Microbiology (2003), 149, 1513–1522
DOI 10.1099/mic.0.26155-0
Response of a strict anaerobe to oxygen: survival
strategies in Desulfovibrio gigas
Paula Fareleira,1,2 Bruno S. Santos,1 Célia António,1
Pedro Moradas-Ferreira,3 Jean LeGall,1,4 António V. Xavier1
and Helena Santos1
1
Instituto de Tecnologia Quı́mica e Biológica, Universidade Nova de Lisboa, Rua da Quinta
Grande, 6 Apartado 127, 2780-156 Oeiras, Portugal
Correspondence
Helena Santos
2
Estação Agronómica Nacional, Instituto Nacional de Investigação Agrária, Quinta do Marquês,
2780-156 Oeiras, Portugal
[email protected]
3
Instituto de Biologia Molecular e Celular, Universidade do Porto, 4150 Porto, Portugal
4
Department of Biochemistry and Molecular Biology, University of Georgia, Athens,
GA 30602, USA
Received 27 November 2002
Revised
24 February 2003
Accepted 25 February 2003
The biochemical response to oxygen of the strictly anaerobic sulfate-reducing bacterium
Desulfovibrio gigas was studied with the goal of elucidating survival strategies in oxic environments.
Cultures of D. gigas on medium containing lactate and sulfate were exposed to oxygen
(concentration 5–120 mM). Growth was fully inhibited by oxygen, but the cultures resumed growth
as soon as they were shifted back to anoxic conditions. Following 24 h exposure to oxygen the
growth rate was as high as 70 % of the growth rates observed before oxygenation. Catalase
levels and activity were enhanced by exposure to oxygen whereas superoxide-scavenging and
glutathione reductase activities were not affected. The general pattern of cellular proteins as
analysed by two-dimensional electrophoresis was altered in the presence of oxygen, the levels of
approximately 12 % of the detected proteins being markedly increased. Among the induced
proteins, a homologue of a 60 kDa eukaryotic heat-shock protein (Hsp60) was identified by
immunoassay analysis. In the absence of external substrates, the steady-state levels of nucleoside
triphosphates detected by in vivo 31P-NMR under saturating concentrations of oxygen were
20 % higher than under anoxic conditions. The higher energy levels developed under oxygen
correlated with a lower rate of substrate (glycogen) mobilization, but no experimental evidence
for a contribution from oxidative phosphorylation was found. The hypothesis that oxygen interferes
with ATP dissipation processes is discussed.
INTRODUCTION
Sulfate-reducing bacteria have been classified as obligate
anaerobes but, in recent years, it has become apparent that
the ability to utilize oxygen is widespread among these
organisms (Dilling & Cypionka, 1990; Dannenberg et al.,
1992; Santos et al., 1993; Kuhnigk et al., 1996; Van Niel et al.,
1996). In particular, aerobic respiration has been examined
in Desulfovibrio species (reviewed by LeGall & Xavier, 1996;
Cypionka, 2000). Previous work using in vivo NMR has
shown that non-growing cells of Desulfovibrio gigas can
utilize oxygen as electron acceptor for the reducing power
generated during glycogen catabolism, while developing
high intracellular levels of ATP (Santos et al., 1993). Subsequently, a novel oxygen-utilizing pathway, linking NADH
Abbreviations: Hsp60, heat-shock protein 60; NDP, nucleoside
diphosphates; NTP, nucleoside triphosphates; TCS, 3,39,49,59tetrachlorosalicylanilide.
0002-6155 G 2003 SGM
oxidation to oxygen reduction, was elucidated in this
organism (Chen et al., 1993a, b). This electron-transfer chain
comprises two unique proteins, an NADH-rubredoxin
oxidoreductase and a new terminal oxygen reductase
(Frazão et al., 2000; Silva et al., 2001a); it also involves
rubredoxin, a well-studied protein to which no physiological function had been previously ascribed. These observations demonstrate the unexpected capacity of this ‘strict
anaerobe’ to profit from the presence of oxygen.
The ability of Desulfovibrio and other sulfate-reducing
bacteria to utilize oxygen may be essential for survival in
their natural habitats. In fact, the presence of communities
of sulfate-reducing bacteria in oxic environments has been
widely reported, especially in the oxic–anoxic interfaces
of microbial mats or sediments, as well as in periodically
oxygenated zones of aquatic environments (Battersby et al.,
1985; Fukui & Takii, 1990; Sass et al., 1997; Manz et al.,
1998; Teske et al., 1998; Minz et al., 1999).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
Printed in Great Britain
1513
P. Fareleira and others
Despite the ability of these organisms to cope with oxygen,
sustainable growth of pure cultures in the presence of
oxygen has never been unequivocally demonstrated
(Cypionka, 2000). Several key enzymes in the pathways
for substrate oxidation are oxygen sensitive (Stams &
Hansen, 1982; Kremer et al., 1989; Hensgens et al., 1993).
Moreover, some Desulfovibrio strains undergo morphological alterations in the presence of oxygen (Sass et al., 1998),
which could result from inactivation of enzymes involved
in cell division (Cypionka, 2000). Several highly reactive
derivatives of oxygen are toxic and able to damage essential
cell components if not scavenged (Fridovich, 1983; Imlay &
Linn, 1988; Farr & Kogoma, 1991). To counteract these
deleterious effects, many anaerobic organisms have developed defence systems similar to those found in aerobes
(Storz et al., 1990). One well-known type of protection
system comprises enzyme activities that diminish or
eliminate oxygen derivatives and radicals. Such enzymes
have been found in many strictly anaerobic microorganisms, including sulfate-reducing bacteria (Hewitt &
Morris, 1975; Hatchikian et al., 1977; Rocha et al., 1996; Dos
Santos et al., 2000; Pan & Imlay, 2001), in which they can
play a major role in oxygen detoxification.
Following our early results on the aerobic metabolism
of D. gigas, it was deemed important to assess oxygen
toxicity and cellular responses to oxygen in this organism.
Cell survival and growth after exposure to oxygen were
examined, as well as the effect of oxygen on substrate
utilization. Energy levels were monitored in real time by
in vivo 31P-NMR in cells exposed to oxygen. Furthermore,
enzyme activities related to the oxidative stress response,
and total protein profiles, were examined to evaluate
alterations in gene expression induced by oxygen.
METHODS
Organism, growth conditions and preparation of resting cell
suspensions. D. gigas (NCIB 9332) was grown on a culture
medium containing lactate and sulfate as carbon and energy sources
in rubber-stoppered glass bottles as previously described (Santos
et al., 1993). Cells were harvested from late-exponential-phase cultures by centrifugation (2000 g, 7 min) under anoxic conditions,
washed with an oxygen-free buffer solution (30 mM MOPS, sodium
salt, 1 mM MgCl2, pH 7?0), and suspended in the same buffer. The
cell density of the resulting suspensions was approximately 30 mg
protein ml21.
NMR spectroscopy. Freshly prepared cell suspensions were trans-
ferred to 10 mm NMR tubes under argon. Before the acquisition of
the initial spectrum, 5 ml antifoam (silicone) and 5 % (v/v) 2H2O
was added to the cell suspensions. Gases were delivered by using an
airlift device in the NMR tube (Santos & Turner, 1986). 31P-NMR
spectra were obtained with a 10 mm broadband probe head in a
Bruker DRX-500 spectrometer operating at 202?45 MHz as previously described (Santos et al., 1994). Spectra were acquired with
proton decoupling under fully relaxing conditions for NTP resonances,
using a 30u flip angle and a repetition delay of 3?1 s. Chemical shifts
were referenced with respect to external 85 % H3PO4. Calibration of
the chemical shift of intracellular Pi as a function of pH was carried
out with a cell suspension of D. gigas treated with a concentration of
1514
100 mM of the protonophore 3,39,49,59-tetrachlorosalicylanilide (TCS).
To calibrate the chemical shift of external Pi, a titration curve of
MOPS buffer containing 1 mM Pi was performed.
1
H-NMR spectra were obtained with presaturation of the water
signal, using a 10 mm broadband probe head in a Bruker AMX-300
spectrometer. Free induction decays were acquired with a 45u flip
angle and a repetition time of 3?3 s. Chemical shifts were referenced
with respect to sodium 3-trimethylsilyl[2,2,3,3-2H]propionate. In all
experiments, the probe head temperature was kept at 33 uC. In vivo
NMR experiments were performed at least twice.
Growth experiments. Cultures were prepared in a medium con-
taining lactate and sulfate and buffered with 30 mM MOPS (sodium
salt), using a 2 litre fermentation vessel at 35 uC with continuous
gassing with argon. Cell growth was monitored by measuring either
the OD450 or the protein concentration. At the middle of the exponential growth phase, the culture was split in two, and fresh
medium was added to each fermentation vessel. Both cultures were
maintained under argon for approximately 5–6 h. One of the cultures was then exposed to defined concentrations of oxygen, by
continuous bubbling with a suitable argon/air mixture, and the
oxygen partial pressure was controlled with an O2 electrode. The
other culture was kept anoxic (control). Each growth experiment
was performed at least twice.
Preparation of cell-free extracts and determination of
enzyme activities. Cell suspensions were disrupted either by ultra-
sonic disintegration in an ice-cold bath or by passing twice through
a French pressure cell at 3?3 MPa; cell debris was removed by
centrifugation (15 min, 16 000 g) in an Eppendorf centrifuge. The
resulting supernatant fraction was dialysed overnight against 10 mM
Tris/HCl pH 7?0.
Superoxide-scavenging activity was assayed spectrophotometrically
by the method described by McCord & Fridovich (1969) for superoxide dismutase (EC 1.15.1.1), with 1 unit (U) of activity causing a
50 % inhibition of the rate of horse heart cytochrome c reduction by
the superoxide anion generated by the system xanthine/xanthine
oxidase. Catalase (EC 1.11.1.6) was measured according to the procedure of Beers & Sizer (1952). Glutathione reductase (EC 1.8.1.7) was
assayed as described by Goldberg & Spooner (1983). One unit of
catalase or glutathione reductase activity was the amount catalysing
the formation of 1 mmol product or the consumption of 1 mmol
substrate per minute. Two independent determinations were
performed for each enzyme activity in all conditions examined.
Protein labelling and sample preparation for electrophoretic
analysis. Anoxic cell suspensions containing 0?5–1 mg protein
ml21 were prepared in MOPS buffer (30 mM, sodium salt, pH 7?7)
from mid-exponential-phase cultures. For protein labelling in vivo,
cell suspensions were split into aliquots (2 ml each) and incubated
for 3–5 min in different conditions: under nitrogen atmosphere at
35 uC (control); under oxygen atmosphere at 35 uC (oxygen shock);
under nitrogen atmosphere at 45 uC (heat shock). [35S]Methionine
(Amersham Life Science) was added to each sample (0?376
106 Bq ml21) and cells were incubated for further 10 min. Cell
suspensions were centrifuged at 3000 g for 10 min at 4 uC, and
washed with 30 mM MOPS buffer, pH 7?7. For one-dimensional
electrophoresis, the cell sediment was immediately suspended in a
lysis solution [7?5 % (v/v) b-mercaptoethanol, 1?5 % (w/v) SDS,
0?75 mM PMSF, 2 mM DTT, 50 mM Tris/HCl pH 6?8]. To prevent
protein degradation, a cocktail of protease inhibitors was added
(Bossier et al., 1993). After vigorous vortexing, the samples were
submitted to three cycles of freeze–thawing; the resulting lysates
were diluted in sample solution [2 % (w/v) SDS, 10 % (v/v) glycerol,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
Microbiology 149
Response to oxygen in Desulfovibrio gigas
0?025 % (w/v) bromophenol blue, 1 % (v/v) b-mercaptoethanol,
16 mM Tris/HCl pH 6?8], and boiled for 5 min before loading onto
the gel.
For two-dimensional electrophoresis, the same procedure was used,
except that after washing with buffer solution, cells were suspended in a
lysis solution containing 9 M urea, 7?6 % (w/v) CHAPS, 2 % (w/v)
DTT, 2 % (v/v) Pharmalytes non-linear pH range 3–10 (PharmaciaBiotechnology), 0?14 % (w/v) PMSF; a cocktail of protease inhibitors
was also added. After the freeze–thawing cycles in liquid nitrogen, the
lysates were diluted in a sample solution [9 M urea, 1 % (w/v) DTT,
2 % (v/v) Pharmalytes non-linear pH range 3–10, 2 % (w/v) CHAPS,
and a trace of bromophenol blue] and centrifuged at 16 000 g for
10 min at 4 uC; cell debris was discarded.
One- and two-dimensional electrophoresis. One-dimensional
electrophoresis was carried out on SDS-polyacrylamide gels according to the method described by Laemmli (1970) (5 % stacking gel
and 10 % separation gel) in a Bio-Rad Mini Protean II apparatus.
For 2D electrophoresis, isoelectric focusing was run using 13 cm long
ready-made gel strips with immobilized pH gradients (pH 3–10,
non-linear) (Immobiline DryStrips, from Pharmacia-Biotechnology),
following previously described protocols (Görg et al., 1995). Isoelectric
focusing was run in a horizontal electrophoresis unit (Multiphor II,
from Pharmacia-Biotechnology) at 20 uC. For improved sample entry,
voltage was set at 300 V for the first 1 h, and at 500 V for the next
2 h. Isoelectric focusing was continued at 3500 V for 6 h. Current
and power settings were limited to 0?5 mA and 2 W, respectively, per
gel strip. After the run, gel strips were equilibrated according to Görg
et al. (1995).
The second dimension was performed as described earlier (Görg
et al., 1995). SDS-PAGE was carried out by the method of Laemmli
(1970), as modified by Studier (1973), using 7?5–20 % polyacrylamide
SDS slab gels in a vertical electrophoresis unit (SE 600 Series, Hoefer
Scientific Instruments). [14C]Methylated proteins (Amersham) were
used as molecular mass standards within the range 14?3–220 kDa.
Electrophoresis was performed at 20 uC for 10 h at 15 mA per gel.
After fixation in 7 % (v/v) acetic acid for 30 min, fluorography was
carried out either by the methods described by Laskey & Mills (1975)
or by incubation with the fluorographic reagent Amplify (PharmaciaBiotechnology) for 20 min. Proteins labelled with [35S]methionine
were detected by exposing the dried gels to Kodak BioMax films at
270u for 5–7 days. Experiments were repeated at least five times, since
for each of the tested conditions, the protein profiles showed some
variability among replicates.
Western immunoblot analysis. Intracellular proteins were sepa-
rated by one- or two-dimensional electrophoresis and electroblotted
from gels to nitrocellulose membranes. Transfers were done overnight at 80 mV. The membranes were treated first with antibodies
(monoclonal mouse anti-heat-shock protein 60, Hsp60, from Sigma;
or polyclonal sheep anti-human catalase, from The Binding Site),
and then with the appropriate horseradish-peroxidase-conjugated
immunoglobulins (anti-mouse IgG, from Amersham; or anti-sheep
IgG, from Sigma). Antigen–antibody associations were revealed by
enhanced chemiluminescence (ECL Detection Reagents, Amersham),
followed by exposure to Kodak BioMax films.
Other analytical methods. Total glutathione (both reduced and
oxidized forms) in cell-free extracts of D. gigas was determined
according to Tietze (1969). Protein concentration was measured by
the method of Bradford (1976) using BSA as standard. For dry cell
mass determination, cells were collected by filtration in nitrocellulose membrane filters (0?2 mm pore size) and dried at 100 uC to
constant mass.
RESULTS
Effect of oxygen on the utilization of carbon
substrates
The consumption of external substrates (lactate or pyruvate) by non-growing cells of D. gigas was monitored
in vivo by 1H-NMR (Table 1). Under argon atmosphere,
the consumption rate of L-lactate (initial concentration
5 mM) was 7?4 nmol min21 (mg dry mass)21 in the presence of thiosulfate, and 8?8 nmol min21 (mg dry mass)21
when nitrite was used as electron acceptor. In both cases, the
utilization of L-lactate was completely inhibited when
oxygen was bubbled through the cell suspensions.
However, the consumption of lactate resumed upon
switching to an anoxic atmosphere. Pyruvate was utilized
at a rate of 7?0 nmol min21 (mg dry mass)21. Under
saturating oxygen concentrations, pyruvate utilization was
slowed to 40 % of the rate measured in anoxic conditions.
When no external substrate was added, the rate of glycogen degradation by cell suspensions supplied with either
argon or pure oxygen was assessed by the production of
acetate. Similar rates of approximately 2 nmol min21 (mg
protein)21 were determined in both cases.
Effect of oxygen on growth
Mid-exponential-phase cultures grown with lactate and
sulfate as substrates were submitted to several concentrations of oxygen (5, 10, 20, 40 or 120 mM) for 8 h, and
Table 1. Effect of oxygen on the utilization rates of external substrates by non-growing cell
suspensions of D. gigas, monitored by in vivo 1H-NMR
Substrate
Electron acceptor
Gas phase
Rate of substrate consumption
[nmol min21 (mg dry mass)21]
L-Lactate
Nitrite
Nitrite
Thiosulfate
Thiosulfate
–
–
Argon
O2
Argon
O2
Argon
O2
8?8
0?0
7?4
0?0
7?0
3?0
Pyruvate
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
1515
P. Fareleira and others
Table 2. Catalase and superoxide-scavenging specific
activities in cell-free extracts of D. gigas prepared from midexponential-phase cultures exposed to different concentrations of oxygen for 8 h
1.0
OD450
Values are means of at least two independent experiments±SD.
ND, Not determined.
O2 concn
(mM)
0.1
0
5
10
15
Catalase
(U mg protein21)
Superoxide-scavenging
(U mg protein21)
130±0?1
206±3?0
2?6±0?3
3?3±0?1
2?6±0?3
2?8±0?3
2?7±0?1
2?6±0?1
0
5
10
20
40
120
ND
248±7?0
291±22
365±0?1
Time (h)
Fig. 1. Development of the OD450 of a culture of D. gigas
subjected to changes in the gas phase. The culture was prepared in a 2 litre fermentation vessel at 35 uC in a modified
Starkey medium containing lactate and sulfate as substrates.
After inoculation (time zero) the culture was incubated under
strictly anoxic conditions until the middle of the exponential
growth phase. Anoxia was maintained by continuous gassing of
the culture with argon. The culture was then exposed for 6 h to
a mixture of argon and air containing 20 mM oxygen. After
exposure to oxygen, anoxia was replenished by flushing the culture with argon.
compared with control cultures maintained under an
argon atmosphere. Upon supply of oxygen to the cultures,
growth ceased immediately (Fig. 1); similar results were
obtained for all the oxygen partial pressures examined.
The values of OD450 and protein concentration measured
during exposure to oxygen were constant. However, upon
switching to an anoxic atmosphere, growth resumed and
the mean growth rate observed in these conditions was
0?14 h21, corresponding to 70 % of the mean value
determined for control cultures not exposed to oxygen
(0?2 h21). The same behaviour was observed when the
length of the exposure to oxygen was extended up to 24 h,
the longest duration examined.
Since saturating concentrations of oxygen did not completely inhibit the utilization of pyruvate by resting cell
suspensions of D. gigas (see above), we deemed it important to investigate whether D. gigas was able to grow in the
presence of oxygen using pyruvate as a carbon source.
Cultures were pre-grown under strictly anoxic conditions
in a MOPS-buffered medium containing pyruvate and
sulfate (growth rate 0?18 h21); however, no growth was
observed when the sulfate concentration was reduced.
Several pyruvate concentrations in the growth medium
(between 15 and 100 mM) and pH values (6?5, 7?0 and 7?5)
were examined without success.
1516
Effect of oxygen on the induction of defence
systems against oxidative stress
The effect of oxygen on enzyme activities involved in
oxidative stress defence and on glutathione levels was
assessed. Assays were carried out in cell-free extracts
prepared from mid-exponential-phase cultures submitted
to the following conditions: (i) different concentrations
of oxygen (0, 5, 10, 20, 40 and 120 mM) for 8 h; or (ii)
saturating concentrations of oxygen for different periods
of time (0, 15, 30, 60 and 120 min).
High levels of superoxide-scavenging activity were found
in all samples assayed (Tables 2 and 3), but no significant
differences were detected between cultures exposed to
different concentrations of oxygen or maintained under
argon atmosphere. Moreover, the duration of exposure to
oxygen had no significant effect on the activity level.
Catalase activity was detected in all samples (Tables 2 and
3), and increasing oxygen concentrations led to higher
levels of activity; furthermore, the samples submitted to
longer exposure times showed higher values of catalase
activity.
Table 3. Catalase and superoxide dismutase specific activities
in cell-free extracts of D. gigas prepared from mid-exponentialphase cultures exposed to saturating concentrations of
oxygen for different periods of time
Values are means of at least two independent experiments±SD.
Time of exposure
Catalase
Superoxide scavenging
to pure O2 (min) [U (mg protein)21] [U (mg protein)21]
0
15
30
60
120
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
183±13
268±19
381±49
375±23
346±19
2?7±0?1
3?0±0?1
2?5±0?2
3?1±0?1
2?8±0?1
Microbiology 149
Response to oxygen in Desulfovibrio gigas
Low levels of glutathione reductase activity were detected
in D. gigas cell-free extracts [0?0025±0?0005 U (mg
protein)21] and no changes were observed under the wide
range of conditions examined.
A mean value of 1?8±0?6 nmol (mg protein)21 was
determined for the total glutathione content (reduced
plus oxidized forms) in D. gigas. No significant differences
were observed upon increasing the oxygen concentration
in the medium or the duration of the oxic period.
the intensity of NTP signals was observed upon switching
from anoxic to oxic conditions (approx. 20 %) (Fig. 3). A
similar enhancement of NTP, dependent on the presence
of oxygen, was observed in crude cell extracts. The level of
NTP in the presence of oxygen was approximately 8 nmol
(mg dry mass)21. The NTP level was not affected when
cells were treated with 100 mM of the protonophore TCS,
in either the presence or absence of oxygen.
Induction of protein synthesis in response
to stress
Effect of oxygen on the energy status of
non-growing cells
The energetic status of cell suspensions saturated with
oxygen (continuous bubbling) was monitored by in vivo
31
P-NMR during 12 h and compared with that of cells
kept under anoxic conditions. NTP and NDP levels were
determined from the intensity of the respective NMR
resonances. Information on the evolution of intracellular
and extracellular pH was obtained from the chemical shifts
of the signals due to intracellular Pi and external Pi,
respectively. High levels of energization were maintained in
whole cells during the first 5–6 h of the experiment (Fig. 2);
after 12 h the intensity of the NTP resonances had decreased
to approximately 40 % of the initial values regardless of
the gas atmosphere used (data not shown). Both external
and intracellular pH decreased due to glycogen catabolism,
which results in the formation of acetic acid.
To obtain further information on the effect of oxygen on
NTP levels, cell suspensions were submitted to sequential
cycles of anaerobiosis/aerobiosis. A systematic increase in
Two-dimensional electrophoresis was used to analyse the
protein profile in response to oxygen or heat. The protein
profiles of D. gigas obtained after oxidative stress induced
by oxygen and under control conditions are shown in
Fig. 4. The spots highlighted on the autoradiographs were
identified by visual inspection and represent proteins whose
levels were clearly increased by oxygen exposure relative
to the control (anoxic conditions), and which appeared
systematically enhanced in all the replicates subjected to
oxic conditions. An alphanumeric designation was arbitrarily assigned to each of these proteins. The results are
summarized in Table 4.
The major proteins synthesized de novo by D. gigas, i.e.
the most intense spots visible in the autoradiographs, were
not significantly affected by a shift from anoxic to oxic
conditions. However, the synthesis of several proteins was
increased in the samples submitted to stress conditions.
Exposure to oxic atmosphere induced an increase in 23
proteins, with molecular masses in the range 15–107 kDa,
and within a pI range of 4?7–7?7. On the other hand, the
Fig. 2. (a) Time sequence of in vivo 31P-NMR spectra of a cell suspension of D. gigas in MOPS buffer (pH 7?5) maintained
under an oxygen atmosphere for several hours. Bubbling oxygen through the cell suspension started at time zero. (b) Plots of
the intensities of the b-NTP signals (squares) and the external pH values of the cell suspension (triangles) (determined from
the chemical shift of the extracellular Pi resonance) as a function of time. NDP, nucleoside diphosphates; NTP, nucleoside
triphosphates; PME, phosphomonoesters.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
1517
P. Fareleira and others
shift from anoxic to oxic conditions caused the disappearance of several proteins.
Time
(min)
80
70
The pattern of proteins induced by heat shock was
considerably different from that obtained upon oxygen
shock. Fourteen proteins were induced (Table 4); all of
them, except two, showed molecular masses and isoelectric
points different from those of the proteins induced by
exposure to oxygen. The common proteins are designated
in Table 4 and Fig. 4 as Ox2/Hs1 and Ox10/Hs6, with
molecular masses of 90–92 kDa and 58 kDa, and pI values
of 4?8–5?1 and 5?8–6?0, respectively.
57
45
32
22
7
0
10
5
0
_20
_5
_15
_10
Chemical shift (p.p.m.)
_25
Fig. 3. Time sequence of in vivo 31P-NMR spectra of a cell
suspension of D. gigas in MOPS buffer (pH 7?5) submitted to
repeated cycles of anaerobiosis (argon atmosphere) and aerobiosis (oxygen atmosphere). NDP, nucleoside diphosphates;
NTP, nucleoside triphosphates; PME, phosphomonoesters.
pI
4.2
Antibodies against mouse Hsp60 and human catalase were
tested on Western blots of D. gigas proteins. Anti-Hsp60
cross-reacted specifically with one protein from D. gigas.
This protein was present in the control sample not submitted to stress conditions, but the intensity of the respective spot increased upon either heat shock or oxygen
shock; it is designated in Fig. 4 and Table 4 as Ox10/Hs6.
The immunoassays for catalase detection were performed
with blots from one-dimensional SDS-PAGE gels; we
were unable to identify the corresponding spots in twodimensional gels. Anti-human catalase recognized a protein
band with a molecular mass estimated at 60–70 kDa. The
cross-reacting band in the sample exposed to oxygen
showed an increase (about 30 %) in intensity compared
with the control. On the other hand, the cross-reacting band
8.0
kDa
97.4
Ox3
pI
4.2
kDa
97.4
Ox3
Ox1
Ox1
Ox2/Hs1 Ox4
Ox5
8.0
Ox6
Ox8
Ox2/Hs1
Ox7
Ox4
66.0
66.0
Ox7
Ox8
Ox5 Ox6
Ox9
Ox10/Hs 6
Ox10/Hs 6
Ox9
Ox12
Ox11
46.0
46.0
Ox12
Ox13
Ox14
Ox14
Ox13
Ox16
Ox16
30.0
30.0
Ox15
Ox15
21.5
14.3
Ox11
Ox18
Ox20
Ox17
Ox21
Ox19
Ox22
21.5
Ox23
Ox18
Ox20
Ox17
Ox19
Ox21 Ox22
Ox23
14.3
Fig. 4. Autoradiographs (sections) of two-dimensional polyacrylamide gel electrophoresis of proteins labelled with
35
L-[ S]methionine from cells of D. gigas exposed to oxygen (left) or under control conditions (right). The spots indicated
by arrows represent either newly induced proteins or clearly increased levels of existing proteins relative to the control.
Identification letters and numbers were given to each of these proteins (see Table 4). Results are based on the comparison of
five replicates.
1518
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
Microbiology 149
Response to oxygen in Desulfovibrio gigas
Table 4. Stress-response proteins detected by two-dimensional electrophoresis in D. gigas cells
exposed to oxygen or to heat shock
Proteins induced by oxygen shock
Identification no.
(arbitrary)
Ox1
Ox2/Hs1
Ox3
Ox4
Ox5
Ox6
Ox7
Ox8
Ox9
Ox10/Hs6
Ox11
Ox12
Ox13
Ox14
Ox15
Ox16
Ox17
Ox18
Ox19
Ox20
Ox21
Ox22
Ox23
Proteins induced by heat shock
Mol. mass*
(kDa)
pI
Identification no.
(arbitrary)
107
92
90
68
66
66
66
65
62
58
45
43
42
36
35
33
20
19
18
18
18
16
15
5?4
4?8
6?8
4?7
5?2
5?3
6?1
5?9
5?4
5?8
5?6
5?5
5?5
5?8
5?7
5?9
5?6
5?5
6?1
5?4
5?7
5?9
6?1
Hs1/Ox2
Hs2
Hs3
Hs4
Hs5
Hs6/Ox10
Hs7
Hs8
Hs9
Hs10
Hs11
Hs12
Hs13
Hs14
Mol. mass*
(kDa)
pI*
90
78
75
75
63
58
56
51
50
48
48
46
38
14
5?1
4?9
6?5
5?4
6?0
6?0
6?3
6?7
6?2
6?7
6?4
6?9
6?5
5?4
*Approximate values, estimated from the position of the correspondent spots on the autoradiographs of the
2D gels.
in the heat-shocked sample showed approximately the
same intensity as the control.
DISCUSSION
D. gigas was unable to grow, with lactate or pyruvate as the
carbon source, in the presence of oxygen even when its
concentration was as low as 5 mM. However, our results
clearly demonstrate the ability of D. gigas cultures to survive
24 h exposure to oxygen and resume growth when shifted
back to anoxic conditions. This observation directed us to
study the cellular responses to oxygen with the goal of
understanding the survival strategies.
The response to oxygen in D. gigas involves an increase
in the synthesis of several proteins, which may alleviate
toxicity and cell injury caused by stress. Most of the induced
proteins could also be detected in control conditions at
lower levels, suggesting that they probably have a cellular
function both in optimal anoxic conditions and in the
presence of oxygen. The major proteins whose levels were
enhanced by oxygen were not induced in heat-shocked
cells, showing that induction of protein synthesis is specific,
depending on the type of stress. Thus far, the nature of
http://mic.sgmjournals.org
most of these proteins is unknown. Among the induced
proteins detected by two-dimensional electrophoresis, only
a homologue of a 60 kDa eukaryotic heat-shock protein
(Hsp60) and catalase were identified by immunoassays. The
global characterization of the induced proteins and the
elucidation of their functional role would be essential for
an understanding of the underlying mechanism of adaptation to oxygen in this organism. However, this goal is
beyond the scope of the present study and would require
information on the amino acid sequence of the induced
proteins. In this context, the availability of the genome
sequence of D. gigas would be very helpful. The genome
sequence determinations of D. desulfuricans (GenBank
accession number NZ_AABN00000000) and D. vulgaris
(GenBank accession number NC_002937) are in progress
and 200 kb in the D. gigas genome have been sequenced
thus far (C. Rodrigues-Pousada, personal communication).
The sequence homology between D. gigas and the other
two organisms is rather low, emphasizing the need for
more specific sequence information and reliable genome
annotation.
The ability of D. gigas to survive in oxic environments
implies the existence of efficient defence systems to
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
1519
P. Fareleira and others
counterbalance the cell damage caused by oxygen-derived
toxic radicals. It has already been demonstrated that two
important enzyme activities involved in the detoxification
of reactive oxygen species, superoxide-scavenging and
catalase, are constitutive in D. gigas, with high levels of
activity (Hatchikian et al., 1977), comparable to those of
canonical aerobes (McCord et al., 1971). Furthermore, the
present study shows that glutathione and glutathione
reductase activity are also present in this organism.
Catalase activity was enhanced by exposure to oxygen,
despite the high levels found in control cells under
anaerobiosis. An iron superoxide dismutase and a catalase
from D. gigas were recently purified and characterized in
detail (Dos Santos et al., 2000). D. gigas and other strict
anaerobes are equipped with alternative protection systems:
neelaredoxin, the blue-coloured protein first isolated from
D. gigas (Chen et al., 1994), present in all known genomes
of anaerobic prokaryotes, has been identified as a scavenger
of the superoxide anion radical (Silva et al., 2001b; Abreu
et al., 2002; Adams et al., 2002). On the other hand,
oxidative stress protection systems involving desulforedoxin
(rubredoxin oxidoreductase) with superoxide reductase
activity have been found in D. vulgaris (Lumppio et al.,
2001). Superoxide scavenging could be accomplished by
one or several proteins that catalyse either the dismutation
or the rubredoxin-coupled reduction of the superoxide
radical, catalase playing a crucial role in the removal of
hydrogen peroxide, the resulting product of superoxide
removal.
D. gigas seems to contain all the necessary enzymic machinery for the efficient detoxification of reactive oxygen
species, yet it is unable to grow in the presence of oxygen.
Studies with Clostridium acetobutylicum showed that exposure of cultures to aerobiosis prevented the net synthesis
of DNA, RNA and protein, suggesting that oxygen could
inhibit critical enzymes involved in those processes and
arrest growth (O’Brien & Morris, 1971). It has been reported
that sulfate-reducing bacteria, including some Desulfovibrio
strains, are able to oxidize organic substrates, such as
pyruvate, formate, ethanol, and even lactate, under microaerophilic concentrations of oxygen (Dannenberg et al.,
1992). Moreover, endogenous reserves can be metabolized
in oxic environments (Santos et al., 1993; Van Niel et al.,
1996). The present study showed that pyruvate could be
oxidized to acetate by non-growing D. gigas cell suspensions
under saturating concentrations of oxygen, but was unable
to sustain growth when provided to either oxic or anoxic
cultures as the sole carbon and energy source. The growth
rates of anoxic cultures in a medium containing pyruvate
and sulfate were not significantly different from those
observed in the presence of lactate and sulfate; therefore,
failure to grow on pyruvate was probably due to the lower
energetic yield associated with the fermentation of pyruvate,
the availability of oxygen as an electron acceptor being
unable to reverse this behaviour.
When cells of D. gigas are deprived of exogenous substrates,
1520
the internal reserve of glycogen that accumulates in high
amounts during growth is mobilized, acting as carbon and
energy source for cell maintenance (Fareleira et al., 1997).
Glycogen plays an essential role in cell survival under oxic
conditions. This is supported by the observation that nongrowing cell suspensions maintain a high cellular energy
charge during exposure to saturating concentrations of
oxygen for long time periods in the absence of external
substrates. Most interestingly, an increase in the intracellular
levels of NTP could be systematically observed after a shift
from anoxic to oxic atmosphere, but similar rates of acetate
production derived from glycogen reserves were found in
both anoxic and oxic conditions (this work). Thus, the
higher steady-state levels of NTP established under oxygen
atmosphere do not correlate with a higher rate of glycogen
mobilization. These observations could be reconciled if a
mechanism other than substrate-level phosphorylation for
ATP formation were operating during oxygen consumption by D. gigas. However, this hypothesis is apparently
not corroborated by our experimental evidence: (i) proton
uncouplers (TCS) did not affect the steady-state levels of
NTP in whole cells metabolizing glycogen in the presence
of oxygen; (ii) the increase in the NTP levels was also
observed upon exposure to oxygen of cell-free extracts.
Recently, a membrane-bound oxygen respiratory chain was
reported in D. gigas cells grown on a culture medium
containing fumarate and sulfate, but this respiratory activity was very poor in membrane fractions derived from
cells grown on lactate as carbon source (Lemos et al., 2001).
The amount of data available strongly indicates that lactategrown D. gigas cells utilize oxygen and synthesize NTP via
cytoplasmic systems (Chen et al., 1993a, b). On the other
hand, the observed enhancement of steady-state NTP levels
under aerobiosis is most likely due to oxygen inhibition of
processes leading to ATP dissipation.
In conclusion, D. gigas appears to be suitably equipped to
cope with oxygen stress: it is endowed with well-known antioxidative stress systems, it can perform very efficient NTP
synthesis from the aerobic metabolism of internal reserves,
and it is able to resume growth even after long periods of
exposure to oxygen. It is still intriguing that this organism
is unable to take advantage of these remarkable metabolic
features to achieve growth in the presence of oxygen. Thus,
it is conceivable that oxygen could have an irreparable effect
on vital cell functions.
ACKNOWLEDGEMENTS
This work was supported by Ministério da Ciência e da Tecnologia
(Portugal) grant PRAXIS/PCNA/P/BIA/130/96. We would like to
thank Professor C. Rodrigues-Pousada for valuable help with the
proteomic analysis and for making available the equipment needed
to perform two-dimensional electrophoresis. Dr H. Soares and Dr
C. Casalou are gratefully acknowledged for advice on two-dimensional
electrophoresis experiments and immunoassays.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
Microbiology 149
Response to oxygen in Desulfovibrio gigas
REFERENCES
Görg, A., Boguth, G., Obermaier, C., Posch, A. & Weiss, W. (1995).
Abreu, I. A., Xavier, A. V., LeGall, J., Cabelli, D. E. & Teixeira, M.
(2002). Superoxide scavenging by neelaredoxin: dismutation and
reduction activities in anaerobes. J Biol Inorg Chem 7, 668–674.
Adams, M. W. W., Jenney, F. E. Jr, Clay M. D. & Johnson, M. K.
(2002). Superoxide reductase: fact or fiction? J Biol Inorg Chem 7,
647–652.
Battersby, N. S., Malcom, S. J., Brown, C. M. & Stanley, S. O. (1985).
Sulphate reduction in oxic and sub-oxic North-East Atlantic
sediments. FEMS Microbiol Ecol 31, 225–228.
Beers, R. F. Jr & Sizer, I. W. (1952). Spectrophotometric method for
Two-dimensional polyacrylamide gel electrophoresis with immobilized pH gradients in the first dimension (IPG-Dalt): the state of
the art and the controversy of vertical versus horizontal systems.
Electrophoresis 16, 1079–1086.
Hatchikian, E. C., LeGall, J. & Bell, G. R. (1977). Significance of
superoxide dismutase and catalase activities in the strict anaerobes,
sulphate-reducing bacteria. In Superoxide and Superoxide Dismutase,
pp. 159–172. Edited by A. M. Michelson, J. M. McCord &
I. Fridovich. London: Academic Press.
Hensgens, C. M. H., Vonk, J., Vanbeeumen, J., Vanbruggen, E. F. J.
& Hansen, T. A. (1993). Purification and characterization of an
measuring the breakdown of hydrogen peroxide by catalase. J Biol
Chem 195, 133–140.
oxygen-labile, NAD-dependent alcohol dehydrogenase from Desulfovibrio gigas. J Bacteriol 175, 2859–2863.
Bossier, P., Fernandes, L., Rocha, D. & Rodrigues-Pousada, C.
(1993). Overexpression of YAP2, coding for a new yAP protein, and
Hewitt, J. & Morris, J. C. (1975). Superoxide dismutase in some
YAP1 in Saccharomyces cerevisiae alleviates growth inhibition caused
by 1,10-phenanthroline. J Biol Chem 31, 23640–23645.
Imlay, J. A. & Linn, S. (1988). DNA damage and oxygen radical
Bradford, M. M. (1976). A rapid and sensitive method for the
Kremer, D. R., Nienhuis-Kuiper, H. E. Timmer C. J. & Hansen, T. A.
(1989). Catabolism of malate and related dicarboxylic acids in
quantitation of microgram quantities of protein utilising the
principle of protein-dye binding. Anal Biochem 72, 248–254.
Chen, M., Liu, M.-Y., LeGall, J., Fareleira, P., Santos, H. & Xavier,
A. V. (1993a). Purification and characterization of an NADH-
rubredoxin oxidoreductase involved in the utilization of oxygen by
Desulfovibrio gigas. Eur J Biochem 216, 443–448.
obligatory anaerobic bacteria. FEBS Lett 50, 315–318.
toxicity. Science 240, 1302–1309.
various Desulfovibrio strains and the involvement of an oxygen-labile
NADPH dehydrogenase. Arch Microbiol 151, 34–39.
Kuhnigk, T., Branke, J., Krekeler, D., Cypionka, H. & König, H.
(1996). A feasible role of sulfate-reducing bacteria in the termite gut.
Syst Appl Microbiol 19, 139–149.
Chen, M., Liu, M.-Y., LeGall, J., Fareleira, P., Santos, H. & Xavier,
A. V. (1993b). Rubredoxin oxidase, a new flavo-hemo protein, is
Laemmli, U. K. (1970). Cleavage of structural proteins during the
the site of oxygen reduction to water by the ‘‘strict anaerobe’’
Desulfovibrio gigas. Biochem Biophys Res Commun 193, 100–105.
Laskey, R. A. & Mills, A. D. (1975). Quantitative film detection of 3H
Chen, L., Sharma, P., LeGall, J., Mariano, A. M., Teixeira, M. &
Xavier, A. V. (1994). A blue non-heme iron protein from
Desulfovibrio gigas. Eur J Biochem 226, 613–618.
Cypionka, H. (2000). Oxygen respiration by Desulfovibrio species.
Annu Rev Microbiol 54, 827–848.
Dannenberg, S., Kroder, M., Dilling, W. & Cypionka, H. (1992).
Oxidation of H2, organic compounds and inorganic sulphur
compounds coupled to reduction of O2 or nitrate by sulphatereducing bacteria. Arch Microbiol 158, 93–99.
Dilling, W. & Cypionka, H. (1990). Aerobic respiration in sulphate-
assembly of the head of bacteriophage T7. Nature 227, 680–685.
and 14C in polyacrylamide gels by fluorography. Eur J Biochem 56,
335–341.
LeGall, J. & Xavier, A. V. (1996). Anaerobes’ response to oxygen: the
sulphate-reducing bacteria. Anaerobe 2, 1–9.
Lemos, R. S., Gomes, C. M., Santana, M., LeGall, J., Xavier, A. V.
& Teixeira, M. (2001). The ‘‘strict’’ anaerobe Desulfovibrio gigas
contains a membrane-bound oxygen-reducing respiratory chain.
FEBS Lett 496, 40–43.
Lumppio, H. L., Shenvi, N. V., Summers, A. O., Voordouw, G., &
Kurtz, D. M., Jr (2001). Rubrerythrin and rubredoxin oxidoreductase
reducing bacteria. FEMS Microbiol Lett 71, 123–128.
in Desulfovibrio vulgaris: a novel oxidative stress protection system.
J Bacteriol 183, 101–108.
Dos Santos, W. G., Pacheco, I., Liu, M.-Y., Teixeira, M., Xavier, A. V.
& LeGall, J. (2000). Purification and characterization of an iron
Manz, W., Eisenbrecher, M., Neu, T. R. & Szewzyk, U. (1998).
Fareleira, P., LeGall, J., Xavier, A. V. & Santos, H. (1997). Pathways
Abundance and spatial organization of Gram-negative sulfatereducing bacteria in activated sludge investigated by in situ probing
with specific 16S rRNA targeted oligonucleotides. FEMS Microbiol
Ecol 25, 43–61.
for utilization of carbon reserves in Desulfovibrio gigas under
fermentative and respiratory conditions. J Bacteriol 179, 3972–3980.
McCord, J. M. & Fridovich, I. (1969). Superoxide dismutase: an
enzymic function for erythrocuprein. J Biol Chem 244, 6049–6055.
Farr, S. B. & Kogoma, T. (1991). Oxidative stress responses
McCord, J. M., Keele, B. B. & Fridovich, I. (1971). An enzyme-based
superoxide dismutase and a catalase from the sulfate-reducing
bacterium Desulfovibrio gigas. J Bacteriol 182, 769–804.
in Escherichia coli and Salmonella typhimurium. Microbiol Rev 55,
561–585.
Frazão, C., Silva, G., Gomes, C. M. & 11 other authors (2000).
Structure of a dioxygen reduction enzyme from Desulfovibrio gigas.
Nat Struct Biol 7, 1041–1045.
theory of obligate anaerobiosis: the physiological function of superoxide dismutase. Proc Natl Acad Sci U S A 68, 1024–1027.
Minz, D., Fishbain, S., Green, S. J., Muyzer, G., Cohen, Y., Rittmann,
B. E. & Stahl, D. A. (1999). Unexpected population distribution in a
Annu Rev Pharmacol Toxicol 23, 239–257.
microbial mat community: sulfate-reducing bacteria localized to the
highly oxic chemocline in contrast to a eukaryotic preference for
anoxia. Appl Environ Microbiol 65, 4659–4665.
Fukui, M. & Takii, S. (1990). Survival of sulphate-reducing bacteria in
O’Brien, R. W. & Morris, J. G. (1971). Oxygen and the growth and
oxic surface sediment of a seawater lake. FEMS Microbiol Ecol 73,
317–322.
metabolism of Clostridium acetobutylicum. J Gen Microbiol 68,
307–318.
Goldberg, D. M. & Spooner, R. (1983). Glutathione reductase. In
Pan, N. & Imlay, J. (2001). How does oxygen inhibit central
metabolism in the obligate anaerobe Bacteroides thetaiotaomicron?
Mol Microbiol 39, 1562–1571.
Fridovich, I. (1983). Superoxide radical: an endogenous toxicant.
Methods of Enzymatic Analysis, 3rd edn, vol. III, pp. 258–265. Edited
by H. U. Bergmeyer. Weinheim: Verlag Chemie.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
1521
P. Fareleira and others
Rocha, E. R., Selby, T., Coleman, J. P. & Smith, C. J. (1996).
Oxidative stress response in an anaerobe, Bacteroides fragilis: a role
for catalase in protection against hydrogen peroxide. J Bacteriol 178,
6895–6903.
Santos, H. & Turner, D. L. (1986). Characterization of the improved
sensitivity obtained using a flow method for oxygenating and mixing
cell suspensions in NMR. J Magn Reson 68, 345–349.
(roo) genes and upstream ORFs. Biochem Biophys Res Commun 280,
491–502.
Silva, G., LeGall, J., Xavier, A. V., Teixeira, M. & Rodrigues-Pousada,
C. (2001b). Molecular characterization of Desulfovibrio gigas
neelaredoxin, a protein involved in oxygen detoxification in
anaerobes. J Bacteriol 183, 4413–4420.
Stams, A. J. M. & Hansen, T. A. (1982). Oxygen-labile L(+)lactate
Santos, H., Fareleira, P., Xavier, A. V., Chen, L., Liu, M.-Y. & LeGall, J.
(1993). Aerobic metabolism of carbon reserves by the ‘‘obligate
dehydrogenase activity in Desulfovibrio desulfuricans. FEMS Microbiol
Lett 13, 389–394.
anaerobe’’ Desulfovibrio gigas. Biochem Biophys Res Commun 195,
551–557.
Storz, G., Tartaglia, L. A., Farr, S. B. & Ames, B. N. (1990). Bacterial
Santos, H., Fareleira, P., LeGall, J. & Xavier, A. V. (1994). In vivo
Studier, F. W. (1973). Analysis of bacteriphage T7 early RNAs and
nuclear magnetic resonance in study of physiology of sulfatereducing bacteria. Methods Enzymol 243, 543–558.
defense against oxidative stress. Trends Genet 6, 363–368.
proteins on slab gels. J Mol Biol 79, 237–248.
Sass, H., Cypionka, H. & Babenzien, H.-D. (1997). Vertical
Teske, A., Ramsing, N. B., Habicht, K., Fukui, M., Küver, J.,
Jorgensen, B. B. & Cohen, Y. (1998). Sulfate-reducing bacteria and
distribution of sulfate reducing bacteria at the oxic–anoxic interface
of the oligotrophic Lake Stechlin. FEMS Microbiol Ecol 22, 245–255.
their activities in cyanobacterial mats of Solar Lake (Sinai, Egypt).
Appl Environ Microbiol 64, 2943–2951.
Sass, H., Bertchold, M., Branke, J., Köning, H., Cypionka, H. &
Babenzien, H.-D. (1998). Psychrotolerant sulfate-reducing bacteria
Tietze, F. (1969). Enzymic method for quantitative determination
from an oxic freshwater sediment, description of Desulfovibrio
cuneatus sp. nov. and Desulfovibrio litoralis sp. nov. Syst Appl
Microbiol 21, 212–219.
Silva, G., Oliveira, S., LeGall, J., Xavier, A. V. & Rodrigues-Pousada,
C. (2001a). Analysis of the Desulfovibrio gigas transcriptional unit
containing rubredoxin (rd) and rubredoxin-oxygen oxidoreductase
1522
of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues. Anal Biochem 27,
502–522.
Van Niel, E. W. J., Pedro Gomes, T. M., Willems, A., Collins, M.,
Prins, R. A. & Gottschal, J. C. (1996). The role of polyglucose in
oxygen-dependent respiration by a new strain of Desulfovibrio
salexigens. FEMS Microbiol Ecol 21, 243–253.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:46:46
Microbiology 149