Download Hyperproduction of Exocellular Levansucrase by

Journal of' General Microhiologj- (1 984), 130, 3 143-3 152. Printed in Great Britain
3143
Hyperproduction of Exocellular Levansucrase by Bacillus subtilis:
Examination of the Phenotype of a s a d h Strain
By R E G I S C H A M B E R T * A N D M A R I E - F R A N C O I S E P E T I T - G L A T R O N
Institut Jacques Monod, CNRS- UniversitP Paris VII, Laboratoire GPnPtique et Membranes,
2 place Jussieu, 75251 Paris Cedex 05, France
(Received 18 June 1984; revised 18 July 1984)
The induction parameters of levansucrase synthesis were the same in Bacillus subtilis strain 168
Marburg and in a derivative, hyperproducing (sacUh)strain. However, only the hyperproducing
strain showed an induction lag period. The kinetics of appearance of functional levansucrase
mRNA was established. Strain 168 did not release levansucrase, but washing the cells with high
ionic strength buffer released different proteins of which levansucrase represented 2 %. In contrast, the great majority of levansucrase synthesized by the hyperproducer was released in a
homogeneous form into the culture medium. In this case high ionic strength treatment caused
the cells to release the remaining levansucrase but not other proteins. A Triton X-100 sensitive
form of levansucrase was isolated by phenol treatment of the sac@ strain; this form was absent
in strain 168. We suggest that the sacU gene product possibly controls the synthesis of cellular
components with which levansucrase is associated and thus its release is normally prevented.
INTRODUCTION
Bacillus subtilis 168 Marburg secretes a large variety of extracellular enzymes (Priest, 1977).
The best known, proteases, a-amylase, nucleases and cell lytic enzymes, are synthesized
constitutively and mainly during the later stage of the growth cycle. In contrast, the synthesis of
levansucrase is initiated in response to the presence of a specific inducer, sucrose, and occurs
only during the exponential phase of growth (Dedonder, 1966). The genetic analysis of sucrose
metabolism in B. subtilis (Lepesant et al., 1972) has led to the identification of several loci related
to the regulation of levansucrase synthesis. Among them, the s a c V locus, which is not linked to
the levansucrase structural gene (sacB) is of particular interest since strains bearing lesions in
this locus belong to a class of pleiotropic mutants affected in the synthesis of several
exoenzymes. These mutants also show defects in several functions related to the cell surface,
such as lack of flagella, inability to become competent for transformation, and morphological
alterations often resulting in filament formation (Sekiguchi et al., 1975; Kunst et al., 1974;
Ayusawa et al., 1975; Yoneda & Maruo, 1975). The relevant mutations, pap, sacU, amyB, were
mapped at the same locus (Steinmetz et al., 1976) but the mode and site of action of this regulatory gene are not understood (Sibakov et al., 1983). However, since these mutations mainly
affect protein secretion, careful comparison of the production and location of exocellular
enzymes in these mutants and in the parent strain should provide further insight into the nature
of this mechanism. In the study described here, we attempted to correlate the hyperproduction of
levansucrase and the synthesis of other exocellular proteins during the exponential phase of
growth.
METHODS
Culture and strains. The two strains used were the wild-type Bacillus subtilis 168 Marburg and the derivative
strain QB 112 ( s a c v 32) (Kunst et al., 1974). They were grown at 37 "C in the medium previously described (PetitGlatron e? af., 1980) supplemented when necessary with 1 mM-tryptophan. The carbon source was glucose
(10 g 1-l). One OD,,, unit of cell suspension corresponded to 1.4 x lo8 cells ml-l and 80 pg protein ml-l estimated according to Bradford (1976).
Downloaded from www.microbiologyresearch.org by
0022-1287/84/0001-2030 $02.00 0 1984 SGM
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
3144
R . CHAMBERT A N D M . - F . PETIT-GLATRON
Leuansucrase acriuity assay. Levansucrase activity was estimated by measuring the initial rate of the exchange
reaction catalysed by this enzyme. One unit of enzyme activity was defined as the amount of enzyme exchanging
1 pmol glucose min-' under the following conditions: 37 "C, 0.005 M-phosphate buffer (pH 6.0), 0.1 M-sucrose and
0.2 M-[ I'C]glucose. One unit of enzyme activity corresponded to 38 pmol, i.e. 2 pg, of enzyme (Chambert &
Treboul, 1976).
Measurements of differential rate of enzyme synthesis. The differential rate of enzyme synthesis was defined as the
slope of the straight line obtained when total enzyme activity was plotted against the OD6s0of the culture. It was
expressed in enzyme activity units (U) per OD,S, unit. During exponential growth, samples of cell suspension
were taken at intervals and measured for total enzyme activity and enzyme activity in the supernatant, as previously described (Petit-Glatron & Chambert, 1981).
Measurements of the lag rime of enzyme synthesis. The lag time, i.e. the time required to obtain a constant differential rate of enzyme synthesis, was determined by the intercept of the extrapolated straight part of levansucrase
production curves, in the presence of the inducer, with the levansucrase production curve in the absence of
inducer.
Induction equation. The operon model (Jacob & Monod, 1961) allowed the construction of equations which
described the induction process as proposed by Yagil(l975). He established a relationship between inducer concentration and the amount of protein ultimately obtained :
log(8 -
Po)
= logPo - logK
+ nlog[inducer]
P and Po are related to the differential rate of enzyme synthesis as follows: fl = (Differential rate of enzyme
synthesis),/[(Differential rate of enzyme synthesis),,, - (Differential rate of enzyme synthesis),], and Po = (Differential rate of enzyme synthesis),/[(Differential rate of enzyme synthesis),,, - (Differential rate of enzyme synthesis),], where (Differential rate of enzyme synthesis),, (Differential rate of enzyme synthesis), and (Differential
rate of enzyme synthesis),,, are respectively the differential rate of enzyme synthesis without inducer, with
inducer at a concentration i and with inducer at a concentration which fully induced the system.
When this equation is applied to experimental data, the derived parameters, n and K , must be interpreted with
respect to the control postulated. In the case of negative induction, n gives the effective stoichiometry of the repressor-inducer molecule interaction and K the dissociation constant of the inducer-repressor complex. Similarly, in
positive induction, n gives the stoichiometry of the inducer interaction with the regulatory macromolecule (receptor) and K the dissociation constant of this interaction. In both cases, K is equal to the effector concentration for
which enzyme synthesis is twice the basal level, to the power n.
Phenol extraction. A sample of cell suspension was chilled in the presence of chloramphenicol(200 pg ml-l ) and
centrifuged to eliminate exocellular levansucrase. The pellet was resuspended in the same volume of fresh culture
medium and the bacteria were lysed with a lysozyme/DNAase mixture (1 mg lysozyme ml-I, 0.01 mg DNAase
mi-', 0-2 mg MgCl, ml-I). An equal volume of water-saturated phenol was added. The mixture was shaken vigorously for 30 s with a vortex mixer and centrifuged for 10 min at 8000 g. The aqueous phase was removed and
dialysed overnight against 10 mM-phosphate buffer (pH 6.0).
Measurements ofprotein synthesis. After a [ 3sS]methionine pulse of the culture, labelled proteins were measured
in the supernatant, in the KCl(2 M) wash supernatant and in lysozyme-lysed bacteria. Samples were precipitated
by adding TCA (lo%, w/v), filtered through Whatman filters, washed with TCA (5%, w/v) containing 10 mMmethionine, dried and counted in a liquid scintillation counter.
SDS-PAGE. Proteins were analysed on SDS 10% (w/v) acrylamide gels. Electrophoreses were performed in a
slab-gel apparatus utilizing the discontinuous system described by Laemmli (1 970). Labelled exoprotein patterns
were quantified by counting 2 mm gel slices in scintillation liquid.
Chemicals.~-[~~SlMethionine
(- 1000 Ci mmol-1 ; 1 TBq mmol-I) was purchased from New England Nuclear.
and [ U-14C]glucose from Amersham. Bovine pancreas DNAase I was from Boehringer. Gel electrophoresis
material was from Bio-Rad.
RESULTS
Induction pattern of levansucrase in B. subtilis strains 168 and Q B 112
We compared levansucrase production in the parent strain B. subtilis 168 and in a hyperproducing mutant strain (QB 112), in the presence of various concentrations of sucrose during
the exponential phase of growth. The results of enzyme activity assays in whole cell suspensions
are shown in Fig. 1. In strain 168, the basal level of levansucrase production (0.1 3 x lo-' U per
ODbs0unit) and the fully induced level (3-8 x lo-' U per ODbs0unit) represented respectively
0-003% and 0.08 % of total protein synthesis. In the hyperproducing strain, the basal level (2 x
lov2 U per OD6,, unit) and the fully induced level (3-9 U per ODGs0unit) represented respectively 0.05% and 8% of total protein synthesis.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
3145
Hyperproduction of exocellular levansucrase
2.0
-
h
1.5
3
3
v
x
.->
.*
Y
1.0
V
v1
2
3
C
0.5
1.0
2.0
3.0
0.5
1.0
1.5
>
3
-2.0
OD650
-2.0
-1.0
log I sucrose1
Fig. 2
Fig. 1
Fig. 1. Levansucrase production by Bacillus subtilis 168 (a) and its derivative strain Q B 112 (b) in the
presence of various sucrose concentrations. Total production of levansucrase was measured in units
ml-' as a function of the cell growth. Sucrose added: 0,
none; 0 ,3.6 mM; A,5-8 mM; +, 17 mM; A,
29 mM; 0,
45 mM. The arrows indicate the beginning of the induction promoted by the addition of
sucrose in exponentially growing cells.
Fig. 2. Induction of levansucrase by sucrose. The lines were constructed according to Yagil's (1975)
Strain 168; 0,
strain Q B 112.
equation (see text). 0,
The hyperproducing strain showed a relatively slow response to the inducer as compared with
its parent strain. The lag period was less than 3 min for any concentration of sucrose in the case
of B . subtilis 168, whereas the lag period increased with the inducer concentration up to a maximum of 25 min in strain QB 112.
Determination of induction parameters
Genetic analysis of the B. subtilis Marburg levansucrase system (Lepesant et al., 1972) has
indicated that the operon model (Jacob & Monod, 1961) is applicable, but it was uncertain
whether the induction was under positive or negative control. From the induction equation
established by Yagil(l975) (see Methods) a straight line is predicted when log ( p - Po) is plotted
versus the logarithm of inducer concentration. Such a correlation was obtained for levansucrase
induction in both our strains (Fig. 2). The values of the induction parameters established by this
quantitative approach were the following. For strain 168, n = 2.5 0.2 and K = 1.1 mM", and
for strain QB 112. n = 2.3
0.1 and K = 1-2 mM". The concentration of inducer for halfmaximum induction was 12.0 mM in both strains. The value of n shows that more than two
molecules of sucrose are necessary to convert the controlling element (repressor or receptor) into
the form which facilitates levansucrase formation. Such a cooperative interaction between
inducer and controlling element has been found for several induced systems (Yagil, 1983).
Distribution of'lei3ansucrasebetween a free exocellular form and a cell-bound form in
strains 168 and QB 112
Fig. 3 shows the total production of the enzyme and the amount of enzyme released into the
culture supernatant after full induction. Levansucrase produced by B. subtilis 168 was not
released into the culture medium, but remained bound to the cell throughout the exponential
phase of growth. In contrast, levansucrase synthesized by B. subtilis QB 1 12 was found both free
in the culture medium and associated with the cells, the ratio of one form to the other varying
with respect to the time after induction. For example, 10% of the total enzyme synthesized was
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
3146
R . CHAMBERT AND
M.-F.
PETIT-GLATRON
I 00
-c
.d
2
0.03
Y
cr
2
0.02
-
.3
&
cd
x
:
0.01
2
cd
s
0
0.2
0
log IKCIl
Fig. 3
Fig. 4
Fig. 3. Levansucrase release into the culture supernatant as a function of cell growth after induction.
and strain QB 1 12 (0).Extracellular levansucrase
Total levansucrase production of strain 168 (0)
released by strain 168 (I
and
)
strain QB 1 12 (a).Full induction was promoted (arrows) by the addition
of 50 mM-sucrose (final concentration).
Fig. 4. Release of cell-bound levansucrase as a function of the washing salt concentration. Strain 168
was fully induced by 50mM-sucrose for 70 min. Growth was then stopped with chloramphenicol
(200 pg ml-' ), and 5 mhi-EDTA and 0.2 mM-phenylmethylsulphonyl fluoride (PMSF) were added.
Cells were harvested by centrifugation and resuspended in the usual medium containing EDTA and
PMSF. The cell suspension was divided into equal parts in flasks containing various concentrations of
KCI, gently shaken for 15 min at 25 "C and centrifuged. Levansucrase activity was measured in the
supernatant (E,) and in the bacterial pellet (Ep).(a) [E,/(E, E,)] x 100, i.e. the percentage of activity
released, is plotted versus the concentration of KCI. Incubation for 30 min in the presence of salt solu- ,531 is plotted against log (KCI concentration, M).
tions did not modify the curve. (b) log [Es/(EIoIa,
+
released 10 min after induction. This ratio became 85% 50 rnin after induction and remained
constant throughout exponential growth. The cell-bound form of the enzyme was not released
when bacteria of both strains were converted to protoplasts prepared by lysozyme treatment in
hypertonic medium according to Schaeffer et al. (1976).
Release of cell-bound levansucrase by washing the bacteria with high ionic strength solution
Joyeux et al. (1971) showed that cell-bound levansucrase in B. subtilis BS5 may be released
from the cells by washing with a high ionic strength buffer. We determined for both induced
strains the extent of release of cell-bound enzyme with respect to the KCI concentration of the
wash buffer. The dissociation process for the induced strain 168 depended strongly on the ionic
strength of the wash buffer but was not time dependent (Fig. 4a). Similar results were obtained
with induced strain QB 112 (not shown). Approximately half the total cell-bound levansucrase
was released in the presence of 1 M-KC1,and 1.5 M-KCl released all the levansucrase. The cooperativity of salt molecules in causing release is illustrated for strain 168 in Fig. 4(b) by plotting
log [enzyme released/(total enzyme - enzyme released)] versus log (KCl concentration, M). The
slope of the straight line obtained is equal to 4.
Pattern of exocellular proteins produced by strains 168 and Q B 112
Both induced and non-induced bacteria were pulse labelled for 3 min with [ 35S]methionine.
The ratioof labelled proteins in the culture supernatant to the total labelled cell proteins, and the
ratio of labelled proteins in the wash supernatant to the total labelled cell proteins were estimated (Table 1). For strain 168, these ratios were of the same order of magnitude and they
were unmodified by the induction process. Non-induced strain QB 112 synthesized much less
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
Hyperproduction of exocellular levansucrase
3147
Fig. 5. SDS-PAGE of labelled protein present in the cell supernatant and in the wash supernatant after
a pulse experiment. Induced and non-induced cells of strains 168 and QB 112 were treated in parallel
(induced cells received 50 mM-SUCrOSe 70 min before the pulse). The four 5 ml suspensions grown to an
(100 pCi ml-I). Incorporation was stopped
OD,S, of 1 were labelled for 3 min with ~-[~~S]methionine
by the addition of chloramphenicol(200 pg ml-l) and 10 mwmethionine. The cells were chilled on ice
and centrifuged. The pellets were resuspended in one-tenth of the initial volume in culture medium containing 2 M-KC~,
incubated for 15 min at 25 "C, then recentrifuged. The culture supernatants and wash
supernatants were dialysed against 10 mM-sodium phosphate buffer, pH 7-0, in the presence of protease
inhibitors and 10 mwmethionine, for 4 h at 4 "C, and then submitted to electrophoresis. Protein
markers were carbonic anhydrase (30 kDal), ovalbumin (46 kDal), bovine serum albumin (68 kDal)
and phosphorylase b (92.5 kDal). Parts A and B represent, respectively, non-induced and induced cells.
Lanes A1 and B1, culture supernatant of strain QB 112; lanes A2 and B2, wash supernatant of strain
QB 112. Lanes A3 and B3, culture supernatant of strain 168; lanes A4 and B4, wash supernatant of
strain 168. Lane L is labelled levansucrase purified as described by Dedonder (1966).
Table 1. Percentage of labelled proteins in the culture supernatant and percentage of labelled
proteins in the wash supernatant, relative to the total labelled cell proteins
The experimental conditions were as described in the legend of Fig. 5. [3SS]Methionineincorporated
into the proteins present in the supernatant, in the high ionic strength (2M-KCl) wash supernatant and
in the cells was determined as described in Methods. Approximately the same amount of
[35S]methionine,i.e. 40% of the labelled methionine added, was incorporated by the cells in all cases.
B. subtilis
strain
168
QB 112
Inducer
-
+
+
Percentage of labelled proteins
in the culture supernatant
Percentage of labelled proteins
in the wash supernatant
0.90
0.90
0.15
3.10
0.90
0.90
0.07
0.40
exocellular protein than strain 168, but when induced it produced about twice as much exocellular protein as strain 168, the greater part of the protein material being released into the culture
supernatant.
The labelled protein released into the culture supernatant or dissociated from the cells by a
high ionic strength wash was subjected to SDS-PAGE (Fig. 5). There were significant
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
3148
R . C H A M B E R T A N D M.-F. P E T I T - G L A T R O N
-
n
7 1.5
E
3
W
x
Y
' 5 1.0
1-
c)
@
OD650
0
2
20
L
45
60
Fig. 6
Fig. 7
Fig. 6 . Appearance of cryptic levansucrase activity in a phenol extract of induced cells as a function of
cell growth (strain QB 112). Cell suspensions were grown to an OD6Soof 0.6. Induction was promoted
by the addition of 50 mM-sucrose.Cell suspension (2 ml) was taken at intervals and treated with phenol
as described in Methods. Levansucrase activity was assayed in the aqueous phase of phenol extract in
the absence
or in the presence ( 0 )of Triton X-100.
(v)
Fig. 7. SDS-PAGE of labelled protein in the phenol extract of the cells during the course of induction.
A cell suspension of strain QB 112 was grown to an OD65oof 0.6. Induction was promoted by the addition of 50mM-sucrose. Cell suspension (2ml) was taken at intervals and pulse labelled with
[3sS]methionine (100 pCi ml-l) for 2 min then treated with phenol as described in Methods. Phenol
extracts of the cells were dialysed overnight at 4 "C against 10 mM-sodium phosphate buffer, pH 7.0.
The samples were lyophilized and resuspended in electrophoresis sample buffer. The numbers below
the photograph are times (in min) after the beginning of the induction. Lane L is exocellular levansucrase purified as described by Dedonder (1 966).
differences between the two strains. Strain 168 showed a variety of different proteins when the
culture supernatant and the wash supernatant were compared. None of these proteins has been
identified, but a major protein (40 kDal) in the wash Supernatant may be flagellin (Simon et al.,
1977). In the wash supernatant of induced strain 168, an additional labelled protein migrating at
the level of levansucrase marker was present. No labelled protein was released by strain QB 1 12
when not induced, whilst only a major protein with the same electrophoretic mobility as the
levansucrase marker was released when the cultures were induced.
Quantitative analysis of sliced gels was used to determine the percentage of radioactivity
recovered in the band corresponding to levansucrase with respect to the total radioactivity distributed in the gel (molecular weight range from 15000 to 200000). In the culture supernatant
of strain QB 112, levansucrase accounted for about 85% of the total radioactivity recovered. In
the wash supernatant, this ratio was 95%, showing that the protein was released in almost
homogeneous form in both supernatant and wash buffer. In induced strain 168, levansucrase
accounted for 2%of the protein in the wash supernatant. All these results indicated that the sac@
mutation seemed to have a dual effect, both inhibiting the synthesis of several exocellular
proteins, and greatly increasing levansucrase synthesis and export.
Appearance of a cryptic levansucrase activity in a phenol extract of the cells during the induction
course in strains 168 and QB 112
We have characterized a macromolecular complex including an inactive protease-sensitive
form of levansucrase isolated by phenol treatment of a constitutive and a hyperproducing strain
of B . subtilis (Petit-Glatron et al., 1980). This form could be activated by Triton X-100. We
examined the production of this complex during the course of induction of levansucrase synthesis in strains 168 and QB 112. The cryptic activity revealed by Triton X-100 appeared only 20
to 30 min after adding the inducer in strain QB 112 (Fig. 6). No cryptic activity was detected in
strain 168 in the same conditions.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
Hyperproduction of exocellular levansucrase
-
3149
0.25
n
I
3
s
0.20
.g
0.15
.Y
m
g 0.10
2
$
0.05
3
5
10
Time (min)
Fig. 8
Fig. 8. Determination of levansucrase mRNA functional half-life in strain Q B 112,7 min after induction. A cell suspension at an OD,,, of 1 was induced for levansucrase synthesis by adding 50 mMsucrose; 7 rnin after induction, 10 ml suspension was transferred to a flask containing rifampicin
(10 pg ml-*)
and shaken at 37 "C.Samples were taken at intervals during the first 10 min and subsequently assayed for levansucrase activity. The mean value and the standard deviation of six
independent assays of levansucrase activity for each sample are plotted as a function of the time after
addition of rifampicin.
Fig. 9. Levansucrase synthesis after inhibition of transcription as a function of induction time. 0 ,
Experimental points; ---, regression curve obtained from the equation AU = AU, e&;. ..., regression
curve obtained from the equation AU = AUo rx.
Table 2. Functional half-life qf levansucrase mRNA as a function of time after addition of
inducer to the culture medium
Time after
addition of sucrose
(min)
7
10
14
21
30
mRNA half-life
(4
88 f 5
100 f 7
85f5
90 f.8
99 f 10
Phenol extracts obtained after pulse labelling strain QB 112 at intervals during the induction
of levansucrase synthesis were analysed by SDS-PAGE (Fig. 7). The main protein which appeared during induction migrated at the level of the levansucrase marker. This protein was not
present during the lag period immediately following induction. This result is in good agreement
with the results presented in Fig. 6 , which show that the Triton X-100 sensitive form of enzyme
activity appeared 20 to 30 rnin following induction in strain QB 112.
Kinetics of appearance of functional levansucrase mRNA, and its half--lifeduring
the induction lag period
The slow response of strain QB 112 to the inducer was in sharp contrast to the relatively rapid
response of strain 168. One explanation for this lag could be a modification of levansucrase
mRNA stability during the lag period. This hypothesis was examined by determining the functional half-life of levansucrase mRNA during the induction phase. In Fig. 8 the results obtained
after 7 rnin induction are shown. Semi-logarithmic plots of these data allowed us to determine
the functional half-life of levansucrase mRNA (Table 2), which apparently remained constant
during the exponential phase of growth. If it is assumed that the amount of enzyme formed after
stopping mRNA synthesis is proportional to the amount of mRNA present when rifampicin
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
3150
R . CHAMBERT A N D M.-F. PETIT-GLATRON
was added, the quantity of specific mRNA present in the cells at various times can be calculated
(Fig. 9). Two functions may be postulated to account for this kinetic behaviour: AU = AUo eb
and AU = AUo t X . From regression adjustment with the least-square method, we obtained
e0.13rand the correlation
the following results: in the first hypothesis, the equation is 4.2 x
t2’2and the correlation
coefficient is 0.908; in the second hypothesis, the equation is 1.0 x
coefficient is 0.983. The second equation fits better with the experimental data (Fig. 9). (The cell
growth curve was exponential during the period of the experiment.)
DISCUSSION
In the present work we have shown that a sac@ derivative (QB 112) of B. subtilis 168 produces 100 times more levansucrase than its parent strain, but greatly reduced quantities of other
exocellular proteins. The sacUh mutation acts as if the potential for exocellular protein synthesis
and the cell’s export machinery are being used exclusively for the synthesis and export of levansucrase. However, this mutation does not modify the functional half-life of levansucrase
mRNA. Both the hyperproducing and the parental strain showed the same induction
parameters for enzyme production, but a lag period for induction was observed only in the
hyperproducing strain. In other inducible enzyme systems such a lag has been attributed to
several factors (Yagil, 1975; Imsande, 1978) such as slow inducer modification, intermediate
steps in the processing of the protein formed, or assembly of cellular components involved in
export machinery. The absence of a lag period in the parent strain allows us to eliminate the
first hypothesis. Two recent studies provide insight into the mode of action of the sacU gene.
Fouet et al. (1984) reported that the sac@ mutation did not affect the level of transcription of
the levansucrase structural gene. Sibakov et al. (1983) observed that the sacUh allele did not
increase the high level of a-amylase synthesis which was obtained by introducing the structural
gene of the enzyme into B . subtilis on a multicopy plasmid, and concluded that B . subtilis has
a translation machinery for exocellular protein which can be saturated. This whole set of results
strongly supports the hypothesis that sac@ mutations either block the synthesis of several exocellular proteins or increase the affinity of levansucrase mRNA for the coupled translation/
secretion machinery. In either case, the consequence would be a major enhancement in the
translation of levansucrase mRNA by the specific machinery for exoprotein synthesis (Horiuchi
et al., 1983) in preference to other mRNAs with which it is normally in competition. Underlying this hypothesis is the idea that the rate enhancement of levansucrase production observed
in the sacUh strain is essentially the result of an increased translation initiation frequency of
levansucrase mRNA. This rate enhancement, unlike that in the parent strain, is coupled with
the release of levansucrase into the culture medium.
In order to explore our general hypothesis it is essential to know the nature of the sacU gene
product. Fouet et al. (1984) showed that this gene directed the synthesis of a 46 kDal protein, but
this finding does not distinguish between a role for this protein as a structural component of a
coupled synthesis/secretion machinery for levansucrase, or via a catalytic activity which
markedly affects putative membrane components engaged in the synthesis or secretion of
several other exocellular proteins. This latter hypothesis has some merit since the pleiotropic
character of sacU mutations involves both the modification of the level of exoprotein synthesis
and altered surface properties. What kind of membrane component might be involved in these
phenomena? We propose that this component is a membrane associated amphiphatic molecule.
The most important and best known cell surface amphiphiles in bacilli are the lipoteichoic acids
(Wicken & Knox, 1980), which are attached to the plasma membrane by simple intercalation of
their lipid moiety into the membrane bilayer. This amphiphile could perhaps act as a modulator
of the activity of the membrane machinery responsible for levansucrase synthesis. In these
terms, we envisage that the sacU gene is involved in the regulation of the cellular pool of lipoteichoic acids. Several data support this hypothesis. Firstly, the polar moiety of lipoteichoic acid
is a linear polymer of some 25 to 30 residues of glycerophosphate which could be a good candidate as a cell surface site for levansucrase attachment, since bound enzyme was not released by
lysozyme hydrolysis of the peptidoglycan network but was released by washing the cells with
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
Hyperproduction of'exocellular levansucrase
3151
buffer of high ionic strength. Moreover, levansucrase has a very high affinity for a polyphosphate matrix such as hydroxyapatite (Gonzy-Treboul et al., 1975). The release of levansucrase
by the s a c V strain would then be the result of an increasing activity of lipoteichoic acid deacylation (Kessler & Shockman, 1979) in this mutant. This activity should lead to the release of the
enzyme-teichoic acid complex into the culture medium. This kind of association could also
explain the fact that levansucrase from the culture supernatant analysed by gel permeation
chromatography had an apparent molecular weight much higher than the purified levansucrase
(Delobbe, 1968). Secondly, by phenol extraction we have isolated from the s a c V strain, but not
its parent, a macrocomplex including an inactive form of levansucrase (Petit-Glatron &
Chambert, 1981). From the mode of extraction of this complex, its molecular weight and its
sensitivity to detergents, we conclude that this macromolecule consists of micelles of lipoteichoic
acid in which are entrapped mainly levansucrase molecules. Lipoteichoic acid may therefore be
hyperproduced in the sac@ strains. Thirdly, lipoteichoic acid may inhibit the activity of autolytic enzymes (Cleveland et al., 1975) and the hyperproduction of this amphiphile in s a c V
strains could explain the decrease of autolysin activity observed in these mutants with
consequent formation of non-septate filaments.
The involvement of membrane components which simultaneously modulate the synthesis of
exocellular proteins and participate in cell surface sites of attachment of these proteins is not yet
clearly settled. Beher et al. (1980) have shown that the cell surface expression of the ompA gene in
Escherichia coli appears to depend on the ability of the gene product to bind Iipopolysaccharides,
the main amphiphiles of the E. coli cell envelope, which are the final attachment site for OmpA
protein. This finding suggests that an interaction between the protein and lipopolysaccharides
plays an essential role in the biosynthesis of this outer membrane protein. Further experiments
will be required to confirm a link between lipoteichoic acid synthesis and biosynthesis of exocellular proteins in B. subtilis. We are now investigating the pool and turnover of this amphiphile
in various strains bearing mutations in the sacU gene.
We are indebted to 1. B. Holland (University of Leicester) for reading the manuscript. This work was supported
by a grant from the Centre National de la Recherche Scientifique.
REFERENCES
AYUSAWA,
D., YONEDA,
Y., YAMANE,
K . & MARUO,B.
(1975). Pleiotropic phenomena in autolytic enzyme(s) content, flagellation and simultaneous
hyperproduction of extracellular a-amylase and
protease in a Bacillus subtilis mutant. Journal of
Bacteriology 124, 459-469.
BEHER,M., PUGSLEY,
A. & SCHNAITMAN,
C. (1980).
Correlation between the expression of an Escherichia
coli cell surface protein and the ability of the protein
to bind to lipopolysaccharide. Journal of Bacteriology
143, 403-410.
BRADFORD,M. M. (1976). A rapid and sensitive
method for the quantitation of microgram quantities
of protein utilizing the principle of protein-dye
binding. Analytical Biochemistry 72, 248-254.
CHAMBERT,
R. & TREBOUL,
G. (1976). Levansucrase of
Bacillus subtilis. Characterization of a stabilized
fructosyl-enzyme complex and identification of an
aspartyl residue as the binding site of the fructosyl
group. European Journal of Biochemistry 71,493-508.
CLEVELAND,
R. F., HOLTJE,J. V., WICKEN,A. J.,
TOMASZ,A., DANEOMOORE,L. & SHOCKMAN,
G . D.
(1975). Inhibition of bacterial wall lysins by lipoteichoic acids and related compounds. Biochemical and
Biophysical Research Communications 67, 1 128-1 135.
DEDONDER,R. (1966). Levansucrase from Bacillus
.sihtili.\. Methods in Enzpologr 8, 500-506.
DELOBBE,
A. (1968). RGle du fer dans la production, la
purification de la Ievane-sucrase. Effets sur l'enzyme
purifie. Bulletin de la Societk de chimie biologique 50,
641-674.
FOUET,A., AUBERT,E., ARNAUD,M., LE Coo, D.,
KLIER,A. & RAPOPORT,G. (1984). The sucrose
system as a model of genetic regulation in Bacillus
subtilis. In Genetics and Biotechnology of Bacilli.
Edited by A. T. Ganesan & J . A. Hoch. New York:
Academic Press (in the Press).
GONZY-TREBOUL,
G., CHAMBERT,
R. & DEDONDER,
R.
(1975). Levansucrase of BuciIIus suhtilis: reexamination of some physical and chemical properties. Biochiniir 57, 17-38.
HORIUCHI,S., TAI, P. C. & DAVIS,D. (1983). A 64
kilodalton membrane protein of Bacillus subtilis
covered by secreting ribosomes. Proceedings of the
National Academy of Sciences of the United States of
America 80, 3287-3291.
IMSANDE,
J . (1 978). Genetic regulation of penicillinase
synthesis in Gram-positive bacteria. Microbiological
Retliews 42, 67-83.
JACOB,F. & MONOD,J . (1961). Genetic regulatory
mechanisms in the synthesis of proteins. Journal qf'
Molecular Biology 3, 3 18-356.
JOYEUX,Y., PUISEUX-DAO,
S. & JOUIN, H. (1971).
Localisation de la levane-sucrase de Bacillus sub&
var. Niger. Annales de I'lnstitut Pasteur 121,295-309.
KESSLER,
R. E. & SHOCKMAN,
G. D. (1979). Precursor-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08
3152
R . CHAMBERT A N D M.-F. PETIT-GLATRON
National Academy of Sciences of the United States of
product relationship of intracellular and extracelluAmerica 73, 2151-2155.
lar lipoteichoic acids of Streptococcus faecium.
SEKIGUCHI,
J ., TAKADA,
N . & OKADA,
H. ( 1975).Genes
Journal of Bacteriology 137, 869-877.
affecting the productivity of a-amylase in Bacillus
KUNST,K., PASCAL,M., LEPESANT-KEJSLAROVA,
J.,
subtilis. Journal of Bacteriologq’ 121, 688-694.
LEPESANT,
J. A., BILLAULT,
A. & DEDONDER,
R.
M., SARVAS,
M. & PALVA,
I. (1983). Increased
(1 974). Pleiotropic mutations affecting sporulation SIBAKOV,
secretion of a-amylase from Bacillus subtilis caused
conditions and the syntheses of extracellular enby multiple copies of a-amylase gene from B.
zymes in Bacillus subtilis 168. Biochimie 56, 1481amyloliquefaciens is not further increased by genes
1489.
LAEMMLI,
U. K. (1970). Cleavage of structural proteins
enhancing the basic level of secretion. FEMS
during the assembly of the head of bacteriophage T4.
Microbiology Letters 17, 8 1-85.
Nature, London 227, 680-685.
SIMON, M. I., EMERSON,S. U., SHAPER,J. H.,
J. A., KUNST,F., LEPESANT-KEJSLAROVA,
J.
LEPESANT,
BERNARD,
P. D. & GLAZER,
A. N. (1977). Classifica& DEDONDER,
R. (1972). Microsomal location of
tion of Bacillus subtilis flagellins. Journal of Bacterimutations affecting sucrose metabolism in Bacillus
ology 130, 200-204.
subtilis Marburg. Molecular and General Genetics 118, STEINMETZ,
M., KUNST,F. & DEDONDER,
R. (1976).
135- 160.
Mapping of mutations affecting synthesis of exoPETIT-GLATRON,
M. F. & CHAMBERT,
R. (1981).
cellular enzymes in Bacillus subtilis. Molecular and
Levansucrase of Bacillus subtilis. Conclusive
General Genetics 148, 28 1-285.
evidence that its production and export are unrelated WICKEN,A. J. & KNOX,K. W. (1980). Bacterial cell
to fatty-acid synthesis but modulated by membranesurface amphiphiles. Biochimica et biophysica acta
modifying agents. European Journal of Biochemistry
604, 1-26.
119, 603-6 1 1.
YAGIL,G. (1975). Quantitative aspects of protein
PETIT-GLATRON,
M. F., CHAMBERT,
R. & STEINMETZ, induction. Current Topics in Cellular Regulation 9,
M. ( 1 980). Levansucrase of Bacillus subtilis: charac183-236.
terization of a form isolated from phenol-treated YAGIL,G. (1983). Enzyme induction. In Mathematical
cells and activated by Triton X-100. European
Models in Molecular and Cellular Biology, pp. 68-83.
Journal of Biochemistry 103, 189-195.
Edited by L. A. Segal. Cambridge: Cambridge
PRIEST,F. G. (1977). Extracellular enzyme synthesis in
University Press.
the genus Bacillus. Bacteriological Reviews 41, 7 11- YONEDA,
Y. & MARUO,B. (1975). Mutation of Bacillus
subtilis causing hyperproduction of or-amylase and
753.
SCHAEFFER,
P., CAMI,B. & HOTCHKISS,
R. D. (1976).
protease, and its synergistic effect. Journal of’
Fusion of bacterial protoplasts. Proceedings of the
Bacteriology 124, 48-54.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:33:08