Biodegradation 9: 133–141, 1998.

© 1998 Kluwer Academic Publishers. Printed in the Netherlands.

133

Aerobic chromate reduction by Bacillus subtilis

Carlos Garbisu∗, Itziar Alkorta, Marı́a J. Llama & Juan L. Serra

Departamento de Bioquı́mica y Biologı́a Molecular, Facultad de Ciencias, Universidad del Paı́s Vasco, Apartado
644, 48080 Bilbao, Spain (∗ author for correspondence)

Accepted 10 June 1998

Key words: aerobic, Bacillus subtilis, chromate, reduction

Summary
We have studied the reduction of hexavalent chromium (chromate) to the less toxic trivalent form by using cell
suspensions and cell-free extracts from the common soil bacterium, Bacillus subtilis. B. subtilis was able to grow
and reduce chromate at concentrations ranging from 0.1 to 1 mM K2 CrO4 . Chromate reduction was not affected by
a 20-fold excess of nitrate-compound that serves as alternate electron acceptor and antagonizes chromate reduction
by anaerobic bacteria. Metabolic poisons including sodium azide and sodium cyanide inhibited chromate reduction.
Reduction was effected by a constitutive system associated with the soluble protein fraction and not with the
membrane fraction. The reducing activity was heat labile and showed a Km of 188 µm CrO4 2− . The reductase can
mediate the transfer of electrons from NAD(P)H to chromate. The results suggest that chromate is reduced via a
detoxification system rather than dissimilatory electron transport.

Introduction tion, ceramic, pyrotechnics, electronics and so on

(Cervantes 1991; Ohtake et al. 1990a).
Conventional methods for removing chromates
Chromium (Cr) is an essential micronutrient required
from effluents include chemical reduction followed by
for the growth of many organisms. However, at high
precipitation, ion exchange and adsorption on coal,
concentrations, Cr is toxic, mutagenic, carcinogenic
activated carbon, alum, kaolinite and flyash (Ohtake
and teratogenic (Komori et al. 1990; Luli et al. 1983).
et al. 1990a). Nevertheless, these methods are expen-
Of the two stable Cr oxidation states, Cr(VI), apart
sive due to their requirements for high energy or large
from being the predominant species involved in mu-
quantities of chemical adsorbents.
tagenicity, carcinogenicity, and teratogenicity (Petrilli
Some chromate resistant bacteria have been shown
& De Flora 1977), is approximately 100 times more
to reduce chromate to the trivalent form, includ-
toxic than Cr(III), a form considered relatively in-
ing strains of Pseudomonas (Bopp & Ehrlich 1988;
nocuous (Luli et al. 1983). Moreover, while Cr(VI)
Horitsu et al. 1987; Ishibashi et al. 1990; Suzuki et
is highly soluble in water and forms divalent oxyan-
al. 1992), Aeromonas (Kvasnikov et al. 1985), En-
ions [chromate (CrO4 2− ) and dichromate (Cr2 O7 2− )],
terobacter (Ohtake et al. 1990a, b; Wang et al. 1989,
Cr(III) readily forms less-soluble chromium hydrox-
1990, 1991), and other species (Gvozdyak et al. 1986).
ides at neutral pH (Bell & Lott 1966). The biological
Chromate reduction occurs both in anaerobic (chro-
effects of chromates appear to be related to its up-
mate being used as final electron acceptor) and aerobic
take, since while Cr(VI) passes easily through cellu-
conditions. Anaerobic chromate reduction occurs with
lar membranes, the trivalent form is unable to cross
a membrane preparation (Komori et al. 1989; Wang et
through biological membranes (Arslan et al. 1987).
al. 1989, 1990). On the other hand, the aerobic chro-
Wastewaters containing chromates are generated in
mate reductase activities found in other bacteria (Bopp
many industrial processes including manufacture of
& Ehrlich 1988; Horitsu et al. 1987) probably involve
metallic alloys (the main use of Cr), chrome leather
soluble proteins (Ishibashi et al. 1990). This bacterial
tanning, metal cleaning processing, wood preserva-

MENNEN/SCHRIKS:Disk/CP: Pips Nr.:182483; Ordernr.:233010-bb (biodkap:bio2fam) v.1.1

biod498.tex; 30/09/1998; 8:51; p.1
134

capacity to reduce Cr(VI) has potential not only for of K2 CrO4 was inoculated with 1 ml of freshly grown
detoxification but also for the removal of chromates inoculum culture and incubated as above.
from water. To study the rate of chromate reduction under
It has also been shown (Efstathion & McKay conditions that do not favor cell growth, resting cell
1977; Novic & Roth 1968) that there are many organ- assays were carried out in the following way: cells
isms that although resistant to high concentrations of were grown overnight in culture medium, harvested
Cr(VI) do no alter the Cr redox state. by centrifugation, washed twice in 1/4 volume of 10
To our knowledge, apart from a brief mention mM Tris/HCl (pH 7.0) plus 2 mM EDTA and resus-
to the capacity of some Gram-positive bacteria (in- pended in 1/2 of this same Tris/EDTA buffer. They
cluding a strain of Bacillus subtilis) to reduce chro- were then incubated as above in an orbital incubator
mate under anaerobic conditions, all studies on chro- and supplemented with the indicated concentrations of
mate reduction have been carried out using Gram- K2 CrO4 .
negative bacteria (mostly, Pseudomonas and Enter- Growth was routinely monitored by measuring op-
obacter strains). Consequently, this is the first re- tical density at 600 nm (OD600 ) in an UVIKON 941
port on the factors affecting chromate reduction by a Plus Spectrophotometer, Milan, Italy. Growth was
Gram-positive organism under aerobic conditions. also monitored by viable cell counts. Viability was as-
sessed by serial dilutions of culture samples in growth
medium lacking chromate, which were immediately
Methods plated on the same medium to determine the number
of viable cells (c.f.u. ml−1 ).
Microorganism and growth conditions
Determination of Cr(VI)
B. subtilis 168t+ , a reverted prototroph of auxotroph
thy− obtained from Marburg strain (Spanish Type Cul- The Cr(VI)-reducing activity was assayed spectropho-
ture Collection, CECT No. 461), was used throughout tometrically at 540 nm (with the supernatant fractions
this study. The organism was cultured on a mini- obtained by centrifuging samples at 15,000 x g for
mal chemically defined liquid medium that contained: 15 min) by measuring the decrease of Cr(VI) using
13.9 g l−1 K2 HPO4 , 6.0 g l−1 KH2 PO4 , 2.0 g l−1 the diphenylcarbazide reagent (Greenberg et al. 1981).
(NH4 )2 SO4 , 1.9 g l−1 C6 H5 Na3 O7 . 2H2 O (tri-sodium In the experiments with cell-free extracts, and due to
citrate dihydrate), 23.6 mg l−1 Ca(NO3 )2 . 4H2 O, the interference caused by some compounds present in
19.8 mg l−1 MnCl2 , 0.28 mg l−1 FeSO4 . 7H2 O and the extract itself on the determination of Cr(VI) using
1% (w/v) glucose. Cultures were grown in 250 ml Er- diphenylcarbazide (Bopp & Ehrlich 1988), chromate
lenmeyer flasks with continuous shaking (250 r.p.m.). reduction was measured spectrophotometrically. Ab-
Minimal medium plates were made by combining sorption spectra were obtained for Cr(VI), Cr(III), and
the above specified ingredients with 1.5% (w/v) agar the cell extracts showing that chromate had a strong
(Difco). The growth temperature was 30 ◦ C. absorbance peak from 350 to 390 nm and that Cr(III)
Growth was initiated in shaked flasks by using in- and the cell extracts caused very little interference (al-
ocula from minimal medium plates. Before initiating though the extracts absorbed in the 350-390 range,
liquid medium experiments, cultures were transferred they did it to a much lesser extent than chromate).
twice into fresh medium, over a period of approxi- Nonetheless, all blanks were prepared using the ex-
mately 24 hours. Reinoculations were timed to ensure tract, thus allowing this interference to be minimized.
that cultures were under excess nutrient conditions, Chromate reduction was thus quantified by measur-
and thus in an environment allowing balanced ex- ing absorbance due to chromate at 375 nm against a
ponential growth. By using this pre-inoculation pro- reagent blank. When the effect of added NAD(P)H
cedure, no lag phase was observed in the control was tested, absorbance was measured at 395 nm in or-
treatments described below. Growth experiments were der to minimize interference of NAD(P)H absorbance
repeated a minimum of three times with consistent with that of chromate. The reaction mixture contained
results. Data from representative experiments are pre- 1 ml of extract and 0.5 – 1 mM K2 CrO4 . To study
sented. the effect of glucose on chromate reduction in the cell-
For growth experiments with chromate, 50 ml of free extracts, the reaction mixtures were supplemented
culture medium containing the indicated concentration with 1% (w/v) glucose.

biod498.tex; 30/09/1998; 8:51; p.2

 135

Cell extract

Crude extract was prepared as follows: cells were

grown overnight at 30 ◦ C in the culture medium, har-
vested by centrifugation at 5,000 x g and 4 ◦ C, and
washed twice (using 1/10 of the initial volume) with
10 mM Tris/HCl (pH 7.0) plus 2 mM EDTA. Sub-
sequently, the cells were resuspended in 1/30 of the
initial volume of this same Tris/EDTA buffer and dis-
rupted in an ice bath by two passages through a French
pressure cell (SLM Aminco, USA) at 24,000 p.s.i.
(165,600 kPa). Unbroken cells were removed by cen-
trifugation at 12,000 x g for 10 min at 4 ◦ C. The super-
natant was then centrifuged a second time at 32,000 x g
for 20 min at 4 ◦ C. Finally, the supernatant fluid of this
second centrifugation was ultracentrifuged at 150,000
x g for 45 min at 4 ◦ C and the new supernatants were
dispensed in 2 ml aliquots and immediately frozen
at −20 ◦ C until use. The pelleted (membrane) frac-
tion was suspended in the same Tris/EDTA buffer and
kept frozen as above. Equivalent amounts of protein Figure 1. B. subtilis growth expressed as OD600 (filled circles) and
of the supernatant fluid (extract) and the resuspended total viable cell count (empty circles) (c.f.u. ml−1 ) for (a) control
(no chromate) culture and (b) 0.5 mM K2 CrO4 culture.
membranes were assayed in all cases.

Protein assay

The protein content was determined according to Pe-

terson (1977) using crystalline bovine serum albumin
as standard.

Results

Chromate reduction by whole cells of B. subtilis

To determine whether B. subtilis was capable of

growth in the presence of chromate, cells grown in
chemically defined minimal medium were used as
inoculum for this same medium containing 0.5 mM Figure 2. Decrease in Cr(VI) in the supernatant fraction of a cul-
K2 CrO4 . While the control cultures (grown in the ture of B. subtilis cells growing in the presence of 0.5 mM K2 CrO4
(filled circles). The decrease in Cr(VI) was also studied in the ab-
absence of Cr) showed no obvious coloration, the cul- sence of cells by means of supplementing the culture medium with
tures grown in the presence of chromate exhibited a the same concentration of chromate (empty circles).
yellowish coloration. Similar final levels of growth
(maximal cell densities were slightly higher for the
control cultures) were observed with and without chro- The data above demonstrated that B. subtilis could
mate as determined by OD600 and total viable counts, tolerate and grow in the presence of 0.5 mM K2 CrO4 ,
but the growth rates differed, being faster for the con- but left open the question whether chromate was me-
trol cultures (Figure 1). The data presented in Figure 1 tabolized. This question was answered by analysis of
indicated that OD600 accurately reflects growth as de- the amount of Cr(VI) remaining in the supernatant
termined by viable counts. The growth data below are, fraction of cultures grown in the presence of chromate
therefore, given in OD600 measurements. (Figure 2). As shown in Figure 2, in the presence

biod498.tex; 30/09/1998; 8:51; p.3

136

Figure 4. Effect of chromate adaptation on (a) B. subtilis growth

Figure 3. (a) Effect of chromate concentration on B. subtilis growth.
and (b) decrease in Cr(VI) in the supernatant fraction of cultures
(b) Decrease in Cr(VI) in the supernatant fraction of B. subtilis
of B. subtilis cells growing in the presence of 0.5 mM K2 CrO4 .
cultures supplemented with different chromate concentrations.
Non-adapted (filled circles) and adapted (empty circles) cultures
were inoculated with a culture grown in the absence or in the
presence of 0.5 mM K2 CrO4 , respectively.
of whole cells, the bulk of the Cr(VI) was removed
most likely through conversion to the trivalent form.
In fact, although a direct demonstration of chromate the well-known toxicity associated with this form of
reduction by means of, for instance, electron paramag- Cr.
netic resonance spectroscopy was not provided, the When minimal medium containing 0.5 mM
reduction of chromates can be judged according to the K2 CrO4 was inoculated with cells previously grown in
color change of the medium: whereas at the beginning the same chromate medium, this ‘chromate-adapted’
the medium appears yellow (as expected due to the culture grew at identical rate than the ‘non-adapted’
color of hexavalent chromium), it becomes more and (control) culture (Figure 4a). In addition, both cul-
more whitish as the reaction proceeds and chromium tures (adapted and non-adapted) followed very similar
hydroxide [Cr(OH)3 ] is being formed. On the other patterns of chromate reduction (Figure 4b). These re-
hand, in the absence of cells (i.e., culture medium sup- sults provide evidence that the system effecting chro-
plemented with 0.5 mM K2 CrO4 ), there is a certain mate reduction is not induced (but constitutive) in B.
degree of purely chemical chromate reduction since subtilis.
the amount of Cr(VI) in the medium decreased from One of the features characterizing anaerobic chro-
0.50 mM K2 CrO4 to 0.44 mM after 24 h of incubation. mate reduction systems is a sensitivity to nitrate and
Nonetheless, when cells were present, the decrease in sulfate ions, both of which compete with chromate
the amount of Cr(VI) was significantly more notice- (CrO4 2− ) as anaerobic electron acceptors. In contrast
able (i.e., from 0.50 mM to 0.04 mM after 24 h of to anaerobic bacteria, the reduction of chromate by B.
incubation) (Figure 2). subtilis was independent of nitrate (since this microor-
Additional experiments revealed that B. subtilis ganism does not perform anaerobic sulfate reduction,
has the ability to grow (Figure 3a) and reduce chro- sulfate was not tested). Neither growth (Figure 5a)
mate (Figure 3b) in the chemically defined liquid nor chromate reduction (Figure 5b) was significantly
medium at chromate concentrations ranging from 0.1 altered by this oxidant (0.5 mM K2 CrO4 ; 10 mM
to 1 mM. Cells failed to grow and reduce chromate NaNO3 ). These findings suggest that chromate is
at 2 mM chromate. When added at 1 mM, chromate not reduced via dissimilatory electron transport in B.
significantly affected cell growth in agreement with subtilis but via a detoxification system.

biod498.tex; 30/09/1998; 8:51; p.4

 137

Figure 7. Effect of kanamycin (30 µg ml−1 ) and low tempera-

ture (4 ◦ C) on the decrease in Cr(VI) in the supernatant fraction
of cultures of B. subtilis resting cells supplemented with 0.5 mM
K2 CrO4 . Control cultures were grown in the absence of kanamycin
and incubated at 30 ◦ C. In the absence of cells (Tris/EDTA buffer
supplemented with 0.5 mM K2 CrO4 ), the decrease in Cr(VI) was
Figure 5. Effect of 10 mM sodium nitrate on (a) B. subtilis growth also determined under the same conditions.
and (b) decrease in Cr(VI) in the supernatant fraction of cultures of
B. subtilis cells growing in the presence of 0.5 mM K2 CrO4 or in
the absence of Cr (control cultures), respectively.
Figure 6 shows the effect of some metabolic poi-
sons (i.e., sodium azide and sodium cyanide) on
growth (Figure 6a) and chromate reduction (Figure 6b)
of B. subtilis cells growing in culture medium supple-
mented with 0.5 mM K2 CrO4 . In fact, the cells were
grown in the absence of these metabolic poisons till
late log phase, moment when 4 mM NaN3 /NaCN was
added to the flasks. As observed in Figure 6(a), growth
was immediately stopped by the addition of these poi-
sons (the control culture, lacking any of these poisons,
kept growing till stationary phase). Chromate reduc-
tion was also markedly inhibited by these metabolic
poisons (Figure 6b). Somewhat different results on
growth and chromate reduction were observed when,
upon reaching late log phase, the incubation temper-
ature of the cultures was lowered from 30 ◦ C to 4 ◦ C
(Figure 6). In fact, the decrease in temperature resulted
in an immediate inhibition of the Cr(VI) reduction,
whereas the cyanide and azide resulted in only partial
inhibition.
To study chromate reduction under conditions
Figure 6. Effect of metabolic poisons (4 mM sodium azide and 4 which do not favor cell growth, resting cell assays
mM sodium cyanide) and low temperature (4 ◦ C) on (a) B. subtilis (cells suspended in Tris/EDTA buffer instead of cul-
growth and (b) decrease in Cr(VI) in the supernatant fraction of cul-
tures of B. subtilis cells growing in the presence of 0.5 mM K2 CrO4 .
ture medium) were carried out. The effect of the inhi-
Control cultures were grown in the absence of metabolic poisons bition of the (1) de novo protein synthesis as well as (2)
and incubated at 30 ◦ C. The cultures were grown under normal metabolic activity on chromate reduction by resting
incubation conditions till late log phase (arrows) when either the cells was studied by supplementing the cultures with
treatment with metabolic poisons or low temperature was applied.
30 µg kanamycin monosulfate ml−1 and incubating

biod498.tex; 30/09/1998; 8:51; p.5

138
Table 1. Decrease in Cr(VI) Catalyzed by Cell-Free Extracts of
B. subtilis∗

Reaction Soluble Boiled Membrane Tris/EDTA

Time Fraction Soluble Fraction Buffer
(h) Fraction

0 100 100 100 100

1 57 100 100 98
2 43 100 100 98
3.5 29 100 99 98
4.5 21 100 98 98
∗ Values are given as a percentage of initial chromate concen-
tration.
100% corresponds to 0.5 mM K2 CrO4 .

Table 2. Decrease in Cr(VI) Catalyzed by Cell-Free Extracts of

Figure 8. Cr(VI) decrease is associated with the presence of cells. B. subtilis Supplemented with 1% (w/v) Glucose∗
An overnight grown culture was centrifuged to separate the cell
fraction from the supernatant fraction. The cells were resuspended Reaction Soluble Boiled Membrane Tris/EDTA
in fresh medium and supplemented with 0.5 mM K2 CrO4 (filled Time Fraction Soluble Fraction Buffer
circles). The supernatant fraction was filtered through a 0.45 µm
(h) Fraction
filter and then supplemented with the same chromate concentration
(empty circles). 0 100 100 100 100
1 3 100 100 99
2 0 100 99 98
them at 4 ◦ C, respectively. As shown in Figure 7, while
3.5 0 100 99 98
incubating the resting cells at 4 ◦ C inhibited chromate 4.5 0 100 99 97
reduction, the presence of kanamycin did not stop
∗ Values are given as a percentage of initial chromate concen-
chromate reduction by resting cells. Nonetheless, the
tration.
rate of chromate reduction in the kanamycin culture
100% corresponds to 0.5 mM K2 CrO4 .
was not as fast as in the control culture (no kanamycin
present, and resting cells incubated at 30 ◦ C) (Fig-
ure 7). In the absence of cells (Tris/EDTA buffer natant fraction) were then supplemented with 0.5 mM
supplemented with 0.5 mM K2 CrO4 ), no chromate K2 CrO4 . Figure 8 shows that the presence of cells is
reduction could be observed under any of the condi- required for the reduction to occur, since there was
tions tested. Similar results were obtained when the no chromate reduction whatsoever in the supernatant
resting cell cultures were supplemented with 1% (w/v) filtered fraction.
glucose (data not shown). However, in this case, the
extent of chromate reduction was greater than when Chromate reduction by cell extracts
the resting cells were incubated in the absence of an
external carbon source [Cr(VI) decreased from 0.50 With the objective of clarifying the mechanism of
mM to less than 0.15 mM in the kanamycin and Cr(VI) decrease (i.e., chromate reduction) from the
control cultures after 10 hours of incubation in the liquid phase of the B. subtilis cultures, we investi-
presence of 1% (w/v) glucose] (data not shown). gated whether or not a cell free extract was able to
To confirm that chromate reduction was associated decrease the Cr(VI) content as well. As shown in
with the cells themselves and not with any possible Table 1, cell extracts of B. subtilis readily reduced
compounds excreted by the cells to the surround- chromate. Actually, the Cr(VI) decreasing activity ap-
ing medium, after growing them overnight in culture pears to be located in the soluble fraction obtained
medium, the cultures were centrifuged. While the pel- after ultracentrifuging the crude cell extracts. No sig-
let (cell fraction) was resuspended in fresh culture nificant reduction was observed with the membrane
medium, the supernatant was subsequently filtered fractions. Similarly, a soluble fraction heated at 100 ◦ C
under sterile conditions through a 0.45 µm filter (Mil- for 10 min did not cause a decrease in Cr(VI) indi-
lipore, Bedford, USA) to remove unpelleted cells and cating that the reducing activity was heat labile. The
suspended cell debris. Both fractions (cell and super- addition of 1% (w/v) glucose to the soluble fraction

biod498.tex; 30/09/1998; 8:51; p.6

 139
Table 3. Effect of NAD(P)H Addition on the Decrease in Cr(VI) Catalyzed by Soluble Cell
Fractions of B. subtilis∗

Reaction Soluble Soluble Soluble Tris/EDTA Tris/EDTA Tris/EDTA

Time Fraction Fraction Fraction +NADH +NADPH
(h) +NADH +NADPH

0 100 100 100 100 100 100

2 99 52 43 98 93 98
4 87 32 27 96 88 96
6 82 27 15 95 85 88
8 57 12 7 90 85 80
∗ Values are given as a percentage of initial chromate concentration. 100% corresponds to 1.0
mM K2 CrO4 . Cultures were supplemented with 3 mM NAD(P)H as indicated.

caused a dramatic increase in the rate of chromate hydroxide [Cr(OH)3] was being formed. Gvozdyak et
reduction (Table 2). Nevertheless, no significant de- al. (1986) already described the ability of a faculta-
crease in Cr(VI) was found when glucose was added tive anaerobic strain of B. subtilis to reduce hexavalent
to the Tris/EDTA buffer supplemented with 0.5 mM Cr though under anaerobic conditions. In this respect,
K2 CrO4 (3% reduction in 4.5 hours) suggesting that B. subtilis differs from strains previously reported by
the reduction does not come from the action of glucose Kvasnikov et al. (1985), Lebedeva & Lyalikova (1979)
itself but some other compound present in the soluble and Romanenko & Koren’kov (1977) in its ability to
fraction of the cell extracts. reduce chromate aerobically.
We also examined whether or not redox cofactors So far, the ability to reduce hexavalent Cr aer-
had an effect on the Cr(VI) decreasing activity of the obically has been detected in species of the genus
extracts. As concluded from Table 3, both NADH and Pseudomonas (Bopp & Ehrlich 1988; Horitsu et al.
NADPH greatly increased the chromate reducing ac- 1987; Ishibashi et al. 1990). However, under anaer-
tivity of the soluble fractions. Again, although some obic conditions, Gvozdyak et al. (1986) showed that
reduction was observed when the Tris/EDTA buffer the reduction of Cr(VI) to Cr(III) is not a strictly spe-
was supplemented with NAD(P)H (around 20% in 8 cific process and can be carried out by many cultures
hours), it was not comparable to that found when the of Gram-positive and Gram-negative microorganisms.
soluble fraction was supplemented with those redox In this context, it is worth mentioning the extensive
cofactors (around 90% in 8 hours). studies carried out with Enterobacter cloacae HO1,
Finally, an apparent Michaelis-Menten constant a strain isolated from activated sludge that uses chro-
(Km ) of 188 µm for CrO4 2− and a maximum veloc- mate as an electron acceptor in an anaerobic mode of
ity (Vmax ) of 1.25 nmol of Cr(VI) reduced min−1 respiration (Komori et al. 1989, 1990; Ohtake et al.
per mg of protein were estimated when data were fit- 1990a, b; Wang et al. 1989, 1990, 1991).
ted by non-linear regression to the Michaelis-Menten Our results support that B. subtilis has the abil-
equation. ity to grow and reduce chromate aerobically in the
chemically defined medium at chromate concentra-
tions ranging from 0.1 to 1 mM while concentrations
Discussion above those values are lethal to growing cells and
prevent the reduction. Other bacteria such as P. fluo-
The present results demonstrate that a well-characteri- rescens LB300 have been described to tolerate much
zed laboratory strain of the Gram-positive com- higher concentrations of chromate and actively re-
mon soil bacterium, B. subtilis, removes aerobically duce Cr(VI) to Cr(III) while growing aerobically in a
hexavalent Cr (chromate) from solution most likely by minimal salts medium (Bopp & Ehrlich 1988). More-
reduction to the less toxic and more insoluble trivalent over, E. cloacae can reduce chromate at levels of 3–5
form. In fact, whereas at the beginning the culture mM, although above 5 mM the cells lose viability
medium appeared yellow (indicative of the presence preventing the reduction as well (Komori et al. 1989).
of hexavalent Cr), it became more and more whitish The B. subtilis system of chromate reduction ap-
as cell growth progressed and theoretically chromium pears to be constitutive. In fact, the lack of significant

biod498.tex; 30/09/1998; 8:51; p.7

140

effect of kanamycin on the chromate-reducing activity The chromate reductase in our B. subtilis strain is
suggests that this activity may not require de novo pro- similar to that reported by Horitsu et al. (1987) and
tein synthesis. Similar results were found by Bopp & Bopp & Ehrlich (1988) in that it can use NADH as
Ehrlich (1988) in their studies of the rates of chromate electron donor for the reduction of chromate. Nev-
reduction by chromate grown cells and cells grown ertheless, although B. subtilis can also mediate the
without chromate. In addition, the finding that the re- transfer of electrons from NAD(P)H to chromate, the
duction of chromate by B. subtilis was independent reducing enzyme of P. ambigua requires NADH but
of nitrate suggests that chromate is not reduced via not NADPH. Unlike P. putida (Ishibashi et al. 1990),
dissimilatory electron transport but via a detoxifica- our strain does not require the addition of exogenous
tion system. Similar results have been reported for P. NAD(P)H to the cell extract for chromate reduction to
putida (Ishibashi et al. 1990). occur, since the soluble fraction of the extract itself
Metabolic poisons (4 mM of either sodium azide or appears to have the cofactors necessary for the reduc-
sodium cyanide) inhibited both cell growth and chro- tase. However, the addition of NAD(P)H significantly
mate reduction. Somewhat different results on growth enhances the rate of chromate reduction.
and chromate reduction were observed when, upon B. subtilis also resembles P. fluorescens in that it
reaching late log phase, the incubation temperature can use exogenous glucose as electron donor too. In
of the cultures was lowered from 30 ◦ C to 4 ◦ C. In this respect, opposite results were found in P. ambigua
fact, the decrease in temperature resulted in an im- G-1 (Horitsu et al. 1987).
mediate inhibition of the Cr(VI) reduction, whereas The Cr (VI)-reducing activity was heat labile and
the cyanide and azide resulted in only partial inhibi- showed a Km of 188 µm for Cr2 O4 2− and a Vmax of
tion. This would be expected as the cyanide and azide 1.25 nmol of Cr(VI) reduced min−1 per mg of pro-
prevent de novo protein synthesis but do not inhibit tein. Similar results were reported by Ishibashi et al.
the activity of any reductase enzymes that are already (1990) in their studies with P. putida, although their
present. On the contrary, the reduced temperature will Km and Vmax values were somewhat different (40 µm
inhibit the activity of any temperature enzymes present for Cr2 O4 2− and 6 nmol of Cr(VI) reduced min−1
in the culture as well as the synthesis of any new per mg of protein).
enzymes. These findings raise the possibility that this or-
Komori et al. (1989) reported that while 100– ganism may be useful for environmental restoration
400 µm sodium cyanide also inhibited chromate re- of chromate polluted waters and liquid waste treat-
duction in E. cloacae, no inhibition was observed with ment. Cr(III) is several orders of magnitude less toxic
0.5-1.0 mM sodium azide. However, as indicated, the than Cr(VI) and forms insoluble hydroxides at neutral
concentrations used by these authors were much lower pH and precipitates, thus making it less bioavail-
than the ones used in our study. From our studies with able (Bopp & Ehrlich 1988). B. subtilis is able to
resting cells, it can be concluded that chromate reduc- form dormant spores, which are resistant to heat and
tion is not dependent upon cell growth but just on the dessication, can be easily stored and transported, and
presence of metabolically active cells. readily germinated into vegetative cells by the addition
The reduction by B. subtilis appears to be of simple nutrients. These attributes make B. subtilis a
enzymatically-driven, since it is catalyzed by cell ex- promising candidate for use in knowledge-based toxic
tracts. Such chromate reductase activity appeared to metal remediation systems. Furthermore, and since B.
be associated with soluble protein rather than with the subtilis is widely distributed in soils and groundwater
membrane fraction. Similar results have been reported systems, the results shown here suggest that these sys-
in P. putida (Ishibashi et al. 1990). On the other hand, tems have a natural capacity for attenuating chromate
it has been described (Bopp & Ehrlich 1988) that some pollution.
or all of the enzymes necessary for the transfer of elec- Experiments are in progress to identify the bio-
trons from NADH to chromate are membrane-bound chemical mechanisms responsible for reducing chro-
in a strain of P. fluorescens grown aerobically. Wang mate in B. subtilis. We are currently assessing the
et al. (1990, 1991) also found that the chromate reduc- proteins involved in the reduction of chromate in this
tase of the anaerobically grown E. cloacae HO1 was strain in an attempt to unravel the nature of the re-
preferentially associated with the membrane fraction ductase catalyzing the reduction of Cr(VI) to Cr(III).
of the cells. In this respect, Suzuki et al. (1992) have already re-
ported on the identification of an NAD(P)H-dependent

biod498.tex; 30/09/1998; 8:51; p.8

 141

chromium reductase of P. ambigua G-1 that reduces Komori K, Rivas A, Toda K & Ohtake H (1990) Biological removal
Cr(VI) to Cr(III) with at least two reaction steps via of toxic chromium using an Enterobacter cloacae strain that re-
duces chromate under anaerobic conditions. Biotechnology and
Cr(V) as an intermediate. Bioengineering 35: 951–954
Kvasnikov EI, Stepanyuk VV, Klushnikova TM, Serpokrylov NS,
Simonova GA, Kasatkina TP & Panchenko LP (1985) A new
Acknowledgments chromium-reducing gram variable bacterium with myxed type
of flagellation. Mikrobiologiya 54: 83–88
Lebedeva EV & Lyalikova NN (1979) Reduction of crocoite by
C.G. and I.A. are the recipients of fellowships from Pseudomonas chromatophila sp. nov. Mikrobiologiya 48: 517–
the Departamento de Educación, Universidades e In- 522
vestigación of the Basque Government. This work was Luli GW, Talnagi JW, Strohl WR & Pfister RM (1983) Hexava-
lent chromiun-resistant bacteria isolated from river sediments.
partially supported by the Basque Government (PGV Applied and Environmental Microbiology 46: 846–854
95-011). Novic RP & Roth C (1968) Plasmid-linked resistance to inor-
ganic salts in Staphylococcus aureus. Journal of Bacteriology 95:
1335–1342
Ohtake H, Fujii E & Toda K (1990a) Reduction of toxic chromate
References in an industrial effluent by use of a chromate-reducing strain of
Enterobacter cloacae. Environmental Technology 11: 663–668
Arslan P, Beltrame M & Tomasi A (1987) Intracellular chromium Ohtake H, Fujii E & Toda K (1990b) Bacterial reduction of
reduction. Biochimica et Biophysica Acta 931: 10–15 hexavalent chromium: kinetic aspects of chromate reduction by
Bell CF & Lott KAK (1966) Modern Approach to Inorganic Enterobacter cloacae HO1. Biocatalysis 4: 227–235
Chemistry, 2nd edn. Butterworth & Co., London Peterson GL (1977) A simplification of the protein assay method
Bopp LH & Ehrlich HL (1988) Chromate resistance and reduc- of Lowry et al. which is more generally applicable. Analytical
tion in Pseudomonas fluorescens strain LB300. Archives of Biochemistry 83: 346–356
Microbiology 150: 426–431 Petrilli FL & De Flora S (1977) Toxicity and mutagenicity of
Cervantes C (1991) Bacterial interactions with chromate. Antonie hexavalent chromium on Salmonella typhimurium. Applied and
van Leeuwenhoek 59: 229–233 Environmental Microbiology 33: 805–809
Efstathiou JD & McKay LL (1977) Inorganic salts resistance associ- Romanenko VI & Koren’kov VN (1977) A pure culture of bacteria
ated with a lactose-fermenting plasmid in Staphylococcus lactis. utilizing chromates and bichromates as hydrogen acceptors in
Journal of Bacteriology 130: 257–265 growth under anaerobic conditions. Mikrobiologiya 46: 414–417
Greenberg AE, Connors JL, Jenkins D & Franson MAH (1981) Suzuki T, Miyata N, Horitsu H, Kawai K, Takamizawa K, Tai Y
Standard Methods for the Examination of Water and Wastewater, & Okazaki M (1992) NAD(P)H-dependent chromium(VI) re-
15th edn. American Public Health Association, Washington, D.C ductase of Pseudomonas ambigua G-1: a Cr(V) intermediate
Gvozdyak PI, Mogilevich NF, Rylśkii AF & Grishchenko NI (1986) is formed during the reduction of Cr(VI) to Cr(III). Journal of
Reduction of hexavalent chromium by collection strains of bac- Bacteriology 174: 5340–5345
teria. Mikrobiologiya 55: 962–965 Wang P-C, Mori T, Komori K, Sasatsu M, Toda K & Ohtake H
Horitsu H, Futo S, Miyazawa Y, Ogai S & Kawai K (1987) (1989) Isolation and characterization of an Enterobacter cloacae
Enzymatic reduction of hexavalent chromium by hexavalent strain that reduces hexavalent chromium under anaerobic condi-
chromium tolerant Pseudomonas ambigua G-1. Agricultural and tions. Applied and Environmental Microbiology 55: 1665–1669
Biological Chemistry 51: 2417–2420 Wang P-C, Mori T, Toda K & Ohtake H (1990) Membrane-
Ishibashi Y, Cervantes C & Silver S (1990) Chromium reduction in associated chromate reductase activity from Enterobacter cloa-
Pseudomonas putida. Applied and Environmental Microbiology cae. Journal of Bacteriology 172: 1670–1672
56: 2268–2270 Wang P-C, Toda K, Ohtake H, Kusaka I & Yabe I (1991) Membrane-
Komori KP, Wang PC, Toda T & Ohtake H (1989) Factors affecting bound respiratory system of Enterobacter cloacae strain HO1
chromate reduction in Enterobacter cloacae strain HO1. Applied grown anaerobically with chromate. FEMS Microbiology Letters
Microbiology and Biotechnology 31: 567–570 78: 11–16

biod498.tex; 30/09/1998; 8:51; p.9