Archives of
Micrebiolegy
Arch. Microbiol. 118, 207-218 (1978)
9 by Springer-Verlag 1978
Pole Cap Formation in Escherichia coli Following Induction of the Maltose-Binding Protein Ingrid Dietzel, Victoria Kolb, and Winfried Boos Department of Biology, University of Konstanz, P.O. Box 7733, D-7750 Konstanz, Federal Republic of Germany
Abstract. After induction with maltose, 3 0 - 40 % of the total protein in the osmotic shock fluid consist of maltose-binding protein while the induction ratio (maltose versus glycerol grown cells) for the amount of binding protein synthesized as well as for maltose transport is in the order of 10. Induction of maltose transport does not occur during all times of the cell cycle, but only shortly before cell division. Electronmicroscopic analysis of cells grown logarithmically on glycerol or maltose revealed in the latter the formation of large pole caps. These pole caps arise from an enlargement of the periplasmic space. Small cells contain one pole cap, large cells contain two. Pulse label studies with strain BUG-6, a mutant that is temperature sensitive for cell division reveal the following: Growth at the non-permissive temperature prevents maltose-binding protein synthesis and formation of new transport capacity. After shifting to the permissive temperature the cells regain both functions. Simultaneously, the newly formed cells exhibit pole caps. We conclude that the induction of maltose-binding protein is responsible for the formation of pole caps. In addition, beside the presence of inducer, cell cycle events occuring during division are necessary for the synthesis of maltose-binding protein. Key words: Periplasm - Maltose-binding protein Maltose transport - Cell division - Pole caps - Cell envelope - Escherichia coli.
Introduction In recent years several substrate-binding proteins have been isolated from the cell envelope from Escherichia Non Standard Abbreviations. GLPT = periplasmic protein, related to transport of glycerolphosphate in Escherichia coli (Silhavy et al., 1976b)
coli and Salmonella typhimurium by the classical osmotic shock procedure of Neu and Heppel (1965). Several lines of evidence indicate that these "shock proteins" are in fact periplasmic, i.e. located outside the cytoplasmic membrane, supposedly beneath the outer membrane (Heppel, 197 i). These binding proteins have been identified as the substrate recognition sites of a particular type of active transport system (Boos, 1974), and some of them as chemoreceptors (Hazelbauer, 1975). One of these transport systems that is specific for maltose and maltodextrins (Wiesmeyer and Cohn, 1960) has become very useful to study for several reasons: It is homogeneous, i.e. there is only one system in E. coli that transports maltose; it is genetically well understood and defined mutants of all gene loci are available (Hofnung, 1974); it uses the 2 receptor of the outer membrane to facilitate the diffusion of the substrate through this diffusion barrier (Szmelcman and Hofnung, 1975; Szmelcman et al., 1976); and its substrate recognition site, the maltose-binding protein (Kellerman and Szmelcman, 1974) is also a chemoreceptor (Hazelbauer, 1975). Two additionally required components that have been defined genetically have not been identified biochemically as yet, but there is strong evidence that at least one of them is a protein located within the cytoplasmic membrane (Silhavy et al., 1976a). The appearance of one of the components of the transport system, the receptor of phage 2, in the outer membrane has been found to be regulated by the cell cycle and according to this study it is clear that, during its in vivo assembly, this component is incorporated in the outer membrane only in certain regions of the bacterial cell surface (Ryter et al., 1975). The present publication deals with the alterations of the cell envelope of E. Coli that are due to the appearance of the maltose-binding protein in the periplasm after induction by maltose.
0302-8933/78/0118/0207/$ 02.40
208
Arch. Microbiol., Vol. 118 (1978)
Material and Methods
Pulse Labelling of the Maltose-Binding Protein
Bacterial Strains and Growth Conditions
From cultures growing logarithmically in minimal medium A with glycerol (0.4%) and maltose (0.2 %) as carbon source 5 ml aliquots were removed at different time intervals. These aliquots were further allowed to grow for 5 min at the same temperature in the presence of [:4C]-labelled amino acids (0.2 btM L-valin, 0.16 ~tM L-leucin and 0.16 ~M IAsoleucin). I n the aliquot of 5 ml each amino acid contributed about 5 . l0 s cpm. Then, a 1000-fold excess of nonlabelled amino acids was added and growth continued for 5 min. Then, chloramphenicol (150 gg/ml, final concentration) was added and the culture stored on ice. The cells were collected by centrifugation. They were washed with 5ml of 10mM Tris-HC1, pH 7.3 and suspended in 5 ml of the same buffer. A French-press extract was prepared as described above. The extract was lyophilized and resuspended in 50 pl of 10 mM Tris buffer containing maltosebinding protein (0.1 mg/ml). This suspension was used directly in the immunodiffusion technique as described (Lengeler et al., 1971). The stained and dried agar-plates were autoradiographed (Kodak Regulix Film) for 1 week.
Strain LA 3400 of E. eoli was used throughout this study. It was obtained by P1 transduction of strain 72 (A [glpR-malA], phoA) to glpR +, mal+,phoA, selecting on maltose (Silhavy et al., 1976b). Strain La 108 was a derivative of BUG-6 (Reeve et al., 1970) that was made phoR by P 1 transduction (Shen and Boos, 1973). This strain normally grows at 35~ but forms long filaments at 42~ The length of these filaments can become at least 50times the normal cell length. However, too long an exposure to 42~ initiates occasional cell division. As growth medium minimal medium A (Mille,r, 1972) was used with 0.4%glycerol or 0.2% maltose as carbon source. Growth temperature was 37 ~C. Induction of the maltose-transport system in cells grown on glycerol was done by the addition of 1 mM maltose (fmal concentration). Synchronization of strain LA 340(1 was done according to Cutler and Evans (1966). This method takes advantage of the tendency of cells to synchronize when they enter the stationary phase of growth. Then they were diluted 20-fold into prewarmed medium. This procedure was not always reproducible; effective synchronisation was obtained in the average of two out of three attempts. In addition, generally only 1 to 2 steps of synchronous growth could be observed. Synchronization was monitored by plating for viable counts on rich medium agar plates.
Transport of maltose was measured in the following way: 1 ml of the growing culture was removed and centrifuged for 2 min at 13000 X g (Eppendorf microcentrifuge), washed twice with minimal medium A and suspended in 1 ml of the same medium. To 100 gl of this suspension 20pl of 50pM [14C] maltose (3.9mCi/mMol, Amersham) were added. After 20 s, 100 pl were removed and filtered through Millipore filter (0.65 gm pore size). The filters were washed once with 10 ml minimal medium A, dried and counted in a toluene based scintillation fluid. All operations were performed at 37~C./~galactosidase activity was determined in toluenized cells as described by Miller (1972). The cold osmotic shock was performed according to Neu and Heppel (1965) using 900 ml culture. The number of cells in this culture was determined by plating for viable cells on rich medium agar plates. The obtained shock fluid was concentrated to about 10 ml by ultrafiltration using Amicon UM-10 filters. Subsequently, the concentrated solutions were dialyzed against 10 mM Tris, pH 7.3, over night. The slightly turbid solution was centrifuged at 100000 x g for 30 min and the supernatant used for determination of the maltosebinding protein and total protein concentration. To prepare cytoplasmic extracts the shocked cells were resuspended in 5 ml of 10 mM Tris, pH7.3 and passed twice through a French pressure cell at 600 atm at 0~ The suspension was centrifuged at 40000 x g for 1 b and the supernatant used for immunodiffusion test. Protein was measured according to Lowry et al. (1951) using serum albumin as standard. Galactose- and maltose-binding protein was estimated in the shock fluid by immunodiffusion technique using anti-galactose- and anti-maltose-binding protein antibodies as described previously (Lengeler et al., 1971). These antibodies were raised in rabbits using homogeneous binding proteins and complete Freud adjuvant. The amount of maltose-binding protein was determined by equilibrium dialysis as described (Schwartz et al:, 1976) using a st)bstrate concentration of 0.1 pM, well below the KD for maltose. 300 pl of shock fluid were put into small sacs of dialysis tubing (Union Carbide) and dialyzed against 10 ml buffer at 4 ~C over night. 100 gl aliquots were counted in a dioxan based scintillation fluid. In shock fluids of maltose grown cells the cpm's obtained from samples in the bag were routinely 2 - 3-fold higher than that of the outside buffer.
Two dimensional polyacrylamide gel electrophoresis was performed as described (Johnson et al., 1975) with modifications described later (Silhavy et al., 1976b).
Electron Microscopy. For negative staining one drop of a logarithmically growing culture was removed and deposited on a formvar and carbon coated grid. Staining was performed with 1% ammonium molybdate, pH 7.3 for 30 s. For embedding, growing cells were pelleted and fixed at 4~ for 4 0 - 6 0 min with 2.5 % glutaraldehyde in 0.l M sodium cacodylate buffer (pH 7.0) and postfixed for 1 - 2 h with 1% osmium tetroxide in the same buffer. Specimens were dehydrated with ethanol and embedded according to Spurr (1969). Sections were made with a Reichert ultramicrotomc OmU2. The thin seclions were stained with lead-citrate (Venable and Coggeshall, 1965). All preparations were examined with a Siemens Elmiskop 101 electron microscope.
Results Induction o f the Maltose-Binding Protein G r o w t h o n t h e E. coli w i l d - t y p e s t r a i n L A 3400 o n maltose results in the induction of a large amount of shock releasable maltose-binding protein. Figure l shows the comparison of crude shock fluids by two d i m e n s i o n a l p o l y a c r y l a m i d e gel e l e c t r o p h o r e s i s o f strain LA 3400 grown on glycerol (A) and on maltose (B). F r o m t h e s e gels it a p p e a r s t h a t a f t e r i n d u c t i o n b y maltose the osmotic shock fluid consists to a large extent of maltose-binding protein. Induction for the s y n t h e s i s o f o t h e r t r a n s p o r t r e l a t e d p r o t e i n s s u c h as t h e g a l a c t o s e - b i n d i n g p r o t e i n b y f u c o s e r e s u l t s in m u c h less dramatic increase of the corresponding protein (Silhavy et al., 1 9 7 6 b ) . A s i m i l a r " n o r m a l " i n d u c t i o n r a t i o c a n b e s e e n i n F i g u r e 1 A f o r G L P T ( s p o t 3) t h a t is specific a l l y i n d u c e d b y g l y c e r o l ( S i l h a v y e t al., 1 9 7 6 b ) . The determination of the total protein content of t h e r e l e a s e d p e r i p l a s m i c p r o t e i n s gives v a l u e s t h a t r a n g e f r o m 2 . 7 - - 3 . 2 m g p r o t e i n i n t h e s h o c k f l u i d f r o m i 0 t2
I. Dietzel et al. : Envelope Changes of E. coli Induced by Maltose
Fig. 1 A and B
Two-dimensional polyacrylamide gel electrophoresis of shock fluid of strain LA 3400 grown in the presence of glycerol (A) and maltose (B). The first dimension consists of electrophoresis in 8 M urea (pH 8.4), followed by electrophoresis in 0.2 ~ sodium dodecylsulfate (pH 6.48). The numbers correspond to the following proteins: I maltose-binding protein; 2 galactosebinding protein; 3 GLPT protein
209
210
Arch. Microbiol., Vol. 118 (1978)
glycerol grown cells and from 3.8-4.3 mg protein in the shock fluid from 1012 maltose grown cells. In addition, as seen in Figure 2, galactose-binding protein isolated from the same number of both types of
cells gives precipitin lines on immunodiffusion plates against anti-galactose-binding protein antibodies that are identical in position and strength. Thus, even though the total yield of shock proteins from one batch to another can vary to some extent and the determination of galactose-binding protein by the immunodiffusion technique only allows a semi-quantitative estimation it would appear that maltose-binding protein is induced in addition to the other periplasmic proteins (for instance the galactose-binding protein), i.e. the required space for these proteins is not limited. The assay for the maltose-binding protein in the crude shock fluid was done by equilibrium dialysis with [14C] maltose at substrate Concentrations below the KD for maltose-binding. Under these conditions the [PL] [P] Scatchard (1949) equation extrapolates to
[LFI [K.]
Fig. 2. Immunodiffusion of crude shock fluids against anti-galactosebinding protein antibodies. Concentrated shock fluids of strain LA3400 grown under the following conditions were used: 1 0.4% glycerol; 2 0.4~ glycerol containing 1 mM D-fucose (for induction of the galactose-binding protein); 30.2% maltose. The crude shock fluids were adjusted in protein concentration as to the identical amount of cells they were derived from. Average concentration was 1 mg/ml. 4 control, purified galactose-binding protein 0.02 mg/ml. The center well contained anti-galactose binding-protein antiserum
(Silhavy et al., 1975) where [P] equals total number of available binding sites, [PL]the protein ligand complex, and [LF] the free ligand concentration. Table I summarizes the results obtained by the binding test in two series of experiments. Substrate concentration in the binding test was 0.1 IxM and the KD for maltose at 4~ C temperature was taken to be 3.5 gM (Schwartz et al., 1976). As can be seen from Table 1, after induction with maltose, the maltose-binding protein comprises between 30 and 40 % of the total shock proteins, while the induction ratio is about 10-fold. With the number of cells in the culture one can calculate that a fully induced
Table 1. Maltose-binding test in crude shock fluid from cells grown in glycerol or maltose Experiment 1 Maltose grown cells
Experiment 2 Glycerol grown cells
Maltose grown cells
Glycerol grown cells
Total protein concentration" [mg/ml]
0.95
1.16
1.16
1.65
[PL]
2.1
0.2
3.3
0.2
1.6 J
0.3
3.9
0.3
7.2 0.29
1.0 0.04
12.1 0.49
1.0 0.04
MBP concentration [laM] [mg/ml] b % of the total protein in the crude shock fluid
30.6
3.6
42
2.4
Number of MBP molecules released per cell ~
22300
1500
24000
1100
" Total protein concentration was determined in the concentrated shock fluid by the method of Lowry et al. (1951) b The coricentration of maltose-binding protein in the crude shock fluid was calculated from the binding assay, using a molecular weight of 40000 (Kellerman and Szmelcman, 1974) ~ The number of maltose-bindingprotein molecules released per cell was calculated from the total amount of maltose-bindingprotein released and the number of cells present in the culture from which the protein was isolated
I. Dietzel et al. : Envelope Changes of E. coli Induced by Maltose
The Kinetic of Maltose Transport Induction in Logarithmically Growing non Synchronized Cells
Fig. 3 A and B
Induction of maltose transport and fl-galactosidase in a logarithmically growing culture of strain LA 3400. When the culture that had been growing on glycerol had reached an optical density of 0.12 (578nm) 10mM maltose and 0.i mM isopropyl-flthiogalactoside were added simultaneously, Maltose uptake is given in pmole substrate taken up per 100 gl culture in 20 s. fl-galactosidase units are given in pMole substrate per min
211
._~"
A
~s00
~ooo ~ ~ r""n ~o o ~ 500 ~ ~n
/ J
/
~_,
A__L~
o B wo
-~ ~ o -~
so
0
Q08 Q12 0.16 02
Q24, 028 0.32
cell mass [OD at 578 nm]
lO: 6
A
T1C E~ ~,o, ld
~__. (23
15
i
i
i
i
i
b
_ ~o.~_o_~_o
Fig. 4 A--D
Induction of maltose transport and fl-galactosidase activity in synchronized cells of strain LA 3400. The culture was synchronized by the dilution method of Cutler and Evans (1966). After dilution, at the time indicated by the arrow 10raM maltose and 0.1 mM isopropyl-fl-thiogalactoside were added simultaneously, and the following parameters were assayed: A cell mass; B viable counts; C maltose uptake given in pmole substrate taken up in 100~tl culture in 20 s; D fl-galactosidase activity
B/I i
The classical studies on the induction of fl-galactosidase in Escherichia coli have shown that after addition of the inducer, synthesis of/Lgalactosidase starts after a short lag period at a rate that is proportional to growth (Cohn and Monod, 1951). In other words, a straight line is obtained when the increase of enzymatic activity is plotted against the increase of cell mass or total protein. This is not the case when maltose transport is measured in a logarithmically growing culture after induction by maltose. Figure 3 shows the simultaneous induction in strain LA3400 of fl-galactosidase by isopropyl-thio-flgalactoside and of maltose transport by maltose. The addition of inducer was done at an optical density of 0.12 (578nm). As can be seen, in contrast to the induction of fl-galactosidase, maltose transport increases initially in a logarithmic fashion and only after an increase to about 150 % of the initial cell mass at the time of inducer addition, maltose transport increases linearly. This suggested that the increase in transport capacity might not occur during all phases of the cell cycle. Therefore, we assayed maltose transport after induction in synchronized cells.
i
i
i
c 2O
E
~
ol 1000
500 2q E
J D
/
/
/g
o
E
I o~o/ ~-L.~ 0 -20 ; 20 ~; ~'o 8'0 ,oo time[rain]
cell releases about 2 x 10~ molecules of maltosebinding protein during osmotic shock. Measuring the initial rate of maltose transport at a substrate concentration of 1 pM in cells grown in glycerol and in maltose, a 10-12-fold increase in transport activity can be observed (not shown).
Fig. 5. Cell division dependence of the maltose-binding protein in strain LA 108. Immunodiffusion of shock fluids and cytoplasmic extracts of strain LA 108 against anti-maltose-binding protein antibodies. 1 pure maltose-binding protein (0.1 mg/ml); 2 shock fluid of cells, grown at 42~ (2 mg protein/ml); 3 cytoplasmic extract of cells grown at 42~ 4 cytoplasmic extract of cells grown at 35~ 5 shock fluid of cells grown at 35~ (2mg protein/ml). The center well contains serum with anti-maltose-binding protein antibodies. The agar plate was stained with Coomassie blue
212
i
Arch. Microbiol., Vol. 118 (1978)
1.0
=
y
0.1
"~
j/ B
160 140
%, -6 E
120
I00
I
80
i =0 40
Fig.6A--C
20
~
/,.0
A &
T % 3.0
!
o 2.0
1.0
20 i
r
;
40
60 i
;
I00 t
;
I/-.0 i
I
180
80 120 160 t i m e [min]
Dependence of maltose transport on cell division in strain LA 108. Strain LA 108 was grown logarithmically at 35~ on 0.4% glycerol and 0.2% maltose to an optical density of 0.3 (at 578 nm). At time 0 half the culture was shifted to 42~ ( 9 the other half remained at 35~ (a A). At the time indicated the cultures were shifted to 35~ and 42~ respectively. The following parameters were measured: ACell mass (OD at 578 nm). B Maltose uptake, given in pmole maltose taken up by 100 ~tl culture in 20 s. C Cell numbers, as viable counts on nutrient plates
Induction of Maltose Transport in Synchronized Cells Cells o f strain L A 3400 were synchronized according to the m e t h o d o f Cutler and Evans (1966). Figure 4 shows the optical density, cell number, and maltose transport o f cells growing in glycerol after induction by maltose at time 0. As can be seen, transport capacity increases only shortly before or during cell division. W h e n the time o f maltose addition within the cell cycle is varied the result remains the same: transport increase only occurs shortly before or during cell division (not shown).
Induction of Maltose Transport Activity and of Maltose-Binding protein in BUG-6, a Mutant that is Temperature Sensitive in Cell Division Two cultures o f L A 108, a phoR derivative o f B U G - 6 , were g r o w n to late logarithmic phase in minimal
Fig.7A and B. Dependence of maltose-binding protein synthesis on cell division in strain LA 108. Cultures of strain LA 108 were pulse labelled with [14C]labelled amino acids as described in the text and under Material and Methods. A Coomassie blue stained immunodiffusion plate of the different extracts (unlabelled maltose-binding protein was added in each aliquot). BAutoradiogram of the dried agar plate shown in A, /control, no extract present; 215rain; 3 25 min; 4 35 rain; 5 45 rain; 6 55 min; 765 rain; 8 75 rain; 9 85 min; 1095rain. The culture was shifted from 42~ to 35~ at 50rain
m e d i u m A containing 0.4 % glycerol and 0.2 % maltose; one, at the permissive temperature o f 35 ~C, the other at 42 ~C. After washing free o f maltose and adjusting to a cell density o f O D (578 nm) 1.0 transport activity for maltose (initial rate of uptake) was 60-fold higher in cells grown at the permissive temperature than in cells grown at the non-permissive temperature. Light microscopic analysis revealed the presence of n o r m a l cells when g r o w n at 35~ and very long filaments when grown at 42 ~C. The periplasmic shock proteins as well as the total cytoplasmic extracts o f these cells were obtained and analyzed by immunodiffusion against anti-maltose-binding protein antibodies (Fig. 5). As can be seen this strain only produces maltose-binding protein when grown at the permissive temperature and it contains it exclusively in the periplasm. As control, we analyzed the growth m e d i u m for the presence o f maltose-binding protein after extensive dialysis against water and subsequent lyophilization. Cultures g r o w n at
I. Dietzel et al. : Envelope Changes of E. coli Induced by Maltose
213
Fig. 8 a - d Negatively stained cells after logarithmic growth in maltose (a, b) and glycerol (e, d). Cells grown in maltose show pole caps, large cells have two (a, arrows), small cells only one (b, arrow). Bars represent 0.5 ~m
both temperatures did not contain material in the medium that crossreacted against anti-maltose-binding protein antibodies. However, when cells were grown over night and were kept at the stationary phase of growth maltose-binding protein could be detected in the growth medium, indicating extensive cell lysis under these conditions. Normal wild-type behaved identical in this respect.
To elucidate further the induction of the maltose transport machinery in this strain the experiment shown in Figure 6 was performed. Cells were grown at 35~ to an optical density of 0.3 in minimal medium A containing 0.4 ~ glycerol and 0.2 ~ maltose. At time 0 half of the culture was shifted to 42~ while the other half was kept at 35~ After 50min the two cultures were shifted to 35~ and 42~ respectively. Viable
214 counts (C), uptake of maltose (B) and cell mass (A) was followed. As can be seen, increase of transport capacity only occurs at 35~ but not at 42~ Moreover, after shift to the permissive temperature the cells divide before transport increase occurs. A control experiment was performed, where chloramphenicol was added to the 42~ culture, at the time of the temperature shift. The cells still divided but no transport increase occurred (not shown). This excludes the possibility that an inactive precursor is formed at 42~ that might be processed after shift to the permissive temperature in the absence of protein synthesis. The latter conclusion is also substantiated by an experiment where maltose-binding protein is pulselabelled by radioactive amino acids. For this purpose the experiment shown in Figure 6 was repeated with a culture (OD0.3) growing at 42~ after time0 and at 35 ~C after 50 min. Aliquots were removed and grown in the presence of radioactively labelled amino acids at the respective temperature for 5 min before chasing with an excess of unlabelled amino acids. The cells were broken by French pressing. Pure maltose-binding protein was added to the cell extracts and immunodiffusion was performed (Fig. 7A). The stained and dried agar plate was subjected to autoradiography (Fig. 7b). Occurrence of radioactivity in the precipitin band indicates maltose-binding protein synthesis. As can be seen, synthesis of maltose-binding protein starts between 25 and 35m in after shift to the permissive temperature well in agreement with the time of increase in maltose transport activity (Fig. 6 B). Thus, it is clear that new synthesis of maltose-binding protein only occurs at the permissive temperature and after one round of division has passed. Apparently, no protein synthesis is necessary for cell division, after shift to the permissive temperature since it occurs in the presence of chloramphenicol (Reeve et al., 1970). In contrast, newly formed transport capacity and formation of maltosebinding protein is completely blocked under these conditions.
Ultrastructure Differences of the Periplasm in Maltose and Glycerol Grown Cells Specimens prepared by negative staining revealed that cells grown on maltose developed an enlargement of their periplasmic space (Fig. 8). These enlargements are analogous to the findings of Wetzel et al. (1970) on Escherichia coli derepressed for alkaline phosphatase and will accordingly be referred to as "pole caps". These pole caps are much less pronounced when the cells are grown in glycerol. The mean distance between 9 the cell wall and the cytoplasmic membrane in these pole caps was about two to four fold larger in maltose
Arch. Microbiol.,Vol. 118 (1978)
25
1
2O u
15-
Fig. 9
102
0
0.2 0.4 06 0.8 1.0 size of pole caps (l~m)
Distribution of the pole cap size in cells of strain LA 3400 grown on glycerol(INN) and maltose ( I ~ ). Pole cap size is given as the distancein jam of cell wall and cytoplasmic membrane at the cell poles
than in glycerol grown cells. The statistical distribution of this distance is shown in Figure 9. In addition, we observed that small, cells developed the pole cap only on one end of the cell, while large cells always developed them on both ends (Fig. 8). Pole caps could also be seen in embedded and ultrathin sectioned cells after growth on maltose but not on glycerol (not shown). In both types of cells the periplasmic space was filled with electron dense material. This pole cap formation cannot be an artifact due to the different osmotic effects of maltose and glycerol that are used as carbon source at 0.4~o and 0.2}/o concentrations. Addition of as little as 5 x 10-4M maltose to a culture growing on glycerol results in the same pole cap formation (not shown). Moreover, cells of strain LA3400 grown on glycerol and maltose respectively were plasmolized with 20 ~o sucrose. As can be seen in Figure 10, the effect of maltose induction can still be seen in plasmolized cells. This indicates that pole cap formation is not an artifact caused by plasmolysis. The observation that maltose induced cells formpole caps is not restricted to strain LA 3400. We observed large amounts of maltose-binding protein in the periplasm of all maltose metabolizing strains that we have tested so far. A particularly interesting example is LA 108. When this strain is grown at 42~ in glycerol no pole caps can be observed on the filamentous cells (Fig. 11). Shifting the culture to 35~ in the presence of 1 mM maltose lets the cells divide but after 20 rain no pole caps have formed (Fig. 11 B). After 70 rain, however, about 30 ~ of the cells have acquired pole caps (Fig. 1 IC,D). It is interesting to note that at this time (when transport of maltose is not yet fully induced) only part of the cells have fully developed pole caps, while others have no pole caps at all. This might indicate that induction of the maltose-binding protein is an all or none phenomenon. Wild-type ceils that had been growing over night in minimal medium A containing 0.2 ~o maltose and were maintained in the stationary phase of growth are smaller than logarithmically growing cells. They do not
I. Dietzel et al. : Envelope Changes of E.
coli
Induced by Maltose
215
Fig. 10a and b
Pole caps after plasmolysis. Strain LA 3400 was grown logarithmically in minimal medium A with 0.2 % maltose (a) or 0.4 ~ glycerol (b) as carbon source. The cells were plasmolized by the addition of 20 (w/v) sucrose for 2 - 4 rain and subsequently prepared for negative staining. Maltose grown cells always showed remnants of pole caps (-~) whereas glycerol grown cells only occasionally showed these structures; most of these cells had their plasma extended into both cell ends. Bars represent 0.2 pm
look significantly different than stationary phase cells that had been growing on 0.4 % glycerol (not shown).
Discussion
The presence of maltose in the growth medium leads to the appearance o f large amounts of maltose-binding protein in the osmotic shock fluid that is obtained by the classical cold osmotic shock procedure of Neu and Heppel (1965). Several lines of evidence indicate that these proteins are in fact periplasmic, that is outside the cytoplasmic membrane (Heppel, 1971).
F r o m the quantitative determination of the maltose-binding protein in these shock fluids it is clear that maltose-binding protein is excreted into the periplasm in addition to the other proteins already there. In other words, the total amount of protein in this space is increased after maltose induction. Simultaneously, after induction with maltose the cells form large pole caps. There is no definite p r o o f that these pole caps are filled with maltose-binding protein. Reaction product staining as performed by Wetzel et al. (1970) and by MacAllister et al. (1972) cannot be used with a protein that does not exhibit enzymatic activity.
216
Arch. Microbiol., Vol. 118 (1978)
Fig.llA--D. Pole cap formation in dependence of cell division in strain LAI08. Cells of strain LA108 were grown at 42~ for 2generations. Aliquots were removed and prepared for negative staining at different time intervals. A 10 min before temperature shift to 35~ B 20min after temperature shift. C and D 75rain after temperature shift. Only at the last time pole caps (~) are visible. Occasionally, cells with and without pole caps were found lying side by side (D)
Also, in contrast to our findings with the maltosebinding protein, alkaline phosphatase seems to distribute itself over the entire periplasmic space and not only the poles (MacAllister et al., 1972). In our hands derepression of alkaline phosphatase does not lead to formation of pole caps. Strain LA 108 carries a phoR mutation and synthesizes alkaline phosphatase constitutively. Yet, pole caps even in this strain are only observed after induction with maltose. It is very unlikely that these pole caps are artifacts of the preparation: the staining solution consisting of 1 a m m o n i u m molebdate had the same osmolarity as the growth medium. In addition, the pole caps were filled with electron dense material that is usually absent in
plasmolized cells. Also, plasmolized cells still retain poles that had formed due to maltose induction. It is not clear what the reason for the pole cap formation is. One possibility would be the excretion of maltosebinding protein (and other periplasmic binding proteins) exclusively on the poles of the cells. Alternatively, the cell wall of growing poles might be particularly yielding to the osmotic pressure created by periplasmic proteins (Stock et al., 1977). Accordingly, periplasmic proteins would be accumulating at the growing poles shortly before cell division. Since stationary cells are essentially not growing, pole caps are no longer formed. In this connection it is interesting that growing cells usually showed one pole cap when they are small but two
I. Dietzelet al. : EnvelopeChangesof E.
coli
Inducedby MaItose
when they are twice the normal cell length. This is possibly related to cell division and one would postulate that the two-pole caped cells were in a stage shortly before cell division while one-pole caped cells had just divided. Maltose transport in synchronized cells increases only shortly before or during cell division. Recently, it had been shown that the receptor for phage 2 is synthesized only during cell division (Ryter et al., 1975). This protein plays an important role in maltose transport (Szmelcman and Hofnung, 1975) by catabolizing the diffusion through the outer membrane (Szmelcman et al., 1976). Therefore, its cell cycle dependent synthesis might be the reason for the corresponding
217
transport of maltose. However, as we have shown in this paper cell division is also a prerequisit for the synthesis of maltose-binding protein, an essential component of maltose transport (Kellerman and Szmelcman, 1974). The morphological evidence only indicates accumulation of periplasmic material at the pole caps. We were as yet unable to obtain morphological evidence for the excretion of periplasmic material at the septum of dividing cells. Few other transport systems have been found to be regulated by events that occur during the bacterial cell cycle (Ohki, 1972; Shen and Boos, 1973). At the present time, it is not clear what the molecular basis of this cell division dependence is. Periplasmic proteins including
218
the maltose-binding protein can be synthesized in cell free systems (Randall and Hardy, 1977). This might indicate that synthesis of these proteins is rather repressed during the cell's elongation than actively stimulated during cell division. Acknowledgements. We wish to thank Miss M. Winter for technical assistance. A sample of pure maltose-binding protein was obtained from Dr. Ferenci. This work was supported by the Deutsche Forschungsgemeinschaft, Sonderforschungsbereich 138 "Biologische Grenzfl~ichen und Spezifit~it", by the North Atlantic Treaty Organization and the "Fonds der Deutschen Chemischen Industrie".
References Boos, W.: Bacterial transport. Annu. Rev. Biochem. 43, 123-146 (1974) Chon, M., Monod, J. : Purification et propri6t6s de la/%galactosidase (lactase) d'Escherichia coli. Biochim. Biophys. Acta 7, 153 - 174 (1951) Cutler, R. G., Evans, J. E.: Synchronization of bacteria by a stationary phase method. J. Bacteriol. 9 1 , 4 6 9 - 4 7 6 (1966) Hazelbauer, G. L. : Maltose chemoreceptor of Escherichia coli. J. Bacteriol. 122, 206-214 (1975) Heppel, L. A. : The concept of periplasmic enzymes. In: Structure and function of biological membranes (L. A. Rothfield, ed.), pp. 223-247. New York: Londong. 1971 Hofnung, M. : Divergent operons and the genetic structure of the maltose B region in Eseherichia coli K-12. Genetics 76, 169-- 184 (1974) Johnson, W. C., Silhavy, T. J., Boos, W. :Two-dimensional polyacrylamide gel electrophoresis of envelope proteins of Escherichia coli. Appl. Microbiol. 29, 405-413 (1975) Kellerman, O., Szmelcman, S. : Active transport of maltose in Escherichia coli K-12. Eur. J. Biochem. 47, 139-149 (1974) Lengeler, J., Hermann, K. O., Uns61d, H. J., Boos, W.: The regulation of the/~-methylgalactoside transport system and of the galactose-binding protein of Escherichia coli K-12. Eur. J. Biochem. 19, 457-470 (1971) Lowry, O. H., Rosebrough, N. J., Farr, A. L., Randall, R. J. : Protein measurements with the folin phenol reagent. J. Biol. Chem. 193, 265-275 (1951) MacAllister, T. J., Costerton, J. W., Thompson, L., Thompson, J., Ingrain, J. M. : Distribution of alkaline phosphatase within the periplasmic space of gram negative bacteria. J. Bacteriol. 111, 827--832 (1972) Miller, J. H. (ed.) : Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York (1972)
Arch. Microbiol., Vol. 118 (1978) Neu, H. C., Heppel, L. A. : The release of enzymes from Escherichia coli by osmotic shock and during the formation of spheroplasts. J. Biol. Chem. 240, 3685-3692 (1965) Ohki, M. : Correlation between metabolism of phosphotidylglycerol and membrane synthesis in Eseherichia coli. J. Mol. Biol. 68, 249-264 (1972) Randall, L. L., Hardy, S. J. S.: Synthesis of exported proteins by membrane bound polysomes from Escherichia eoli. Eur. J. Biochem. 75, 4 3 - 5 3 (1977) Reeve, J. N., Groves, D. J., Clark, D. J. : Regulation of cell division in Escherichia coli; characterization of temperature sensitive division mutants. J. Bacteriol. 104, 1052-1064 (1970) Ryter, A., Shuman, H., Schwartz, M. : Integration of the receptor for bacteriophage lambda in the outer membrane of Escherichia co li, coupling with cell division. J. Bacteriol. 122, 295-301 (1975) Scatchard, G. : The attraction of proteins for small molecules and ions. Ann. N. Y. Acad. Sci. 51, 660-672 (1949) Schwartz, M., Kellerman, O., Szmelcman, S., Hazelbauer, G.: Further studies on the binding of maltose to the maltose-binding protein of Escherichia coli. Eur. J. Biochem. 71, 167-170 (1976) Shen, B. H. P., Boos, W.: Regulation of the/~-methylgalactoside transport system and the galactose-binding protein by the cell cycle of Escheriehia coli. Proc. Natl. Acad. Sci. USA 70, 1481 1485 (1973) Silhavy, T. J., Casadaban, M. J., Shuman, H. A., Beckwith, J. R.: Conversion of/%galactosidase to a membrane bound state by gene fusion. Proc. Nat. Acad. Sci. USA 73, 3423 - 3427 (1976a) Silhavy, T. J., Hartig-Beecken, I., Boos, W. : Periplasmic protein related to the sn-glycerol-3-phosphate transport system of Escherichia coli. J. Bacteriol. 126, 951-958 (1976b) Silhavy, T. J., Szmelcman, S., Boos, W., Schwartz, M. : On the significance of the retention of ligand by protein. Proc. Nat. Acad. Sci. U.S.A. 72, 2120-2124 (1975) Spurt, A. : A low viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastr. Res. 26, 3 1 - 4 5 (1969) Stock, J. B., Rauch, B., Roseman, S. : Periplasmic space in Salmonella typhimurium and Escherichia coli. J. Biol. Chem. 252, 78507861 (1977) Szmelcman, S., Hofnung, M. : Maltose transport in Escherichia coli K-12: Involvement of the bacteriophage lambda receptor. J. Bacteriol. 124, 112-118 (1975) Szmelcman, S., Schwartz, M., Silhavy, T. J., Boos, W.: Maltose transport in Escherichia coli K-t2. Eur. J. Biochem. 65, 1 3 - 1 9 (1976) Venable, J. H., Coggeshatl, R. : A simplified lead citrate stain for use in electron microscopy. J. Cell. Biol. 25, 407-409 (1965) Wetzel, B. K., Spicer, S. S., Dvorak, H. F., Heppel, L. A.: Cytochemical localization of certain phosphatases in Escherichia coli. J. Bacteriol. 104, 529-542 (1970) Wiesmeyer, H., Cohn, M. : The characterization of the pathway of maltose utilization by Escheriehia eoli. III. A description of the concentrating mechanism. Biochim. Biophys. Acta 39, 440 - 447 (1960) Received March 29, 1978
Micrebiolegy
Arch. Microbiol. 118, 207-218 (1978)
9 by Springer-Verlag 1978
Pole Cap Formation in Escherichia coli Following Induction of the Maltose-Binding Protein Ingrid Dietzel, Victoria Kolb, and Winfried Boos Department of Biology, University of Konstanz, P.O. Box 7733, D-7750 Konstanz, Federal Republic of Germany
Abstract. After induction with maltose, 3 0 - 40 % of the total protein in the osmotic shock fluid consist of maltose-binding protein while the induction ratio (maltose versus glycerol grown cells) for the amount of binding protein synthesized as well as for maltose transport is in the order of 10. Induction of maltose transport does not occur during all times of the cell cycle, but only shortly before cell division. Electronmicroscopic analysis of cells grown logarithmically on glycerol or maltose revealed in the latter the formation of large pole caps. These pole caps arise from an enlargement of the periplasmic space. Small cells contain one pole cap, large cells contain two. Pulse label studies with strain BUG-6, a mutant that is temperature sensitive for cell division reveal the following: Growth at the non-permissive temperature prevents maltose-binding protein synthesis and formation of new transport capacity. After shifting to the permissive temperature the cells regain both functions. Simultaneously, the newly formed cells exhibit pole caps. We conclude that the induction of maltose-binding protein is responsible for the formation of pole caps. In addition, beside the presence of inducer, cell cycle events occuring during division are necessary for the synthesis of maltose-binding protein. Key words: Periplasm - Maltose-binding protein Maltose transport - Cell division - Pole caps - Cell envelope - Escherichia coli.
Introduction In recent years several substrate-binding proteins have been isolated from the cell envelope from Escherichia Non Standard Abbreviations. GLPT = periplasmic protein, related to transport of glycerolphosphate in Escherichia coli (Silhavy et al., 1976b)
coli and Salmonella typhimurium by the classical osmotic shock procedure of Neu and Heppel (1965). Several lines of evidence indicate that these "shock proteins" are in fact periplasmic, i.e. located outside the cytoplasmic membrane, supposedly beneath the outer membrane (Heppel, 197 i). These binding proteins have been identified as the substrate recognition sites of a particular type of active transport system (Boos, 1974), and some of them as chemoreceptors (Hazelbauer, 1975). One of these transport systems that is specific for maltose and maltodextrins (Wiesmeyer and Cohn, 1960) has become very useful to study for several reasons: It is homogeneous, i.e. there is only one system in E. coli that transports maltose; it is genetically well understood and defined mutants of all gene loci are available (Hofnung, 1974); it uses the 2 receptor of the outer membrane to facilitate the diffusion of the substrate through this diffusion barrier (Szmelcman and Hofnung, 1975; Szmelcman et al., 1976); and its substrate recognition site, the maltose-binding protein (Kellerman and Szmelcman, 1974) is also a chemoreceptor (Hazelbauer, 1975). Two additionally required components that have been defined genetically have not been identified biochemically as yet, but there is strong evidence that at least one of them is a protein located within the cytoplasmic membrane (Silhavy et al., 1976a). The appearance of one of the components of the transport system, the receptor of phage 2, in the outer membrane has been found to be regulated by the cell cycle and according to this study it is clear that, during its in vivo assembly, this component is incorporated in the outer membrane only in certain regions of the bacterial cell surface (Ryter et al., 1975). The present publication deals with the alterations of the cell envelope of E. Coli that are due to the appearance of the maltose-binding protein in the periplasm after induction by maltose.
0302-8933/78/0118/0207/$ 02.40
208
Arch. Microbiol., Vol. 118 (1978)
Material and Methods
Pulse Labelling of the Maltose-Binding Protein
Bacterial Strains and Growth Conditions
From cultures growing logarithmically in minimal medium A with glycerol (0.4%) and maltose (0.2 %) as carbon source 5 ml aliquots were removed at different time intervals. These aliquots were further allowed to grow for 5 min at the same temperature in the presence of [:4C]-labelled amino acids (0.2 btM L-valin, 0.16 ~tM L-leucin and 0.16 ~M IAsoleucin). I n the aliquot of 5 ml each amino acid contributed about 5 . l0 s cpm. Then, a 1000-fold excess of nonlabelled amino acids was added and growth continued for 5 min. Then, chloramphenicol (150 gg/ml, final concentration) was added and the culture stored on ice. The cells were collected by centrifugation. They were washed with 5ml of 10mM Tris-HC1, pH 7.3 and suspended in 5 ml of the same buffer. A French-press extract was prepared as described above. The extract was lyophilized and resuspended in 50 pl of 10 mM Tris buffer containing maltosebinding protein (0.1 mg/ml). This suspension was used directly in the immunodiffusion technique as described (Lengeler et al., 1971). The stained and dried agar-plates were autoradiographed (Kodak Regulix Film) for 1 week.
Strain LA 3400 of E. eoli was used throughout this study. It was obtained by P1 transduction of strain 72 (A [glpR-malA], phoA) to glpR +, mal+,phoA, selecting on maltose (Silhavy et al., 1976b). Strain La 108 was a derivative of BUG-6 (Reeve et al., 1970) that was made phoR by P 1 transduction (Shen and Boos, 1973). This strain normally grows at 35~ but forms long filaments at 42~ The length of these filaments can become at least 50times the normal cell length. However, too long an exposure to 42~ initiates occasional cell division. As growth medium minimal medium A (Mille,r, 1972) was used with 0.4%glycerol or 0.2% maltose as carbon source. Growth temperature was 37 ~C. Induction of the maltose-transport system in cells grown on glycerol was done by the addition of 1 mM maltose (fmal concentration). Synchronization of strain LA 340(1 was done according to Cutler and Evans (1966). This method takes advantage of the tendency of cells to synchronize when they enter the stationary phase of growth. Then they were diluted 20-fold into prewarmed medium. This procedure was not always reproducible; effective synchronisation was obtained in the average of two out of three attempts. In addition, generally only 1 to 2 steps of synchronous growth could be observed. Synchronization was monitored by plating for viable counts on rich medium agar plates.
Transport of maltose was measured in the following way: 1 ml of the growing culture was removed and centrifuged for 2 min at 13000 X g (Eppendorf microcentrifuge), washed twice with minimal medium A and suspended in 1 ml of the same medium. To 100 gl of this suspension 20pl of 50pM [14C] maltose (3.9mCi/mMol, Amersham) were added. After 20 s, 100 pl were removed and filtered through Millipore filter (0.65 gm pore size). The filters were washed once with 10 ml minimal medium A, dried and counted in a toluene based scintillation fluid. All operations were performed at 37~C./~galactosidase activity was determined in toluenized cells as described by Miller (1972). The cold osmotic shock was performed according to Neu and Heppel (1965) using 900 ml culture. The number of cells in this culture was determined by plating for viable cells on rich medium agar plates. The obtained shock fluid was concentrated to about 10 ml by ultrafiltration using Amicon UM-10 filters. Subsequently, the concentrated solutions were dialyzed against 10 mM Tris, pH 7.3, over night. The slightly turbid solution was centrifuged at 100000 x g for 30 min and the supernatant used for determination of the maltosebinding protein and total protein concentration. To prepare cytoplasmic extracts the shocked cells were resuspended in 5 ml of 10 mM Tris, pH7.3 and passed twice through a French pressure cell at 600 atm at 0~ The suspension was centrifuged at 40000 x g for 1 b and the supernatant used for immunodiffusion test. Protein was measured according to Lowry et al. (1951) using serum albumin as standard. Galactose- and maltose-binding protein was estimated in the shock fluid by immunodiffusion technique using anti-galactose- and anti-maltose-binding protein antibodies as described previously (Lengeler et al., 1971). These antibodies were raised in rabbits using homogeneous binding proteins and complete Freud adjuvant. The amount of maltose-binding protein was determined by equilibrium dialysis as described (Schwartz et al:, 1976) using a st)bstrate concentration of 0.1 pM, well below the KD for maltose. 300 pl of shock fluid were put into small sacs of dialysis tubing (Union Carbide) and dialyzed against 10 ml buffer at 4 ~C over night. 100 gl aliquots were counted in a dioxan based scintillation fluid. In shock fluids of maltose grown cells the cpm's obtained from samples in the bag were routinely 2 - 3-fold higher than that of the outside buffer.
Two dimensional polyacrylamide gel electrophoresis was performed as described (Johnson et al., 1975) with modifications described later (Silhavy et al., 1976b).
Electron Microscopy. For negative staining one drop of a logarithmically growing culture was removed and deposited on a formvar and carbon coated grid. Staining was performed with 1% ammonium molybdate, pH 7.3 for 30 s. For embedding, growing cells were pelleted and fixed at 4~ for 4 0 - 6 0 min with 2.5 % glutaraldehyde in 0.l M sodium cacodylate buffer (pH 7.0) and postfixed for 1 - 2 h with 1% osmium tetroxide in the same buffer. Specimens were dehydrated with ethanol and embedded according to Spurr (1969). Sections were made with a Reichert ultramicrotomc OmU2. The thin seclions were stained with lead-citrate (Venable and Coggeshall, 1965). All preparations were examined with a Siemens Elmiskop 101 electron microscope.
Results Induction o f the Maltose-Binding Protein G r o w t h o n t h e E. coli w i l d - t y p e s t r a i n L A 3400 o n maltose results in the induction of a large amount of shock releasable maltose-binding protein. Figure l shows the comparison of crude shock fluids by two d i m e n s i o n a l p o l y a c r y l a m i d e gel e l e c t r o p h o r e s i s o f strain LA 3400 grown on glycerol (A) and on maltose (B). F r o m t h e s e gels it a p p e a r s t h a t a f t e r i n d u c t i o n b y maltose the osmotic shock fluid consists to a large extent of maltose-binding protein. Induction for the s y n t h e s i s o f o t h e r t r a n s p o r t r e l a t e d p r o t e i n s s u c h as t h e g a l a c t o s e - b i n d i n g p r o t e i n b y f u c o s e r e s u l t s in m u c h less dramatic increase of the corresponding protein (Silhavy et al., 1 9 7 6 b ) . A s i m i l a r " n o r m a l " i n d u c t i o n r a t i o c a n b e s e e n i n F i g u r e 1 A f o r G L P T ( s p o t 3) t h a t is specific a l l y i n d u c e d b y g l y c e r o l ( S i l h a v y e t al., 1 9 7 6 b ) . The determination of the total protein content of t h e r e l e a s e d p e r i p l a s m i c p r o t e i n s gives v a l u e s t h a t r a n g e f r o m 2 . 7 - - 3 . 2 m g p r o t e i n i n t h e s h o c k f l u i d f r o m i 0 t2
I. Dietzel et al. : Envelope Changes of E. coli Induced by Maltose
Fig. 1 A and B
Two-dimensional polyacrylamide gel electrophoresis of shock fluid of strain LA 3400 grown in the presence of glycerol (A) and maltose (B). The first dimension consists of electrophoresis in 8 M urea (pH 8.4), followed by electrophoresis in 0.2 ~ sodium dodecylsulfate (pH 6.48). The numbers correspond to the following proteins: I maltose-binding protein; 2 galactosebinding protein; 3 GLPT protein
209
210
Arch. Microbiol., Vol. 118 (1978)
glycerol grown cells and from 3.8-4.3 mg protein in the shock fluid from 1012 maltose grown cells. In addition, as seen in Figure 2, galactose-binding protein isolated from the same number of both types of
cells gives precipitin lines on immunodiffusion plates against anti-galactose-binding protein antibodies that are identical in position and strength. Thus, even though the total yield of shock proteins from one batch to another can vary to some extent and the determination of galactose-binding protein by the immunodiffusion technique only allows a semi-quantitative estimation it would appear that maltose-binding protein is induced in addition to the other periplasmic proteins (for instance the galactose-binding protein), i.e. the required space for these proteins is not limited. The assay for the maltose-binding protein in the crude shock fluid was done by equilibrium dialysis with [14C] maltose at substrate Concentrations below the KD for maltose-binding. Under these conditions the [PL] [P] Scatchard (1949) equation extrapolates to
[LFI [K.]
Fig. 2. Immunodiffusion of crude shock fluids against anti-galactosebinding protein antibodies. Concentrated shock fluids of strain LA3400 grown under the following conditions were used: 1 0.4% glycerol; 2 0.4~ glycerol containing 1 mM D-fucose (for induction of the galactose-binding protein); 30.2% maltose. The crude shock fluids were adjusted in protein concentration as to the identical amount of cells they were derived from. Average concentration was 1 mg/ml. 4 control, purified galactose-binding protein 0.02 mg/ml. The center well contained anti-galactose binding-protein antiserum
(Silhavy et al., 1975) where [P] equals total number of available binding sites, [PL]the protein ligand complex, and [LF] the free ligand concentration. Table I summarizes the results obtained by the binding test in two series of experiments. Substrate concentration in the binding test was 0.1 IxM and the KD for maltose at 4~ C temperature was taken to be 3.5 gM (Schwartz et al., 1976). As can be seen from Table 1, after induction with maltose, the maltose-binding protein comprises between 30 and 40 % of the total shock proteins, while the induction ratio is about 10-fold. With the number of cells in the culture one can calculate that a fully induced
Table 1. Maltose-binding test in crude shock fluid from cells grown in glycerol or maltose Experiment 1 Maltose grown cells
Experiment 2 Glycerol grown cells
Maltose grown cells
Glycerol grown cells
Total protein concentration" [mg/ml]
0.95
1.16
1.16
1.65
[PL]
2.1
0.2
3.3
0.2
1.6 J
0.3
3.9
0.3
7.2 0.29
1.0 0.04
12.1 0.49
1.0 0.04
MBP concentration [laM] [mg/ml] b % of the total protein in the crude shock fluid
30.6
3.6
42
2.4
Number of MBP molecules released per cell ~
22300
1500
24000
1100
" Total protein concentration was determined in the concentrated shock fluid by the method of Lowry et al. (1951) b The coricentration of maltose-binding protein in the crude shock fluid was calculated from the binding assay, using a molecular weight of 40000 (Kellerman and Szmelcman, 1974) ~ The number of maltose-bindingprotein molecules released per cell was calculated from the total amount of maltose-bindingprotein released and the number of cells present in the culture from which the protein was isolated
I. Dietzel et al. : Envelope Changes of E. coli Induced by Maltose
The Kinetic of Maltose Transport Induction in Logarithmically Growing non Synchronized Cells
Fig. 3 A and B
Induction of maltose transport and fl-galactosidase in a logarithmically growing culture of strain LA 3400. When the culture that had been growing on glycerol had reached an optical density of 0.12 (578nm) 10mM maltose and 0.i mM isopropyl-flthiogalactoside were added simultaneously, Maltose uptake is given in pmole substrate taken up per 100 gl culture in 20 s. fl-galactosidase units are given in pMole substrate per min
211
._~"
A
~s00
~ooo ~ ~ r""n ~o o ~ 500 ~ ~n
/ J
/
~_,
A__L~
o B wo
-~ ~ o -~
so
0
Q08 Q12 0.16 02
Q24, 028 0.32
cell mass [OD at 578 nm]
lO: 6
A
T1C E~ ~,o, ld
~__. (23
15
i
i
i
i
i
b
_ ~o.~_o_~_o
Fig. 4 A--D
Induction of maltose transport and fl-galactosidase activity in synchronized cells of strain LA 3400. The culture was synchronized by the dilution method of Cutler and Evans (1966). After dilution, at the time indicated by the arrow 10raM maltose and 0.1 mM isopropyl-fl-thiogalactoside were added simultaneously, and the following parameters were assayed: A cell mass; B viable counts; C maltose uptake given in pmole substrate taken up in 100~tl culture in 20 s; D fl-galactosidase activity
B/I i
The classical studies on the induction of fl-galactosidase in Escherichia coli have shown that after addition of the inducer, synthesis of/Lgalactosidase starts after a short lag period at a rate that is proportional to growth (Cohn and Monod, 1951). In other words, a straight line is obtained when the increase of enzymatic activity is plotted against the increase of cell mass or total protein. This is not the case when maltose transport is measured in a logarithmically growing culture after induction by maltose. Figure 3 shows the simultaneous induction in strain LA3400 of fl-galactosidase by isopropyl-thio-flgalactoside and of maltose transport by maltose. The addition of inducer was done at an optical density of 0.12 (578nm). As can be seen, in contrast to the induction of fl-galactosidase, maltose transport increases initially in a logarithmic fashion and only after an increase to about 150 % of the initial cell mass at the time of inducer addition, maltose transport increases linearly. This suggested that the increase in transport capacity might not occur during all phases of the cell cycle. Therefore, we assayed maltose transport after induction in synchronized cells.
i
i
i
c 2O
E
~
ol 1000
500 2q E
J D
/
/
/g
o
E
I o~o/ ~-L.~ 0 -20 ; 20 ~; ~'o 8'0 ,oo time[rain]
cell releases about 2 x 10~ molecules of maltosebinding protein during osmotic shock. Measuring the initial rate of maltose transport at a substrate concentration of 1 pM in cells grown in glycerol and in maltose, a 10-12-fold increase in transport activity can be observed (not shown).
Fig. 5. Cell division dependence of the maltose-binding protein in strain LA 108. Immunodiffusion of shock fluids and cytoplasmic extracts of strain LA 108 against anti-maltose-binding protein antibodies. 1 pure maltose-binding protein (0.1 mg/ml); 2 shock fluid of cells, grown at 42~ (2 mg protein/ml); 3 cytoplasmic extract of cells grown at 42~ 4 cytoplasmic extract of cells grown at 35~ 5 shock fluid of cells grown at 35~ (2mg protein/ml). The center well contains serum with anti-maltose-binding protein antibodies. The agar plate was stained with Coomassie blue
212
i
Arch. Microbiol., Vol. 118 (1978)
1.0
=
y
0.1
"~
j/ B
160 140
%, -6 E
120
I00
I
80
i =0 40
Fig.6A--C
20
~
/,.0
A &
T % 3.0
!
o 2.0
1.0
20 i
r
;
40
60 i
;
I00 t
;
I/-.0 i
I
180
80 120 160 t i m e [min]
Dependence of maltose transport on cell division in strain LA 108. Strain LA 108 was grown logarithmically at 35~ on 0.4% glycerol and 0.2% maltose to an optical density of 0.3 (at 578 nm). At time 0 half the culture was shifted to 42~ ( 9 the other half remained at 35~ (a A). At the time indicated the cultures were shifted to 35~ and 42~ respectively. The following parameters were measured: ACell mass (OD at 578 nm). B Maltose uptake, given in pmole maltose taken up by 100 ~tl culture in 20 s. C Cell numbers, as viable counts on nutrient plates
Induction of Maltose Transport in Synchronized Cells Cells o f strain L A 3400 were synchronized according to the m e t h o d o f Cutler and Evans (1966). Figure 4 shows the optical density, cell number, and maltose transport o f cells growing in glycerol after induction by maltose at time 0. As can be seen, transport capacity increases only shortly before or during cell division. W h e n the time o f maltose addition within the cell cycle is varied the result remains the same: transport increase only occurs shortly before or during cell division (not shown).
Induction of Maltose Transport Activity and of Maltose-Binding protein in BUG-6, a Mutant that is Temperature Sensitive in Cell Division Two cultures o f L A 108, a phoR derivative o f B U G - 6 , were g r o w n to late logarithmic phase in minimal
Fig.7A and B. Dependence of maltose-binding protein synthesis on cell division in strain LA 108. Cultures of strain LA 108 were pulse labelled with [14C]labelled amino acids as described in the text and under Material and Methods. A Coomassie blue stained immunodiffusion plate of the different extracts (unlabelled maltose-binding protein was added in each aliquot). BAutoradiogram of the dried agar plate shown in A, /control, no extract present; 215rain; 3 25 min; 4 35 rain; 5 45 rain; 6 55 min; 765 rain; 8 75 rain; 9 85 min; 1095rain. The culture was shifted from 42~ to 35~ at 50rain
m e d i u m A containing 0.4 % glycerol and 0.2 % maltose; one, at the permissive temperature o f 35 ~C, the other at 42 ~C. After washing free o f maltose and adjusting to a cell density o f O D (578 nm) 1.0 transport activity for maltose (initial rate of uptake) was 60-fold higher in cells grown at the permissive temperature than in cells grown at the non-permissive temperature. Light microscopic analysis revealed the presence of n o r m a l cells when g r o w n at 35~ and very long filaments when grown at 42 ~C. The periplasmic shock proteins as well as the total cytoplasmic extracts o f these cells were obtained and analyzed by immunodiffusion against anti-maltose-binding protein antibodies (Fig. 5). As can be seen this strain only produces maltose-binding protein when grown at the permissive temperature and it contains it exclusively in the periplasm. As control, we analyzed the growth m e d i u m for the presence o f maltose-binding protein after extensive dialysis against water and subsequent lyophilization. Cultures g r o w n at
I. Dietzel et al. : Envelope Changes of E. coli Induced by Maltose
213
Fig. 8 a - d Negatively stained cells after logarithmic growth in maltose (a, b) and glycerol (e, d). Cells grown in maltose show pole caps, large cells have two (a, arrows), small cells only one (b, arrow). Bars represent 0.5 ~m
both temperatures did not contain material in the medium that crossreacted against anti-maltose-binding protein antibodies. However, when cells were grown over night and were kept at the stationary phase of growth maltose-binding protein could be detected in the growth medium, indicating extensive cell lysis under these conditions. Normal wild-type behaved identical in this respect.
To elucidate further the induction of the maltose transport machinery in this strain the experiment shown in Figure 6 was performed. Cells were grown at 35~ to an optical density of 0.3 in minimal medium A containing 0.4 ~ glycerol and 0.2 ~ maltose. At time 0 half of the culture was shifted to 42~ while the other half was kept at 35~ After 50min the two cultures were shifted to 35~ and 42~ respectively. Viable
214 counts (C), uptake of maltose (B) and cell mass (A) was followed. As can be seen, increase of transport capacity only occurs at 35~ but not at 42~ Moreover, after shift to the permissive temperature the cells divide before transport increase occurs. A control experiment was performed, where chloramphenicol was added to the 42~ culture, at the time of the temperature shift. The cells still divided but no transport increase occurred (not shown). This excludes the possibility that an inactive precursor is formed at 42~ that might be processed after shift to the permissive temperature in the absence of protein synthesis. The latter conclusion is also substantiated by an experiment where maltose-binding protein is pulselabelled by radioactive amino acids. For this purpose the experiment shown in Figure 6 was repeated with a culture (OD0.3) growing at 42~ after time0 and at 35 ~C after 50 min. Aliquots were removed and grown in the presence of radioactively labelled amino acids at the respective temperature for 5 min before chasing with an excess of unlabelled amino acids. The cells were broken by French pressing. Pure maltose-binding protein was added to the cell extracts and immunodiffusion was performed (Fig. 7A). The stained and dried agar plate was subjected to autoradiography (Fig. 7b). Occurrence of radioactivity in the precipitin band indicates maltose-binding protein synthesis. As can be seen, synthesis of maltose-binding protein starts between 25 and 35m in after shift to the permissive temperature well in agreement with the time of increase in maltose transport activity (Fig. 6 B). Thus, it is clear that new synthesis of maltose-binding protein only occurs at the permissive temperature and after one round of division has passed. Apparently, no protein synthesis is necessary for cell division, after shift to the permissive temperature since it occurs in the presence of chloramphenicol (Reeve et al., 1970). In contrast, newly formed transport capacity and formation of maltosebinding protein is completely blocked under these conditions.
Ultrastructure Differences of the Periplasm in Maltose and Glycerol Grown Cells Specimens prepared by negative staining revealed that cells grown on maltose developed an enlargement of their periplasmic space (Fig. 8). These enlargements are analogous to the findings of Wetzel et al. (1970) on Escherichia coli derepressed for alkaline phosphatase and will accordingly be referred to as "pole caps". These pole caps are much less pronounced when the cells are grown in glycerol. The mean distance between 9 the cell wall and the cytoplasmic membrane in these pole caps was about two to four fold larger in maltose
Arch. Microbiol.,Vol. 118 (1978)
25
1
2O u
15-
Fig. 9
102
0
0.2 0.4 06 0.8 1.0 size of pole caps (l~m)
Distribution of the pole cap size in cells of strain LA 3400 grown on glycerol(INN) and maltose ( I ~ ). Pole cap size is given as the distancein jam of cell wall and cytoplasmic membrane at the cell poles
than in glycerol grown cells. The statistical distribution of this distance is shown in Figure 9. In addition, we observed that small, cells developed the pole cap only on one end of the cell, while large cells always developed them on both ends (Fig. 8). Pole caps could also be seen in embedded and ultrathin sectioned cells after growth on maltose but not on glycerol (not shown). In both types of cells the periplasmic space was filled with electron dense material. This pole cap formation cannot be an artifact due to the different osmotic effects of maltose and glycerol that are used as carbon source at 0.4~o and 0.2}/o concentrations. Addition of as little as 5 x 10-4M maltose to a culture growing on glycerol results in the same pole cap formation (not shown). Moreover, cells of strain LA3400 grown on glycerol and maltose respectively were plasmolized with 20 ~o sucrose. As can be seen in Figure 10, the effect of maltose induction can still be seen in plasmolized cells. This indicates that pole cap formation is not an artifact caused by plasmolysis. The observation that maltose induced cells formpole caps is not restricted to strain LA 3400. We observed large amounts of maltose-binding protein in the periplasm of all maltose metabolizing strains that we have tested so far. A particularly interesting example is LA 108. When this strain is grown at 42~ in glycerol no pole caps can be observed on the filamentous cells (Fig. 11). Shifting the culture to 35~ in the presence of 1 mM maltose lets the cells divide but after 20 rain no pole caps have formed (Fig. 11 B). After 70 rain, however, about 30 ~ of the cells have acquired pole caps (Fig. 1 IC,D). It is interesting to note that at this time (when transport of maltose is not yet fully induced) only part of the cells have fully developed pole caps, while others have no pole caps at all. This might indicate that induction of the maltose-binding protein is an all or none phenomenon. Wild-type ceils that had been growing over night in minimal medium A containing 0.2 ~o maltose and were maintained in the stationary phase of growth are smaller than logarithmically growing cells. They do not
I. Dietzel et al. : Envelope Changes of E.
coli
Induced by Maltose
215
Fig. 10a and b
Pole caps after plasmolysis. Strain LA 3400 was grown logarithmically in minimal medium A with 0.2 % maltose (a) or 0.4 ~ glycerol (b) as carbon source. The cells were plasmolized by the addition of 20 (w/v) sucrose for 2 - 4 rain and subsequently prepared for negative staining. Maltose grown cells always showed remnants of pole caps (-~) whereas glycerol grown cells only occasionally showed these structures; most of these cells had their plasma extended into both cell ends. Bars represent 0.2 pm
look significantly different than stationary phase cells that had been growing on 0.4 % glycerol (not shown).
Discussion
The presence of maltose in the growth medium leads to the appearance o f large amounts of maltose-binding protein in the osmotic shock fluid that is obtained by the classical cold osmotic shock procedure of Neu and Heppel (1965). Several lines of evidence indicate that these proteins are in fact periplasmic, that is outside the cytoplasmic membrane (Heppel, 1971).
F r o m the quantitative determination of the maltose-binding protein in these shock fluids it is clear that maltose-binding protein is excreted into the periplasm in addition to the other proteins already there. In other words, the total amount of protein in this space is increased after maltose induction. Simultaneously, after induction with maltose the cells form large pole caps. There is no definite p r o o f that these pole caps are filled with maltose-binding protein. Reaction product staining as performed by Wetzel et al. (1970) and by MacAllister et al. (1972) cannot be used with a protein that does not exhibit enzymatic activity.
216
Arch. Microbiol., Vol. 118 (1978)
Fig.llA--D. Pole cap formation in dependence of cell division in strain LAI08. Cells of strain LA108 were grown at 42~ for 2generations. Aliquots were removed and prepared for negative staining at different time intervals. A 10 min before temperature shift to 35~ B 20min after temperature shift. C and D 75rain after temperature shift. Only at the last time pole caps (~) are visible. Occasionally, cells with and without pole caps were found lying side by side (D)
Also, in contrast to our findings with the maltosebinding protein, alkaline phosphatase seems to distribute itself over the entire periplasmic space and not only the poles (MacAllister et al., 1972). In our hands derepression of alkaline phosphatase does not lead to formation of pole caps. Strain LA 108 carries a phoR mutation and synthesizes alkaline phosphatase constitutively. Yet, pole caps even in this strain are only observed after induction with maltose. It is very unlikely that these pole caps are artifacts of the preparation: the staining solution consisting of 1 a m m o n i u m molebdate had the same osmolarity as the growth medium. In addition, the pole caps were filled with electron dense material that is usually absent in
plasmolized cells. Also, plasmolized cells still retain poles that had formed due to maltose induction. It is not clear what the reason for the pole cap formation is. One possibility would be the excretion of maltosebinding protein (and other periplasmic binding proteins) exclusively on the poles of the cells. Alternatively, the cell wall of growing poles might be particularly yielding to the osmotic pressure created by periplasmic proteins (Stock et al., 1977). Accordingly, periplasmic proteins would be accumulating at the growing poles shortly before cell division. Since stationary cells are essentially not growing, pole caps are no longer formed. In this connection it is interesting that growing cells usually showed one pole cap when they are small but two
I. Dietzelet al. : EnvelopeChangesof E.
coli
Inducedby MaItose
when they are twice the normal cell length. This is possibly related to cell division and one would postulate that the two-pole caped cells were in a stage shortly before cell division while one-pole caped cells had just divided. Maltose transport in synchronized cells increases only shortly before or during cell division. Recently, it had been shown that the receptor for phage 2 is synthesized only during cell division (Ryter et al., 1975). This protein plays an important role in maltose transport (Szmelcman and Hofnung, 1975) by catabolizing the diffusion through the outer membrane (Szmelcman et al., 1976). Therefore, its cell cycle dependent synthesis might be the reason for the corresponding
217
transport of maltose. However, as we have shown in this paper cell division is also a prerequisit for the synthesis of maltose-binding protein, an essential component of maltose transport (Kellerman and Szmelcman, 1974). The morphological evidence only indicates accumulation of periplasmic material at the pole caps. We were as yet unable to obtain morphological evidence for the excretion of periplasmic material at the septum of dividing cells. Few other transport systems have been found to be regulated by events that occur during the bacterial cell cycle (Ohki, 1972; Shen and Boos, 1973). At the present time, it is not clear what the molecular basis of this cell division dependence is. Periplasmic proteins including
218
the maltose-binding protein can be synthesized in cell free systems (Randall and Hardy, 1977). This might indicate that synthesis of these proteins is rather repressed during the cell's elongation than actively stimulated during cell division. Acknowledgements. We wish to thank Miss M. Winter for technical assistance. A sample of pure maltose-binding protein was obtained from Dr. Ferenci. This work was supported by the Deutsche Forschungsgemeinschaft, Sonderforschungsbereich 138 "Biologische Grenzfl~ichen und Spezifit~it", by the North Atlantic Treaty Organization and the "Fonds der Deutschen Chemischen Industrie".
References Boos, W.: Bacterial transport. Annu. Rev. Biochem. 43, 123-146 (1974) Chon, M., Monod, J. : Purification et propri6t6s de la/%galactosidase (lactase) d'Escherichia coli. Biochim. Biophys. Acta 7, 153 - 174 (1951) Cutler, R. G., Evans, J. E.: Synchronization of bacteria by a stationary phase method. J. Bacteriol. 9 1 , 4 6 9 - 4 7 6 (1966) Hazelbauer, G. L. : Maltose chemoreceptor of Escherichia coli. J. Bacteriol. 122, 206-214 (1975) Heppel, L. A. : The concept of periplasmic enzymes. In: Structure and function of biological membranes (L. A. Rothfield, ed.), pp. 223-247. New York: Londong. 1971 Hofnung, M. : Divergent operons and the genetic structure of the maltose B region in Eseherichia coli K-12. Genetics 76, 169-- 184 (1974) Johnson, W. C., Silhavy, T. J., Boos, W. :Two-dimensional polyacrylamide gel electrophoresis of envelope proteins of Escherichia coli. Appl. Microbiol. 29, 405-413 (1975) Kellerman, O., Szmelcman, S. : Active transport of maltose in Escherichia coli K-12. Eur. J. Biochem. 47, 139-149 (1974) Lengeler, J., Hermann, K. O., Uns61d, H. J., Boos, W.: The regulation of the/~-methylgalactoside transport system and of the galactose-binding protein of Escherichia coli K-12. Eur. J. Biochem. 19, 457-470 (1971) Lowry, O. H., Rosebrough, N. J., Farr, A. L., Randall, R. J. : Protein measurements with the folin phenol reagent. J. Biol. Chem. 193, 265-275 (1951) MacAllister, T. J., Costerton, J. W., Thompson, L., Thompson, J., Ingrain, J. M. : Distribution of alkaline phosphatase within the periplasmic space of gram negative bacteria. J. Bacteriol. 111, 827--832 (1972) Miller, J. H. (ed.) : Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York (1972)
Arch. Microbiol., Vol. 118 (1978) Neu, H. C., Heppel, L. A. : The release of enzymes from Escherichia coli by osmotic shock and during the formation of spheroplasts. J. Biol. Chem. 240, 3685-3692 (1965) Ohki, M. : Correlation between metabolism of phosphotidylglycerol and membrane synthesis in Eseherichia coli. J. Mol. Biol. 68, 249-264 (1972) Randall, L. L., Hardy, S. J. S.: Synthesis of exported proteins by membrane bound polysomes from Escherichia eoli. Eur. J. Biochem. 75, 4 3 - 5 3 (1977) Reeve, J. N., Groves, D. J., Clark, D. J. : Regulation of cell division in Escherichia coli; characterization of temperature sensitive division mutants. J. Bacteriol. 104, 1052-1064 (1970) Ryter, A., Shuman, H., Schwartz, M. : Integration of the receptor for bacteriophage lambda in the outer membrane of Escherichia co li, coupling with cell division. J. Bacteriol. 122, 295-301 (1975) Scatchard, G. : The attraction of proteins for small molecules and ions. Ann. N. Y. Acad. Sci. 51, 660-672 (1949) Schwartz, M., Kellerman, O., Szmelcman, S., Hazelbauer, G.: Further studies on the binding of maltose to the maltose-binding protein of Escherichia coli. Eur. J. Biochem. 71, 167-170 (1976) Shen, B. H. P., Boos, W.: Regulation of the/~-methylgalactoside transport system and the galactose-binding protein by the cell cycle of Escheriehia coli. Proc. Natl. Acad. Sci. USA 70, 1481 1485 (1973) Silhavy, T. J., Casadaban, M. J., Shuman, H. A., Beckwith, J. R.: Conversion of/%galactosidase to a membrane bound state by gene fusion. Proc. Nat. Acad. Sci. USA 73, 3423 - 3427 (1976a) Silhavy, T. J., Hartig-Beecken, I., Boos, W. : Periplasmic protein related to the sn-glycerol-3-phosphate transport system of Escherichia coli. J. Bacteriol. 126, 951-958 (1976b) Silhavy, T. J., Szmelcman, S., Boos, W., Schwartz, M. : On the significance of the retention of ligand by protein. Proc. Nat. Acad. Sci. U.S.A. 72, 2120-2124 (1975) Spurt, A. : A low viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastr. Res. 26, 3 1 - 4 5 (1969) Stock, J. B., Rauch, B., Roseman, S. : Periplasmic space in Salmonella typhimurium and Escherichia coli. J. Biol. Chem. 252, 78507861 (1977) Szmelcman, S., Hofnung, M. : Maltose transport in Escherichia coli K-12: Involvement of the bacteriophage lambda receptor. J. Bacteriol. 124, 112-118 (1975) Szmelcman, S., Schwartz, M., Silhavy, T. J., Boos, W.: Maltose transport in Escherichia coli K-t2. Eur. J. Biochem. 65, 1 3 - 1 9 (1976) Venable, J. H., Coggeshatl, R. : A simplified lead citrate stain for use in electron microscopy. J. Cell. Biol. 25, 407-409 (1965) Wetzel, B. K., Spicer, S. S., Dvorak, H. F., Heppel, L. A.: Cytochemical localization of certain phosphatases in Escherichia coli. J. Bacteriol. 104, 529-542 (1970) Wiesmeyer, H., Cohn, M. : The characterization of the pathway of maltose utilization by Escheriehia eoli. III. A description of the concentrating mechanism. Biochim. Biophys. Acta 39, 440 - 447 (1960) Received March 29, 1978
