
1.00
5.33
SC Trial 3
5.9
1.00
5.86
1
V
V
The observed As transformation of an isolate was assigned based on the ratio of As in isolate media:As in sterile
control media. The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% = oxidation
and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical
density (OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
Isolate Strain
Name
111
Table A.8. Data from As phenotype screening of isolates cultured from Amsterdam
column studies withstanding no As pressure. Isolates treated with 2.9 mg L-1 (38.7 µM) of
both AsIII and AsV and grown in duplicate for approximately 8 days in SSE/SSM media.
Isolate Strain
Name
Measured
[As]
(mg/L)
Dilution
Factor
V
V
Final
[As]
(mg/L)
As in Isolate Media:As in
Sterile Control (SC) Media
(%)
Observed As
1
Transformation
OD
(A 500)
0.43
0.42
0.37
0.39
0.46
0.40
2.22
0.58
0.38
0.43
0.40
0.36
0.48
0.52
3.75
2.83
12
11
10
10
16
14
59
15
10
11
11
10
13
14
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
0.265
0.228
0.14
0.347
1.212
1.198
0.186
0.445
0.745
0.734
0.451
0.596
0.978
0.96
V
Samples speciated to include As only.
DM2G
DM2G
DM2H
DM2H
DM2I
DM2I
DM2J
DM2J
DM2K
DM2K
DM2L
DM2L
DM2M
DM2M
SC3 Trial 1
3
SC Trial 2
0.3
0.3
0.3
0.3
0.3
0.3
1.6
0.4
0.3
0.3
0.3
0.3
0.3
0.4
2.7
2.0
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
Total arsenic samples; run to ensure no loss of arsenic via methylation, etc.
DM2G
6.2
1.00
6.20
DM2G
6.2
1.00
6.16
DM2H
6.1
1.00
6.11
DM2H
5.9
1.00
5.92
DM2I
5.4
1.00
5.38
DM2I
5.4
1.00
5.40
DM2J
6.2
1.00
6.24
DM2J
6.1
1.00
6.10
DM2K
6.2
1.00
6.18
DM2K
6.2
1.00
6.17
DM2L
6.1
1.00
6.08
DM2L
6.1
1.00
6.11
DM2M
6.2
1.00
6.22
DM2M
6.2
1.00
6.20
SC3 Trial 1
6.3
1.00
6.26
3
SC Trial 2
5.3
1.00
5.33
1
V
V
The observed As transformation of an isolate was assigned based on the ratio of As in isolate media:As in sterile control media.
The following ratio ranges were assigned the corresponding result: <75% = reduction, >125% = oxidation and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly low optical density
(OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
Column 8A
Column 7B
Column 7A
Column 4A
Column 3B
Column 3A
Column 2A
Column 1B
Column 1A
AsV
Treatment Columns
Column 9A
AsV
Sterile Controls
Column 9B
AsIII
Treatment Columns
Column 10A
AsIII
Sterile Controls
Column 10B
Column 8B
Column 4B
Column 2B
Figure A.3. DGGE gel demonstrating microbial banding patterns of soil columns treated with 2 mg L-1
(26.7 µM) arsenite or arsenate and broken-down at the experiment’s conclusion (57 days). An “A” in the
column name designates the sample corresponds to the top sampling port of the column, while a “B”
denotes a bottom port sample.
112
113
28 days
Column 9B
Column 9A
Column 10B
Column 10A
57 days
Column 12B
Column 12A
Column 11B
Column 11A
20 days
Figure A.4. An example DGGE gel demonstrating little change over time or region in
microbial banding patterns of soil columns treated with 2 mg L-1 (26.7 µM) arsenate.
Columns 11 and 12 were “sacrificed” and molecularly analyzed 20 and 28 days,
respectively, after experiment commenced, while columns 10 and 9 endured the entire
experiment duration of 57 days. An “A” in the column name designates the sample
corresponds to the top sampling port of the column, while a “B” denotes a bottom port
sample. A similar result was observed in the 2 mg L-1 AsIII-treated columns.
114
Column 12B
Column 12A
Column 11B
Column 11A
AsV
Treatment Columns
Column 14B
Column 14A
Column 13B
AsV
Sterile Controls
Column 13A
Column 2B
Column 2A
Column 1B
AsIII
Treatment Columns
Column 1A
Column 6B
Column 6A
Column 5B
Column 5A
AsIII
Sterile Controls
Figure A.5. DGGE gel demonstrating microbial banding patterns of soil columns treated
with 20 mg L-1 (267 µM) arsenite or arsenate and broken-down at the experiment’s
conclusion (26 days). An “A” in the column name designates the sample corresponds to
the top sampling port of the column, while a “B” denotes a bottom port sample.
115
Column 12B
Column 12A
Column 11B
Column 11A
Column 10B
Column 10A
Column 9B
Column 9A
26 days
16 days
5 days
Figure A.6. An example DGGE gel demonstrating little change over time or region in
microbial banding patterns of soil columns treated with 20 mg L-1 (267 µM) arsenate.
Columns 9 and 10 were “sacrificed” and molecularly analyzed 5 and 16 days, respectively,
after experiment commenced, while columns 11 and 12 endured the entire experiment
duration of 26 days. An “A” in the column name designates the sample corresponds to the
top sampling port of the column, while a “B” denotes a bottom port sample. A similar
result was observed in the 20 mg L-1 AsIII-treated columns.
116
Column 8A
AsV
Treatment Columns
Column 7A
Column 6A
AsV
Sterile Controls
Column 5A
Column 4A
Column 3A
Column 2A
Column 1A
AsIII
AsIII
Sterile Controls Treatment Columns
Figure A.7. DGGE gel demonstrating microbial banding patterns of soil columns treated
with 200 mg L-1 (2.7 mM) arsenite or arsenate and broken-down at the experiment’s
conclusion (25 days). An “A” in the column name designates the sample corresponds to
the top sampling port of the column, while a “B” denotes a bottom port sample.
117
Column 8B
Column 8A
Column 7B
Column 7A
Non-sterile Control Columns
Figure A.8. DGGE gel demonstrating microbial banding patterns of soil columns that
endured no As pressure (non-sterile controls) and were broken-down at the experiment’s
conclusion (26 days). An “A” in the column name designates the sample corresponds to
the top sampling port of the column, while a “B” denotes a bottom port sample.
Isolate DM1L
Isolate DM1K
Isolate DM1J
Band #:
AsIII-treated Column 2-2
Isolate DM1BB
Isolate DM1AA
Isolate DM1T
Isolate DM1S
Isolate DM1R
Band #:
AsIII-treated Column 2-3
118
2-C3-1
2-C2-1
2-C2-2
2-C3-2
2-C3-3
2-C3-5
Figure A.9. DGGE gel showing co-migration of microbial signatures between two isolates
(DM1S and DM1BB) cultivated from 2 mg L-1 (26.7 µM) arsenite-treated columns and the
environmental sample from which they originated.
Figure A.10. DGGE gel showing no co-migration of microbial signatures between isolates cultivated from 2 mg
L-1 (26.7 µM) arsenite or arsenate-treated columns and the environmental sample from which they originated.
AsIII-treated Column 2-4
Isolate DM1A
Isolate DM1B
Isolate DM1E
Isolate DM1M
Isolate DM1N
Isolate DM1O
Isolate DM1Q
AsV-treated Column 2-9
Isolate DM1F
Isolate DM1G
Isolate DM1U
Isolate DM1V
AsV-treated Column 2-10
Isolate DM1
Isolate DM1I
Isolate DM1W
Isolate DM1X
Isolate DM1Y
119
20-C12-1
Band #:
Figure A.11. DGGE gel showing no co-migration of microbial signatures between isolates cultivated from 20
mg L-1 (267 µM) arsenite or arsenate-treated columns and the environmental sample from which they
originated.
AsV-treated Column 20-12
Isolate DM2N
Isolate DM2O
Isolate DM2P
Isolate DM2Q
Isolate DM2R
AsV-treated Column 20-13
Isolate DM2S
AsIII-treated Column 20-2
Isolate DM2A
Isolate DM2B
Isolate DM2C
Isolate DM2D
Isolate DM2T
Isolate DM2U
AsIII-treated Column 20-5
Isolate DM2E
Isolate DM2F
120
Isolate DM3F
Isolate DM3E
Isolate DM3D
Isolate DM3C
Isolate DM3B
AsIII-treated Column 200-3
121
Band #:
200-C3-1
200-C3-2
200-C3-3
200-C3-4
200-C3-5
Figure A.12. DGGE gel showing co-migration of microbial signatures between three
isolates (DM3C, DM3E, and DM3F) cultivated from 200 mg L-1 (2.7 mM) arsenite-treated
columns and the environmental sample from which they originated.
200-C7-1
Band #:
200-C6-2
200-C6-1
Band #:
200-C5-3
200-C5-4
200-C5-2
Band #:
Isolate DM3R
Isolate DM3P
AsV-treated Column 200-5
Isolate DM3O
Isolate DM3N
Isolate DM3M
AsV-treated Column 200-6
Isolate DM3L
Isolate DM3K
Isolate DM3J
Isolate DM3I
Isolate DM3H
Isolate DM3G
AsV-treated Column 200-7
Figure A.13. DGGE gel showing co-migration of microbial signatures between three isolates (DM3M,
DM3N, and DM3R) cultivated from 200 mg L-1 (2.7 mM) arsenate-treated columns and the
environmental sample from which they originated.
122
Isolate DM2M
Isolate DM2L
Isolate DM2K
Isolate DM2J
Isolate DM2I
Isolate DM2H
Isolate DM2G
Non-sterile Control 7
123
Figure A.14. DGGE gel showing no co-migration of microbial signatures between isolates
cultivated from a non-sterile control column and the environmental sample from which
they originated.
124
APPENDIX B:
SUPPLEMENTARY MATERIAL FOR CHAPTER 3
125
A) 50 µM P - 1000 µM As Experiment
B) 1000 µM P - 1000 µM As Experiment
C) 50 µM P - 100 µM As Experiment
D) 500 µM P - 100 µM As Experiment
E) 1000 µM P - 100 µM As Experiment
Figure B.1. Concentration of total arsenic over time in experiments initially spiked with
either 1000, 100, 10 or 5 µM As.
126
F) 50 µM P - 5 µM As Experiment
G) 50 µM P - 10 µM As Experiment
Figure B.1 con’t. Concentration of total arsenic over time in experiments initially spiked
with either 1000, 100, 10 or 5 µM As.
127
A) Agrobacterium tumefaciens str. 5A
B) Variovorax sp. str. RM1
C) Agrobacterium tumefaciens str. 5B
D) Arthrobacter sp. str. S6
E) Bacillus sp. str. S18
50 µM P – 100 µM As
50 µM P – 5 µM As
500 µM P – 100 µM As
50 µM P – 10 µM As
1000 µM P – 100 µM As
50 µM P – 1000 µM As
1000 µM P – 1000 µM As
Figure B.2. Comparison of isolate growth curves in all P:As experiments.
128
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
Figure B.3. Correlation of phosphate concentration and optical density (OD500) over
experiment duration in all Agrobacterium tumefaciens str. 5A trials.
129
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
Figure B.4. Correlation of phosphate concentration and optical density (OD500) over
experiment duration in all Variovorax sp. str. RM1 trials.
130
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
Figure B.5. Correlation of phosphate concentration and optical density (OD500) over
experiment duration in all Agrobacterium tumefaciens str. 5B trials.
131
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
G) 1000 µM P - 1000 µM As
Figure B.5 con’t. Correlation of phosphate concentration and optical density (OD500) over
experiment duration in all Agrobacterium tumefaciens str. 5B trials.
132
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
Figure B.6. Correlation of phosphate concentration and optical density (OD500) over
experiment duration in all Arthrobacter sp. str. S6 trials.
133
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
G) 1000 µM P - 1000 µM As
Figure B.6 con’t. Correlation of phosphate concentration and optical density (OD500) over
experiment duration in all Arthrobacter sp. str. S6 trials.
134
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
Figure B.7. Correlation of phosphate concentration and optical density (OD500) over
experiment duration in all Bacillus sp. str. S18 trials.
135
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
G) 1000 µM P - 1000 µM As
Figure B.7 con’t. Correlation of phosphate concentration and optical density (OD500) over
experiment duration in all Bacillus sp. str. S18 trials.
136
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
Figure B.8. Correlation of phosphate concentration and the arsenite (AsIII) fraction over
experiment duration in all Agrobacterium tumefaciens str. 5A trials.
137
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
Figure B.9. Correlation of phosphate concentration and the arsenite (AsIII) fraction over
experiment duration in all Variovorax sp. str. RM1 trials.
138
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
Figure B.10. Correlation of phosphate concentration and the arsenate (AsV) fraction over
experiment duration in all Agrobacterium tumefaciens str. 5B trials.
139
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
G) 1000 µM P - 1000 µM As
Figure B.10 con’t. Correlation of phosphate concentration and the arsenate (AsV) fraction
over experiment duration in all Agrobacterium tumefaciens str. 5B trials.
140
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
Figure B.11. Correlation of phosphate concentration and the arsenate (AsV) fraction over
experiment duration in all Arthrobacter sp. str. S6 trials.
141
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
G) 1000 µM P - 1000 µM As
Figure B.11 con’t. Correlation of phosphate concentration and the arsenate (AsV) fraction
over experiment duration in all Arthrobacter sp. str. S6 trials.
142
A) 50 µM P - 100 µM As
B) 50 µM P - 5 µM As
C) 500 µM P - 100 µM As
D) 50 µM P - 10 µM As
Figure B.12. Correlation of phosphate concentration and the arsenate (AsV) fraction over
experiment duration in all Bacillus sp. str. S18 trials.
143
E) 1000 µM P - 100 µM As
F) 50 µM P - 1000 µM As
G) 1000 µM P - 1000 µM As
Figure B.12 con’t. Correlation of phosphate concentration and the arsenate (AsV) fraction
over experiment duration in all Bacillus sp. str. S18 trials.
144
A) 50 µM P - 1000 µM As
B) 50 µM P - 100 µM As
C) 50 µM P - 10 µM As
D) 50 µM P - 5 µM As
E) 500 µM P - 100 µM As
F) 1000 µM P - 100 µM As
Figure B.13. Correlation of arsenate (AsV) reduction and optical density (OD500) over
experiment duration in Agrobacterium tumefaciens str. 5B trials.
145
A) 50 µM P - 1000 µM As
B) 50 µM P - 100 µM As
C) 50 µM P - 10 µM As
D) 50 µM P - 5 µM As
E) 500 µM P - 100 µM As
F) 1000 µM P - 100 µM As
Figure B.14. Correlation of arsenate (AsV) reduction and optical density (OD500) over
experiment duration in Arthrobacter sp. str. S6 trials.
146
A) 50 µM P - 1000 µM As
B) 50 µM P - 100 µM As
C) 50 µM P - 10 µM As
D) 50 µM P - 5 µM As
E) 500 µM P - 100 µM As
F) 1000 µM P - 100 µM As
Figure B.15. Correlation of arsenate (AsV) reduction and optical density (OD500) over
experiment duration in Bacillus sp. str. S18 trials.
147
Table B.1. Univariate analysis of variance (ANOVA) data depicting no significant
differences (α=0.05) in the arsenite fraction at the final sampling in Agrobacterium
tumefaciens str. 5A P:As trials, as shown in Figure 3.1. Line color corresponds to P:As
ratio as follows: 50 µM P: 5 µM As – blue; 1000 µM P: 100 µM As – cyan; 50 µM P: 10
µM As – green; 50 µM P: 100 µM As – pink; and 500 µM P: 100 µM As – red.
(I) line
blue
cyan
green
pink
red
(J) line
cyan
green
pink
red
blue
green
pink
red
blue
cyan
pink
red
blue
cyan
green
red
blue
cyan
green
pink
Mean
Difference
(I-J)
-.9915
-4.7400
-5.6480
-1.7340
.9915
-3.7485
-4.6565
-.7425
4.7400
3.7485
-.9080
3.0060
5.6480
4.6565
.9080
3.9140
1.7340
.7425
-3.0060
-3.9140
Based on observed means.
Std. Error
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
3.81604
Sig.
.805
.269
.199
.669
.805
.371
.277
.853
.269
.371
.821
.467
.199
.277
.821
.352
.669
.853
.467
.352
95% Confidence Interval
Lower Bound Upper Bound
-10.8010
8.8180
-14.5495
5.0695
-15.4575
4.1615
-11.5435
8.0755
-8.8180
10.8010
-13.5580
6.0610
-14.4660
5.1530
-10.5520
9.0670
-5.0695
14.5495
-6.0610
13.5580
-10.7175
8.9015
-6.8035
12.8155
-4.1615
15.4575
-5.1530
14.4660
-8.9015
10.7175
-5.8955
13.7235
-8.0755
11.5435
-9.0670
10.5520
-12.8155
6.8035
-13.7235
5.8955
148
Table B.2. Univariate analysis of variance (ANOVA) data depicting significant
differences (“*”; α=0.05) in the arsenite fraction at the final sampling in Variovorax sp. str.
RM1 P:As trials, as shown in Figure 3.1. Line color corresponds to P:As ratio as follows:
50 µM P: 5 µM As – blue; 1000 µM P: 100 µM As – cyan; 50 µM P: 10 µM As – green;
50 µM P: 100 µM As – pink; and 500 µM P: 100 µM As – red.
(I) line
blue
cyan
green
pink
red
(J) line
cyan
green
pink
red
blue
green
pink
red
blue
cyan
pink
red
blue
cyan
green
red
blue
cyan
green
pink
Mean
Difference
(I-J)
.9890
-7.0540*
1.5245
2.6165
-.9890
-8.0430*
.5355
1.6275
7.0540*
8.0430*
8.5785*
9.6705*
-1.5245
-.5355
-8.5785*
1.0920
-2.6165
-1.6275
-9.6705*
-1.0920
Std. Error
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
1.26052
Sig.
.468
.003
.281
.093
.468
.001
.689
.253
.003
.001
.001
.001
.281
.689
.001
.426
.093
.253
.001
.426
Based on observed means.
*. The mean difference is significant at the .05 level.
95% Confidence Interval
Lower Bound Upper Bound
-2.2513
4.2293
-10.2943
-3.8137
-1.7158
4.7648
-.6238
5.8568
-4.2293
2.2513
-11.2833
-4.8027
-2.7048
3.7758
-1.6128
4.8678
3.8137
10.2943
4.8027
11.2833
5.3382
11.8188
6.4302
12.9108
-4.7648
1.7158
-3.7758
2.7048
-11.8188
-5.3382
-2.1483
4.3323
-5.8568
.6238
-4.8678
1.6128
-12.9108
-6.4302
-4.3323
2.1483
149
Table B.3. Univariate analysis of variance (ANOVA) data depicting significant
differences (“*”; α=0.05) in the arsenate fraction at the final sampling in Agrobacterium
tumefaciens. str. 5B P:As trials, as shown in Figure 3.2. Line color corresponds to P:As
ratio as follows: 50 µM P: 5 µM As – blue; 1000 µM P: 100 µM As – cyan; 50 µM P: 10
µM As – green; 50 µM P: 100 µM As – pink; and 500 µM P: 100 µM As – red.
(I) color
blue
cyan
green
pink
red
(J) color
cyan
green
pink
red
blue
green
pink
red
blue
cyan
pink
red
blue
cyan
green
red
blue
cyan
green
pink
Mean
Difference
Std. Error
(I-J)
-87.8563*
1.85163
4.5447
1.85163
4.5602
1.85163
-86.0008*
1.85163
87.8563*
1.85163
92.4010*
1.85163
92.4165*
1.85163
1.8555
1.85163
-4.5447
1.85163
-92.4010*
1.85163
.0155
1.85163
-90.5455*
1.85163
-4.5602
1.85163
-92.4165*
1.85163
-.0155
1.85163
-90.5610*
1.85163
86.0008*
1.85163
-1.8555
1.85163
90.5455*
1.85163
90.5610*
1.85163
Sig.
.000
.058
.057
.000
.000
.000
.000
.362
.058
.000
.994
.000
.057
.000
.994
.000
.000
.362
.000
.000
Based on observed means.
*. The mean difference is significant at the .05 level.
95% Confidence Interval
Lower Bound Upper Bound
-92.6161
-83.0966
-.2151
9.3045
-.1996
9.3200
-90.7606
-81.2410
83.0966
92.6161
87.6413
97.1608
87.6567
97.1763
-2.9042
6.6153
-9.3045
.2151
-97.1608
-87.6413
-4.7443
4.7752
-95.3053
-85.7857
-9.3200
.1996
-97.1763
-87.6567
-4.7752
4.7443
-95.3207
-85.8012
81.2410
90.7606
-6.6153
2.9042
85.7857
95.3053
85.8012
95.3207
150
Table B.4. Univariate analysis of variance (ANOVA) data depicting significant
differences (“*”; α=0.05) in the arsenate fraction at the final sampling in Arthrobacter sp.
str. S6 P:As trials, as shown in Figure 3.2. Line color corresponds to P:As ratio as follows:
50 µM P: 5 µM As – blue; 1000 µM P: 100 µM As – cyan; 50 µM P: 10 µM As – green;
50 µM P: 100 µM As – pink; and 500 µM P: 100 µM As – red.
(I) line
blue
cyan
green
pink
red
(J) line
cyan
green
pink
red
blue
green
pink
red
blue
cyan
pink
red
blue
cyan
green
red
blue
cyan
green
pink
Mean
Difference
(I-J)
-91.6645*
-2.6900
-2.5980
-78.6305*
91.6645*
88.9745*
89.0665*
13.0340*
2.6900
-88.9745*
.0920
-75.9405*
2.5980
-89.0665*
-.0920
-76.0325*
78.6305*
-13.0340*
75.9405*
76.0325*
Std. Error
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
3.22528
Sig.
.000
.442
.457
.000
.000
.000
.000
.010
.442
.000
.978
.000
.457
.000
.978
.000
.000
.010
.000
.000
Based on observed means.
*. The mean difference is significant at the .05 level.
95% Confidence Interval
Lower Bound Upper Bound
-99.9553
-83.3737
-10.9808
5.6008
-10.8888
5.6928
-86.9213
-70.3397
83.3737
99.9553
80.6837
97.2653
80.7757
97.3573
4.7432
21.3248
-5.6008
10.9808
-97.2653
-80.6837
-8.1988
8.3828
-84.2313
-67.6497
-5.6928
10.8888
-97.3573
-80.7757
-8.3828
8.1988
-84.3233
-67.7417
70.3397
86.9213
-21.3248
-4.7432
67.6497
84.2313
67.7417
84.3233
151
Table B.5. Univariate analysis of variance (ANOVA) data depicting significant
differences (“*”; α=0.05) in the arsenate fraction at the final sampling in Bacillus sp. str.
S18 P:As trials, as shown in Figure 3.2. Line color corresponds to P:As ratio as follows:
50 µM P: 5 µM As – blue; 1000 µM P: 100 µM As – cyan; 50 µM P: 10 µM As – green;
50 µM P: 100 µM As – pink; and 500 µM P: 100 µM As – red.
(I) line
blue
cyan
green
pink
red
(J) line
cyan
green
pink
red
blue
green
pink
red
blue
cyan
pink
red
blue
cyan
green
red
blue
cyan
green
pink
Mean
Difference
(I-J)
-65.5490*
-32.0570*
-1.1500
-56.7515*
65.5490*
33.4920*
64.3990*
8.7975*
32.0570*
-33.4920*
30.9070*
-24.6945*
1.1500
-64.3990*
-30.9070*
-55.6015*
56.7515*
-8.7975*
24.6945*
55.6015*
Std. Error
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
3.12447
Sig.
.000
.000
.728
.000
.000
.000
.000
.037
.000
.000
.000
.001
.728
.000
.000
.000
.000
.037
.001
.000
Based on observed means.
*. The mean difference is significant at the .05 level.
95% Confidence Interval
Lower Bound Upper Bound
-73.5807
-57.5173
-40.0887
-24.0253
-9.1817
6.8817
-64.7832
-48.7198
57.5173
73.5807
25.4603
41.5237
56.3673
72.4307
.7658
16.8292
24.0253
40.0887
-41.5237
-25.4603
22.8753
38.9387
-32.7262
-16.6628
-6.8817
9.1817
-72.4307
-56.3673
-38.9387
-22.8753
-63.6332
-47.5698
48.7198
64.7832
-16.8292
-.7658
16.6628
32.7262
47.5698
63.6332
152
Michaelis-Menten Equation Fitting for Agrobacterium str. 5B
v=Vmax/(1+(Km/[S1]))
estimated Vmax
5.00E-11
estimated Km
0.0550
Extrapolation
[S1] =
v = amt. reduction
corresponding [As]
mmol/hr/cell
(in mM)
(at max rate interval)
0.52
4.46E-11
0.057
2.48E-11
0.0043
3.25E-12
0.0017
1.09E-12
Var:
4.E-22
r2:
Y'
4.52E-11
2.54E-11
3.63E-12
1.50E-12
Sum:
(Y'-Y)2
3.80E-25
4.02E-25
1.38E-25
1.71E-25
1.09E-24
x
1.000
0.100
0.010
0.007
0.005
y
4.74E-11
3.23E-11
7.69E-12
5.65E-12
4.17E-12
0.999
5.00E-11
4.50E-11
v (mmol/hr/cell)
4.00E-11
3.50E-11
3.00E-11
Actual
Fitted
2.50E-11
2.00E-11
1.50E-11
1.00E-11
5.00E-12
0.00E+00
0
0.2
0.4
0.6
0.8
1
1.2
[S] (mM)
Figure B.16. Michaelis-Menten kinetic modeling for Agrobacterium tumefaciens str. 5B.
The model was used to estimate the maximum velocity of the reduction reaction (VMAX)
and the Michaelis constant (KM) based on the known maximum arsenate reduction velocity
(υ) and corresponding arsenate concentration (S1) of the four P:As experiments that
included an initial phosphate concentration of 50 µM.
153
Michaelis Competitive Inhibition Equation Fitting for Agrobacterium str. 5B
v=Vmax[S2]/Km(1+([I]/Ki)+[S2]
estimated Vmax
5.00E-11
estimated Km
0.0550
initial [As]
0.1000
estimated Ki
0.0236
[S2] =
corresponding
[As]
(in mmol)
0.057
0.1
0.1
[I] =
corresponding [P] v (@ max v (@ max rate
(in mmol)
rate interval)
interval)
X
(Y'-Y)2
Y
Y'
0.03
2.48E-11
2.27E-11
4.41E-24
0.48
7.92E-13
3.97E-12
1.01E-23
0.95
5.64E-13
2.11E-12
2.40E-24
Var:
2
r:
Sum: 1.69E-23
2.E-22
Extrapolation
x
y
0.05
1.84E-11
0.50
3.79E-12
0.80
2.48E-12
1.00
2.01E-12
0.971
3.00E-11
v (mmol/hr/cell)
2.50E-11
2.00E-11
Actual
1.50E-11
Fitted
1.00E-11
5.00E-12
0.00E+00
0.00
0.20
0.40
0.60
0.80
1.00
1.20
[I] (mM)
Figure B.17.
Michaelis-Menten enzyme inhibition modeling for Agrobacterium
tumefaciens str. 5B. The model was used to estimate the phosphate inhibition constant
(Ki) based on the calculated VMAX and KM, in addition to the known maximum arsenatereduction velocity (υ) and the corresponding phosphate (I) and arsenate (S2) concentrations
of the three P:As experiments that included an initial arsenate concentration of 100 µM.
154
Michaelis-Menten Equation Fitting for Arthrobacter str. S6
v=Vmax/(1+(Km/[S1]))
estimated Vmax
1.90E-11
estimated Km
0.0700
Extrapolation
[S1] =
v = amt. reduction
corresponding [As]
mmol/hr/cell
(in mM)
(at max rate interval)
0.83
1.64E-11
0.06
9.35E-12
0.006
1.92E-12
0.0025
1.17E-12
Var:
5.E-23
r2:
Y'
1.75E-11
8.77E-12
1.50E-12
6.55E-13
Sum:
(Y'-Y)2
1.30E-24
3.38E-25
1.77E-25
2.64E-25
2.08E-24
x
1.000
0.100
0.010
0.007
0.005
y
1.78E-11
1.12E-11
2.38E-12
1.73E-12
1.27E-12
0.990
1.80E-11
v (mmol/hr/cell)
1.60E-11
1.40E-11
1.20E-11
1.00E-11
Actual
8.00E-12
Fitted
6.00E-12
4.00E-12
2.00E-12
0.00E+00
0
0.2
0.4
0.6
0.8
1
1.2
[S] (mM)
Figure B.18. Michaelis-Menten kinetic modeling for Arthrobacter sp. str. S6. The model
was used to estimate the maximum velocity of the reduction reaction (VMAX) and the
Michaelis constant (KM) based on the known maximum arsenate reduction velocity (υ) and
corresponding arsenate concentration (S1) of the four P:As experiments that included an
initial phosphate concentration of 50 µM.
155
Michaelis Competitive Inhibition Equation Fitting for Arthrobacter str. S6
v=Vmax[S2]/Km(1+([I]/Ki)+[S2]
estimated Vmax
1.90E-11
estimated Km
0.0700
initial [As]
0.1000
estimated Ki
0.0900
[S2] =
[I] =
corresponding corresponding [P] v (@ max
[As]
(mmol)
rate interval)
X
(mmol)
Y
0.06
0.04
9.35E-12
0.1
0.50
3.05E-12
0.1
1.00
6.29E-13
Var:
2
r:
2.E-23
v (@ max
rate
interval)
(Y'-Y)2
Y'
7.14E-12 4.87E-24
3.40E-12 1.21E-25
2.00E-12 1.89E-24
Sum: 6.88E-24
Extrapolation
x
y
0.05
9.10E-12
0.50
3.40E-12
0.80
2.40E-12
1.00
2.00E-12
0.887
1.00E-11
9.00E-12
v (mmol/hr/cell)
8.00E-12
7.00E-12
6.00E-12
Actual
Fitted
5.00E-12
4.00E-12
3.00E-12
2.00E-12
1.00E-12
0.00E+00
0.00
0.20
0.40
0.60
0.80
1.00
1.20
[I] (mM)
Figure B.19. Michaelis-Menten enzyme inhibition modeling for Arthrobacter sp. str. S6.
The model was used to estimate the phosphate inhibition constant (Ki) based on the
calculated VMAX and KM, in addition to the known maximum arsenate-reduction velocity
(υ) and the corresponding phosphate (I) and arsenate (S2) concentrations of the three P:As
experiments that included an initial arsenate concentration of 100 µM.
156
Michaelis-Menten Equation Fitting for Bacillus str. S18
v=Vmax/(1+(Km/[S1]))
estimated Vmax
1.75E-09
estimated Km
7.50E-02
Extrapolation
[S1] =
v = amt. reduction
mmol/hr/cell
corresponding [As]
(at max rate interval)
(in mM)
1
1.75E-09
0.07
8.39E-10
0.0076
2.68E-10
0.0035
1.85E-11
Var:
6.E-19
r2:
Y'
1.63E-09
8.45E-10
1.61E-10
7.80E-11
Sum:
(Y'-Y)2
1.44E-20
3.73E-23
1.14E-20
3.54E-21
2.93E-20
x
1.000
0.100
0.010
0.007
0.005
y
1.63E-09
1.00E-09
2.06E-10
1.49E-10
1.09E-10
0.988
2.00E-09
1.80E-09
v (mmol/hr/cell)
1.60E-09
1.40E-09
1.20E-09
Actual
1.00E-09
Fitted
8.00E-10
6.00E-10
4.00E-10
2.00E-10
0.00E+00
0
0.2
0.4
0.6
0.8
1
1.2
[S] (mM)
Figure B.20. Michaelis-Menten kinetic modeling for Bacillus sp. str. S18. The model was
used to estimate the maximum velocity of the reduction reaction (VMAX) and the Michaelis
constant (KM) based on the known maximum arsenate reduction velocity (υ) and
corresponding arsenate concentration (S1) of the four P:As experiments that included an
initial phosphate concentration of 50 µM.
157
Michaelis Competitive Inhibition Equation Fitting for Bacillus str. S18
v=Vmax[S2]/Km(1+([I]/Ki)+[S2]
estimated Vmax
1.75E-09
estimated Km
0.0750
initial [As]
0.1000
estimated Ki
0.1050
[S2] =
corresponding
[As]
(mmol)
0.07
0.1
0.1
[I] =
corresponding
[P] (mmol)
X
0.04
0.50
1.00
Var:
2
r:
v (@ max
rate interval)
Y
8.39E-10
2.00E-10
2.64E-10
v (@ max
rate
interval)
Y'
6.94E-10
3.29E-10
1.97E-10
1.E-19
2
(Y'-Y)
2.08E-20
1.66E-20
4.52E-21
Sum: 4.20E-20
Extrapolation
x
y
0.05
8.31E-10
0.50
3.29E-10
0.80
2.35E-10
1.00
1.97E-10
0.887
9.00E-10
8.00E-10
v (mmol/hr/cell)
7.00E-10
6.00E-10
5.00E-10
Actual
Fitted
4.00E-10
3.00E-10
2.00E-10
1.00E-10
0.00E+00
0.00
0.20
0.40
0.60
0.80
1.00
1.20
[I] (mM)
Figure B.21. Michaelis-Menten enzyme inhibition modeling for Bacillus sp. str. S18. The
model was used to estimate the phosphate inhibition constant (Ki) based on the calculated
VMAX and KM, in addition to the known maximum arsenate-reduction velocity (υ) and the
corresponding phosphate (I) and arsenate (S2) concentrations of the three P:As experiments
that included an initial arsenate concentration of 100 µM.
158
APPENDIX C:
PRELIMINARY ANALYSIS OF SOIL SAMPLES COLLECTED
FROM ANACONDA-DEER LODGE COUNTY, MONTANA
159
Introduction
A total of 11 soil samples were obtained on November 9, 2004 from various locations
within Anaconda-Deer Lodge County, Montana, an area known for its vast metal
contamination. In fact, several of the sampling points were located on areas assigned to the
National Priorities List (NPL) as severely contaminated Superfund sites. The subsequent
sections include thorough descriptions of each sample site and images of each site are
included in Figure C.1.
All samples were collected as a composite of several sub-samples on each respective
locale using a shovel sterilized with 10% bleach and homogenized in sterilized plastic
buckets. The soils were sieved (2 mm) under a laboratory sterile air hood and stored at 4oC
until future use. A homogenized soil extract was immediately collected from each sample
after sieving and submitted for chemical analysis, Table C.1. A second homogenized sample
was collected and stored at -80oC for future molecular analysis.
Initial separation of
microbial signatures was performed on all samples (Figure C.2) via denaturing gradient gel
electrophoresis (DGGE), as detailed in the Materials and Methods section of Chapter 2.
Miles Crossing - Silver Bow Creek NPL Site
Soil samples were collected at a location within the Silver Bow Creek NPL site known as
Miles Crossing. The site has a long history of metal contamination, likely the result of
fluvial deposited mine, mill, and smelter wastes washed downstream from Butte and
Anaconda mining operations in the summer of 1908 during the greatest flood event in
160
western Montana’s recorded history (Smith et al., 1998). Evaporative salts and tailings
deposits, predominantly with the appearance of copper and iron, were visible on the soil
surface.
Visible vegetation included: Baltic rush (Juncus balticus), Kochia (Kochia
scoparia), Redtop (Agrostis stolonifera), Saltgrass (Distichlis stricta), Spotted knapweed
(Centaurea maculosa), Western wheatgrass (Pascopyrum smithii), Willow (Salix sp.), and
Tufted hairgrass (Deschampsia caespitosa), which is a metal tolerant species and an indicator
of tailings contamination (D. Neuman, personal communication, November 9, 2004). Soil
texture and vegetation gradients were visible on the site (Figure C.1A); therefore, composite
samples were collected from three spatial landscape positions to elucidate any corresponding
chemical and microbiological gradients. A description of each sampling point follows.
Miles Crossing Pit 1 (MC1)
• Coordinates: N 46.01172 o W 112.72380 o
• Elevation: 1,615 m
• Sub-location: mid landscape position (Figure C.1B)
• Vegetative cover: none
• Visible soil characteristics: reddish tailings on soil surface and subsoil stratification
• Sample depth range from surface: 10 – 51 cm
161
Miles Crossing Pit 2
• Coordinates: N 46.01195 o W 112.72327 o
• Elevation: 1,613 m
• Sub-location: low landscape position, adjacent to creek bank (Figure C.1C)
• Vegetative cover: none
• Visible soil characteristics: two distinct soil layers and groundwater encroachment in pit
led to the collection of 3 separate samples:
Upper Horizon (MC2-5)
ƒ Soil characteristics: red, orange and yellow well-mixed tailings materials
ƒ Sample depth range from surface: 13 – 25 cm
Lower Horizon (MC2-10)
ƒ Soil characteristics: organic-rich; dark black; saturated, at groundwater layer
ƒ Sample depth range from surface: 25 – 46 cm
Groundwater (MC2-18)
ƒ Characteristics: aqueous and suspended sediment
ƒ Sample depth range from surface: 46 cm
162
Miles Crossing Pit 3 (MC3)
• Coordinates: N 46.01137 o W 112.72372 o
• Elevation: 1,617 m
• Sub location: upper landscape position (Figure C.1D)
• Vegetative cover: ~50%, mostly Saltgrass (Distichlis stricta)
• Visible soil characteristics: tailings material just beneath the slightly crusted, salt-rich
surface, subsoil fibrous throughout, appeared to be an A horizon buried beneath tailings
deposits covered by second A horizon
• Sample depth range from surface: 0 – 15 cm
Near Substation - East of Washoe Smelter Stack
The site’s proximity to an electrical substation east of the Washoe smelter stack (Figure
C.1E) yielded the designation of the soil samples collected within. The site has a history of
aerial metal contamination from fallout originating from copper smelters (Anaconda,
Montana). Not surprisingly, no visible tailings were present. The site boasted a fair species
richness with visible vegetation composed of: Baltic rush (Juncus balticus), Basin wildrye
(Leymus cinereus), Redtop (Agrostis stolonifera), Saltgrass (Distichlis stricta), Spotted
knapweed (Centaurea maculosa), Western wheatgrass (Pascopyrum smithii), and Wood’s
rose (Rosa woodsii).
163
Substation Pit
• Coordinates: N 46.10313 o W 112.87296 o
• Elevation: 1573 m
• Vegetative cover: ~65%
• Visible soil characteristics: two distinct soil layers (Figure C.1F) led to separate sampling
of each layer and an additional sample was taken of the top 5 cm of the surface soil:
Surface Soil (SUB-2)
ƒ Sample depth range from surface: 0 – 5 cm
Upper Horizon (SUB-6)
ƒ Sample depth range from surface: 0 – 15 cm
Lower Horizon (SUB-12)
ƒ Sample depth range from surface: 15 – 30 cm
The soil samples from the Substation site were chosen for additional molecular and
microbial cultivation analysis due to their high relative concentrations of arsenic and low
concentrations of copper (Table C.1).
The initial DGGE separation attempt on the
Anaconda-Deer Lodge soils (Figure C.2) did not produce a clear microbial signature for the
three Substation samples; therefore additional DGGE analysis employing various DNA
extract dilutions prior to PCR amplification was conducted on these samples in an attempt to
produce clearer banding patterns (Figure C.3). Microbial isolates were cultivated from
homogenized soil samples that had previously been stored at 4oC. The colony morphology
of each purified isolate was recorded (Table C.2). DNA was extracted from two isolates
164
attributed to the 0 – 5 cm depth range, six from the 0 – 15 cm range and seven belonging to
the 15 – 30 cm range, submitted for sequencing by TGen (Phoenix, AZ) and compared to
known microbial sequences in the GenBank database (Table C.3). Isolates were then tested
for their ability to oxidize or reduce As and assigned a corresponding As phenotype (Table
C.4). (See the Materials and Methods section of Chapter 2 for the detailed methodology of
all microbiological techniques mentioned herein.) Several additional isolates were cultivated
from the Substation soil samples; however, DNA significant for sequencing was
unattainable, thus they were dropped from further experimentation. An isolate from the
Substation 0 – 15 cm depth range, Arthrobacter-like str. S6, and an isolate from the
Substation 15 – 30 cm depth range, Bacillus drentensis-like str. S18, were selected for
inclusion in subsequent phosphate experiments, detailed in Chapter 3.
Near Gardner Ditch – Anaconda Smelter Site
Gardner Ditch sampling point was located on state land having previously been aerially
contaminated with metal fallout from the Anaconda-area copper smelters (Figure C.1G).
Visible vegetation on the site included: Basin wildrye (Leymus cinereus), Fescue (Festuca
ovina), Rubber rabbitbrush (Chrysothamnus nauseosus), Spotted knapweed (Centaurea
maculosa), and Western wheatgrass (Pascopyrum smithii).
sampling point follows.
Further description of the
165
Gardner Ditch Pit (GD)
• Coordinates: N 46.19208 o W 112.86601 o
• Elevation: 1532 m
• Vegetative cover: ~50%, dominated by Spotted knapweed as a result of severe
overgrazing (D. Neuman, personal communication, November 9, 2004)
• Visible soil characteristics: very rocky, 30% coarse fragments present in soil sample, no
visible tailings (Figure C.1H)
• Sample depth range from surface: 0 – 8 cm
Clark Fork River NPL Site
The Clark Fork sample was collected along the bank of the Clark Fork River north of
Perkins Lane near Warm Springs, Montana (Figure C.1I). The area is listed as a NPL site
due to its high arsenic and copper concentrations and phytotoxic soils, the result of fluvial
deposition of nearby mine, mill, and smelter wastes. Mine tailings were not visible from the
soil surface, but were uncovered at depth. Vegetative cover consisted mostly of grasses
highly grazed to stem base by cattle and was dominated by Basin wildrye (Leymus cinereus)
and Willow (Salix sp.).
166
Clark Fork Pit (CF)
• Coordinates: N 46.21024o W 112.76768 o
• Elevation: 1,453 m
• Vegetative cover: ~90%
• Visible soil characteristics: buried tailings (Figure C.1J)
• Sample depth range from surface: 20 – 30 cm
167
A) Landscape at Miles Crossing
B) MC1 Pit
C) MC2 Pit
D) MC3 Pit
E) Landscape at Substation
F) SUB Pit
Figure C.1. Pictures of the landscape and soil pits at each sampling point.
168
G) Landscape at Gardner Ditch
H) GD Pit
I) Landscape at Clark Fork
J) CF Pit
Figure C.1 con’t. Pictures of the landscape and soil pits at each sampling point.
7.3
123
1329
1279
4591
<2
<2
<1
<0.48
<0.37
228
<0.18
<3
5613
<15
pH1
K1
Mg1
Na1
Ca1
Cu
Fe
As
Pb
Al
Si
Cd
P
S
Zn
1
30.7
39.6
% H2O (at 105 C)
<15
1185
19
<0.18
420
<0.37
<0.48
5
<2
14
3693
196
769
1043
6.4
Analysis performed on saturated paste of sample.
O
231
77
Alkalinity (mgCaCO3/L)
1.04
1.21
EC1 (mmhos/cm)
GD
CF
Analyte
15201
34924
10
39.680
1153
1019
3.861
1
4
8557
2246
2466
1876
184
4.2
28.7
<10
3.92
MC1
1774
15747
6
6.851
1748
2
<0.48
1
517
2832
2570
2123
1831
8
2.8
34.7
<10
2.74
MC2-5
15
19239
16
0.445
1011
11
<0.48
5
2833
3
9232
2962
3373
711
6
72.3
29
2.89
MC2-10
<15
15061
10
0.356
1289
7
<0.48
5
2281
<2
6986
2845
2530
621
6.1
75.9
48
2.46
MC2-18
2615
26224
6
5.783
1410
204
0.483
1
5
1213
13548
21083
2505
4013
4.6
44.4
<10
4.67
MC3
184
655
16
1.690
915
15
<0.48
12
4
69
1073
209
181
1008
4.6
49.5
<10
0.58
SUB-2
92
655
32
0.356
1
4
<0.48
40
<2
11
1397
157
189
624
5.4
44.3
29
0.51
SUB-6
<15
405
10
<0.18
509
<0.37
<0.48
85
<2
<2
1148
178
218
15
7.5
44.9
77
0.30
SUB-12
Table C.1. Chemical characterization of soil samples collected in Anaconda-Deer Lodge County, Montana. (Analytes
reported in µM unless otherwise noted.)
169
1:10 Amsterdam
1:10 SUB-12
1:10 SUB-6
1:10 SUB-2
MC2-18
MC2-10
MC2-5
MC3
MC1
CF
1:10 GD
GD
170
Figure C.2. Initial DGGE separation of microbial signatures in soils collected from
Anaconda-Deer Lodge County, Montana. A 1:10 dilution of DNA extract was employed
in the PCR amplification of samples exhibiting particularly high quantities of DNA based
on quantification after electrophoresis on a 1.2% agarose gel.
SUB-12 1:5000
SUB-12 1:1000
SUB-12 1:100
SUB-6 1:5000
SUB-6 1:1000
SUB-6 1:50 dup
SUB-6 1:50
SUB-2 1:5000
SUB-2 1:1000
SUB-2 1:100
SUB-2 1:10
171
Figure C.3. DGGE gel displaying results of employing various dilutions of the DNA
extract from the three Substation soil samples prior to PCR amplification in an attempt to
produce clearer banding patterns. This further analysis was in response to the poor
banding patterns obtained in the initial DGGE separation attempt, Figure C.2.
172
Table C.2. Colony morphology of isolates cultivated from Substation soil samples. (Plated
on R2A media.)
Isolate Strain
Name
Description
S1
whitish-cream, opaque, shiny, convex, punctiform colony with smooth
margin
S2
cream, translucent, shiny, punctiform colony with smooth margin
S6
cream, opaque, shiny, irregular colony with undulate margin; tends to
smear
S7
cream, opaque, shiny, concentric, moderately umbonate, irregular
colony with smooth margin
S8
cream, opaque, shiny, convex, punctiform colony with smooth margin
S9
cream, opaque, shiny, low convex, irregular colony with translucent
smooth margin
S13
peachish-cream, opaque, shiny, irregular colony with smooth margin;
tends to smear
S14
peachish-cream, opaque, shiny, irregular colony with smooth margin;
tends to smear
S15
cream, opaque, shiny, convex, irregular colony with clear, translucent
outer ring and white fuzzy top; tends to smear
S16
cream, opaque, shiny, punctiform colony with smooth margin
S17
brown, wavy interlaced, irregular colony with diffuse translucent thick
cream outer ring and convex rugose fuzzy white top
S18
cream, opaque, shiny, irregular colony , diffusing to translucent smooth
outer ring/margin; tends to smear
S19
cream, opaque, shiny, irregular colony , diffusing to translucent smooth
thick outer ring/margin; tends to smear
S21
cream, opaque, shiny, concentric, irregular colony, diffusing to lighter
undulate margin
S22
brown, shiny, circular colony with moderately convex rugose white
fuzzy top and smooth margin
173
Table C.3. Pure cultures of bacteria isolated from Substation soils collected in AnacondaDeer Lodge County, Montana. Isolates are identified based on strain number, closest
cultivated relative in GenBank, major phylogenetic division, and experimentally
determined As phenotype (i.e. arsenite oxidizer or arsenate reducer).
Isolate
Strain
Name
Closest GenBank Neighbor (% sim.)
Phylogenic Group
As
Phenotype
Isolates cultivated from the 0-5 cm soil depth range.
S1
S2
Arthrobacter oxydans (98.57) 2
Arthrobacter polychromogenes (98.2) 3
Isolates cultivated from the 0-15 cm soil depth range.
S6 Arthrobacter sp. str. 108 (99.5) 7
S7 Bacillus simplex str. WN570 (99.9) 5
S8 Arthrobacter oxydans str. c306 (99.7) 2
S9 Paenibacillus sp. str. DS-1 (97.4) 6
S13 Arthrobacter polychromogenes (98.2) 3
S14 Arthrobacter ramosus (99.5) 4
Isolates cultivated from the 15-30 cm soil depth range.
S15 Phyllobacterium brassicacearum str. STM (99.9)
S16 Arthrobacter ramosus (99.1) 4
S17 Saccharothrix sp. str. LM 150 (99.5)
S18 Bacillus drentensis str. WN575 (99.6) 5, 7
S19 Paenibacillus sp. str. DS-1 (98.0) 6
S21 Paenibacillus sp. str. DS-1 (97.7) 6
S22 Streptomyces resistomycificus str. NBRC 12814 (98.9)
1
Isolate showed neither oxidation or reduction capability.
Actinobacteria
Actinobacteria
reducer
reducer
Actinobacteria
Firmicutes
Actinobacteria
Firmicutes
Actinobacteria
Actinobacteria
reducer
reducer
reducer
reducer
reducer
reducer
Alphaproteobacteria
Actinobacteria
Actinobacteria
Firmicutes
Firmicutes
Firmicutes
Actinobacteria
2
Arthrobacter oxydans -like strs. DMS1 and DMS8 are 98.57% similar.
3
Arthrobacter polychromogenes -like strs. S2 and S13 are 100% similar.
4
Arthrobacter ramosus -like strs. S14 and S16 are 99.44% similar.
5
Bacillus -like strs. S7 and S18 are 95.24% similar.
1
reducer
reducer
reducer
reducer
reducer
reducer
6
Paenibacillus -like str. S9 is 99.55% similar to str. S19 and 99.17% similar to str. S21. And
Paenibacillus -like strs. S19 and S21 are 99.31% similar.
7
Isolate used in the experimental screening in Chapter 3: Inhibition of Microbial Arsenate
Reduction by Phosphate .
174
Table C.4. Data from As phenotype screening of isolates cultured from Substation soil
samples. Isolates treated with 2.9 mg L-1 (38.7 µM) of both AsIII and AsV and grown in
duplicate for approximately 8 days in SSE/SSM media.
Isolate Strain
Name
Measured
[As]
(mg/L)
Dilution
Factor
Final
[As]
(mg/L)
AsV in Isolate Media:AsV in
Sterile Control (SC) Media
(%)
Observed As
Transformation1
OD
(A 500)
0.57
0.64
0.48
0.58
0.70
0.65
3.29
4.22
17
19
14
17
20
19
96
123
reduction
reduction
reduction
reduction
reduction
reduction
none
none
0.707
0.951
0.188
0.183
1.174
1.141
0.212
0.264
Samples speciated to include As V only.
S1
0.4
1.40
S1
0.5
1.40
S2
0.3
1.40
S2
0.4
1.40
S6
0.5
1.40
S6
0.5
1.40
S7
2.4
1.40
S7
3.0
1.40
S72
1.4
1.40
1.93
59
reduction
0.287
2
S7
S8
S8
S9
S9
S13
S13
S14
S14
S15
S15
1.6
0.5
0.5
0.5
0.6
0.4
0.4
0.5
0.5
1.8
2.0
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
2.22
0.66
0.67
0.64
0.79
0.51
0.63
0.67
0.68
2.46
2.77
67
19
20
19
23
15
18
20
20
76
85
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
none
none
0.315
1.001
1.118
0.416
0.373
0.448
0.362
1.301
1.303
0.251
0.209
S152
2.1
1.40
2.94
89
none
0.202
S152
S16
S16
S17
S17
S18
S18
S19
S19
S20
S20
2.2
0.7
0.8
0.3
0.4
0.7
0.7
0.8
1.1
1.1
1.2
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
3.15
1.03
1.11
0.48
0.53
0.92
1.04
1.11
1.57
1.54
1.63
96
32
34
15
16
28
32
34
48
47
50
none
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
reduction
0.185
1.163
0.999
0.036
0.024
0.464
0.489
0.26
0.246
0.389
0.283
1
The observed As transformation of an isolate was assigned based on the ratio of AsV in isolate media:AsV in
sterile control media. The following ratio ranges were assigned the corresponding result: <75% = reduction,
>125% = oxidation and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly
low optical density (OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.
175
Table C.4 con't. Data from As phenotype screening of isolates cultured from Substation
soil samples. Isolates treated with 2.9 mg L-1 (38.7 µM) of both AsIII and AsV and grown
in duplicate for approximately 8 days in SSE/SSM media.
Isolate Strain
Name
Measured
[As]
(mg/L)
Dilution
Factor
Final
[As]
(mg/L)
AsV in Isolate Media:AsV in
Sterile Control (SC) Media
(%)
Observed As
Transformation1
OD
(A 500)
S21
S21
S22
S22
0.7
0.7
0.5
0.5
1.40
1.40
1.40
1.40
1.03
0.96
0.73
0.70
32
29
22
21
reduction
reduction
reduction
reduction
0.305
0.254
0.371
0.222
SC3 Trial 1
2.5
1.40
3.44
SC3 Trial 2
2.3
1.40
3.26
3
SC Trial 3
2.3
Total arsenic samples; run to
arsenic via methylation, etc.
S1
5.9
S2
5.9
S6
6.0
S7
6.1
1.40
3.29
ensure no loss of
1.00
1.00
1.00
1.00
5.87
5.87
6.01
6.12
S72
S8
S9
S13
S14
6.0
6.1
6.1
6.1
6.2
1.00
1.00
1.00
1.00
1.00
5.95
6.14
6.15
6.08
6.19
S152
S16
S17
S18
S19
S20
S21
S22
6.0
6.0
6.7
5.7
5.9
5.7
6.0
6.1
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
5.96
6.05
6.69
5.68
5.88
5.74
6.00
6.09
SC Trial 1
6.1
1.00
6.11
SC Trial 2
SC3 Trial 3
6.0
5.9
1.00
1.00
6.04
5.86
1
The observed As transformation of an isolate was assigned based on the ratio of AsV in isolate media:AsV in
sterile control media. The following ratio ranges were assigned the corresponding result: <75% = reduction,
>125% = oxidation and 75 - 125% = none.
2
Data from re-screening. Isolates with conflicting duplicate observed transformations or with an extrememly
low optical density (OD) were re-treated in duplicate.
3
Average of two duplicate sterile controls.