FEMS Microbiology Letters 78 (1991) 1-6
© 1991 Federation of European Microbiological Societies 0378-1097/91/$03.50
ADONIS 037810979100091U
FEMSLE 04287
Pyoverdine-mediated iron transport in Pseudomonas aeruginosa:
involvement of a high-molecular-mass outer membrane protein
Keith Poole, Shadi Neshat and David Heinrichs
Department of Microbiology and Immunology, Queen's University, Kingston, Ontario, Canada
Received 17 September 1990
Accepted 24 September 1990
Key words: Pseudomonas aeruginosa; Pyoverdine; Iron transport; Outer membrane protein
1. SUMMARY
Reduced expression of an iron-regulated outer
membrane protein (IROMP) of approximate
molecular mass 90000 was observed in Pseudomonas aeruginosa concomitant with a loss of
pyoverdine production in a wild type strain grown
at 43°C and in a mutant deficient in pyoverdine
production. Consistent with an implied role in
pyoverdine-mediated iron transport a mutant lacking the 90 kDa protein transported barely detect,
able levels of ferri-pyoverdine. Interestingly, the
mutant still exhibited pyoverdine-dependent
growth in an iron-deficient medium containing the
synthetic iron chelator ethylene diamine-di(o-hydroxyphenol acetic acid) (EDDHA) suggesting
that a second uptake system for ferri-pyoverdine
may exist in P. aeruginosa.
2. INTRODUCTION
That iron plays a critical role in infection is
now well established [1-3]. The availability of iron
Correspondence to: K. Poole, Department of Microbiology and
Immunology, Queen's University, Kingston, Ontario, Canada
K7L 3N6.
and ability of microorganisms to acquire it have
been correlated extensively with the ability to cause
and the severity of disease [1-3]. Indeed, the ironrestricted nature of the host, owing in part to the
presence of the iron binding glycoproteins transferrin (in serum) and lactoferrin (in secretions) [4],
forces invading microorganisms to compete for
iron in order to grow and, ultimately, cause infection. To acquire iron under these conditions many
successful pathogens synthesize siderophores, a
class of low-molecular-mass high-affinity iron chelators, together with the corresponding cell surface
receptors [5]. These receptors tend, with one reported exception [6], to be of high molecular mass
and are regulated by medium iron levels [7].
Pseudomonas aeruginosa, an important opportunistic pathogen [8], synthesizes two known
siderophores under iron-limiting conditions,
pyochelin [91 and pyoverdine [10]. Pyoverdine is
capable of removing transferrin-bound iron in
vitro [11], probably accounting for the ability of
pyoverdine-producing strains to grow in the presence of human serum [12]. The receptor for ferripyoverdine has yet to be unambiguously identified
although there is a report suggesting that an
IROMP of 80 kDa is the receptor (see discussion
in ref. 13). In the present report we describe the
identification of a 90 kDa iron-regulated outer
2
membrane protein (IROMP) which functions in
ferri-pyoverdine uptake.
3. MATERIALS AND METHODS
3.1. Bacterial strains and growth media
P. aeruginosa strain K376 (manuscript in preparation) is a pyochelin-deficient derivative of
pyoverdine-deficient strain PAO6609 (met-9011,
amiE200, rpsL, pod9) [14]. P. aeruginosa strain
K437 is a derivative of K376 deficient in an ironregulated 90-kDa outer membrane protein (this
study). It was obtained as a tiny colony on irondeficient succinate minimal plates containing dipyridyl (0.25 mM) and citrate (1.5 mM) following
N-methyl-N'-nitro-n-nitrosoguanidine mutagenesis [15] of K376. P. aeruginosa PAO1 is a wild-type
strain.
The iron-deficient medium used throughout
consisted of BM2 minimal medium [16] supplemented with 0.5 mM MgSO4 and either 20 mM
potassium succinate (pH 8.0) (succinate medium)
or 0.2% (w/v) glucose (glucose medium). Methionine (1 mM) and FeSO4 (50 mM) were included
as required. L-broth [17] was employed as the rich
medium. Solid media were obtained by the addition of 1.5% (w/v) Bactoagar (Difco).
3.2. Outer membrane preparation and SDS-polyacrylamide gel electrophoresis (PAGE)
Outer membranes were prepared by differential
Triton X-100 solubilization of isolated cell envelopes as described [18]. SDS-PAGE was performed as described previously [19] using 7% (w/v)
polyacrylamide in the running gel and omitting
2-mercaptoethanol from the sample loading buffer.
3.3. Pyoverdine isolation
Pyoverdine was recovered from the culture supernatants of iron-limited PAO1 cells using the
methodology described [20] with modifications.
Briefly, an overnight culture (500 ml) of PAO1 in
iron-deficient glucose medium was harvested by
centrifugation (10000 × g; 10 min) and the supernatant decanted and lyophilized. The lyophilized
material was subsequently resuspended in 12 ml
H20 and, following centrifugation to remove in-
solubles (15 000 x g; 10 min), extracted once with
an equal volume of ethyl acetate. The organic
phase was discarded and solid NaCI added to the
aqueous phase to saturation. The aqueous phase
was then extracted twice with 0.5 volume of phenol:chloroform (1 g/ml) and the organic phases
pooled. Following addition of 2 volumes of diethyl ether to the organic phase, the precipitated
pyoverdine was pelleted by centrifugation in a
Hermle model Z21 table-top centrifuge (4000 rpm;
10 min) and subsequently washed 3 times with 3
ml diethyl ether. The pellet was air-dried and
resuspended in 10 ml H20. For long term storage
of the pyoverdine 1 ml aliquots were lyophilized
and stored at - 20 ° C.
3.4. Transport assays
P. aeruginosa cells used in 55FEC13 uptake assays were grown in iron-deficient succinate
medium to an absorbance at 600 n m (A60o) of 1.0.
Five ml of cell culture were harvested by filtration
on 0.45 ~tm membrane filters, washed once with
an equal volume of iron-deficient succinate
medium and resuspended in an equal volume of
the same medium. Washed cells were then incubated with shaking for 15 min at 37°C prior to
the start of the assay. Uptake was initiated by the
addition of 20 ttl of a solution of pyoverdine (4
mg/ml), ethylene diamine di-(a-hydroxyphenyl
acetic acid) (EDDHA) (150 /~g/ml) and various
concentrations of 55FEC13 to 1 ml of cells. Aliquots
(200 #1) were removed at various time intervals,
filtered on 0.45-/~m filters (GN-6, Gelman) and
washed with 10 ml H20. Following drying, filters
were counted in scintillation fluid using the tritium channel of a LKB model 1215 Raekbeta
liquid scintillation counter (Pharmacia-LKB). In
some experiments an equal volume of H20 replaced pyoverdine or KCN (10 mM) was included.
3.5. Growth promotion experiments
Stationary phase cells of P. aeruginosa were
diluted into fresh iron-deficient succinate medium
containing EDDHA (150/~g/rnl) and pyoverdine
(80/zg/ml) to A600 = 0.1. Cultures were shaken at
37°C and cell density (measured at A6o0) determined at various time intervals. Experiments
performed in the absence of pyoverdine confirmed
the inability of the pyoverdine-deficient strains
used in this study to grow in EDDHA-containing
iron-deficient minimal medium [14].
4. RESULTS A N D DISCUSSION
Wild-type cells of P. aeruginosa synthesize a
number of high molecular mass outer membrane
proteins in response to iron limitation [21] (Fig. 1,
lane 2, cf. lane 1). Interestingly, reduced levels of
an approximately 90-kDa I R O M P were observed
in the outer membranes of cells cultured at 43 ° C
(Fig. 1, lane 3) as compared to 37 ° C (Fig. 1, lane
2), concomitant with a failure of the cells to produce pyoverdine at this elevated temperature (unpublished observation). Moreover, this protein was
also reduced in outer membranes of a mutant,
strain PAO6609 [14], unable to synthesize pyoverdine (Fig. 1, lane 4). In light of these data and
because iron-siderophore receptors are generally
high-molecular-mass IROMPs, we decided to test
the obvious hypothesis that the 90-kDa protein
functions as the ferri-pyoverdine receptor.
A mutant deficient in production of this protein (K437) (Fig. 1, lane 6; cf. parent strain, lane
5) was subsequently isolated and its capacity for
pyoverdine-dependent iron uptake examined. As
expected, the mutant showed a marked decrease in
pyoverdine-mediated iron transport as "compared
Fig. 1. SDS-polyacrylamide gel electrophoretogram of outer
membranes prepared from strains of P. aeruginosa grown in
iron-deficientsuccinate medium with (lane 1) or without (lanes
2-6) added FeSO4. Lanes: 1-3, PAO1; 4; PAO6609; 5, K376;
6, K437. Membranes were derived from cells grown at 37 °C
(lanes 1, 2, 4-6) or 43°C 0ane 3).
with the parent (Fig. 2), uptake being just detectable above the background observed in the absence of siderophore (Fig. 2). This supports our
earlier suspicion that the 90-kDa protein is indeed
involved in ferri-pyoverdine uptake, presumably
as the receptor.
Interestingly, Cornelis et al. [13] suggest that
the ferri-pyoverdine receptor is an 80 kDa IROMP,
although data showing this have yet to be published. As such, it remains to be seen whether the
two proteins are identical and the reported differences in molecular mass simply reflect difficulties in accurately sizing proteins in this molecular
mass range or whether these really are two distinct
proteins, both of which function in ferri-pyoverdine uptake.
Despite the almost total loss of ferri-pyoverdine
uptake in K437, the mutant still exhibited
pyoverdine-dependent growth in iron-deficient
succinate medium containing E D D H A (Fig. 3),
consistent with pyoverdine promoting iron transport into the mutant. One possibility is that a
second uptake system for ferri-pyoverdine exists
in P. aervginosa which is inefficient at the iron
concentrations used. Nonetheless, increasing the
iron level 5-fold in uptake assays produced no
detectable increase in pyoverdine-dependent iron
transport in the mutant, despite the fact that
pyoverdine was in vast molar excess over iron
(data not shown). Alternatively, the low level of
pyoverdine-mediated iron uptake seen in K437
(Fig. 2) may reflect a second uptake system which
is of similar affinity to the major uptake system
present in K376 but of lower capacity, and which
may be responsible for the observed growth of the
mutant in EDDHA-containing medium. Certainly, the reduced rate of pyoverdine-mediated
iron transport in K437 (as compared to K376)
(Fig. 2) is reflected in a noticeably slower rate of
pyoverdine-dependent growth in EDDHA-containing medium (Fig. 3). Growth of the mutant in
the presence of pyoverdine did not enhance
pyoverdine-dependent iron uptake (data not
shown) indicating that if pyoverdine-dependent
growth of the mutant in EDDHA-containing
media was due to a second uptake system, it was
not pyoverdine-inducible. We are currently attempting to isolate a mutant of K437 unable to
20
o
15
0
E
10
0
~
/
5
C
--II
0
1
2
Time
3
4
5
(minutes)
Fig. 2. Pyoverdine-mediated iron (55Fe3+) transport by P.
aeruginosa K376 ( • •
• ) and K437 ( •
• ) grown in
iron-deficient succinate medium. The uptake mixture contamed pyoverdine (80 /~g/ml), EDDHA (3 /tg/ml), 5~FeSOa
(110 nM) and 1 ml cells at A600 = 1.0. Transport in the absence
of added pyoverdine for both strains is indicated (Ira
II).
Data are representative of 3 separate experiments.
transport ferri-pyoverdine in order to assess
whether a second system for ferri-pyoverdine uptake does, indeed, exist in P. aeruginosa.
The mutant K437 was isolated using a selection
procedure originally devised to obtain mutants
defective in ferric citrate uptake which, perhaps,
explains its recovery on iron-deficient succinate
plates containing dipyridyl and citrate (see
MATERIALS AND M~rnoDs). Isolated as a tiny colony on this medium, the mutant was subsequently
demonstrated, in contrast to the parent strain
K376, to be incapable of growth on iron-deficient
succinate medium containing only dipyridyl (0.25
raM) (Fig. 4). In light of the demonstrated role of
the 90-kDa IROMP in ferri-pyoverdine uptake
one is forced to conclude that K376, despite the
reported pyoverdine-deficiency of PAO6609 [14]
from which it was derived (manuscript in preparation), must still be producing some pyoverdine (or
related product) which is responsible for the observed ability of K376 to grow in the presence of
dipyridyl (Fig. 4). Loss of this 'residual' iron
transport capability in the 90,kDa-protein-deficient strain K437 would explain, then, the mutant's
inability to grow in iron-deficient succinate
medium containing dipyridyl. Besides suggesting
that the pod phenotype of PAO6609 (from which
K376 and, ultimately, K437 were derived) is leaky,
the ability of K376 and failure of K437 to grow in
the presence of dipyridyl provides what may be a
1.00
c
O
O
(~
0.90
A
i
0.80
0
0
0.70
~
0
o
0.60
£0
0.50
'~
.!=
0.40
e-
0.30
~"
@
"0
0.40
0
o
0.30
I0
0.20
~
0.20
c
0.10
/ • J •
0.10
0
0.50
0.00
0
i
i
i
i
i
i
1
2
3
4
5
6
0.00
Time
(hours)
Fig. 3. Growth of P. aeruginosa strain K376 (III
It,
•
• ) and K437 (e
e, •
,
• ) in iron-deficient
succinate medium containing EDDHA (3 pg/rnl) and pyoverdine (80 pg/ml) (m
II, •
• ) or no siderophore
(-
- , ,,
,,).
0
i
i
i
i
i
i
i
t
1
2
3
4
5
8
7
8
Time
9
(hours)
Fig. 4. Growth of P. aeruginosa K376 ( e
e) mad 1(437
(• ) in iron-deficient succinate medium containing
dipyridyl (0.25 mM).
ready means for cloning the corresponding gene-by complementation of the mutant's inability to
grow on dipyridyl-containing medium. We are
currently pursuing this.
ACKNOWLEDGEMENTS
We thank J.-M. Meyer and A. Kropinski for
strains. This work was supported by an operating
grant from the Medical Research Council of
Canada.
REFERENCES
11] Kluger, M.J. (1987) in Iron and Infection: Molecular,
Physiological and Clinical Aspects (Bullen, J.J. and Griffiths, E., eds.), pp. 243-281, John Wiley and Sons Ltd.,
New York.
[21 Weinberg, E.D. (1984) Physiol. Rev. 64, 65-93.
[3] Crosa, J.H. (1984) Ann. Rev. Microbiol. 38, 69-89.
[4] Sawatzi, G. (1987) in Iron Transport in Microbes, Plants
and Animals (Winkelmann, G., van der Helm, D. and
Neilands, J.B., eds.), pp. 477-489, VCH Verlaagsgesellschaft mbH, Weinheim, F.R.G.
[5] Neilands, J.B. (1981) Annu. Rev. Nutr. 1, 27-46.
[6] Sokol, P.A. and Woods, D.E. (1983) Infect. Immun. 40,
665-669.
[7] Neilands, J.B. (1982) Annu. Rev. Microbiol. 36, 285-309.
[8] Bozenhart, K. and Ruden, H. (1987) in Basic Research
and Clinical Aspects of Pseudomonas aeruginosa (Doting,
G., Holder, I.A. and Bozenhart, K., eds.), pp. 1-15, Karger,
Basel.
19] Cox, C.D. (1980) J. Bacteriol. 142, 581-587.
[101 Cox, C.D. and Adams, P. (1985) Infect. lmmun. 48,
130-138.
[11] Sriyosachati, S. and Cox, C.D. (1986) Infect Immun. 52,
885-891.
[12] Ankenbauer, R., Sriyosachati, S. and Cox, C.D. (1985)
Infect. Immun. 49, 132-140.
[13] Cornelis, P., Hohnadel, D. and Meyer, J.-M. (1989) Infect.
Immun. 57, 3491-3497.
[141 Hohnadel, D., Haas, D. and Meyer, J.-M. (1986) FEMS
Microbiol. Lett. 36, 195-199.
[15] Goldberg, J.B. and Ohman, D.E. (1984) J. Bacteriol. 158,
1115-1121.
[16] Gilleland, H.E., Jr., Stinnet, J.D. and Eagon, R.G. (1974)
J. Bacteriol. 117, 302-311.
[17] Poole, K. and Braun, V. (1988) Infect Immun. 56, 29672971.
[18] Schnaitman, C.A. (1974) J. Bactetiol. 118, 442-453.
[19] Lugtenberg, B., Mrijers, J., Peters, R., van der Hoek, P.
and van Alphen, L. (1975) FEBS Lett. 58, 254-258.
[20] Meyer, J.-M. and Abdallah, M.A. (1978) J. Gen. Microbiol. 107, 319-328.
[211 Meyer, J.-M., Mock, M. and Abdallah, M.A. (1978) FEMS
Microbiol. Lett. 5, 395-398.