Biochimica et Biophysica Acta 1656 (2004) 148 – 155
www.bba-direct.com
Arsenite oxidation by the heterotroph Hydrogenophaga sp. str. NT-14:
the arsenite oxidase and its physiological electron acceptor
Rachel N. vanden Hoven, Joanne M. Santini *
Department of Microbiology, La Trobe University, 3086, Victoria, Melbourne, Australia
Received 29 December 2003; received in revised form 3 March 2004; accepted 4 March 2004
Available online 23 March 2004
Abstract
Heterotrophic arsenite oxidation by Hydrogenophaga sp. str. NT-14 is coupled to the reduction of oxygen and appears to yield energy for
growth. Purification and partial characterization of the arsenite oxidase revealed that it (1) contains two heterologous subunits, AroA (86
kDa) and AroB (16 kDa), (2) has a native molecular mass of 306 kDa suggesting an a3h3 configuration, and (3) contains molybdenum and
iron as cofactors. Although the Hydrogenophaga sp. str. NT-14 arsenite oxidase shares similarities to the arsenite oxidases purified from NT26 and Alcaligenes faecalis, it differs with respect to activity and overall conformation. A c-551-type cytochrome was purified from
Hydrogenophaga sp. str. NT-14 and appears to be the physiological electron acceptor for the arsenite oxidase. The cytochrome can also
accept electrons from the purified NT-26 arsenite oxidase. A hypothetical electron transport chain for heterotrophic arsenite oxidation is
proposed.
D 2004 Elsevier B.V. All rights reserved.
Keywords: Arsenite; Arsenite oxidase; Heterotroph; Cytochrome; Electron transport
1. Introduction
Arsenic is ubiquitous in the environment and has the
following four oxidation states, 0 (elemental), 3 (arsine),
+ 3 (arsenite) and + 5 (arsenate) [1,2]. The two soluble
forms of arsenic, arsenite [As(III)] and arsenate [As(V)], are
toxic to biological systems, with arsenite considered 100
times more toxic [3]. As(V) is an analogue of phosphate and
can enter cells through the phosphate transport system
where it has the ability to replace phosphate in ATP
synthesis and thus disrupts normal phosphorylation processes [3]. The mechanism by which arsenite enters the cell is
still under investigation, although two modes have been
proposed: (1) it can enter cells by diffusion [2] or (2) it can
enter cells through the glycerol transport system [4]. The
toxicity of As(III) is attributed to its ability to bind sulfhydryl groups of proteins, thus impairing their function [2].
Since the first report of bacterial arsenite oxidation by
Green in 1918 [5], several arsenite-oxidizing bacteria have
been isolated. These organisms have been isolated from
* Corresponding author. Tel.: +61-3-94792206; fax: +61-3-94791222.
E-mail address: j.santini@latrobe.edu.au (J.M. Santini).
0005-2728/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbabio.2004.03.001
different environments including, gold mines [6,7], sewage
[8], hyper-saline lake [9], soil [10], cattle dip fluid [5,11],
geothermal springs [12,13] and arsenic contaminated ground
water [14]. The oxidation of As(III) to As(V) when coupled to
the reduction of oxygen to water is an exergonic process [6].
The arsenite-oxidizing bacteria isolated to date can be
divided into two groups: (i) chemolithoautotrophs (aerobes
or anaerobes, using arsenite as the electron donor and CO2/
HCO3- as the sole carbon source) or (ii) heterotrophs (growth
in the presence of organic matter). There are currently two
chemolithoautotrophic arsenite-oxidizing bacteria that have
been studied in detail: the aerobe NT-26 [6] and the
facultative anaerobe MLHE1 [9]. NT-26, a member of the
a-Proteobacteria, was isolated from a goldmine and oxidizes As(III) to As(V) using oxygen as the terminal electron
acceptor [6]. MLHE1, a member of the g-Proteobacteria,
was isolated from a hyper-saline lake and oxidizes As(III) to
As(V) using NO3 as the terminal electron acceptor [9].
Several heterotrophic arsenite-oxidizing bacteria have
been isolated and include: Alcaligenes faecalis [8]; ULPAs1
[14]; Agrobacterium albertimagni [13]; and three Thermus
species [12,15]. It is thought that the oxidation of As(III) by
these bacteria is a detoxification process. Hydrogenophaga
R.N. vanden Hoven, J.M. Santini / Biochimica et Biophysica Acta 1656 (2004) 148–155
sp. str. NT-14, a member of the h-Proteobacteria [7], can
oxidize arsenite to arsenate heterotrophically and appears to
gain energy from this process as the final cell yield when
grown in the presence of arsenite is significantly higher
(OD600 0.132 F 0.005) than when grown in its absence
(OD600 0.154 F 0.01) [R.N. vanden Hoven and J.M. Santini, unpublished data].
Of the arsenite-oxidizing bacteria isolated, A. faecalis [8]
(a member of the h-Proteobacteria) [7] and NT-26 have been
studied in detail and their arsenite oxidases purified and
characterized [16,17]. The crystal structure of the A. faecalis
arsenite oxidase has been resolved [18] and the NT-26
arsenite oxidase genes have been sequenced [17]. The NT26 arsenite oxidase is soluble and located in the periplasm [6]
whereas that from A. faecalis is anchored to the periplasmic
face of the inner membrane [16]. The NT-26 Aro and A.
faecalis arsenite oxidase consist of two heterologous subunits, a large molybdenum-containing catalytic subunit (a)
and a small Rieske-type subunit (h) [16 – 18]. Recently, the
arsenite oxidase genes (aox) of ULPAs1, also a member of
the h-Proteobacteria, have been identified and sequenced
[19]. The two genes, aoxA and aoxB, encode a putative
arsenite oxidase that consists of two subunits, a small
Rieske-type protein (AoxA) and a larger molybdenum-containing protein (AoxB). All three arsenite oxidases share
sequence similarities at the amino acid level with the two
enzymes from ULPAs1 and A. faecalis sharing the highest
sequence identities (i.e., large subunit 72% and small subunit
65%); NT-26 Aro subunits share V 52% identity with the
respective subunits of the arsenite oxidases [17]. This is not
surprising given that ULPAs1 and A. faecalis are both
members of the h-Proteobacteria.
This report describes the purification and partial characterization of the Hydrogenophaga sp. str. NT-14 arsenite
oxidase and its physiological electron acceptor. This information has for the first time led to a proposed electron
transport chain for the generation of energy from heterotrophic arsenite oxidation.
2. Materials and methods
2.1. Purification of the arsenite oxidase and cytochrome c
Cells were grown in 5-l batch cultures containing
minimal salt medium (MSM) [17], 5 mM As(III) and
0.04% yeast extract. Late logarithmic phase (15 h) cells
were harvested by centrifugation at 21,000 g for 20 min
(4 jC). The cell pellet was suspended in 40-ml ice-cold 10
mM Tris/HCl (pH 8) and centrifuged at 27,000 g for 20
min (4 jC). Spheroplasts were formed as described previously [6] with the following modifications: (1) the cell
pellet was resuspended in ice-cold 750 mM sucrose/30 mM
Tris/HCl (pH 8), (2) the concentration of lysozyme added
was 7.7 mg/g, (3) cells were incubated with lysozyme for 4
min (0 jC), (4) after the addition of EDTA, the cells were
149
incubated for a further 3 min (0 jC), and (5) the cell
suspension was centrifuged at 30,000 g for 30 min (4 jC).
The proteins in the periplasmic fraction were precipitated
using ammonium sulfate at 50% and 80% saturation. The
precipitated proteins were suspended in 3-ml 50 mM MES
(pH 5.5), centrifuged at 21,000 g for 5 min and the
supernatant added to a previously equilibrated (50 mM
MES pH 5.5) PD-10 desalting column (Amersham Pharmacia Biotech). The proteins were eluted in 50 mM MES (pH
5.5) and concentrated using a Centricon 30 (YM-30 centrifugal filter device, Millipore). The sample was filtered
through a 0.22-Am filter and loaded onto a SP Sepharose
Fast Flow cation exchange column (Amersham Parmacia
Biotech) that had been previously equilibrated with 50 mM
MES (pH 5.5). The chromatography was carried out at room
temperature and the Aro and cytochrome were eluted from
the column using a 0– 1 M NaCl gradient in 50 mM MES
(pH 5.5). It is worth noting that the Hydrogenophaga sp. str.
NT-14 Aro could not be purified using a HIC column or
DEAE column as was done for the NT-26 [17] and A.
faecalis [16] arsenite oxidases, respectively. The Aro eluted
at a concentration of 0.14 M NaCl and the cytochrome
eluted at 0.09 M NaCl. Fractions containing Aro activity or
the cytochrome were pooled and concentrated and the
samples were loaded separately onto a Superdex 200 gel
filtration column (Amersham Pharmacia Biotech) previously equilibrated with 50 mM MES (pH 5.5)/100 mM NaCl.
The fractions containing Aro activity or the cytochrome
were pooled and concentrated.
2.2. Enzyme assays, absorbance spectra and protein
determination
Assays for arsenite oxidase activity were performed as
described previously [6] in the determined optimum buffer,
50 mM MES (pH 5.5). The concentration of arsenite
included in the assay was 2.5 mM (except for experiments
to determine the Km and Vmax, in which the arsenite
concentration ranged from 0.01 to 2.5 mM).
The oxidized and reduced states of the purified cytochrome were recorded with a UV absorbance wavelength
spectrum (nm) using a Cary 1E UV –Visible spectrophotometer. The cytochrome was diluted in 50 mM MES (pH
5.5) in a glass cuvette in the absence and presence of
purified Aro and the oxidized spectrum recorded. Reduction
of the cytochrome was initiated upon addition of 2.5 mM
arsenite to the cuvette.
Bradford reagent [20] was used to determine protein
concentrations. Bovine serum albumin served as the
standard.
2.3. Protein gel electrophoresis, transfer and cytochrome c
ingel activity staining
SDS-PAGE and transfer of the proteins to PVDF membranes was performed as described previously [21]. TMBZ-
150
R.N. vanden Hoven, J.M. Santini / Biochimica et Biophysica Acta 1656 (2004) 148–155
peroxidase activity staining of the heme c was performed as
described previously [22].
2.4. N-terminal sequencing and cofactor analyses
The N-terminal sequences and cofactor analyses of the
Aro and the N-terminal sequence of the cytochrome were
determined as described previously [21].
3. Results
3.1. Purification of the Aro
The Aro was purified by a two step (50% and 80%)
ammonium sulfate precipitation of the periplasm, followed
by cation-exchange and gel filtration chromatography,
which resulted in a 57.5-fold enrichment of the Aro
(Table 1). It is worth noting that in the purification table,
an increase in total activity (U) was observed between the
total cell extract and periplasmic fraction; this also occurred for the NT-26 purification table [17]. An explanation for this finding is that another protein(s) in the total
cell extract may be accepting electrons from the oxidation
of arsenite catalyzed by the Aro resulting in a decrease in
the amount of artificial electron acceptor (i.e., DCPIP)
reduced. The purified Aro contains two heterologous
subunits with molecular masses of 86 kDa (AroA) and
16 kDa (AroB), and appeared to be >99% pure (Fig. 1).
The native molecular mass determined by gel filtration
chromatography was 306 kDa and, assuming the presence
of both subunits in a stochiometric ratio of 1:1, the NT-14
Aro has an a3h3 configuration. The Km for arsenite was
35 AM and the V max 6.1 Amol of arsenite oxidizedmin 1mg protein 1, corresponding to an enzyme
turnover of 30.4 s 1 (based on the native molecular mass
of 309 kDa). The purified Aro was analyzed for the
presence of cofactors and found to contain iron
(10.6 F 0.4 mol/a3h3) and molybdenum (2.1 F 0.6 mol/
a3h3). The quantity of iron and molybdenum is lower
than expected and this may have been due to incomplete
hydrolysis of the enzyme or the loss of cofactors during
the purification procedure.
The N-terminal sequences of the AroA (38 amino acids)
and AroB (40 amino acids) subunits were determined (Fig.
2a and b, respectively). The N-terminal sequence of AroA
Purification step Total
Total
Specific
Purification
protein (mg) activity (U) activity (U/mg) fold (-fold)
2.0
10.2
9.5
2.3
1.5
was found to be similar to the molybdenum-containing
subunits of other arsenite oxidases [17 – 19]. The AroA Nterminal sequence showed three conserved cysteines, CysXaa2-Cys-Xaa3-Cys, which are involved in binding a
[3Fe4S] cluster in the A. faecalis arsenite oxidase [18]; this
conserved motif is also present in the other arsenite oxidases
[17,19]. Based on a BLASTP search, the Hydrogenophaga
sp. str. NT-14 AroA (Fig. 2a) displayed sequence identity to
the ULPAs1 putative AoxB (71%), A. faecalis a-subunit
(58%) and NT-26 AroA (55%). The Hydrogenophaga sp.
str. NT-14 AroB N-terminal sequence, based on a BLASTP
search, displayed sequence identity to the small (Riesketype) subunits of the other arsenite oxidases and includes the
ULPAs1 putative AoxA (48%), A. faecalis h-subunit (43%)
and NT-26 AroB (15%) (Fig. 2b). These subunits contain a
[2Fe2S] cluster similar to the Rieske subunits of the cytochrome bc1 and b6f complexes [17], which play an important role in electron transport chains of prokaryotes, plants
and animals. It is worth noting that the Hydrogenophaga sp.
str. NT-14, ULPAs1 and A. faecalis (all members of the hProteobacteria) small subunits share a higher sequence
identity to each other than to that of NT-26 (a member of
the a-Proteobacteria).
3.2. Purification of the cytochrome c
Table 1
Purification of the NT-14 arsenite oxidase
Cell extract
21.2
Periplasm
16.1
Desalting
7.7
Cation exchange 0.4
Gel filtration
0.3
Fig. 1. SDS-polyacrylamide gel of the purified arsenite oxidase stained with
Coomassie Blue R350. Lane 1, purified Aro; lane 2, molecular weight
markers: phosphorylase b (94 kDa), albumin (67 kDa), ovalbumin (43
kDa), carbonic anhydrase (30 kDa), trypsin inhibitor (20 kDa) and alactalbumin (14 kDa).
0.09
0.6
1.23
5.0
5.6
1
6.7
13.7
52.4
57.5
A cytochrome was detected during the purification of the
Aro. Based on SDS-polyacrylamide gel electrophoresis, the
cytochrome was 6 –6.5 kDa (Fig. 3a) but was not completely pure. The native molecular mass of the cytochrome,
based on gel filtration chromatography, was 11 kDa, suggesting that the cytochrome may be present as a homodimer.
The presence of a c-type heme was detected by ingel
R.N. vanden Hoven, J.M. Santini / Biochimica et Biophysica Acta 1656 (2004) 148–155
151
Fig. 2. Sequence alignments of the Hydrogenophaga sp. str. NT-14 arsenite oxidase subunits and cytochrome. (a) Hydrogenophaga sp. str. NT-14 AroA,
putative arsenite oxidase ULPAs1 AoxB (acc. no. AAN05581.1), A. faecalis (A.f) arsenite oxidase a (acc. No. nrp 1G8J_A), NT-26 arsenite oxidase AroA
(acc. no. AY345225). (b) Hydrogenophaga sp. str. NT-14 AroB, putative arsenite oxidase ULPAs1 AoxA (acc. no. AAN05580.1), A. faecalis (A.f) arsenite
oxidase h (acc. no. nrp 1G8J_B), NT-26 arsenite oxidase AroB (acc. no. AY345225). (c) Hydrogenophaga sp. str. NT-14 cytochrome c-551, Hydrogenophilus
thermotuteolus (H.t.) cytochrome c-552 (acc. no. JN0251), P. denitrificans (P.d) cytochrome c-551 (acc. no. P00103), H. thermophilus (H.th.) cytochrome c552 (acc. no. CAA40902.1), ULPAs1 putative AoxD (acc. no. AAN05583.1).
activity staining for peroxidase activity (characteristic of
cytochrome heme c) (Fig. 3b).
The cytochrome appears to be the physiological electron
acceptor to the Hydrogenophaga sp. str. NT-14 Aro. The
cytochrome wavelength spectrum showed the following
oxidized states (diffused a and h regions and Soret-409
nm) and reduced states in the presence of the purified Aro
and arsenite (a-551 nm, h-521 nm and Soret-416 nm) (Fig.
4). The Aro alone displayed no absorbance spectrum as the
concentration used was too low to detect the Fe-S clusters
and the cytochrome alone showed only an oxidized spectrum; neither of these spectra could be altered upon addition
of arsenite (data not shown). The NT-14 cytochrome could
Fig. 3. SDS-polyacrylamide gel and heme-activity stain of the cytochrome
c. (a) Coomassie blue R350 stain. Lane 1, molecular weight standards (see
Fig. 2); lane 2: purified cytochrome c-551. (b) Ingel activity stain of the
purified cytochrome c-551.
also act as an electron acceptor to the purified NT-26 Aro,
resulting in the same oxidized and reduced spectra seen with
the Hydrogenophaga sp. str. NT-14 Aro (Fig. 5). The
wavelength spectra and activity stain suggest that the
cytochrome is a small c-type cytochrome belonging to the
c-551 family of cytochromes.
The N-terminal sequence of the cytochrome (20 amino
acids) was determined (Fig. 2c) and found to contain a
conserved cysteine motif, Cys-Xaa2-Cys-His, which is typical of heme binding sites [23]. As determined by a
BLASTP search, the cytochrome displayed sequence iden-
Fig. 4. Oxidized and reduced spectra of the purified cytochrome in the
presence of the Hydrogenophaga sp. str. NT-14 Aro. The cytochrome
wavelength spectra was recorded using 20-Ag cytochrome and 6.2-Ag Aro.
The oxidized spectrum displayed diffused a- and h-regions and a Soret
peak of 409 nm. The reaction was initiated with the addition of 2.5 mM
arsenite, which resulted in the reduction of the cytochrome with an a-peak
of 551 nm, a h-peak of 521 nm and a Soret peak of 416 nm.
152
R.N. vanden Hoven, J.M. Santini / Biochimica et Biophysica Acta 1656 (2004) 148–155
Fig. 5. Oxidized and reduced spectra of the purified Hydrogenophaga sp.
str. NT-14 c-551 cytochrome in the presence of the purified NT-26 Aro. The
cytochrome wavelength spectra were recorded using 10-Ag cytochrome and
10.2-Ag of NT-26 Aro. The oxidized and reduced spectra are identical to
those in Fig. 4.
tity to the, Hydrogenophilus thermoluteolus (previously
named Pseudomonas hydrogenothermophila) c-552 cytochrome (80%) [24] (Fig. 3c), Pseudomonas denitrificans c551 cytochrome (60%) [25] and Hydrogenobacter thermophilus c-552 cytochrome (45%) [26]. These cytochromes
appear to be slightly larger than that of Hydrogenophaga sp.
str. NT-14, ranging from approximately 7.6 to 9 kDa. A
fourth open reading frame in the ULPAs1 aox gene sequence was identified and designated aoxD. The putative
AoxD appears to be a c-type cytochrome sharing 35%
sequence identity with that of NT-14. A BLASTP search
of the entire AoxD sequence also revealed identity to the H.
thermophilus cytochrome c-552 (52%) [26].
4. Discussion
The Hydrogenophaga sp. str. NT-14 Aro is indirectly
associated with the cytoplasmic membrane, as 100% of Aro
activity was associated with the membranes isolated from
total cell extracts [R.N. vanden Hoven and J.M. Santini,
unpublished data]. The Aro was released into the periplasm
by the preparation of spheroplasts, suggesting that the Aro
was indirectly associated with the membranes via a membrane-bound protein. The properties of the purified Hydrogenophaga sp. str. NT-14, NT-26 [17] and A. faecalis
[16,18] arsenite oxidases are compared in Table 2. Both
the Hydrogenophaga sp. str. NT-14 and NT-26 Aro’s are
periplasmic while the arsenite oxidase from A. faecalis is
membrane-bound. The native molecular weights of the
enzymes vary: the Hydrogenophaga sp. str. NT-14 Aro is
309 kDa (a3h3), the NT-26 Aro is 219 kDa (a2h2) and the
A. faecalis arsenite oxidase is 100 kDa (a1h1) (Table 2). The
redox cofactors are identical, with all three enzymes con-
taining molybdenum and iron. Sulfur was detected in the A.
faecalis arsenite oxidase by EPR [16]. The presence of
sulfur in the Hydrogenophaga sp. str. NT-14 Aro was not
confirmed but when denatured the enzyme exhibited a
strong sulfide smell. The Vmax for the Hydrogenophaga
sp. str. NT-14 Aro was almost double that of the NT-26 and
A. faecalis arsenite oxidases. The significance of these
differences is as yet not known. Perhaps with further
investigation of these enzymes and other arsenite oxidases
the reasons for the variations may be revealed.
Cytochromes are electron carrier proteins that contain
heme prosthetic groups bound to the peptide by the following cysteine motif, Cys-Xaa2-Cys-His [23]. They exhibit
different UV – Visible spectra with three main absorbance
peaks: a, h and g (Soret) [27]. c-type cytochromes generally
have a reduced a region absorbing between 550– 557 nm
[27]. The Hydrogenophaga sp. str. NT-14 cytochrome has
an a region absorbance of 551 nm, suggesting that it
belongs to the c-551 type cytochromes [28].
A cytochrome c and azurin were detected during the
DEAE-anion exchange chromatography step in the purification of the A. faecalis arsenite oxidase [16]. The absorbance spectrum of the A. faecalis cytochrome c, which was
not purified, indicated that the cytochrome could be reduced
by arsenite in the presence of the A. faecalis arsenite oxidase
and, appeared to have a similar a-region to c-type cytochromes (550 – 557 nm) [16]. The A. faecalis arsenite
oxidase crystal structure shows a small hydrophobic cleft
in the Rieske-type h subunit, where a small globular protein
such as a c-type cytochrome or azurin could bind [18]. It is
possible that the Hydrogenophaga sp. str. NT-14 AroB
subunit may contain a similar cleft to allow binding of the
c-551 dimer, further supporting the involvement of the c551 cytochrome in the arsenite oxidation electron transport
chain. Most c-551- and c-552-type cytochromes can donate
electrons directly to (i) terminal oxidases (e.g., cytochrome c
oxidase aa3) [29] and (ii) terminal reductases (e.g., nitrate
reductases) [30]. The aa3 oxidase can act as a proton pump
creating a proton gradient across the membrane that can be
used to generate ATP [31,32], suggesting that heterotrophic
arsenite oxidation could be linked to an electron transport
chain.
Table 2
Comparisons of the arsenite oxidases from Hydrogenophaga sp. str. NT-14,
NT-26 and A. faecalis
NT-14
Location
Native molecular weight
Subunits + size
Structure
Co-factors
Km (AM)
Vmax (U/mg)
a
a
periplasm
309 kDa
a 86 kDa
h 16 kDa
a3h3
Fe, Mo
35
6.1
Based on preparation of spheroplasts.
NT-26
A. faecalis
periplasm
219 kDa
a 98 kDa
h 14 kDa
a2h2
Fe, Mo
61
2.4
membrane
100 kDa
a 85 kDa
h 15 kDa
ah
Fe, S, Mo
8
2.88
R.N. vanden Hoven, J.M. Santini / Biochimica et Biophysica Acta 1656 (2004) 148–155
Based on the crystal structure of the A. faecalis arsenite
oxidase, it is possible to map the flow of electrons through
this enzyme [18]. The A. faecalis arsenite oxidase and
presumably the other arsenite oxidases have a molybdenum atom at the enzyme active site that in A. faecalis is
coordinated by two pyranopterin MGD-type rings [18].
The arsenite oxidase lacks an amino acid ligand coordinated to the molybdenum atom at the active site, making it
the first member of a new subgroup of the DMSO
reductase family of molybdoenzymes [18]. The arsenite
oxidase lacks the typical serine, cysteine or selenocysteine
as found in other enzymes of the DMSO reductase family.
This may be due to the ability of arsenite to irreversibly
bind thiol groups of cysteines and instead the corresponding residue in the A. faecalis arsenite oxidase is an
alanine (position 199) [18]. The arsenite (substrate) binding site of the arsenite oxidase is comprised of His (195),
Glu (203), Arg (419) and His (423) residues [18]. The
amino acids involved in substrate binding along with the
potential molybdenum binding alanine are conserved in the
amino acid sequences of the large subunits of the arsenite
oxidases from NT-26, A. faecalis and ULPAs1 [17]. The
positioning of arsenite in the substrate binding site exposes
the lone electron pair to the oxidized MoMO atom [18,33].
Electrons are transferred from arsenite to the molybdenum
atom reducing it from MoVI to MoIV. This process is rapid
with the molybdenum atom becoming re-oxidized by
deprotonation of water at the active site [18,33]. The A.
faecalis arsenite oxidase contains two high potential ironsulfur clusters (> + 300 mV) [18]. The first is a [3Fe4S]
cluster present in the molybdenum-containing subunit and
153
the second is a Rieske-type [2Fe2S] cluster in the small
subunit [18].
With reference to the A. faecalis crystal structure and the
identification of the physiological electron acceptor of the
Hydrogenophaga sp. str. NT-14 Aro, an electron transport
chain for arsenite oxidation is for the first time proposed
(Fig. 6). Two electrons can be transferred from arsenite to
the molybdenum atom that are subsequently passed to the
[3Fe4S] cluster of the AroA subunit, where only one
electron at a time can be passed to the [2Fe2S] cluster of
the AroB subunit [34,35]. The [3Fe4S] cluster can potentially accept two electrons and may possibly store one of
these during electron transfer to the [2Fe2S] cluster of the
Rieske subunit. Re-oxidation of the molybdenum atom
occurs rapidly, supporting the hypothesis that the [3Fe4S]
cluster accepts both electrons simultaneously [34,35]. The
electrons are then passed to the small soluble c-551
cytochrome, which appears to be present as a dimer,
possibly accepting two electrons from the Rieske subunit.
The c-551 cytochrome can pass electrons to an aa3
cytochrome c oxidase (Fig. 6). This would be the most
likely terminal oxidase as it is common under normal
atmospheric oxygen conditions and is the most common
terminal oxidase in prokaryotes [36]. An electron transport
chain using a c-551 type cytochrome and an aa3 terminal
oxidase is feasible based on other examples of aerobic
electron transport chains (e.g., those in mitochondria and
ammonia oxidation by Nitrosomonas europaea) [37,38].
Electrons can be transferred from the c-551 cytochrome to
the aa3 at the CuA subunit [32]. The electrons are then
transferred to the cytoplasmic face of the aa3 where four
Fig. 6. Putative electron transport chain for heterotrophic arsenite oxidation. Arsenite is oxidized to arsenate at the Mo active site of the AroA subunit. The two
electrons are then transferred to the [3Fe4S] centre of AroA where they are then passed one at a time to the [2Fe2S] centre of the Rieske (AroB) subunit. The
electrons are passed to the cytochrome c-551 bound as a dimer to the AroB and then to the CuA and heme a of the cytochrome c oxidase aa3 complex. A total
of four electrons are required to reduce O2 to H2O in the cytoplasm of the cell and four protons are then pumped to the periplasm creating a proton motive
gradient across the membrane, which can subsequently be used to generate ATP.
154
R.N. vanden Hoven, J.M. Santini / Biochimica et Biophysica Acta 1656 (2004) 148–155
electrons are required to reduce O2 to H2O [32]. This
process generates a proton gradient across the membrane,
as the aa3 complex facilitates the translocation of four
protons from the cytoplasm to the periplasm [32]. The
proton gradient can be subsequently used to synthesize
ATP via an ATP synthase.
NT-14 is the first heterotrophic arsenite-oxidizing bacterium thought to gain energy from the oxidation of arsenite.
Purification and preliminary characterization of the NT-14
Aro has led to the discovery of the physiological electron
acceptor, a c-551 cytochrome. This suggests that heterotrophic arsenite oxidation may in fact lead to the generation of
energy and, as such, an electron transport chain for heterotrophic arsenite oxidation has been proposed.
[12]
[13]
[14]
[15]
[16]
Acknowledgements
This work was supported by an Australian Research
Council Grant (DP0209802) to JMS. JMS is a recipient of
an ARC Australian Postdoctoral Fellowship. RNV is a
recipient of an Australian Postgraduate Award. We would
like to thank I. Streimann for assistance with graphics and
D. Flood for technical assistance. We would also like to
extend our thanks to the reviewers of this manuscript for
their constructive comments.
[17]
[18]
[19]
[20]
[21]
References
[22]
[1] H.L. Ehrlich, Bacterial oxidation of As(III) compounds, in: W.T.
Frankenberger Jr. (Ed.), Environmental Chemistry of Arsenic, Marcel
Dekker, New York, 2002, pp. 313 – 327.
[2] R.S. Oremland, J.F. Stolz, The ecology of arsenic, Science 300 (2003)
939 – 944.
[3] B.P. Rosen, Biochemistry of arsenic detoxification, FEBS Lett. 529
(2002) 86 – 92.
[4] R. Wysocki, C.C. Chéry, D. Wawrzycka, M. Van Hulle, R. Cornelis,
J.M. Thevelein, M.J. Tamás, The glycerol channel Fps1p mediates the
uptake of arsenite and antimonite in Saccharomyces cerevisiae, Mol.
Microbiol. 40 (2001) 1391 – 1401.
[5] H.H. Green, 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 (1918) 465 – 467.
[6] J.M. Santini, L.I. Sly, R.D. Schnagl, J.M. Macy, A new chemolithoautotrophic arsenite-oxidizing bacterium isolated from a gold mine:
phylogenetic, physiological and preliminary biochemical studies,
Appl. Environ. Microbiol. 66 (2000) 92 – 97.
[7] J.M. Santini, L.I. Sly, A. Wen, D. Comrie, P. De Wulf-Durand, J.M.
Macy, New arsenite-oxidizing bacteria isolated from Australian goldmining environments—phylogenetic relationships, Geomicrobiol.
J. 19 (2002) 67 – 76.
[8] S.E. Phillips, M.L. Taylor, Oxidation of arsenite to arsenate by Alcaligenes faecalis, Appl. Environ. Microbiol. 32 (1976) 392 – 399.
[9] R.S. Oremland, S.E. Hoeft, J.M. Santini, N. Bano, R.A. Hollibaugh,
J.T. Hollibaugh, Anaerobic oxidation of arsenite in Mono lake water
and by a facultative, arsenite-oxidizing chemoautotroph, strain
MLHE-1, Appl. Environ. Microbiol. 68 (2002) 4795 – 4802.
[10] F.H. Osborne, H.L. Ehrlich, Oxidation of arsenite by a soil isolate of
Alcaligenes, J. Appl. Bacteriol. 41 (1976) 295 – 305.
[11] A.W. Turner, Bacterial oxidation of arsenite: I. Description of bacteria
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
isolated from arsenical cattle-dipping fluids, Aust. J. Biol. Sci 7
(1954) 452 – 478.
T.M. Gihring, J.F. Banfield, Arsenite oxidation and arsenate respiration by a new Thermus isolate, FEMS Microbiol. Lett. 204 (2001)
335 – 340.
T.M. Salmassi, K. Venkateswaren, M. Satomi, K. Nealson, D.K. Newman, J.G. Hering, Oxidation of arsenite by Agrobacterium albertimagni, AOL 15, sp nov., isolated from Hot Creek, California, Geomicrobiol. J. 19 (2002) 53 – 66.
W. Weeger, D. Lièvremont, M. Perret, F. Lagarde, J.-C. Hubert,
M. Leroy, M.-C. Lett, Oxidation of arsenite to arsenate by a
bacterium isolated from an aquatic environment, BioMetals 12
(1999) 141 – 149.
T.M. Gihring, G.K. Druschel, R.B. Mccleskey, R.J. Hamers, J.F. Banfield, Rapid arsenite oxidation by Thermus aquaticus and Thermus
thermophilus: field and laboratory investigations, Environ. Sci. Technol. 35 (2001) 3857 – 3862.
G.L. Anderson, J. Williams, R. Hille, The purification and characterization of arsenite oxidase from Alcaligenes faecalis, a molybdenumcontaining hydroxylase, J. Biol. Chem. 267 (1992) 23674 – 23682.
J.M. Santini, R.N. vanden Hoven, Molybdenum-containing arsenite
oxidase of the cheomolithoautotrophic arsenite oxidizer NT-26,
J. Bacteriol. 186 (2004) 1614 – 1619.
P.J. Ellis, T. Conrads, R. Hille, P. Kuhn, 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 (2001) 125 – 132.
D. Muller, D. Lièvremont, D.D. Simeonova, J.-C. Hubert, M.-C. Lett,
Arsenite oxidase aox genes from a metal-resistant h-Proteobacterium,
J. Bacteriol. 185 (2003) 135 – 141.
M.M. Bradford, A rapid and sensitive method for the quantitation of
microgram quantities of proteins utilizing the principle of protein-dye
binding, Anal. Biochem. 72 (1976) 248 – 254.
T. Krafft, J.M. Macy, Purification and characterization of the respiratory arsenate reductase of Chrysiogenes arsenatis, Eur. J. Biochem.
255 (1998) 647 – 653.
C.F. Goodhew, K.R. Brown, G.W. Pettigrew, Heme staining in gels, a
useful tool in the study of bacterial c-type cytochromes, Biochim.
Biophys. Acta 852 (1986) 288 – 294.
T. Yamanaka, Group c cytochromes, in: T. Yamanaka (Ed.), The
Biochemistry of Bacterial Cytochromes, Springer-Verlag, New York,
1992, pp. 91 – 168.
Y. Sanbongi, S.Y. Chung, K. Yokoyama, Y. Igarashi, T. Kodama,
Thermostability of Pseudomonas hydrogenothermophila cytochrome
c-552, Biosci. Biotechnol. Biochem. 56 (1992) 990 – 991.
R.P. Ambler, Bacterial cytochromes c and molecular evolution, Syst.
Zool. 22 (1973) 554 – 565.
Y. Sanbongi, J.H. Yang, Y. Igarashi, T. Kodama, Cloning, nucleotide
sequence and expression of the cytochrome c-552 gene from Hydrogenobacter thermophilus, Eur. J. Biochem. 198 (1991) 7 – 12.
L. Thöny-Meyer, Biogenesis of respiratory cytochromes in bacteria,
Microbiol. Mol. Biol. Rev. 61 (1997) 337 – 376.
G.W. Pettigrew, G.R. Moore, Resolution, characterization and classification of c-type cytochrome, in: A. Rich (Ed.), Cytochromes c:
Biological Aspects, Springer-Verlag, London, 1987, pp. 1 – 28.
M.F. Otten, J. Van Der Oost, W.N.M. Reijnders, H.V. Westerhoff, B.
Ludwig, R.J.M. Van Spanning, Cytochromes c550, c552 and c1 in the
electron transport network of Paracoccus denitrificans: redundant or
subtly different in function, J. Bacteriol. 183 (2001) 7017 – 7026.
B.C. Berks, S.J. Ferguson, J.W.B. Moir, D.J. Richardson, Enzymes
and associated electron transport systems that catalyse the respiratory
reduction of nitrogen oxides and oxyanions, Biochim. Biophys. Acta
1232 (1995) 97 – 173.
C.V. Frost, K. Schulten, Phylogenetic analysis of metabolic pathways,
J. Mol. Evol. 52 (2001) 471 – 489.
D.A. Mills, S. Ferguson-Miller, Understanding the mechanism of
proton movement linked to oxygen reduction in cytochrome c oxidase: lessons from other proteins, FEBS Lett. 545 (2003) 47 – 51.
R.N. vanden Hoven, J.M. Santini / Biochimica et Biophysica Acta 1656 (2004) 148–155
[33] T. Conrads, C. Hemann, G.M. George, I.J. Pickering, R.C. Prince, R.
Hille, The active site of the arsenite oxidase from Alcaligenes faecalis,
J. Am. Chem. Soc. 124 (2002) 11276 – 11277.
[34] D.O. Hall, M.C.W. Evans, Iron-sulphur proteins, Nature 223 (1969)
1342 – 1348.
[35] J.M. Berg, R.H. Holm, Structure and reactions of the iron-sulfur
protein clusters and their synthetic analogs, in: T.G. Spiro (Ed.),
Iron-Sulfur Proteins, Wiley, New York, 1982, pp. 1 – 66.
[36] G.W. Pettigrew, G.R. Moore, The role of mitochondrial cytochrome c
155
in electron transport, in: A. Rich (Ed.), Cytochromes c: Biological
Aspects, Springer-Verlag, London, 1987, pp. 29 – 112.
[37] G.W. Pettigrew, G.R. Moore, The function of bacterial and photosynthetic cytochrome c, in: A. Rich (Ed.), Cytochromes c: Biological
Aspects, Springer-Verlag, London, 1987, pp. 113 – 230.
[38] M. Wittaker, D. Bergmann, D. Arciero, A.B. Hooper, Electron transfer during the oxidation of ammonia by the chemolithotrophic bacterium Nitrosomonas europaea, Biochim. Biophys. Acta 1459 (2000)
346 – 355.