FEMS Microbiology Letters 121 (1994) 141-146
© 1994 Federation of European Microbiological Societies 0378-1097/94/$07.00
Published by Elsevier
141
FEMSLE 06102
Phosphatase-mediated heavy metal accumulation
by a Citrobacter sp. and related enterobacteria
L.E. Macaskie *, K.M. B o n t h r o n e and D.A. R o u c h 1
School of Biological Sciences, The Unit,ersity of Birmingham, Edgbaston, Birmingham B15 2T7] UK
(Received 6 May 1994; revision received 1 June 1994; accepted 3 June 1994)
Abstract: A Citrobacter sp. was reported previously to accumulate heavy metals as cell-bound heavy metal phosphates. Metal
uptake is mediated by the activity of a periplasmic acid-type phosphatase that liberates inorganic phosphate to provide the
precipitant ligand for heavy metals presented to the cells. Amino acid sequencing of peptide fragments of the purified enzyme
revealed significant homology to the phoN product (acid phosphatase) of some other enterobacteria. These organisms, together
with Klebsiella pneumoniae, previously reported to produce acid phosphatase, were tested for their ability to remove uranium and
lanthanum from challenge solutions supplemented with phosphatase substrate. The coupling of phosphate liberation to metal
bioaccumulation was limited to the metal accumulating Citrobacter sp.; therefore the participation of species-specific additional
factors in metal bioaccumulation was suggested.
Key words: Citrobacter sp.; Salmonella typhimurium; Morganella morganii; Providencia stuartii; Klebsiella pneumoniae; Phosphatase;
PhoN; Heavy metal accumulation
Introduction
A strain of a species of Citrobacter, isolated
from metal-polluted soil [1] was resistant to cadmium and accumulated this and other metals as
ceil-bound metal phosphates [2-4]. Metal accumulation is mediated by an acid-type phosphatase
which liberates inorganic phosphate (HPO42-)
from a suitable substrate molecule, with stoichiometric precipitation of metal phosphates as cell-
* Corresponding author. Tel: (021) 414-5889; Fax: (021) 4146557.
i Present address: New Chemistry Laboratory, The University of Oxford, South Parks Road, Oxford OX1 3QT, UK
SSDI 0 3 7 8 - 1 0 9 7 ( 9 4 ) 0 0 2 5 3 - N
bound MHPO 4 ([3,4]; M = heavy metal cation).
This reaction has been harnessed to the continuous removal of heavy metals from challenge solutions presented to immobilized cells held within a
flow-through column [2]; application to the decontamination of metal-laden industrial wastes
has been suggested [2].
The insensitivity of the acid-type phosphatase
to cadmium [5,6] and some other heavy metals
[61, together with the ability of growing cells to
accumulate Cd [7], suggests that metal accumulation may have evolved as a resistance mechanism
whereby metals are desolubilized away from cellular sensitive sites; indeed, lead was accumulated
extensively by another Citrobacter sp. isolated
from a metal-contaminated soil [8], while strains
142
of Citrobacter freundii obtained from a culture
collection gave only low phosphatase activity and
very little metal bioaccumulation (D. Montgomery, A.C.R. Dean and L.E. Macaskie, unpublished).
The acid-type phosphatase has been purified
to homogeneity [6,9] and extensively characterized [6]. During this earlier study peptide mapping was done, and amino acid sequences of the
peptide fragments were obtained [6]. A significant homology was seen with some of the sequences of the phoN product (acid phosphatase)
of some other enterobacteria (this investigation).
It might be anticipated that these enterobacterial
species would also accumulate heavy metals and
may find a similar application in metal waste
treatment bioprocesses. These organisms, together with Klebsiella pneumoniae (previously Kl.
(Aerobacter) aerogenes) with reported acid phosphatase activity [10], were tested for their ability
to accumulate uranium and lanthanum from challenge solutions.
Materials and Methods
Citrobacter sp. strain N14 was isolated from
metal-polluted soil [1], and was as described previously [11]. The phosphatase deficient mutant
lp4a was as described previously; phosphatase
deficiency was confirmed by immunogold labelling experiments [6]. Strains of Salmonella typhimurium were a gift from Dr. J. Stephen (School
of Biological Sciences, University of Birmingham).
Other enterobacteria ( Klebsiella pneumoniae
subsp, pneumoniae NCIMB 8806, Morganella
morganii NCIMB 10466 and Providencia rettgeri
NCIMB 10457) were from the National Collection of Industrial and Marine Bacteria (Aberdeen,
Scotland). The organisms were routinely grown in
glycerol-glycerol 2-phosphate minimal medium as
described previously [11]; in this case phosphatase activity assists in the utilization of glycerol 2-phosphate as the sole phosphorus source.
The growth temperature was 30°C (Citrobacter
sp., M. morganii, P. rettgeri, KI. pneumoniae) or
37°C (S. typhimurium, Kl. pneumoniae); with the
latter strain similar results were obtained at both
temperatures. M. morganii and some strains of S.
typhimurium grew poorly in the minimal medium;
yeast extract (Difco) was routinely incorporated
to 0.1% (wt/vol), as appropriate. For determination of phosphatase activity, samples were removed from mid-logarithmic phase cultures (after
3-4 generations: OD at 600 nm (OD600) = 0.30.4) or from overnight cultures, as specified, and
the phosphatase activity was determined by the
liberation of p-nitrophenol from p-nitrophenyl
phosphate, as described previously [11]. Although
the pH optimum of acid phosphatase is typically
in the region of 5.0-6.5 [6], this is extended to pH
7.0 in whole cells [12]; phosphatase activity was
assayed at the latter pH since this is the pH of
maximum metal accumulation [12].
The Citrobacter phosphatase was purified to
homogeneity via ammonium sulphate fractionation, ion exchange chromatography (SP-Sephadex
C50 cation exchange and QAE Sephadex A-50
anion exchange), hydroxyapatite separation and
chromatography using phenyl sepharose as described previously [6,9]. Enzyme purity was confirmed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis, and immunoblotting [6].
Amino acid sequencing of peptide fragments was
done commercially (Department of Biochemistry,
University of Oxford, UK) using an Applied
Biosystems (ABI) 470 A protein sequencer with
on-line analysis (PTH analyser) using the standard 03CPTH program from Applied Biosystems.
Data analysis utilized 'Export Ease' software from
Millipore (UK) Ltd.
For evaluation of the metal uptake capacity of
the Citrobacter sp. and other enterobacteria, cultures were harvested as for the phosphatase assay, washed in isotonic saline (8.5 g / l of NaC1)
and resuspended at an OD60o of 0.3-0.4 in metal
challenge solution (2 mM citrate buffer/20 mM
MOPS (3-[morpholino]propanesulfonic acid)NaOH buffer, pH 7, supplemented with either 1
mM uranyl nitrate or lanthanum nitrate (B.D.H.
Ltd.)). The suspensions were shaken gently for 6
h or 18-24 h as specified, and the residual metal
in the supernatant of centrifuged samples was
determined by assay.
Uranyl ion (UO 2+) and lanthanum were assayed using arsenazo III by a modification of the
143
method of Fritz and Bradford [13]. A saturated
solution of arsenazo I I I was p r e p a r e d by the
addition of 0.038 g of arsenazo I I I (B.D.H. Ltd.)
to 25 ml of distilled water (1 h, with stirring) and
filtration through W h a t m a n No 1 filter paper.
Metal challenge solution or spent supernatant
was diluted 20-fold (uranyl ion) or 10-fold
(lanthanum) in metal-unsupplemented challenge
solution to a total volume of 2 ml. The samples
were acidified (0.3 ml of 0.75 M HC1), 0.1 ml of
arsenazo III solution was added and the bluemauve metal-arsenazo III complex was visualized
and quantified at A652. Metal removal was expressed as a percentage of the initial metal in the
challenge solution.
Some experiments utilized immobilized cells.
Five g wet weight of Citrobacter sp. N14 or P.
rettgeri N C I M B 10457 was immobilized in polyacrylamide gel as described previously [9,12]. Each
preparation provided material for five replicate
columns of volume 25 ml (approx). The columns
were challenged with metal-supplemented flows
(composition as above) with metal removal determined by assay and comparison with the input
solution.
Strains of Morganella, Providencia and Salmonella were obtained and examined for phosphatase activity and metal bioaccumulation. A
strain of Klebsiella pneumoniae was also tested.
Although not apparently containing nueleotide
sequences that hybridize to phoN [16], substantial biochemical evidence was found for acid
phosphatase activity in this species [10]. Conversely, the Citrobacter sp. N14 clearly contains a
PhoN-type phosphatase (Fig. 1), but no evidence
for hybridization of Salmonella phoN fragments
to D N A from the related C. freundii was found
[16].
Various enterobacteria were screened for
phosphatase activity and ability to accumulate
La 3+ and U O 2+ as shown in Table 1. As reported previously [11] the phosphatase activity of
Citrobacter sp. N14 was highly variable. T h r e e
high-activity cultures removed most of the presented metals, while the phosphatase-deficient
mutant lp4a removed no UOzz÷ but some La 3+.
This control established that at least some of the
lanthanum uptake was attributable to non-specific
binding, or to precipitation on the cells. Low-activity cultures of strain N14 removed more than
85% of the presented La 3+ and 2 5 - 5 0 % of the
UO22+ after 18-24 h. None of the other entero-
Results
The amino acid sequence of five peptide fragments of purified Citrobacter sp. N14 acid phosphatase was c o m p a r e d with known polypeptide
sequences using the ' S W E E P ' program [14] to
search the O W L composite database, release 17.0
[14]. Highly significant alignments were seen between two of the fragments (1 and 3) with the
phoN product of Morganella morganii, Providencia stuartii [16] and Salmonella typhimurium [17]
(Fig. 1). These strong similarities indicate the
Citrobacter enzyme to be a class PhoN phosphatase; the low similarities of fragments 2, 4 and
5 are consistent with the expected evolutionary
divergence among these enzymes. The PhoN protein of S. typhimurium was identified as a nonspecific acid phosphatase of subunit molecular
mass of 26 k D a [17,18], which is similar to the the
subunit molecular mass of the Citrobacter phosphatase h o m o t e t r a m e r (27 kDa; [6]).
c i t r o b a c t e r sp. NI4 [FI]
M. morganii
P. stuartii
S. typhimurium
Consensus
C i t r o b a c t e r sp. NI4 [F2)
M. morganii
P. stuartii
S. typhimuriu2n
Consensus
c i t r o b a c r e r sp. NI4 IF4]
M. morganii
P. stuartii
S. typhimurium
Consensus
AR~'PDFY%q~KEAQSXDSLSIdSPPPPAVDSIDI'U~D
25 G N D A % ~ g K P D L ~ E Q A X D S L K I , L p p I ~ E U G S Z Q I m . A I D
63
25 GNDV'I"~KPDLY'~51~SQAIDSIJ~Id~PPPPEVGSILI~ID 63
15 A K Y T S A E T V Q P F H S P E E ~ / N S Q F Y L P P P P G N D D P A Y R Y D
53
............
S---LPPPP ........ D
LVGAAZVAI~
185 A N Q D A I L E R 193
185 E N Q D K I L K R 193
175 E R A Q E L A ~
183
.....
R
DVTTTPDFYYLK
225 T L H S D P A I Q A Q L
225 T L H S N ~ E F Q K Q L
215 R L Q T I P A F Q K S L
.... P-F . .
VICGYHW0SD%'TAG
[F3)
202 V I C G ~ S ~
215
202 VICGYHW0SD~DAA215
192 V X C G A H W O B D V D A O 205
V I C G -H W Q S D V - A
[F5]
236
236
226
. .
.
.
QAYNDAQIYGD
242 K E D H P K L N Y
. . . .
250
Fig. 1. Amino acid sequences were determined as described in
the text, and compared to the OWL database using 'SWEEP'
[14]; OWL file names for M. morganii, P. stuartii and S.
typhimurium were S19187, PHON-PROST and PHONSALTY, respectively. The homologous sequences found, the
PhoN products of M. morganii, P. stuartff and S. typhimurium,
were aligned with PILEUP [15], and the most likely probable
match for each of the peptide fragments to the alignment was
determined with PROFILEGAP [15]. Homologous residues
(residues identical between the Citrobacter and other sequences) are in boldface and positional identity among all
four sequences is shown in the consensus lines.
144
b a c t e r i a gave c o m p a r a b l e m e t a l r e m o v a l . In general, a l t h o u g h p h o s p h a t a s e activity was g e n e r a l l y
only a b o u t 3 0 - 5 0 % of th e lowest activity s h o w n
by strain N14, significant r e m o v a l o f l a n t h a n u m
was given only by P. rettgeri a n d S. typhimurium
(strain 11027). T h e l a t t e r illustrates t h a t m e t a l
r e m o v a l is strain-specific. I n d e e d , S. typhimurium
strain M206 gave n o L a 3 + r e m o v a l b u t r e m o v e d
5 - 6 % o f t h e U O 2+. Kl. pneumoniae gave a simi-
lar result. O v e r a l l P. rettgeri s h o w e d p r o m i s e ,
r e m o v i n g b o t h metals, a l t h o u g h uranyl r e m o v a l
was similarly low ( T a b l e 1).
C o l u m n s c o n t a i n i n g p o l y a c r y l a m i d e gel-imm o b i l i z e d cells w e r e t e s t e d at flow r at es o f 15 to
230 m l / h with c h a l l e n g e solution s u p p l e m e n t e d
with 1 m M uranyl ion. E v e n at t h e slowest flow
rate (15 m l / h ) b r e a k t h r o u g h o f UO22÷ was alm o s t i m m e d i a t e with P. rettgeri-supplemented
Table 1
Phosphatase activity and metal uptake by Citrobacter sp. N14 and other enterobacteria
Organism
Batch
Phosphatase specific activity
(nmol product/min/mg protein)
Percentage of challenge metal removed
La (6 h)
La (18 h)
U (18-24 h)
Citrobacter sp.
I
II
III
IV
V
65.6
60.4
550.O
596.0
938.7
NT
NT
26
48
100
100
85
92
NT
NT
NT
N14
100
96
97
98
Citrobacter sp.
I
1.7
0
NT
0
lp4a
II
2.0
14
NT
NT
IC1.pneumoniae
I
II
III
IV
V
VI
9.6 (30°C)
49.3 (37°C)
29.6 (30°C)
37.6 (37°C)
51.9 (30°C)
20.8 (37°C)
NT
NT
NT
NT
NT
NT
0
0
6
0
0
4
7
7
7
8
3
4
I
II
25.6
24.9
NT
NT
72
76
5
4
I
II
26.7
29.7
NT
NT
NT
0
2
1
8.1
NCIMB 8066
P. rettgeri
NCIMB 10457
M. rnorganii
NCIMB 10466
S. typhimurium
M206
thax 1
II
III
IV
V
VI
10.0
9.5
6.O
2.6
9.5
NT
NT
NT
NT
NT
NT
NT
0
0
0
0
0
4
14
5
0
5
I
21.7
20.7
NT
NT
0
2
0
1
7.6
14.6
NT
NT
1
1
II
0
1
I
II
24.6
35.9
NT
NT
86
93
2
1
I
II
LT7
11027
I
6
The micro-organisms were grown to the mid-logarithmic phase as described in Materials and Methods. Samples were assayed for
phosphatase activity or challenged with lanthanum or uranyl solution for times as shown. Metal removal is expressed as percentage
of the initial metal removed. NT: not tested.
145
columns, confirming negligible retention. In accordance with previous results [2,20,21] in excess
of 90% of the input UO 2+ was removed by the
Citrobacter-supplemented columns, at flow rates
up to 200 m l / h (10 columns volumes/h).
Discussion
A survey of some enterobacteria having acid
phosphatase activity has given no evidence for a
role for the enzyme alone in heavy metal resistance via phosphatase-mediated metal bioaccumulation. Although some lanthanum uptake was
observed by several strains, the significance of
this is questionable, being observed also with the
phosphatase-deficient mutant Citrobacter sp.
lp4a; indeed, other experiments using the chemically similar yttrium have shown that yttrium removal was largely attributable to rapid precipitation on the cells (K. Bonthrone and L.E.
Macaskie, unpublished). In contrast to lanthanum
the uranyl ion is highly toxic to both cell growth
[19] and phosphatase activity [20], although phosphate production by the enzyme confers a resistance effect [20]. No evidence for extensive UO 2+
removal was seen with any strain other than the
metal-accumulating Citrobacter sp. although several strains did produce significant levels of phosphatase. The presence of phoN was not sought in
the strains tested; indeed, this may not be a
prerequisite for acid phosphatase activity (see
above) but was taken as an indication of which
enterobacterial genera might display metal bioaccumulation. Acid phosphatase activity was found
in all the strains examined (Table 1), while other
studies have shown that of 150 Salmonella strains,
representing 28 serotypes, 98% had phosphatase
activity [16].
Evidence obtained from other studies has also
suggested that the presence of acid phosphatase
per se is not sufficient to confer metal bioaccumulation properties [21]; current evidence indicates that it is the juxtaposition of phosphate
release to suitable nucleation sites on the cell
surface which permits initiation of uptake
(nucleation) prior to sustained metal phosphate
crystal growth [21]. These findings suggest that
the metal accumulating Citrobacter sp., if not
unique, is highly unusual; the appropriate combination of phosphatase overproduction and cell
surface architecture (nucleation sites) may have
been selected during prolonged stress in the
metal-polluted environment.
It is worth noting that no phoN sequences
have been found in Citrobacter freundii [16];
Groisman et al. [16] point out that Salmonella
and related enterobacteria (Escherichia, Citrobacter) have a high DNA G + C content, while
Morganella and Providencia are of low G + C
content and were proposed to have split from the
Salmonella and Citrobacter groups early in the
evolution of the enterobacteriaceae. The phoN
sequences are of low G + C content and it was
proposed that phoN was passed by lateral transfer early in the evolutionary sequence, and retained chromosomally [16]. By this argument, the
Citrobacter strain N14 is atypical of the genus;
indeed, attempts to classify the organism further
in the laboratory using the API 20E system (L.E.
Macaskie, unpublished), and commercially (Torry
Research Station, Aberdeen, UK) have failed to
resolve the strain identity to beyond genus level.
Acknowledgements
The comparative evaluation of Citrobacter sp.
N14 and Providencia rettgeri for the removal of
uranyl ion was supported by a grant from the
European Community (No. BE-5350). DAR was
supported by MRC grant G9025236CB to N.L.
Brown. The authors are indebted to B.C. Jeong
and A. Willis (Dept. of Biochemistry, University
of Oxford) for the provision of purified phosphatase and amino acid sequencing, respectively.
For software support they thank staff of both the
University of Birmingham, Academic Computing
Service, and the Science and Engineering Research Council SEQNET facility at Daresbury.
References
1 Macaskie, L.E. and Dean, A.C.R. (1982) Cadmium removal by micro-organisms.Environ.Technol. Lett. 3, 4956.
146
2 Macaskie, L.E. (1990) An immobilized cell bioprocess for
the removal of heavy metals from aqueous flows. J. Chem.
Technol. Biotechnol. 49, 357-379.
3 Macaskie, L.E., Dean, A.C.R., Cheetham, A.K., Jakeman,
R.J.B. and Skarnulis, A.J. (1987). Cadmium accumulation
by a Citrobacter sp.: the chemical nature of the accumulated metal precipitate and its location on the bacterial
cells. J. Gen. Microbiol. 133, 539-544.
4 Macaskie, L.E., Empson, R.M., Cheetham, A.K., Grey,
C.P. and Skarnulis, A.J. (1992) Uranium bioaccumulation
by a Citrobacter sp. as a result of enzymically-mediated
growth of polycrystalline HUO2PO 4. Science 257, 782-784.
5 Hambling, S.G., Macaskie, L.E. and Dean, A.C.R. (1987)
Phosphatase synthesis in a Citrobacter sp. growing in continuous culture J. Gen. Microbiol. 133, 2743-2749.
6 Jeong, B.C. (1992) Studies on the atypical phosphatase of
a metal accumulating Citrobacter sp. D. Phil thesis, University of Oxford, U.K.
7 Macaskie, L.E. and Dean, A.C.R. (1984) Cadmium accumulation by a Citrobacter sp. J. Gen. Microbiol. 130,
53-62.
8 Aickin, R.M., Dean, A.C.R., Cheetham, A.K. and Skarnulls, A.J. (1979) Electron microscope studies on the uptake of lead by a Citrobacter sp. Microbios Lett. 9, 7-15.
9 Owen, S.J., Jeong, B.C., Poole, P.S. and Macaskie, L.E.
(1992) Tributyl phosphate degradation by immobilized cells
of a Citrobacter sp. Appl. Biochem. Biotechnol. 34/35,
693-707.
10 Bolton, P.G. and Dean, A.C.R. (1972) Phosphatase synthesis in Klebsiella (Aerobacter) aerogenes. Biochem. J.
127, 87-96.
11 Macaskie, L.E., Blackmore, J.D. and Empson, R.M. (1988)
Phosphatase overproduction and enhanced uranium accumulation by a stable mutant of a Citrobacter sp. isolated
by a novel method. FEMS Microbiol. Lett. 55, 157-162.
12 Tolley, M.R., Strachan, L.F. and Macaskie, L.E. (1994)
Lanthanum accumulation from acidic solutions using a
13
14
15
16
17
18
19
20
21
Citrobacter sp. immobilized in a flow-through bioreactor.
J. Ind. Microbiol. (In press).
Fritz, J.S. and Bradford, E.C. (1958) Detection of thorium
and uranium. Analyt. Chem. 30, 1021-1022.
Akrigg, D., Bleasby, A.J., Dix, N.I.M., Findlay, J.B.C.,
North, A.C.T., Parry-Smith, D., Wooton, J.C., Blundell,
T.L., Gardner, S.P., Hayes, F., Islam, S., Sternberg, M.J.E.,
Thornton, J.M., Tickle, l.J. and Murray-Rust, P. (1988) A
protein sequence/structure database. Nature 335, 745746.
Devereaux, J., Hasberli, P. and Smithies, O. (1984) A
comprehensive set of sequence analysis programs for the
VAX. Nucl. Acids Res. 12, 387-395.
Groisman, E.A., Saier, Jr., M.H. and Ochman, H. (1992)
Horizontal transfer of a phosphatase gene as evidence for
mosaic structure of the Salmonella genome. EMBO J. 11,
1309-1316.
Kasahara, M., Nakata, A. and Shinagawa, H. (1991)
Molecular analysis of the Salmonella typhimurium phoN
gene which encodes a nonspecific acid phosphatase. J.
Bacteriol. 173, 6760-6765.
Kier, L.D., Weppelman, R.M. and Ames, B.N. (1977)
Resolution and purification of three periplasmic phosphatases in Salmonella typhimurium. J. Bacteriol. 130,
399-410.
Plummet, E. and Macaskie, L.E. (1990) Actinide and
lanthanum toxicity towards a Citrobacter sp.: uptake of
lanthanum and a strategy for the biological treatment of
liquid wastes containing plutonium. Bull. Environ. Contam. Toxicol. 44, 173-180.
Tolley, M.R. (1993) The biological treatment of liquid
wastes containing heavy metals D. Phil thesis, University
of Oxford, U.K.
Macaskie, L.E., Jeong, B.C. and Tolley, M.R. (1994) Enzymically-accelerated biomineralization of heavy metals:
application to the removal of americium and plutonium
from aqueous flows. FEMS Microbiol. Rev., 14, (in press).