FEMS MicrobiologyLetters 93 (I 992) 87-92
© 1992 Federation of European MicrobiologicalSocieties 0378-10~7/q2/$05.(1~)
Published by Elsevier
FEMSLE 04880
Oxidation and reduction of arsenic
by Sulfolobus acidocaldarius strain BC
H. M i k a e l S e h l i n a n d E. B 6 r j e L i n d s t r 6 m
Unitfor Applied Celland MolecularBiolo~%Uni~ersityof UmeJ, Umet~.Sweden
Received 29 January ItS2
Revision received 21 February 1992
~,,.cepted 28 February 1992
Key words: Sulfolobus acidocaldarius; Enzymatic arsenite oxidation; Arsenate reduction; Bioleaching:
Arsenic toxicity; Arsenic resistance
1. S U M M A R Y
The mineral leaching archaebacterium Sulfolobus acidocaldarius strain BC is shown to have
the ability to oxidise arsenite to arsenate. Arsenite oxidation activity was 8-fold higher for cells
grown in the presence of arsenite when compared
with cells grown without arsenite. In cell-free
extracts, the arsenite oxidation activity was found
in the m e m b r a n e fraction. The arsenite oxidation
activity was sensitive to proteinase K and showed
the highest activity at acidic pH. A tetrathionated e p e n d e n t arsenate reduction activity was also
observed,
2. I N T R O D U C T I O N
During bioleaching of arsenopyrite concentrates, high amounts of arsenic are solubilized
Correspondence to: E.B. Lindstr6m, Unit for Applied Cell and
Molecular Biology, University of UmeL S-901 87 Ume'~. Sweden.
into the culture medium. These high levels of
arsenic are thought to inhibit the bioleaching
process. In solution, arsenic is found either as
arsenite, As(Ill). or as arsenate, As(V). Arsenite
has been shown to be more toxic than arsenate to
Thiobacillus ferrooxidans and Thiobacillus thiooxidarts [1] and to Rhodococcus erythropolis [2]. Oxidation of arsenite is considered to be a resistance
mechanism towards arsenite [3] and such oxidation has been reported to occur in soil bacteria
[4-6]. An enzyme that oxidises arsenite to arsenate has been isolated and purified from Alcaligenes faecalis [7]. Oxidation of arsenite has not,
however, been reported for mineral leaching bacteria.
In this work, we have studied the growth of the
mineral leaching archaebacterium Sulfolobus acidocaldarius strain BC in defined medium in the
presence of iltorganic arsenic compounds. We
present evidence for oxidation of arsenite and
reduction of arsenate u n d e r these growth conditions as well as evidence for the er.zymatic nature
of arsenite oxidation by S. acidocaldarius strain
BC.
88
3. M A T E R I A L S A N D M E T H O D S
3.1. Organism and growth conditions
Sulfolobus acidocaldarius strain BC [8] was
grown at 65°C in the defined medium 9K, without
ferrous iron, at pH 2.0 [9]. Potassium tetrathionate was used as energy source. Growth was measured by recording the optical density (O.D.) at
440 nm using a Beckman spectrophotometer
model 34.
3.2. Preparation of cell-free extracts
5 × 10 t° ceils were resuspended in 10 ml 50
mM Tris-HCI, pH 7.5 and broken in a French
press 3 times. This starting material was centrifuged at 20000 × g for 15 min. The pellet, Pz0,
was resuspended in 50 mM acetate buffer, pH
2.0, and the supernatant, $20, was dialysed
overnight against the same buffer and then centrifuged at 5 0 0 0 0 × g for 15 rain. The supernatant, Ss0, was collected and the pellet, Ps0, was
resuspended in 50 mM acetate buffer, pH 2.0.
3.3. Analytical techniques
The concentration of tetrathionate was determined by the method of Kelly et al. [10] after
adaptation to a Tcchnicon Autoanalyzer.
Arsenate was determined by ion chromatography [11.12] using a L D C / M i i t o n Roy chromatograph equipped with a conductivity detector and
Dionex H P I C A G 4 A as pre-column and
D I O N E X HPIC AS4A as separation column. The
elution buffer was 5.0 mM NaHCO3-0.5 mM
Na2CO 3 and the flow rate was 2.0 m l / m i n . The
samples were filtered through a 0.2-p.m sterile
filter (Sartorius Minisart N) before analysis.
The total arsenic concentration was determined after chemical oxidation to arsenate in the
following manner: 2 mi of sample (filtered through
a 0.2-p.m sterile filter) was incubated with 200 p.i
of 3% H 2 0 2 in screw-capped tubes at 70°C for 30
min. Excess H z O 2 was then evaporated at 95°C
until dryness and 2.0 ml of distilled water was
added and evaporated again at 95°C. Finally, 2.0
ml of distilled water was added to dissolve the
remaining salts. The arsenate concentration was
then determined as above.
Protein was determined using the BCA Protein Assay Reagent Kit from Pierce (USA). BSA
was used as standard.
3.4. Determination of arsenite oxidation activity
Arsenite oxidation activity was determined by
incubating whole cells or extracts for 1 or 6 h
together with arsenite in 50 mM acetate buffer at
65°C at pH 2.0. unless stated otherwise. After
incubation, the concentration of arsenate formed
was determined. One unit is defined as the
amount of enzyme that oxidizes 1 nmol As(liD to
A s ( V ) / h at pH 2.0 and 65°C. Activity is expressed as units/mi extract. Specific activity is
expressed as units//.tg protein.
3.5. Determination o f arsenate reduction
Arsenate reduction was determined in 3 ways:
(1) samples withdrawn from the growing culture
containing arsenate were filtered through a 0.2p.m filter. After filtration, each sample was split
into two halves of which one was analysed for
arsenate content and the other was analysed for
total arsenic content; (2) the samples indicated in
Table 2 were incubated together with 1 mM
arsenate for 24 h at 65°C. After incubation, the
total arsenic and the remaining arsenate was determined; (3) the pooled samples noted in Fig. 1
were divided into two aliquots and mixed with an
equal volume of 2 mM arsenate, pH 2.0. One
aliquot was frozen until analysis for arsenate and
the other aliquot was incubated for 6 h at 65°C
before the arsenate was measured.
3.6. Treatment with proteinase K
The Ps0 extract was incubated with proteinase
K (50/~g/ml) in 50 mM Tris. HCI, pH 7.5 for 1 h
at 37°C and then dialysed to pH 2.0 before determining the arsenite oxidation activity. Pso extract,
incubated under the same conditions but without
protcinase K, was used as control.
3.7 The effect o f pH on arsenite oxidation actit'ity
The Pso extract was divided into 3 aliquots:
one was kept at pH 2.0; one was dialysed
overnight against acetate buffer, pH 4.0; and one
was dialysed overnight against acetate buffer, pH
7.0. The sample dialysed to pH 7.0 was again split
into two aliquots, of which one was kept at pH
7.0 and the other part was dialysed overnight
against acetate buffer, pH 2.0. These 4 samples
were then assayed for arsenite oxidation activity.
1.00~
~
.
"~"0.10
4. RESULTS
4.1. Arsenite oxidation acticity b~ growing cells
We have shown earlier [13] that if arsenite is
added to a culture of Sulfolobus acidocaldarius
strain BC growing in mid-exponential phase, a
linear increase of arsenate is obtained in late
exponential or early-stationary phase. If the arsenite is added in stationary phase, no oxidation
of arsenite to arsenate is seen.
In order to investigate whether the time period
during which the oxidation of arsenite takes place
could be prolonged, S. acidocaldarius strain BC
was grown in the presence of 10 mM tetrathionate until mid-exponential phase at which time I
mM arsenite was added. Then, approximately 24
h after the depletion of the energy source, ano t h e r 5 m M tetrathionate was added. As seen
previously, arsenate was not detected until lateexponential growth (Fig. 1). T h e concentration of
arsenate increased linearly until fresh tetrathionate was added. At that time, the optical density
increased again and the arsenate concentration
decreased to background levels within 2 - 3 h (Fig.
1). W h e n this added tetrathionate was depleted,
arsenate appeared again. Almost all added arsenic was recovered from the growth medium
after filtration and oxidation with H_,O.~ which
indicates that the arsenate was reduced and not
taken up by or bound to the cells.
In order to evaluate if growth of S. acidocaldarius strain BC in the presence of arsenite could
enhance the arsenite oxidation activity, 0.5 mM
arsenite was added to a culture in exponential
growth phase. The specific arsenite oxidation activity of whole cells was found to be 6.9 units/~tg
protein for the culture grown in the presence of
arsenite for 2.2 generations compared to 0.85
units//.tg protein for the culture grown without
arsenite. This 8-fold difference in specific arsenite oxidation activity indicates that growth in the
o
0.0
1.0
B
0.~
0.61
~-4
0.40.2'
40
80
120
160
200
240
lime (hours)
Fig. I. Oxidation of arsenite and reduction of arsenate during
growth of Sulfolobus achlocaldarius strain BC in defined
medium. The filled arrow indicates the addition of I mM
arsenite and the unfilled arrow shows the addition of 5 mM
tetrathionate to the culture medium. A: t,, optical density: [2,
tetrathionate concentration. B: ©, arsenate concentration: e.
total arsenic concentration.
presence of arseo.!te induced the arsenite oxidation activity.
4.2. Ar:enite oxidation acticity in cell-free extract
For further characterisation of the arsenite
oxidation activity and to purify the enzyme(s), a
cell-free extract was made from uninduced cells.
After French press treatment, 50% of the total
arsenite oxidation activity of the starting material
was found in the S.~n-fraction (Table 1). U p o n
dialysis of the Sz0-fraction and centrifugation at
50000 x g, all the arsenite oxidation activity was
recovered in the pellet PsoTo evaluate whether the oxidation of arsenite
is enzymatic or chemical, the Pso fraction was
treated with proteinase K. After this treatment,
the protein concentration was 2-fold less and the
arsenite oxidation activity was lost completely
(data not shown). Therefore, we conclude that
the oxidation is enzymatic.
The effect of pH on the arsenite oxidation
activity in cell-free extract was studied next by
measuring the activity of the Ps0 fraction at three
different pH-values. The highest activities were
found at pH 2.0 and 4.0 while the activity at pH
7.0 was 40-50-fold lower (data not shown). However, by lowering the oH of this sample from 7.0
to 2.0 by dialysis, the arsenite oxidation activity
could be restored.
4.3. Redl, ction of arsenate
The reduction of arsenate in the experiment
presented in Fig. 1 could be caused by the cells or
by the added tetrathionate. To distinguish between these two possibilities, S. acidocaldarius
strain BC was grown in absence of arsenite to
mid-exponential phase at which time the culture
was harvested. These uninduced cells and spent
culture medium were separated by centrifugation
and analysed separately for arsenate-reducing capacity. The results in Table 2 show that washed
cells in presence of tetrathionate (sample 1) could
reduce arsenate while cells incubated without tetrathionate (sample I1) could not. Also, spent
culture medium alone (sample I11) had the capacity to reduce arsenate. Fresh medium 9K, with
(sample IV) or without tetrathionate (sample V),
could not reAuce arsenate. The arsenate-reducing
capacity of the spent culture medium was lost
after dialysis overnight against 9K-medium (sam-
Table 1
Separation of the arsenite oxidation activity by fractional
centrifugation
Fraction
Total activity(units)
Starting material '~
10000
S2o
4 600
P21~
3 700
Ss.
--: It)
Pso
?,2tR]
a After French press treatment but before centrifugation. The
starting material was dialysed overnight against 51) mM
acetate buffer, pH 2.0.
Table 2
Determination of arsenate reduction capacity
Sample
Addition
of 5 mM
tetrathionate
+
-
1. Washed cells
I1. Washed cells
Ill. Spent culture
medium
IV. 9K-medium
+
V. 9K-medium
VI. Dialysed culture
medium
-
Arsenate Total arsenic
in solution in solution
(mM)
mM
< 0.02
1.01
0.94
1.10
0.03
1.02
1.04
0.98
0.96
1.12
I.I13
1.15
pie Vl). This shows that it is not the cells directly
that are responsible for the reduction of arsenate
but some low molecular mass, transient metabolite released from the cells during growth. The
arsenate-reducing capacity for the pooled samples in Fig. l confirmed this. The reduction activity was present when there was still tetrathionate
left in the medium (Pool l) or when the tetrathionate had been re-added to the culture
medium (Pool 3). W h e n the tetrathionate was
depleted (Pool 2), no arsenate reduction activity
was found in the medium (data not shown).
To study the time course of the reduction of
arsenate, S. acidocaldarius strain BC was grown
to mid-exponential phase at which time 2 mM
arsenate was added (Fig. 2). After 9 h, 76% of
the arsenate added was reduced and when the
energy source was depleted, the concentration of
arsenate in the growth medium started to rise
again. A t the end of the experiment, 4 0 - 5 0 % of
the arsenate added was recovered.
5. DISCUSSION
The reason for not observing the oxidation of
arsenite until late-exponential phase is probably
two-fold: (1) the induction time needed for arsenite oxidation activity to appear and (2) the reduction activity that is observed during cellular
growth. The reduction of arsenate is 20-fold faster
than the oxidation of arsenite (estimated from
~
A
~
0.10 t
'10
a
s
~
2
0,0
,
,
2.4
2.o~
:~
v
.-~
1.6
1.2
o.a I
11.4
0
,
0
',
',
40
',
',
80
~
: : ~ I f I
120
160
200
240
Any conclusive statement about resistance mechanisms mus t await further experiments.
After the Ff.:nch press treatment and pelleting
of the unbroken cells and large cell debris to
obtain a cell-frce extnact, we found that all of the
arsenite oxidation activity in the remaining supernatant, S_,., could be recovered in the p~. fraction. This suggests that the enzyme(s) responsible
for the arsenite oxidation activity is located in the
membrane. Several methods, including treatment
with different detergents, were used to try to
solubilize the ar'~cnitc oxidation activity. None of
these treatments worked and therefore no further
purification of the activit~ was achieved.
The scnsitivity of the arscnite oxidation activity
to proteinasc K shows that the oxidation reaction
is enzymatic. Since acidophilic bacteria have been
shown to have a cytoplasmic pH of 5.5 to 6.6 [14],
the results from the pH-dependence experiment
suggest that the membrane bound enzyme(s) responsible for the oxidation of arsenite is located
near the outside of the cell, where the pH is
normally low for these archaebacteria.
Time (hours)
Fig. 2. Time course of the reduction of arsenate. The arrow
shows the addition of 2 mM arsenate. A: ,a, optical density:
E], tetrathionate concentration. .,~,arsenate concentration:
total arsenic concentration.
O,
B:
data in Fig. 2), but proper rate determinations
have to await the separation of the two activities.
The toxicity of soluble arsenic compotmds is
related to the oxidation state of the arsenic. Since
arsenate is the less toxic ion of the two, oxidation
of arsenite to arsenate is considered to be a
resistance mechanism towards arsenite [3]. Addition of 1 mM arsenite to the growth medium
reduces the growth rate of both ti~e uninduced
and the induced culture. Even though Sulfolobus
acidocaldarius strain BC can oxidize arsenite to
arsenate, no significant difference in growth rates
for the two cultures was seen (data not shown).
One possible explanation for this lack of effect on
growth rate can be the rapid reduction of the
arsenate formed back to the toxic ion arsenite.
ACKNOWLEDGEMENTS
Kcvin Hallbcrg is gratefully acknowledged for
critically reviewing the manuscript. This work was
supported by grants from the Nordic Industrial
Foundation and the Swedish National Board for
Technical Development.
REFERENCES
[1] ('ollinet, M.-N and Morin, D. (1990) Antonie yon
Lceuwenhoek 57. 237-2.14.
[2] Dabbs, E.R. and Sole, G.J. (19,~8) Mol. Gcn. Genet. 211.
148-154.
[3] Williams, J.W and Silver, S. (1984) Enz. Microbiol.
Technol. 6. 53t|-537.
[4] Osborne, F.H and Ehrlich, H.I.. (1976) J. Appl. Bactcri01. 41, 2t)5-3q)5.
[5] Phillips, S.E. and Taylor, M.L. (197~) Appl. Environ.
MicrobioL 32. 392-3t)9.
16] Abdrashitova, S.A., Ilyaletdinov, A.N.. Mynbaeva. B.N.
and Abdulina, G.G. (1982) Mikrobiologiya51, 27-30.
[7] Rinderle, SJ., Schrier. J. and Williams,J.W. (1984) Fed.
Proc. 43, 21)(~0.
[8] Marsh. R.M. and Norris, P.R. 119831 FEMS Microbiol.
Lett. 17. 311-315.
[9] Silverman. M.P. and Lundgren. D.G. 119591 J. Bacteriol.
77. 642-647.
[111] Kelly. D.P.. Chambers, L.A. and Trudinger, P.A. 11969)
Anal. Chem. 41. 898-9111.
[11] Tan. L.K. and Dulrizac. J.E. 119851 Anal. Chem. 57,
11)27-11132.
[12] Tan, L.K. and Dutrizac. J.E. 119851 Anal. Chem. 57,
2615 -2620.
[13] LindstrSm, E.B. and Sehlin. H.M. (1089) In: Biohydrometallur~y 1989, (Salley, J., McCready, R.G.L. and
Wichlacz, P.L., Eds.), pp. 59-70. Canmet, Ottawa.
[14] Matin, A. (1990)FEMS Microbiol. Rev. 75. 307-318.