Archives of

Arch Microbiol (1988) 150 : 523- 528

Microbiology
9 Springer-Verlag1988

Characterization of a pyoverdine-deficient mutant

of Pseudomonasfluorescens
impaired in the secretion of extracellular lipase*
L. Fernandez 1, C. San Jos6 l, H. Cholette 2, and R. C. McKellar 2
1 Departamento de Tecnologia y Bioquimica de los Alimentos, Facultad de Veterinaria, Universidad Complutense, Ciudad Universitaria,
E-28040 Madrid, Spain
2 Food Research Centre, Research Branch, Agriculture Canada, Ottawa K1A OC6, Canada

Abstract. A mutant of Pseudomonas fluorescens strain B52 1987). Proteinase and lipase secretion were stimulated by
deficient in the synthesis of the fluorescent pigment, the addition of pyoverdine, and this effect was attributed to
pyoverdine, was isolated. Absence of pyoverdine and other the ability of pyoverdine to chelate iron(III) and thus reduce
siderophores was confirmed by gel filtration, a specific the repressive effects of free iron(III). The same study also
siderophore assay, and inhibition studies with the iron suggested that lipase secretion was more sensitive to iron(III)
chelator EDDA. Both parent and mutant synthesized ad- than was proteinase. In the present study, we use a mutant
ditional outer membrane proteins in response to iron-limi- of P. fluorescens deficient in pyoverdine production to dem-
tation. Mutant cells cultured in the absence of iron(III) onstrate that lipase secretion is dependent on the presence
accumulated SSFe-labeled pyoverdine. The mutant prod- of an iron-chelating siderophore.
uced extracellular proteinase normally on various media,
but was deficient in lipase secretion. Growth of the mutant Materials and methods
with partially-purified pyoverdine resulted in a 2.5-fold
Strains and culture conditions
stimulation of lipase secretion. The mutant grew poorly
in deferrated medium; however, the addition of iron(III) Pseudomonas fluorescens B52 was kindly provided by Dr.
stimulated growth. Proteinase secretion in deferrated me- B. C. Richardson. Strain A21 was a pyoverdine-deficient
dium was stimulated over a narrow range of iron(III) con- mutant isolated from strain B52. Strain 32A and a pro-
centration, while lipase secretion was only slightly affected. teinase- and lipase-deficient mutant (RM 14) derived from
The data suggest that separate regulatory mechanisms exist it were described previously (Torrie et al. 1983). Strains were
for the control of proteinase and lipase secretion by iron(III). maintained on nutrient agar (Difco, Madison, WI, USA) at
5~ and transferred monthly.
Key words: Pseudomonas fluorescens - Proteinase - Li- Cells were routinely cultured from a 1.5% inoculum in
pase - Pyoverdine-deficient mutant - Iron - Repression Erlenmeyer flasks containing one-fifth volume of pyruvate
mineral salts medium plus l mM CaCI2 (PMSTCa;
McKellar and Cholette 1985). Flasks were shaken at
180 rpm at 20~ for 2 4 - 4 8 h in a gyrotory waterbath
Pseudomonas fluorescens produces a fluorescent pigment, shaker (New Brunswick Scientific, New Brunswick, N J,
pyoverdine (Meyer and Abdallah 1978). This pigment has USA, Model G86).
been identified as an iron-chelating siderophore, and has
been implicated in a high-affinity transport system for
Mutant isolation
iron(III) functioning under conditions of iron stress (Meyer
and Hornsperger 1978). Other Pseudomonas spp. produce Strain B52 was grown for 18 h on PMSTCa, washed and
fluorescent pigments (Neilands 1983) which permit growth resuspended in the same medium to an optical density at
at low iron concentrations. The requirement for iron- 600 nm (A600) of 0.42 as determined on a Bausch and
chelating pigments has been confirmed by mutant studies; Lomb Spectronic 21 spectrophotometer (Fisher Scientific,
pigment-deficient strains fail to grow in medium containing Pittsburg, PA, USA). One tenth millilitre ethylmethane
chelated iron (Ankenbauer et al. 1986). In other strains, sulfonate (EMS; Sigma Chemical Co., St Louis, MO, USA)
non-fluorescent pigments such as nocardamine (Meyer and was added to 5 ml of cell suspension, and incubated at
Abdallah 1980) and ferribactin (Philson and Llinas 1982) 37~ for I h. Treated cells were washed in PMS~Ca and
serve as the siderophores. inoculated into PMSvCa containing 0.1% each of yeast ex-
In a previous study from our laboratory, a relationship tract and casamino acids. After incubation at 20 ~C for 24 h,
between extracellular enzyme secretion and pyoverdine pro- cells were diluted in fresh medium to 100 cfu/ml and 0.5 ml
duction in P. fluorescens was established (McKellar et al. plated on PMSTCa agar containing 10 mM asparagine.
Plates were incubated at 30~ (a temperature at which P.
fluorescens produces limited pyoverdine) and several non-
* Contribution No. 768 from the Food Research Centre
fluorescent colonies were selected. One of these (designated
Offprint requests to: R. C. MeKellar P. fluorescens A21) was selected for further study.
524

Purification of pyoverdine 50 ml of 50 mM BES pH 7.0. The cell pellet was resuspended

in the original volume of BES and used to inoculate (1%
Pyoverdine was purified by two passages through a Biogel
v/v) 10 ml volumes of PMS7Ca in 50 ml Erlenmeyer flasks.
P2 column (Biorad Labs, Richmond, CA, USA) as described
Flasks were shaken at 180 rpm and at prescribed intervals,
previously (McKellar et al. 1987). Freeze-dried culture fluid
0.7 ml samples were taken, centrifuged at 12,000xg for
from strain A21 was also fractionated with Biogel P2 after
3 rain and the supernatant fractions were kept as the source
being reconstituted in water containing 1 mM FeC13.
of enzyme. In some experiments, PMSvCa was replaced by
Pyoverdine was determined by measuring the A4oo of suit-
Nutrient Broth containing 1 mM CaC12 (NBCa), Tryptic
able dilutions of column fractions prepared in 50 mM N,N-
Soy Broth (TSB), Brain Heart Infusion (BHI) and Plate
his [2-hydroxyethyl]-2-aminoethane sulfonic acid (BES;
Count Broth (PCB), all obtained from Difco, Detroit, MI,
Sigma Chemical Co.) buffer pH 7 (eM = 16500; Meyer and
USA.
Abdallah 1978).

Preparation of labeledpyoverdine Transport of Fe(III)-pyoverdine complex

The 55Fe(III)-pyoverdine complex was prepared by mixing Cells were grown in PMSTCa for 48 h as described pre-
viously. These cells were washed twice with deferrated
1.46 gmol of filter-sterilized pyoverdine with an equal amount
of SSFeC13 in 0.1 N HC1 (31 mCi/mg; New England Nu- PMSTCa containing 100 tag/ml chloramphenicol and resus-
clear, Boston, MA, USA). Unlabeled FeC13 was then added pended in the same media to an A60o of 1.0. Ten millilitres
of cell suspension were shaken at 180 rpm at 20~ To
to the mixture to assure that all the pyoverdine was
these cells, 1.5 nmol of 55Fe(iIi)_pyoverdine were added and
complexed with Fe(III). The mixture was loaded on a
2.5 • 30 cm Biogel P2 column previously equilibrated with samples (0.5 ml) were periodically withdrawn and rapidly
filtered through nylon membrane filters (0.45 gin; MSI)
0.1 M pyridine/acetate buffer pH 6.5 and eluted with the
under vacuum. The filters were washed twice with 5 ml of
same buffer. The material absorbing at 450 nm was pooled
deferrated PMSvCa containing 100 gg/ml chloramphenicol
and concentrated with a rotary evaporator.
and counted in 5 ml of scintillation fluid (Scinti Verse I;
Fisher Scientific Co.) in a Beckman Scintillation Counter
Extraction of iron Model LS 3801 (Beckman Instruments, Palo Alto, CA,
Iron was extracted from PMSTCa as described previously USA).
(McKellar et al. 1987).
Assays
Isolation of cytoplasmic membranes Proteinase. Activity was determined using the hide powder
azure method as described by McKellar and Cholette (1985).
Cells were grown in one-litre Erlenmeyer flasks containing
Activity was expressed as HPA units/ml cell-free culture
200 ml of PMSvCa with shaking at 180 rpm at 20~ for 3
fluid, where one HPA unit was the amount of enzyme re-
days. When required, FeC13 was added to a final concen-
quired to produce an increase in absorbance at 595 nm of
tration of 20 gM. The isolation of the inner and outer mem-
1.0 per h at 35~
branes was carried out as described by Mizuno and
Kageyama (1978). Inner and outer membranes were separ-
ated on discontinuous sucrose gradient (55-40%), and
Lipase. Activity was determined as described by McKellar
(1986), and was expressed as gmol fl-naphthol released/ml
15.15 gl fractions were collected.
enzyme.
In order to show that outer membrane preparations were
free from contamination with inner membranes, lysis of
spheroplasts during stripping of outer membranes was
Siderophores. The presence of siderophore in cell-free culture
fluid was assayed by the method of Schwyn and Neilands
monitored by assaying for the intracellular enzyme, glucose-
(1986). Strains B52 and A21 were grown in PMSTCa. Dupli-
6-phosphate dehydrogenase using a test kit (Sigma Chemical
cate flasks containing 20 ~M of FeC13 were used as controls.
Co.). Samples (0.5 ml) of the culture supernatants were mixed with
an equal volume of CAS shuttle solution which contained
SDS Polyacrylamide electrophoresis 0.6raM H D T M A (hexadecyltrimethylammonium bro-
SDS-PAGE of membrane proteins was performed using the mide), 0.015 mM Fe(III), 0.15 mM CAS (Chrome azurol S),
PhastSystem T M (Pharmacia, Uppsala, Sweden). Gradients 4.3% (w/v) anhydrous piperazine (adjusted to pH 5.6 with
of 1 0 - 1 5 % (w/v) acrylamide were run with Phase gel SDS 12 N HC1), and 4 mM sulfosalicylic acid. When required, the
buffer strips. Samples were solubilized in 10 mM Tris/HC1 supernatant fractions were diluted to adjust the siderophore
(pH 8.0), 1 mM EDTA, 2.5% SDS and 5% 2-mercaptoetha- concentration to less than 15 ktM. After 20 min the A63o was
nol at 100~ for 5 rain. Electrophoresis was carried out at measured.
15~C with a current of 10 mA per gel. The gels were stained
with Coomassie Brilliant Blue as described in the
PhastSystem Manual. Low molecular weight standards were Results
obtained from Pharmacia.
A non-fluorescent mutant (A21) of strain B52 was isolated
after treatment with EMS. Cell-free culture fluid from A21
Growth experiments obtained after growth on PMSTCa, an iron-limiting me-
Cells cultured as described above for 48 h were collected by dium, was fractionated on Biogel P2. A small peak absorbing
centrifugation at 5,000 x g for 10 rain, and washed with at 400 nm was noted; this peak corresponded to the
 525

pyoverdine peak found with the parent strain B52 (Fig. 1). E D D A inhibited the growth of siderophore-deficient mu-
This p e a k did not a p p e a r to be a siderophore, since an assay tants of Pseudomonas aeruginosa ( A n k e n b a u e r et al. 1986).
specific for siderophores failed to reveal the presence of these We examined the effects o f this iron chelator on the ability
c o m p o u n d s in the culture fluid o f strain A21 (data not o f strains A21 and B52 to grow and produce extracellular
shown). proteinase and lipase. Strain B52 grew normally in up to
125 g M E D D A , and suffered no loss of ability to produce
extracellular enzymes (Table 1). In contrast, growth of strain
A21 was inhibited by the chelator; m a r k e d loss o f growth
was observed at 25 g M and complete inhibition was noted
at 125 g M (Table i). Strain A21 continued to secrete pro-
teinase at up to 50 ~tM E D D A . Lipase secretion by A21 was
stimulated by E D D A (Table 1).
1.0- The ability o f the two strains to produce extracellular
3
2 enzymes in various m e d i a was c o m p a r e d (Table 2). Protein-
ase secretion by strain B52 in TSB was superior to that
in PMS7Ca, while secretion in N B C a and BHI was p o o r
.5-
c o m p a r e d with mineral media. These findings are in agree-
ment with earlier studies ( M c K e l l a r 1982). Strain A21 p r o d -
uced 50% more proteinase on P M S v C a than did B52
(Table 2), and on other media was as effective as B52 in
producing proteinase. These results suggest that loss o f
k i ,
10 15 20 25 pyoverdine had little effect on the ability to secrete protein-
TUBE ase. Extracellular lipase secretion by B52 on complex media
Fig. 1. Fractionation of cell-free culture fluid from B52 ( 9 and was significantly lower than that found on the mineral me-
A21 ( 0 ) on Biogel P2 dium (Table 2). Lipase secretion by A21 on complex media

Table 1. Effect of EDDA on growth, proteinase and lipase secretion by a pyoverdine-deficient mutant (A21) of Pseudomonasfluorescens
B52

EDDA a B52 A21

(pM)
Growth P.A. b L.A. c Growth P.A. b L.A. c
(A6oo) (A6oo)

0 1.58 17.9 58.2 1.22 18.2 17.4

"25 1.45 15.0 68.9 0.226 8.30 35.9
50 1.64 16.0 69.7 0.015 12.7 64.4
125 1.55 16.5 59.4 0 0 0

a Cells were grown on pyruvate mineral salts medium containing 1 mM CaC12 (PMSvCa) in the presence of EDDA for 48 h at 20~
b Proteinase activity expressed as HPA unit/ml - A 6 o o
c Lipase activity expressed as gmol/ml - A 6 o 0

Table 2. Growth, proteinase and lipase secretion by B52 and A21 on various media

Medium ~ Growth b Proteinase c Lipase a

(A6oo)
Activity % Control Activity % Control
B52
PMS7Ca 1.14 64.3 100 74.6 100
NBCa 1.08 12.0 18.7 18.0 23.4
TSB 2.18 79.3 123.0 8.29 10.8
BHI 2.18 28.3 44.0 6.00 7.80
PCB 2.10 67.3 105.0 11.0 14.4
A21
PMSTCa 0.768 98.3 153.0 5.61 7.70
NBCa 1.06 17.0 25.5 7.90 10.6
TSB 2.18 59.8 89.3 3.38 4.88
BHI 2.18 37.5 56.1 2.30 3.32
PCB 2.02 64.3 96.1 10.60 15.1

Growth media were: pyruvate mineral salts with 1.0 mM CaClz (PMSvCa); Nutrient Broth with 1.0 raM CaCI~ (NBCa); Tryptic Soy
Broth (TSB); Brain Heart Infusion (BHI); Plate Count Broth (PCB)
u Growth was determined after 48 h at 20 ~C
c Proteinase activity was expressed as HPA units/ml. A6oo; control was activity of B52 on PMSvCa
d Lipase activity was expressed as gmol/ml. A6oo; control was activity of B52 on PMSvCa
 526

1.2 A
1.28 ~ ~
1.1

A
E
1.0
.9
1,21" ~ ~
c
0.96 120 30
.8 ==
<
.7
0.80" 9 100 25 ~:
"1-
I-
.6
.5 0.64 ~0 20
0
nr
.4
0.48 9 9 50 15 '<
z
.3 u.I

0.32- 40 10
S
.2
.1
0.16- 20
0 i i = i

140 B 8 1~2 1'6 20

Fe (111) (#rn)

Fig. 3. Effect of iron concentration on growth ( 9 proteinase ( 9

120,
and lipase ( I ) secretion by A21 in deferrated PMS7Ca. Growth
and enzyme activities were determined after 48 h at 20~
0
E 100 -
>- was further reduced compared with the wild type strain.
I-
80- Most significantly, lipase secretion by A21 on PMS7Ca was
I- only 7.7% of that found with B52 (Table 2).
tj
The ability of B52 and A21 to secrete extracellular en-
LU 60- zymes was also examined as a function of growth. The
,,r results presented in Fig. 2 show that both strains grew well in
a,,
,-i 40- PMS7Ca at 20~ Initially, the mutant appeared to grow
more rapidly than the parent; however, this was due to
extensive aggregation of B52 cells to the walls of the culture
20- flasks during the logarithmic growth phase. After 48 h of
growth, B52 cell density exceeded that of A21. Proteinase
0
secretion by the two strains was in proportion to growth;
~, i i ! i
A21 produced 40% more enzyme after 48 h than did the

/
140- C parent strain. Lipase secretion by A21 was limited after 25 h;
however, the parent strain continued to secrete this enzyme
over the entire period of the experiment.
E 120 -
The results suggest that lipase secretion by A21 was
C
impaired, possibly due to the absence of pyoverdine. This
".1'- 100 - was confirmed in studies in which growth and extracellular
enzyme secretion were determined in the presence and ab-
>-
I--
sence of partially purified pyoverdine. Data presented in
80- Table 3 show that growth, proteinase and lipase secretion
I-- by B52 on PMSTCa was unaffected by the addition of 10 gM
,< pyoverdine. In contrast, growth of A21 on PMSTCa was
ua 60-
stimulated 35% by the inclusion of the siderophore. Ad-
,<

_j
z
dition of the siderophore had little effect on proteinase se-
ua 40 cretion by A21 ; however, lipase secretion increased 2.7-fold
I-- on PMSTCa in the presence of added pyoverdine.
O
t~ Growth, proteinase and lipase secretion by A21 was also
e~
20 determined in PMSTCa treated with 8-hydroxyquinoline to
remove iron. Maximum growth was achieved when the me-
dium was supplemented with 5 pM iron(III) (Fig. 3). A nar-
- ' 10 '
20 30' 4'o row optimum for proteinase secretion was found at 2 gM
iron(III), and activity decreased with increased iron(III) con-
T I M E (h) centration to a minimum at 15 gM.
In contrast to these findings, increasing the iron(III)
Fig. 2 A - C. Growth (A), lipase (B), and proteinase (C) secretion concentration in the growth medium had little stimulatory
by A21 ( 9 and B52 ( 9 on PMSTCa effect on lipase secretion (Fig. 3). A broad optimum was
 527

Table 3. Effect ofpyoverdine addition on growth, proteinase and lipase secretion by B52 and A21 on PMS7Ca

Strain Pyoverdine Growth" Proteinase u Lipase ~

(10 gM) (A6oo)
Activity % Control Activity % Control

B52 - 1.17 42.5 100 91.5 100

+ 1.22 44.3 t 04.4 82.3 90.4
A21 - 0.84 58.0 100 20.1 100
+ 1.13 69.5 119.8 54.5 271.1

a Growth was determined after 48 h at 20~C

b Proteinase activity was expressed as HPA unit/ml " A6o0
c Lipase activity was expressed as p,mol/ml - A600

3.0-

2.5-

~2
E 2.0

1.5-

ua
>
o 1,0-

O
0.5

,; 2; 3; 4b ~'0 ~;
TIME (rain)

Fig. 5. Accumulation of 5SFe-labeled pyoverdine by A21 after

growth in the presence (O) and absence ( 0 ) of 20 gM added
iron(III)

Fig. 4A, B. Outer (A) and inner (B) membrane profiles of B52, A21,
32A, and RM14 on SDS-PAGE. Lane 1, standard proteins (from the 30 k D a protein (band B) from the outer membrane after
bottom); phosphorylase, 94 kDa; bovine serum albumin, 67 kDa; growth in the either the presence or absence of added
carbonic anhydrase, 30 kDa; soybean trypsin inhibitor, 20.1 kDa; iron(III).
c~-lactalbumin, 14 kDa; Lanes 2 and 3, B52; Lanes 4 and 5, A21 ; Pyoverdine-mediated iron uptake by A21 grown in the
Lanes 6 and 7, 32A; Lanes 8 and 9, RM14; Lanes 2, 4, 6, 8, growth presence and absence iron(III) was examined. Figure 5
with 20 ~M iron(III); Lanes 3, 5, 7, 9, growth with no added iron(III) shows that A21 cells grown without added iron(III) accumu-
lated the iron(III)-pyoverdine complex much more rapidly
than cells grown with excess iron, suggesting that the specific
transport system for the iron(III)-pyoverdine complex was
obtained at between 5 and 10 gM concentrations of the
present.
cation, and the sharp peak of activity observed with the
proteinase was not apparent.
S D S - P A G E profiles of the inner and outer cytoplasmic
membranes isolated from cells growing in the presence and Discussion
absence o f iron(III) are shown in Fig. 4. Growth o f B52 The use of mutants deficient in the secretion of extracellular
without added iron(III) resulted in the loss of a 20 k D a products has facilitated the understanding of processes in-
protein (band A) and the appearance of a 90 k D a protein volved in their regulation (Ankenbauer et al. 1986; Sokol et
(band C) in the outer membrane. Several inner membrane al. 1982). In the present study, a mutant (A21) impaired in
proteins (bands D - H ) were also induced by growth under the production of pyoverdine was isolated from Pseudo-
limited iron(III). A21 produced essentially the same profile monasfluorescens B52. On a variety of complex and defined
as B52. For comparison purposes, membranes from strain media, growth and proteinase secretion by the mutant were
32A and its proteinase-and lipase-deficient mutant (RM14) similar to that of the parent; however, lipase secretion was
were also examined. Figure 4 shows that RM14 had lost a reduced.
528

The suggestion that the loss oflipase in A21 is an indirect it is impaired in the synthesis of one or more enzymes re-
effect resulting from increased repression by free iron(III) quired for pyoverdine synthesis or secretion.
rather than iron limitation is supported by several pieces of
evidence: (1) Growth in the presence of pyoverdine resulted
References
in recovery of lipase secretion, while iron(III) addition failed
to restore activity; (2) Lipase secretion was reduced in sev- Ankenbauer R, Hanne LF, Cox CD (1986) Mapping of mutations
eral complex media which would not likely provide iron- in Pseudomonas aeruginosa defective in pyoverdin production.
limiting conditions; (3) In a previous study (McKellar et al. J Bacteriol 167: 7 - It
1987), lipase secretion by the parent strain was inhibited Bagg R, Neilands JB (1987) Molecular mechanism of regulation of
siderophore-mediated iron assimilation. Microbiol Revs 51:
during the early stages of secretion by added iron(III), while
509--518
pyoverdine effectively prevented this inhibition. McKellar RC (1982) Factors influencing the production of extra-
Bagg and Neilands (1987) have established that sidero- cellular proteinase by Pseudomonasfluorescens. J Appl Bacteriol
phore production by Escherichia coli is controlled by the fur 53 : 305- 316
gene which codes for an iron repressor. Extracellular enzyme McKellar RC (1986) A rapid colorimetric assay for the extracellular
secretion may also be under the control of an iron repressor; lipase of Pseudomonasfluorescens B52 using fl-naphthyl capry-
Sokol et al. (1982) observed independent regulation of extra- late. J Dairy Res 53 : 117 -- 127
cellular proteinases in P. aeruginosa and postulated that McKellar RC (1988) Regulation and control of enzyme synthesis.
this microorganism produces separate repressors differing in In: McKellar RC (ed) Enzymes of psychrotrophs in raw food.
CRC Press, Boca Raton, FL (in press)
their sensitivity to iron. McKellar RC, Cholette H (1985) Inhibition by chelating agents of
An alternative mechanism linking extracellular enzyme the formation of active extracellular proteinase by Pseudomonas
secretion to intracellular energy levels has been proposed. fluorescens 32A. J Dairy Res 52:91 - 100
Whooley and McLoughlin (1982) and McLoughlin et al. McKellar RC, Shamsuzzaman K, San Jos+ C, Cholette H (1987)
(1986) have demonstrated that uncouplers of oxidative Influence of iron(III) and pyoverdine, a siderophore produced
phosphorylation and ATP levels control siderophore and by Pseudomonasfluorescens B52, on its extracellular proteinase
extracellutar proteinase secretion in Pseudomonas aerug- and lipase production. Arch Microbiol 147 : 225- 230
inosa. McKellar (1988) has discussed the possible role of McLoughlin AJ, Whooley MA, O'Callaghan JA (1986) Implication
of adenine nucleotide levels in the regulation of exoprotease
energy levels as the primary controlling factor in extracellu-
production by Pseudomonas aeruginosa. Irish J Food Sci Tech-
lar enzyme secretion by psychotrophic bacteria, and has nol 10:127-134
suggested that iron(III) availability may be involved in regu- Meyer JM, Abdallah MA (1978) The fluorescent pigment of
lation. Pseudomonasfluorescens: Biosynthesis, purification and physi-
Proteinase and pyoverdine appear to respond indepen- cochemical properties. J Gen Microbiol 107:319-328
dently, but with similar sensitivity, to iron(III). Proteinase Meyer JM, Abdallah MA (1980) The siderochromes of non-fluor-
appears to respond to total iron(III) concentration rather escent pseudomonads: Prodution of nocardamine by Pseudo-
than free iron(III), since its response to this cation is similar monas stutzeri. J Gen Microbiol 118:125-129
in the presence (B52) and absence (A21) of the siderophore Meyer JM, Hornsperger JM (1978) Role of pyoverdine, the iron-
binding fluorescent pigment of Pseudomonasfluorescens, in iron
(McKellar et al. 1987; this study). Therefore, pyoverdine
transport. J Gen Microbiol 107:329-331
and proteinase may be regulated indirectly by the influence Meyer JM, Mock M; Abdallah MA (1979) Effect of iron on the
of iron(III) on energy levels. protein composition of the outer membrane of fluorescent
Lipase appears to be regulated in a different manner. pseudomonads. FEMS Microbiol Lett 5:395-398
Lipase secretion by the parent strain was extensive, and was Mizuno T, Kageyama M (1978) Separation and characterization of
proportional to pyoverdine synthesis (McKellar et al. 1987). the outer membrane of Pseudomonas aeruginosa. J Biochem
In the mutant, however, activity was low at all iron(III) 84:179-191
concentrations. If lipase were responding to energy levels, Neilands JB (1983) Siderophores. In: Theil EC, Eichhorn GL,
iron(III)-pyoverdine and free iron(HI) would be expected to Marzilli LG (eds) Iron binding proteins without cofactors or
sulfur clusters. Elsevier, New York, pp 137-166
have a similar effect. Since lipase secretion is clearly more Philson SB, Llinas M (1982) Siderochromes from Pseudomonas
sensitive to free iron(III) than is the proteinase, it is proposed fluorescens I. Isolation and characterization. J Biol Chem
that lipase secretion is regulated primarily by a specific re- 257:8081 - 8085
pressor. Schwyn B, Neilands JB (1981) Universal chemical assay for the
Meyer et al. (1979) showed that when P. fluorescens detection and determination of siderophores. Anal Biochem
was grown under iron limitation, the synthesis of outer 160:47-56
membrane proteins was induced. Our studies have con- Sokol PA, Cox CD, Iglewski BA (1982) Pseudomonas aeruginosa
firmed this observation, and have shown that the mutant, mutants altered in their sensitivity to the effect of iron on toxin
while deficient in pyoverdine production, was still able to A or elastase yields. J Bacteriot 151 : 783 - 787
Torrie JP, Cholette H, Froehlich DA, McKellar RC (1983) Growth
synthesize outer membrane proteins and accumulate of an extracellular proteinase-deficient strain of Pseudomonas
iron(III)-pyoverdine in response to iron-limitation. Nei- fluorescens on milk and milk proteins. J Dairy Res 50:365-
lands (1983) has proposed a model in which the struc- 374
tural genes for the enzymes involved in pyoverdine secretion Whooley MA, McLoughlin AJ (1982) The regulation of pyoverdine
are coordinately regulated with the gene for the membrane production in Pseudornonas aeruginosa. Eur J Appl Microbiol
receptor. Since the mutant was able to accumulate iron(III)- Biotechnol 15 : 161 - 166
pyoverdine after growth in limiting iron(III), it does not
seem to be deficient in the membrane receptor. More likely, Received April 28, 1988/Accepted July 14, 1988