|
|
|
|||||||
|
|||||||||
(Received for publication, August 30, 1994; and in revised form, December 12,
1994) From the
The mannose transporter complex acts by a mechanism which
couples translocation with phosphorylation of the substrate. It
consists of a hydrophilic subunit (IIAB A new plasmid pTSHIC9 for the controlled
overexpression of the cytoplasmic phosphoryl carrier proteins, enzyme
I, HPr, and IIA
Uptake of hexoses and hexitols in bacteria is mediated by
membrane protein complexes, the so-called enzymes II of the bacterial
phosphotransferase system (PTS). ( The mannose transporter of the PTS has a
broad substrate specificity for mannose, glucose, and related hexoses.
In addition, it acts as a host factor required for infection of E.
coli by bacteriophage
Figure 1:
Hypothetical model of the mannose
transporter and schematic representation of its action. A,
vectorial transport and phosphorylation. Translocation and
phosphorylation are tightly coupled. The transported solute (
Due to the complexity
of the system comprising two cytoplasmic phosphoryl carrier proteins
(enzyme I and HPr) in addition to the transporter, sugar
phosphorylation rather than vectorial transport activity is measured in
routine assays (Kundig and Roseman, 1971). Of all the purified PTS
transporters only the mannitol transporter could be reconstituted into
phospholipid vesicles, so far (Elferink et al., 1990). The
mannose transporter differs from the mannitol transporter in subunit
composition and stoichiometry, chemistry of the active site, and
protein structure. The molecular mechanism of vectorial solute
transport and its coupling to phosphorylation are not understood. As a
first step toward the elucidation of this process the mannose
transporter was reconstituted into phospholipid vesicles and the
vectorial transport and phosphorylation activity was measured. Also
addressed was the question of substrate channeling (Srivastava and
Bernhard, 1986; Srere, 1987; Welch and Easterby, 1994) between the
transport and phosphorylation moieties, the hypotheses of transport
regulation by the membrane potential (Reider et al., 1979;
Robillard and Konings, 1981, 1982), and the proposition of a coupling
of solute uptake with proton extrusion (Scarborough, 1985). To provide
the soluble phosphoryl carrier proteins enzyme I and HPr, which are
required in large amounts for the loading of proteoliposomes, a
recombinant plasmid for their overexpression was constructed and
methods for their purification are described.
In parallel experiments the membrane potential formed by
K
Figure 2:
Plasmid map of pTSHIC9 (A) and
Coomassie-stained polyacrylamide gels of enzyme I, HPr, and IIA
Figure 3:
Proteolysis of the
IIC
Import of [
Figure 4:
Uptake and phosphorylation of glucose. A and B, IIC
Observations with intact bacteria indicated that PTS
transporters not only catalyze vectorial transport with phosphorylation
but also phosphorylation of substrates in the cytoplasm (nonvectorial
phosphorylation, e.g. of glucose generated in the cell from
maltose). When [
Figure 5:
Nonvectorial phosphorylation of glucose. A, P-enolpyruvate and the phosphoryl carrier proteins are
added to the outside of IIC
This nonvectorial phosphorylation
raises two questions. First, passive diffusion of solute followed by
nonvectorial phosphorylation could in principle be misinterpreted as
true vectorial phosphorylation in the reconstituted system. Second,
transport coupled to phosphorylation and nonvectorial phosphorylation
could be competing reactions. The two problems were addressed by
measuring the efflux of [ To test for passive and
IIC
Figure 6:
Nonspecific carry over (adsorption) and
passive diffusion of Glc out of proteoliposomes. Proteoliposomes were
loaded with [
Figure 7:
Export
and phosphorylation of glucose. IIC
Are nonvectorial
phosphorylation, as described above (Fig. 5) and vectorial
transport and/or phosphorylation of transported substrate (Fig. 7) competing reactions? To address this question export
and phosphorylation of trapped [
The purified mannose transporter of the bacterial
phosphotransferase system was reconstituted by detergent dilution into
proteoliposomes and two activities, vectorial transport concomitant
with phosphorylation and nonvectorial phosphorylation were measured.
Transport is coupled to phosphorylation in a 1:1 ratio. The K Although nonvectorial
phosphorylation on the one hand and phosphorylation coupled with solute
translocation on the other hand are mediated by the same protein
complex, no cross-inhibition between the two reactions could be
observed. Neither the export of [ The results
summarized above differ in three respects from recent observations with
the mannitol transporter which indicated facilitated diffusion,
exchange, and uncoupling of transport and phosphorylation. (i) Purified
IICBA The proteolysis data are not easily reconciled with the
proposition of Beneski et al.(1982) that the II The present
reconstitution method is useful whenever nonvectorial phosphorylation
and vectorial transport must be investigated in parallel and when
interference with PTS activity by other membrane components has to be
excluded. It will be used as an assay complementary to in vivo uptake studies and in vitro measurement of nonvectorial
phosphorylation.
Volume 270,
Number 10,
Issue of March 10, 1995 pp. 5258-5265
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
) and two
transmembrane subunits (IIC, IID). The
purified complex was reconstituted into phospholipid vesicles by octyl
glucoside dilution. Glucose export was measured with proteoliposomes
which were loaded with radiolabeled glucose and to which purified
IIAB, cytoplasmic phosphorylcarrier proteins, and
P-enolpyruvate were added from the outside. Vectorial transport was
accompanied by stoichiometric phosphorylation of the transported sugar.
Glucose added to the outside of the proteoliposomes was also
phosphorylated rapidly but did not compete with vectorial export and
phosphorylation of internal glucose. Glucose uptake was measured with
proteoliposomes which were loaded with the cytoplasmic phosphoryl
carrier proteins and P-enolpyruvate and to which glucose was added from
the outside. Vectorial import and phosphorylation occurred with a
higher specificity (K
30 ± 6
µM, k
401 ± 32 pmol of
Glc/µg of IICD/min) than nonvectorial phosphorylation (K 201 ± 43 µM, k 975 ± 88 pmol of Glc/µg of
IICD/min).
, and a simplified procedure for the
purification of these proteins are also described.
)They couple vectorial
translocation with phosphorylation of the transported solute. Coupling
is tight unlike in eukaryotic cells where transport and phosphorylation
of glucose are mediated by a sequential pair of enzymes (transporter
and kinase). However, phosphorylation of cytoplasmic substrates without
transport, e.g. of glucose derived from maltose within the
cell, has also been observed (Thompson and Chassy, 1985; Thompson et al., 1985; Nuoffer et al., 1988). There exist 16
enzymes II of different substrate specificity in Escherichia
coli. They are composed of three functional units (IIA, IIB, IIC,
or IICD; for nomenclature see Saier and Reizer(1992)) which occur
either as protein subunits or domains of a multidomain protein. The IIC
unit spans the membrane several times. It contains the substrate
binding site. The IIA and IIB units are hydrophilic. They sequentially
transfer phosphoryl groups from the ``high-energy''
phosphoryl carrier protein phospho-HPr to the substrate on the IIC
subunit. Phospho-HPr in turn is regenerated by P-enolpyruvate in a
reaction catalyzed by enzyme I. The collective of phosphoproteins which
have additional functions in metabolite transport, chemotaxis, and
metabolic regulation constitute the bacterial phosphotransferase system
(for reviews, see Meadow et al.(1990), Erni(1992), and Postma et al.(1993)).
(Elliott and Arber, 1978). It consists
of three subunits (Fig. 1). The IIC and
IID subunits form a tight transmembrane complex which is
sufficient for penetration of phage DNA (Erni et al.,
1987). For transport and phosphorylation of sugars, the third subunit,
IIAB, is also required. IIAB is reversibly
associated with the IIC-IID complex (K
5-10 nM). (
)It consists of two hydrophilic domains IIA and IIB which are linked through a hinge
peptide-rich in Ala and Pro (Erni, 1989). IIA and
IIB each contain one phosphorylation site (His
and His
) which relay the phosphoryl groups from HPr
to the substrate on the IIC-IID complex.
The isolated IIB domain together with the IIC-IID complex is sufficient for equilibrium phosphoryl exchange between
Glc-6-P and Glc (Erni et al., 1989).
Glc) does not exchange with free solute (Glc). B,
nonvectorial phosphorylation. Thin lines indicate the flow of
phosphoryl groups from P-HPr via the IIA domain (dotted) and the IIB domain (diagonal
hatching) to Glc. The two domains of IIAB are linked
by an Ala-Pro-rich hinge. One IIC subunit (vertical
hatching) and two IID subunits (horizontal
hatching) form the transmembrane
complex.
Bacterial Strains and Construction of Expression
Plasmids
E. coli K-12 WA2127 is ptsG
manX
YZ.
WA2127
HIC is ptsGmanXYZptsHptsIcrr.
It was constructed by conjugative transfer into WA2127 of the ptsH
ptsI crr deletion from strain UE7 (Hfr KL16 thi (ptsH
ptsI crr) galR galP::Tn10; gift of W. Boos, University of
Konstanz). Plasmid pTSPM8 encodes manY and manZ under
the control of Ptac (Erni et al., 1987). The
recombinant plasmid pTSHIC9 carries behind the Ptac promoter ptsH, ptsI, and crr encoding in this order
HPr, enzyme I, and the IIA subunit of the glucose
transporter. pTSHIC9 was constructed from the expression vector
pJF119HE (Fürste et al., 1986) and plasmid
pDIA3206 (contains 11 kilobase E. coli chromosomal DNA
carrying ptsH ptsI crr; de Reuse and Danchin(1988)) as
follows. (i) A 9.1-kilobase SalI-BamHI fragment
containing ptsH ptsI crr from pDIA3206 was inserted into the
polylinker of pJF119HE. (ii) Extra DNA at the 3` end of the ptsH
ptsI crr coding region was deleted by BamHI and partial HpaI digestion of the plasmid followed by self-ligation of the
desired 9.4-kilobase fragment. (iii) 5` noncoding chromosomal DNA was
deleted unidirectionally starting from the polylinker toward the 5` end
of ptsH (SalI and SphI digestion in the
polylinker, unidirectional deletion with ExoIII nuclease for different
time intervals, trimming with S1 nuclease, and self-ligation of the
truncated fragments). (iv) Strain WA2127HIC was transformed with
the self-ligated plasmids, plated on McConkey indicator plates (0.4%
Glc, 100 µg/ml ampicillin), and colonies which weakly fermented Glc
(complementation of IICB dependent Glc transport) were
selected. (v) The selected colonies were replica plated on IPTG
containing plates, and non-fermenting colonies were selected (strong
overexpression of ptsH ptsI crr was found to inhibit Glc
fermentation for unknown reasons). (vi) Cell lysates from
IPTG-sensitive colonies were analyzed on polyacrylamide gels for
overexpression of enzyme I, HPr, and IIA by Coomassie
staining and autoradiography following incubation of the extracts with
[
P]P-enolpyruvate. The plasmid from a
transformant displaying maximal overexpression was isolated and the
noncoding region upstream of ptsH was sequenced by the dideoxy
chain termination method using the Sequenase
kit (U. S.
Biochemical Corp.).Overexpression and Purification of the
IIC
The IIC-IID Complex and of the IIAB Subunit-IID complex
was overexpressed in E. coli WA2127 transformed with plasmid
pTSPM8 (Erni et al., 1987). Cells were grown at 37 °C to A
= 1.5. Protein expression was induced
with 50 µM IPTG and incubation continued for 5 h. The
IIC-IID complex was solubilized with 2%
octyl-
-glucopyranoside (Sigma) and purified by isoelectric
focusing in a U-tube in the presence of the same detergent (Erni et
al., 1987). The IIC-IID complex from
the peak fraction was concentrated in a Centricon 10 (Amicon) to 1 mg
of protein/ml. The concentrated protein solution was used for
reconstitution experiments. Protein concentrations were determined
using the assay kit of Bio-Rad and bovine serum albumin as standard.
The wild-type IIAB subunit was purified as described
(Stolz et al., 1993).Overexpression and Purification of Enzyme I, HPr, and
IIA
Enzyme I and HPr were overexpressed in E.
coli WA2127 transformed with plasmid pTSHIC9. Cells were grown at
37 °C to A = 1.5, protein expression
was induced with 0.5 mM IPTG, and incubation continued
overnight. Cells were collected by centrifugation, resuspended in a
buffer A (20 mM Tris-HCl, pH 7.5, 1 mM EDTA, 0.5
mM DTT, and 100 µg/ml phenylmethylsulfonyl fluoride), and
ruptured by two passages through a French pressure cell (15,000 psi).
Cell debris were sedimented by low speed centrifugation (10 min, 12,000
g) and the membrane fraction by ultracentrifugation
(90 min, 150,000 g). Protamine sulfate was added to
the high-speed supernatant to a final concentration of 0.33%. After
removal of the precipitate by low speed centrifugation, enzyme I,
IIA, and HPr account for 50% of protein in the clarified
supernatant. The supernatant was applied to a DE-52 column (Whatman,
70-ml bed volume, equilibrated with buffer A), and eluted with a
0-700 mM NaCl gradient (buffer A, 2 ml/min flow rate).
HPr eluted between 50 and 80 mM NaCl, IIA between 0.20 and 0.25 M NaCl, and enzyme I between 0.4
and 0.45 M NaCl. The HPr, IIA, and enzyme I
containing fractions were pooled and concentrated by precipitation with
80% ammonium sulfate. The precipitates were dissolved in a small volume
of buffer A and dialyzed against buffer A. IIA and HPr
were further purified by gel filtration on Superdex 75 (Pharmacia
Biotech Inc.), and enzyme I on Superdex 200. Protein concentrations
were determined using a modified Lowry assay (Markwell et al.,
1978) and bovine serum albumin as standard.Reconstitution of the IIC
E. coliL--IID Complex
-phosphatidylethanolamine (Type IX, Sigma) was
dissolved in chloroform, dried in a round-bottom flask, and resuspended
at a concentration of 20 mg/ml in reconstitution buffer (20 mM imidazole/HCl, pH 6.0, 150 mM KCl, 1 mM MgCl, 1 mM DTT). The suspension was
homogenized by brief sonication in a bath-type sonicator and stored at
-70 °C until use. 1 ml of phospholipid suspension (20 mg/ml)
and 0.11 ml of 10% octyl--D-glucopyranoside (Sigma) were
added to a screw capped test tube, flushed with N, and
vortexed for 30 s at room temperature. 0.05 ml of the isoelectrofocused
IIC-IID complex (approximately 50 µg of
protein) was added to the phospholipid/detergent solution. The solution
was mixed gently, incubated for 1 h at 4 °C, pipetted into 30 ml of
reconstitution buffer, mixed, and proteoliposomes were collected by
centrifugation (1 h, 150,000 g, 4 °C). The
proteoliposomes were resuspended in 0.5 ml of phosphate buffer (50
mM KP
, pH 7.5, 5 mM MgCl, 1
mM DTT, 40 mg/ml phospholipids, 100 µg/ml
IIC-IID; molar lipid/protein ratio 30,000)
and then stored at -70 °C until use.Assay for Vectorial Import and Phosphorylation of
Glc
The frozen proteoliposomes were thawed at room temperature.
P-enolpyruvate and purified enzyme I, HPr, and IIAB were
added to 250 µl of concentrated proteoliposomes to final
concentrations of 4.5 mM, 0.3 µM, 2.5
µM, and 0.7 µM, respectively. The mixture was
sonicated for 2 min at room temperature in a bath-type sonicator
(TEC-40, TECSONIC, Switzerland). The sonicated proteoliposomes were
frozen in liquid N and thawed in a 22 °C water bath.
Freezing and thawing was repeated 8 times. After sonication for 3 s in
a bath-type sonicator, the proteoliposomes were purified by spin column
gel filtration (Chonn et al.(1991). 1-ml tuberculin syringes
plugged with glass wool were filled with Sephacryl S-300 (High
Resolution, Pharmacia) equilibrated with buffer (50 mM KP, pH 7.5, 5 mM MgCl, 1 mM DTT) and hanging in 13 100-mm test tubes centrifuged for 2
min at 2000 rpm. Fills and centrifugations were repeated until the bed
volume was 1 ml. A final spin of 5 min at 2000 rpm assured that the gel
was uniformly packed and that excess buffer was removed. 80-µl
aliquots of proteoliposomes were then applied and spun (1 min, at 1000
rpm). To drive the elution of proteoliposomes to completion, the column
was washed with 80 µl of phosphate buffer. The proteoliposomes were
diluted 4-fold into phosphate buffer to a final volume of 2 ml. The
import reaction was started by adding [
C]Glc
(DuPont NEN, diluted to 10 mM, 4.5 mCi/mmol) to the desired
concentration. Aliquots of 100 µl were withdrawn at the indicated
time points and diluted into 2 ml of buffer (50 mM KP, pH 7.5, 5 mM MgCl, 1 mM DTT). To measure active import the diluted aliquots were
immediately filtered through glass microfiber filters (GF/F, Whatman)
under suction, the filters washed with 2 1 ml of phosphate
buffer, and the radioactivity retained on the filters determined by
liquid scintillation counting. To measure Glc phosphorylation, aliquots
were diluted into 2 ml of 0.1% Triton X-100 and Glc-6-P was separated
from Glc by anion-exchange chromatography as described (Erni et
al., 1982). The kinetic constants were obtained by nonlinear
fitting of the Michaelis-Menten equation to data points.Assay for Vectorial Export and Phosphorylation, and
Nonvectorial Phosphorylation of Glc
100 µl of D-[U-C]Glc (DuPont NEN, diluted to 10
mM, 4.5 mCi/mmol) was added to 250 µl of concentrated
proteoliposomes. The mixture was sonicated and freeze-thawed as
described above. To remove extravesicular [C]Glc
the suspension was incubated for 15 min at 22 °C with hexokinase
(4.5 mg/ml) and ATP (4.0 mM) and then spin column purified and
4-fold diluted as described above (5 mg/ml phospholipid, 0.18
µM IIC-IID). Enzyme I, HPr,
and IIAB (final concentrations of 0.3, 2.5, and 0.7
µM) were added to the proteoliposomes and at 22 °C the
transport reaction was started by adding P-enolpyruvate to a final
concentration of 4.5 mM. Aliquots of 100 µl were withdrawn
and [C]Glc retained in the proteoliposomes as
well as [C]Glc-6-P formed were determined as
described above. Proteoliposomes mock-loaded with Glc-free buffer were
used to measure nonvectorial phosphorylation of external Glc. The
reaction was started by adding enzyme I, HPr, IIAB, and
P-enolpyruvate together with [U-C]Glc to the
outside. Glc phosphorylation was measured by the ion-exchange method
(Kundig and Roseman, 1971; Erni et al., 1982). To measure
competition between vectorial export of internal
[C]Glc and nonvectorial phosphorylation of
external Glc, [
C]Glc was added together with
P-enolpyruvate. The kinetic constants were obtained by nonlinear
fitting of the Michaelis-Menten equation to data points.Generation of K
The proteoliposomes collected by ultracentrifugation
were suspended in buffer A (150 mM KP Diffusion
Potential, pH 7.5, 5
mM MgSO
, 1 mM DTT) and loaded with
[C]Glc as described. External K was removed together with external [C]Glc
by spin column gel filtration through Sephacryl 300 equilibrated in
buffer B (149 mM NaP, pH 7.5, 1 mM KP, 5 mM MgSO, 1 mM DTT). In this experiment, SO
was
used to replace Cl in order to minimize the passive
anion permeation. SO has no effect on
the phosphorylation activity of the mannose transporter (data not
shown). The K diffusion potential was started by
adding valinomycin (Sigma) to a final concentration of 40 nM.
5 min after the addition of valinomycin, the PTS-dependent extrusion of
[C]Glc was started by adding P-enolpyruvate. -efflux from K-loaded
proteoliposomes was measured with the fluorescence indicator
1,3,3,1`,3`,3`-hexamethylindoldicarbocyanine (NK529, Nippon Kankoh
Shikiso Kenkyusho, Okayama, Japan) as described by Apell et
al.(1985). 170 µl of K containing purified
proteoliposomes (without [C]Glc) and 530 µl
of buffer (149 mM NaP, pH 7.5, 1 mM KP) containing 37 µM NK529 (added from
2.5 mM stock solution in 1:9 (v/v) ethanol/water) were mixed
in a 1-ml cuvette. After the fluorescence signal reached a constant
value (F
), valinomycin was added to a final concentration
of 40 nM whereupon the fluorescence signal rapidly decreased
to a new constant value (F). The K concentration was
increased by small increments and the fluorescence change (F
= F-F) was monitored. There is a linear relationship
between the membrane potential, which was calculated from the Nernst
equation, and the fluorescence change: 0.64 F/mV. The fluorescence
measurements were carried out with a Perkin-Elmer LS-5B luminescence
spectrometer. The excitation wavelength was set to 620 nm (slit width 5
mm) and the emission wavelength to 680 nm (slit width 5 mm).Protease Treatment of the Proteoliposomes
40
µl of spin column-purified proteoliposomes were incubated with the
indicated amounts of trypsin (Sigma) or subtilisin (Serva) in potassium
phosphate buffer (50 mM KP, 5 mM MgCl, 1 mM DTT, pH 7.5) with and without 2.7%
Triton X-100 at room temperature in a final volume of 60 µl.
Proteolysis was stopped by adding phenylmethylsulfonyl fluoride (3.5
mM, final concentration) at the indicated time intervals.
Proteins were precipitated with 15% trichloroacetic acid, washed with
acetone to remove detergent and phospholipid, and treated with formic
acid/ethanol (1:2, v/v) to completely unfold the IIC subunit. The lyophilized protein samples were solubilized in 25
µl of sample buffer and separated by polyacrylamide gel
electrophoresis. The proteins were stained with alkaline silver (Wray et al., 1981).
Overexpression and Purification of Enzyme I, HPr, and
IIA
To maximize protein expression and minimize
plasmid loss during fermentation, the pts operon encoding
enzyme I, HPr, and IIA was put under the control of the Ptac promoter. The optimal distance between Ptac and
the coding region was determined by progressively truncating the
noncoding region upstream of ptsH and the screening of
transformants for protein overexpression. Plasmid pTSHIC9 (Fig. 2A) supporting the strongest overexpression was
isolated and the 5` region between ptsH and Ptac was
sequenced. It comprises 148 noncoding nucleotides. It ends 16
nucleotides downstream of the -10 region of the pts P1
promoter (de Reuse et al., 1992). Enzyme I is the most
prominent of all E. coli proteins in whole cell lysates, and
IIA and HPr also appear as prominent bands on a
Coomassie-stained electropherogram (Fig. 2B, lane 1).
The proteins were purified by procedures adapted from the Roseman
laboratory (Kukuruzinska et al., 1982; Nakazawa and Weigel,
1982; Meadow, 1982). After only two steps of purification the three
proteins appear homogeneous as judged by polyacrylamide gel
elecrophoresis (Fig. 2B), and 90% of the PTS activity
present in the cytoplasmic fraction is recovered (Schunk, 1992).
However, the IIA preparation was cross-contaminated with
HPr activity as indicated by the strong background of PTS activity of
an assay mixture from which exogenous HPr was omitted. This residual
HPr activity could be completely removed by two extra steps of
IIA purification: anion exchange chromatography on Mono-Q
(Pharmacia) and gel filtration on Superdex 75. No mutual
cross-contamination of enzyme I and HPr and no contamination of these
proteins by IIA could be detected using elevated
background activity in the sugar phosphotransferase assay as the
criterium.
(B). Lane 1, whole cell lysate from
IPTG-induced cells; lane 2, purified HPr; lane 3,
purified enzyme I; lane 4, purified IIA (the
double band represents full-length and NH terminally
processed forms of IIA (Meadow et al., 1986)).
The molecular mass markers are phosphorylase b (94 kDa),
bovine serum albumin (68 kDa), ovalbumin (43 kDa), carbonic anhydrase
(30 kDa), soybean trypsin inhibitor (20.1 kDa), and -lactalbumin
(14.4 kDa). The 8-25% polyacrylamide gel was prepared according
to Fling and Gregerson (1986).
Preparation and Physical Characterization of
Proteoliposomes
Proteoliposomes were formed by dilution of the
purified, -octyl glucoside solubilized
IIC-IID complex into a detergent-free
buffer containing a 30,000-fold molar excess of E. coli phospholipids. 80-90% of the PTS activity present in the
IIC-IID preparation could be recovered in
the sedimented proteoliposome preparation. Detergent dialysis also
afforded proteoliposomes but their IIC-IID activity was below 20%. The uranyl acetate-stained vesicles had
diameters of between 200 and 500 nm (results not shown). The
orientation of IIC-IID in the membrane is
random as concluded from proteolysis experiments. Treatment of
proteoliposomes with subtilisin resulted in a 50% conversion of the
29-kDa IID subunit into a new 28-kDa fragment (Fig. 3A). Trypsin removed 50% of the IID (Fig. 3B) from the intact vesicles. No tryptic
fragments of IID could be detected by alkaline silver
stain (Fig. 3B) and immune blots (results not shown).
When the proteoliposomes were solubilized with detergent prior to
proteolysis, conversion of the IID subunit was complete (Fig. 3). The IIC subunit behaved differently. It
was cleaved into smaller fragments by subtilisin and trypsin, but
cleavage did not go to completion after detergent solubilization of the
proteoliposomes (Fig. 3). The reason for this behavior is not
clear. After detergent solubilization, 50% of (nonvectorial)
phosphotransferase activity could be recovered from the trypsin-treated
proteoliposomes. Only 20% activity could be recovered from the
subtilisin-treated proteoliposomes, suggesting that the
IIC-IID complex is accessible to subtilisin
in both orientations. This results in more than 50% inactivation but
cleavage from only one side afforded a product of different
electrophoretic mobility.
/IID complex in proteoliposomes. A, 50% of the IID (D) subunits are
processed to an intermediate (D*) upon incubation of intact
proteoliposomes with subtilisin (S), close to 100% after
solubilization of proteoliposomes with Triton X-100. Digestion of
IIC (C) to C* is not complete. B, 50% of the IID (D) subunits are
processed upon incubation of intact proteoliposomes with trypsin (T). No distinct IID intermediate is formed and
digestion of IIC (C) to C* is not
complete. The 20% polyacrylamide gels were stained with alkaline
silver.
Transport and Phosphorylation Activity
The
IIC-IID complex is oriented randomly in the
proteoliposomes. However, by adding the soluble phosphoryl carrier
proteins and P-enolpyruvate to only one of the two compartments a
unique direction of transport can be imposed upon the reconstituted
system. Uptake can be measured if the proteoliposomes are loaded with
the phosphoryl carrier proteins and [C]Glc is
added to the outside (``right-side out''), efflux if they are
loaded with [C]Glc and the soluble
phosphotransferase proteins are added to the outside
(``inside-out''). The inside-out orientation allows for
manipulation of protein-protein interactions occurring at the
cytoplasmic face of the membrane but is of limited use for kinetic
studies because intravesicular glucose is quickly depleted. The
right-side out orientation allows assay for time-dependent solute
accumulation but intravesicular conditions are not easily manipulated. C]Glc was measured with
proteoliposomes which were loaded with purified soluble PTS proteins
and P-enolpyruvate. [C]Glc was added to start
the uptake reaction (Fig. 4). 80% of the glucose originally
present at a concentration of 5 µM was concentrated in the
proteoliposomes and accumulation was accompanied by phosphorylation (Fig. 4A). The k and K
of vectorial transport by proteoliposomes were
401 ± 32 pmol of Glc/µg of
IIC-IIDD/min and 30 ± 6
µM, respectively (Fig. 4, B and C). Assuming that the molecular mass of the functional unit of
the mannose transporter is 90 kDa and that only 50% are correctly
oriented and therefore active, a turnover number of 1.2 s
can be calculated. No uptake of [C]Glc
could be detected, when P-enolpyruvate, enzyme I, HPr,
IIAB, and [C]Glc were added to the
outside of the proteoliposomes (results not shown), indicating that
transport occurs only if the solute ([C]Glc) and
the phosphoryl transfer components are on opposite sides of the
membrane.
-IID containing proteoliposomes are loaded with IIAB,
the phosphoryl carrier proteins, and P-enolpyruvate. The uptake
reaction is started (arrow) by addition of
[C]Glc to the outside. Open symbols,
uptake; closed symbols, phosphorylation. A background of 5
pmol of Glc has been subtracted. A, 0.1 µM IIC-IID,
IIC-IID:phospholipid ratio =
1:30,000, 5 µM Glc (
, 
), 50 µM Glc (
, 
). B, 0.04 µM IIC-IID,
IIC-IID:phospholipid ratio 1:75,000, 3.8
µM (), 7.5 µM (
), 15 µM (
), 28 µM (
), 56 µM (), and 112 µM Glc (
). C,
Lineweaver-Burk plot of initial rates of uptake activity from B. k = 1.2 ± 0.1
s and K = 30
± 6 µM. Shown are means and S.D. of three
independent transport experiments done with a single batch of
reconstituted proteoliposomes.
C]Glc is added together with the
soluble phosphoryl carrier proteins to the outside of
IIC-IID containing proteoliposomes, Glc-6-P
is formed rapidly without concommitant transport (Fig. 5A). The k and K of nonvectorial phosphorylation by
proteoliposomes are 975 ± 88 pmol of Glc-6-P/µg of
IIC-IID/min and 201 ± 43
µM, respectively (Fig. 5B).
Phosphorylation without transport is therefore 2-3 times faster
than vectorial phosphorylation.
-IID containing
proteoliposomes (0.1 µM IIC-IID,
IIC-IID:phospholipid ratio 1:30,000). The
reaction was started by addition of [C]Glc to
15.1 µM (), 38.5 µM (), 76.3
µM (), 151.0 µM (), 298.5
µM (), and 579.0 µM (
) final
concentration. B, Lineweaver-Burk plot of initial rates of
phosphorylation activity from A. k = 2.9 ± 0.3 s and K = 201 ± 43
µM. Shown are means and S.D. of three independent
experiments done with a single batch of reconstituted
proteoliposomes.
C]Glc in the presence
and absence of nonlabeled external glucose.-IID-mediated facilitated diffusion,
(proteo)liposomes were loaded with [C]Glc and
purified by spin column chromatography (without the hexokinase
treatment described under ``Materials and Methods''). By
adding hexokinase and ATP, a constant amount of Glc could be converted
to Glc-6-P in the external compartment of pure liposomes as well as of
proteoliposomes containing the IIC-IID complex. This amount did not change when the soluble phosphoryl
carrier proteins were added as long as either P-enolpyruvate or the
IIAB subunit were omitted (Fig. 6). The amount of
Glc-6-P did, however, increase immediately above this background when
P-enolpyruvate (Fig. 7B) was added, and this increase
was accompanied by a stoichiometric decrease of trapped glucose (Fig. 7A). This result indicates that (i) export of
[C]Glc concomitant with phosphorylation is
catalyzed only by the complete phosphotransferase system, (ii) passive
and IIC-IID-dependent facilitated diffusion
of [C]Glc do not occur on the time scale of up
to 1 h, and (iii) a small amount of [C]Glc which
was not removed by spin column purification alone remains accessible
from the outside. Henceforth, residual nontrapped glucose was removed
with hexokinase prior to spin column purification as described under
``Materials and Methods.'' When IIAB, enzyme I,
HPr, and P-enolpyruvate were added to the hexokinase-treated and spin
column-purified proteoliposomes, rapid export of trapped
[C]Glc concomitant with phosphorylation could be
measured (Fig. 7). Export of glucose and phosphorylation are
strictly coupled in a 1:1 molar ratio.
C]Glc and purified by spin column
centrifugation (time 0). Hexokinase (1 mg/ml) and ATP (4.0 mM)
were added after 20 min (arrow) and the formation of Glc-6-P
was monitored by ion exchange chromatography. Pure liposomes without
IIC-IID (); proteoliposomes with
IIAB, enzyme I, HPr, but without P-enolpyruvate ()
proteoliposomes with enzyme I, HPr, P-enolpyruvate but without
IIAB (
). A constant amount of Glc-6-P is formed
from external glucose that does not further increase with
time.
-IID containing proteoliposomes (0.1 µM IIC-IID,
IIC-IID:phospholipid ratio 1:30,000) are
loaded with [C]Glc (2.85 mM). The
export reaction is started (arrow) by addition to the outside
of IIAB, HPr, enzyme I, and P-enolpyruvate without
(, ) and with (, ) 17.5 mM unlabeled
Glc. A, export of trapped [C]Glc. B, vectorial phosphorylation, formation of Glc-6-P. External
glucose does not compete with export and phosphorylation of
[C]Glc (compare triangles and circles). , diffusion of [C]Glc
observed when P-enolpyruvate is omitted from the incubation mixture.
Shown are means and S.D. of three independent experiments done with a
single batch of reconstituted
proteoliposomes.
C]Glc was
measured in the presences of increasing amounts of external
[C]Glc. As shown in Fig. 7, a 6-fold
molar excess of external [C]Glc neither slows
down the rate of [C]Glc export (Fig. 7A) nor its phosphorylation (Fig. 7B). This indicates that (i) nonvectorial
phosphorylation and vectorial transport are not competing reactions and
(ii) that [C]Glc does not equilibrate with
[C]Glc between the translocation and
phosphorylation reaction and coupling between these two steps therefore
must be tight (Fig. 1). No difference was observed whether
[C]Glc was added simultaneously with the
cytoplasmic phosphoryl carrier proteins (Fig. 7) or whether it
was added before (results not shown). Reconstituted
IIC-IID does catalyze exchange
equilibration between internal and external glucose.Phosphotransferase Activity and Membrane
Potential
It has been proposed that the activity of PTS
transporters is regulated by the transmembrane electrochemical
potential (Robillard and Konings, 1981, 1982). The rate and extent of
solute accumulation by intact cells increased when the proton
electrochemical gradient was abolished by uncouplers or respiratory
chain inhibitors (Reider et al., 1979; Singh et al.,
1985; Nuoffer et al., 1988). To rationalize this inverse
correlation between membrane potential and transport activity, a
mechanism involving the release of a proton on the periplasmic side
concomitant with sugar phosphorylation was suggested (Scarborough,
1985). Reconstitution of the purified mannose transporter allowed
testing for a possibly direct effect of the membrane potential on the
transport activity. The extrusion from and the import of
[C]Glc into K-loaded
proteoliposomes was measured in the absence and in the presence of
valinomycin. The calculated K diffusion potential was
120 mV, positive outside. According to the predictions the transport
activity should have increased in the extrusion experiment (negative
polarization with respect to the direction of transport, inside-out
orientation) and decreased in the uptake experiment (positive
polarization, right-side out orientation). However, in neither case
could a difference be detected (results not shown). Similarly the
addition of the uncoupler dinitrophenol had no effect on extrusion or
uptake of [C]Glc.
of the transport reaction is 30 ± 6
µM. A turnover number of 1.2 ± 0.1 s can be calculated, assuming that the minimal functional unit
consists of one IIC and two IID subunits
(Rhiel et al., 1994), and that at most 50% of the complexes
are correctly oriented and active (probably an overestimate). The
turnover number of the PTS transporter for mannitol is 4 s (Elferink et al., 1990) and similar numbers can be
calculated from reported experiments with the purified
H/lactose transporter (3.5 s,
Newman et al.(1981); 0.6 s, Consler et
al.(1993)), the intestinal Na/glucose transporter
(4 s, Peerce and Clarke(1990)), and the histidine
transporter (1.1 min, Bishop et al.(1989)).
Nonvectorial phosphorylation has an approximately 7-fold higher K and a 3-fold faster k than the transport reaction. It is likely that solute
translocation requires more extensive protein motion than the binding
and release reaction during nonvectorial phosphorylation, and that the
former reaction therefore is slower than the latter. A similar
difference of catalytic constants was also observed with the mannitol
transporter (Elferink et al., 1990).C]Glc out of
proteoliposomes nor phosphorylation of the exported solute could be
inhibited by Glc added to the outside of the proteoliposomes.
Noncompetition between nonvectorial and vectorial phosphorylation
indicates that the transported glucose is not in diffusion equilibrium
with glucose in the bulk phase. Transport and phosphorylation are
either mechanistically coupled or the rate of phosphorylation of bound
glucose is much faster than the rate by which unphosphorylated glucose
could dissociate (dissociation being a precondition for exchange; Fig. 1). This is a clear example of tight channeling by a
multifunctional enzyme which catalyzes two sequential reactions
(Srivastava and Bernhard, 1986; Srere, 1987). No competition between
translocation and nonvectorial phosphorylation could be detected within
the experimental error of the export assay. However, the initial rate
of export could not be measured accurately because export is fast and
due to the small intravesicular volume of very short duration. Small
differences might therefore have gone unnoticed.
reconstituted into proteoliposomes mediated
facilitated diffusion in a way which was saturable and specific for
mannitol (Elferink et al., 1990). (ii) Addition of unlabeled
Mtl to the exterior of inside-out membrane vesicles resulted in the
complete exchange of [
H]Mtl bound to the interior
(Lolkema et al., 1991). (iii) Over 50% of the radiolabeled
mannitol bound to the periplasmic side of IICBA containing inside-out oriented membrane vesicles exchanged with
unlabeled mannitol before it became phosphorylated, indicating that
mannitol dissociates after translocation at a rate comparable to that
of phosphorylation. The differences could have several reasons. (i) The
two transporters are very different with respect to amino acid
sequence, active site residues (histidine versus cysteine in
the second phosphorylation site; Pas and Robillard(1988), Erni et
al.(1989), and Pas et al.(1991)), and subunit composition
(4 domains in 3 subunits versus 3 domains in 1 subunit) and
the molecular mechanism of their action might be correspondingly
different. (ii) Facilitated diffusion catalyzed by the
IIC-IID complex went unnoticed because it
was indistinguishable from passive diffusion of L-glucose and
diffusion of D-glucose out of protein-free liposomes (results
not shown). complex is symmetrically oriented in the membrane. They observed
that II interacts with 2-deoxyglucose and phosphorylate
this sugar when the phosphoryl carrier proteins are located either
inside of right-side out membrane vesicles (resulting in uptake and
phosphorylation) or outside the vesicles (nonvectorial
phosphorylation). If the two surfaces exposed toward the aqueous phases
were identical, proteolysis should proceed to 100% and not only 50% as
observed (Fig. 3). However, functional symmetry found by Beneski et al.(1982) does not necessitate a structural symmetry of the
complex. Both faces of an asymmetrical complex could in principle have
independent sugar phosphorylation activity, and noncompetition between
vectorial transport and nonvectorial phosphorylation could then be
explained. However, it appears more likely to us that some
randomization of sidedness might have occurred during preparation of
membrane vesicles by Beneski et al.(1982).
) and IID, transmembrane subunits
of the mannose transporter; IIAB, hydrophilic subunit of
the mannose transporter; IICBA, mannitol transporter;
HPr, histidine-containing phosphocarrier protein of the PTS;
IICB, transmembrane subunit of the glucose transporter;
IIA, cytoplasmic subunit of the glucose transporter; ptsH ptsI crr, genes coding for HPr, enzyme I and
IIA; ptsG, gene encoding IICB;
IPTG, isopropyl--D-thiogalactopyranoside; DTT,
dithiothreitol.)
We thank R. Lanz for expert technical assistance, Dr.
C. Nuoffer for the construction of plasmid pTSHIC9, Drs. H. de Reuse
and A. Danchin for plasmid pDIA3206, Prof. W. Boos for strain UE7, and
Prof. H. Oetliker for assistance in measuring of the membrane
potential. Electron microscopy was performed by T. Wyler.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Hu, K. Hu, D. C. Williams Jr., M. E. Komlosh, M. Cai, and G. M. Clore Solution NMR Structures of Productive and Non-productive Complexes between the A and B Domains of the Cytoplasmic Subunit of the Mannose Transporter of the Escherichia coli Phosphotransferase System J. Biol. Chem., April 18, 2008; 283(16): 11024 - 11037. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. B. Diep, M. Skaugen, Z. Salehian, H. Holo, and I. F. Nes Common mechanisms of target cell recognition and immunity for class II bacteriocins PNAS, February 13, 2007; 104(7): 2384 - 2389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Bachler, K. Flukiger-Bruhwiler, P. Schneider, P. Bahler, and B. Erni From ATP as Substrate to ADP as Coenzyme: FUNCTIONAL EVOLUTION OF THE NUCLEOTIDE BINDING SUBUNIT OF DIHYDROXYACETONE KINASES J. Biol. Chem., May 6, 2005; 280(18): 18321 - 18325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Aboulwafa and M. H. Saier Jr. Characterization of Soluble Enzyme II Complexes of the Escherichia coli Phosphotransferase System J. Bacteriol., December 15, 2004; 186(24): 8453 - 8462. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. F. Garcia-Alles, K. Flukiger, J. Hewel, R. Gutknecht, C. Siebold, S. Schurch, and B. Erni Mechanism-based Inhibition of Enzyme I of the Escherichia coli Phosphotransferase System. CYSTEINE 502 IS AN ESSENTIAL RESIDUE J. Biol. Chem., February 22, 2002; 277(9): 6934 - 6942. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Gutknecht, K. Flukiger, R. Lanz, and B. Erni Mechanism of Phosphoryl Transfer in the Dimeric IIABMan Subunit of the Escherichia coli Mannose Transporter J. Biol. Chem., March 5, 1999; 274(10): 6091 - 6096. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Gutknecht, M. Manni, Q. Mao, and B. Erni The Glucose Transporter of Escherichia coli with Circularly Permuted Domains Is Active in Vivo and in Vitro J. Biol. Chem., October 2, 1998; 273(40): 25745 - 25750. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Gutknecht, R. Lanz, and B. Erni Mutational Analysis of Invariant Arginines in the IIABMan Subunit of the Escherichia coli Phosphotransferase System J. Biol. Chem., May 15, 1998; 273(20): 12234 - 12238. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Lanz and B. Erni The Glucose Transporter of the Escherichia coli Phosphotransferase System. MUTANT ANALYSIS OF THE INVARIANT ARGININES, HISTIDINES, AND DOMAIN LINKER J. Biol. Chem., May 15, 1998; 273(20): 12239 - 12243. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Bouma and S. Roseman Sugar Transport by the Marine Chitinolytic Bacterium Vibrio furnissii. MOLECULAR CLONING AND ANALYSIS OF THE MANNOSE/GLUCOSE PERMEASE J. Biol. Chem., December 27, 1996; 271(52): 33468 - 33475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Mukhija and B. Erni Purification by Ni2+ Affinity Chromatography, and Functional Reconstitution of the Transporter for N-Acetylglucosamine of Escherichia coli J. Biol. Chem., June 21, 1996; 271(25): 14819 - 14824. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Mao, T. Schunk, B. Gerber, and B. Erni A String of Enzymes, Purification and Characterization of a Fusion Protein Comprising the Four Subunits of the Glucose Phosphotransferase System of Escherichia coli J. Biol. Chem., August 4, 1995; 270(31): 18295 - 18300. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Mao, R. G. Deeley, and S. P. C. Cole Functional Reconstitution of Substrate Transport by Purified Multidrug Resistance Protein MRP1 (ABCC1) in Phospholipid Vesicles J. Biol. Chem., October 27, 2000; 275(44): 34166 - 34172. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |