|
|
|
|||||||
|
|||||||||
(Received for publication, January 18, 1996) From the
Digestion of vesicle-bound SecA by trypsin entrapped within the
vesicles showed that refolding precursor ribose-binding protein (pRBP)
of Escherichia coli retards the lipid bilayer penetration by
SecA while the signal peptide enhances it. This discrepancy was found
to be due to reduced SecA binding to the vesicles in the presence of
the pRBP while the signal peptide induced a tight binding. Studies on
the binding of 1-anilino-8-naphthalene sulfonate (ANS) to SecA
indicated that SecA assumes more closed conformation upon interaction
with pRBP and signal peptide induces more open structure of SecA.
Kinetic studies of ANS binding to SecA upon dilution of unfolded pRBP
with SecA solution showed an initial fast ANS binding, which was
followed by a slow release of ANS. This suggests that first the signal
peptide portion of the pRBP binds with the SecA making its structure
more open and then the subsequent binding of the mature domain makes
the SecA structure more compact. The pRBP enhanced the digestion of
SecA added to the E. coli inverted vesicles, suggesting an
inhibition of SecA penetration while the signal peptide had an opposite
effect, agreeing with the results from the model systems above. When
the pRBP and ATP were present together, however, the penetration of
SecA increased dramatically underlining the importance of the SecY/E
complex for the membrane insertion of SecA.
Proteins translocated across the plasma membrane of Escherichia coli are synthesized as precursors with
amino-terminal signal peptides which contain the information for the
membrane targeting. It has been demonstrated that signal peptides
interact with many components of the export machineries (Akita et
al., 1990; Altman et al., 1990; Bieker et al.,
1990). Targeting of various precursor proteins to the plasma membrane
of E. coil seems to be carried out through two routes (Wolin,
1994). One subgroup of precursor proteins, such as maltose-binding
protein, are post-translationally translocated and initially binds with
SecB. Proteins of another subgroup, including the ribose-binding
protein (RBP), ( SecA protein, which has a pivotal role
among a number of protein components of the translocation machinery, is
an ATPase (Lill et al., 1989) and displays some unusual
physical properties. In particular, it can exist as a water-soluble
form, a peripheral protein, and an integral protein (Cabelli et
al., 1991). The basal ATPase activity is stimulated by its
interaction with signal peptide and the mature domain of the precursor
proteins, negatively charged phospholipids, and SecY/E complex (Lill et al., 1990). Since most of the ATPases involved in the
membrane transport or locomotion have ATP/ADP cycle closely linked to
their cyclic change in conformation (and also membrane topology for the
integral protein ATPases), it has been speculated that the ATP
hydrolysis cycle of the SecA is also coupled to its cyclic
conformational and topological changes, which are closely linked to the
translocation of precursor proteins (Economou and Wickner, 1994; Kim et al., 1994; Ahn and Kim, 1994). In a preliminary
communication (Ahn and Kim, 1994), we reported that the SecA protein
traverses the lipid bilayer and its membrane topology depends on the
kind of the nucleotide present, indicating the possibility of topology
change coupled to the ATP/ADP cycle of the SecA. Here, we extended this
study to investigate the effect of the precursor RBP (pRBP) and its
signal peptide on the membrane topology of SecA. Since the isolated
wild-type signal peptide has a low solubility in water, we used a
revertant signal peptide (SF). The first mutation of L(-17)P in
the wild-type signal peptide abolishes the translocation, but an
additional mutation within the signal peptide of S(-15)F to form
the SF peptide largely recovers the translocation capability (Park et al., 1988). The structural studies showed that both
wild-type and revertant peptides have a high propensity for
For the
iodide quenching of Trp fluorescence, a series of KI solutions with the
concentration up to 0.1 M were added to the reaction mixtures
to obtain the same total concentration of KI plus KCl but different KI
concentration. Changes in the exposure of hydrophobic patches on the
SecA upon interaction with either refolding pRBP or the signal peptides
were also monitored with fluorescence from 200 µM ANS,
which was added to reaction samples. The fluorescence emission caused
by excitation at 370 nm was measured between 420 and 600 nm.
The kinetic progress of
conformational change of SecA brought about by refolding pRBP was also
studied by time-dependent SecA binding to phenyl-Sepharose. 0.1
µM
Figure 1:
Effects of pRBP and SF signal peptide
on the topology of SecA in the lipid bilayer. SecA was incubated (1 h,
30 °C) with phospholipid vesicles entrapped with trypsin, trypsin
inhibitor, and either unfolded pRBP or signal peptide SF. The digestion
products were analyzed by SDS-PAGE and stained with Coomassie Brilliant
Blue.
An
appreciable reduction in the density of the intact SecA can be seen
when signal peptide SF was present (lane b), indicating an
enhanced penetration of SecA. Here, we used a revertant signal peptide
in lieu of the wild-type signal peptide because the latter peptide is
insoluble in buffer solutions. Compared with the case when SecA alone
is present (lane a), there is no change in the band positions
but the density of the 30-kDa band increased while those of the other
fragments remained about the same. A Met-rich 25-residue peptide of the
Ffh protein of E. coli, which showed a higher propensity of
forming amphiphilic A dramatic
reduction in the digestion of SecA was seen when refolding pRBP
interacted with SecA (lane e). The thick band around 30 kDa is
the pRBP. It is possible that a small amount of 30-kDa fragment of SecA
may be also present. In any case, it is clear that unfolded pRBP
reduced the lipid bilayer crossing of SecA, unlike the signal peptide,
which increased the membrane penetration of SecA. The refolding mRBP (lane h) and folded pRBP (data not shown) had no effect. It is
interesting that the refolding mature domain of pRBP overshadowed the
effect of the signal peptide, which enhances the bilayer penetration of
the SecA. It is likely that the pRBP forms a complex with SecA while it
is being folded. The presence of either ATP (lane f) or
ATP The retardation of
SecA penetration by refolding pRBP was also demonstrated by the
reduction of the fluorescence quenching of the Trp in SecA by iodide
entrapped within the vesicles. SecA has seven Trp, but there is no Trp
in pRBP. Fig. 2(line a) gives the Trp fluorescence
spectrum of SecA bound to vesicles, and line b shows the
quenching by vesicle-entrapped iodide. These have also been shown in a
previous report (Ahn and Kim, 1994). When refolding pRBP is also
present, a tremendous reduction of quenching occurred (line d)
even after an allowance for the increase in the fluorescence intensity
when pRBP interacts with SecA bound to vesicles without iodide was made (line c). It is clear that the pRBP reduces the SecA crossing
of the lipid bilayer, thus lessening the exposure of Trp to the iodide
present within the internal space of the vesicles.
Figure 2:
Effect
of pRBP on the quenching of SecA Trp fluorescence by iodide entrapped
within vesicle. 0.4 µM SecA was incubated with 0.5 mM (phospholipids) of vesicles entrapped with iodide for 15 min at 30
°C. 1.2 µM unfolded pRBP was added to this mixture and
further incubated for 15 min. Trp fluorescence emission spectra of SecA
were obtained with 295 nm excitation wavelength. I
Figure 3:
Inhibition of SecA binding to liposomes by
pRBP.
Fig. 4shows that the bilayer bound SecA is harder to extract
with high concentration of KCl or urea in the presence of the signal
peptide. From these binding results, we may conclude that the reduced
amount of SecA fragments in the presence of refolding pRBP (Fig. 1) originated from the inhibitory effect of pRBP on the
SecA binding to vesicles.
Figure 4:
The extractability of membrane-bound SecA
protein. Membrane-bound
Figure 5:
Intrinsic tryptophan fluorescence spectra
of SecA. All solutions contained 0.2 µM native SecA.
Solutions with added pRBP or signal peptide are indicated. SecA was
incubated with either 0.8 M pRBP or signal peptide for 1 h at
30 °C prior to the tryptophan fluorescence emission spectra were
recorded as described under ``Experimental Procedures.'' The
final concentration (0.02 M) of GdnHCl was adjusted to be the
same in all reaction samples.
Fig. 6gives quenching
of the SecA Trp fluorescence by iodide in the presence of either
refolding pRBP or the SF signal peptide. The fluorescence intensity at
the emission wavelength of 338 nm was measured. The fluorescence data
are plotted according to the Stern-Volmer equation (Eftink and Ghiron,
1981), where F
Figure 6:
Quenching of tryptophan fluorescence of
SecA by iodide. 0.5 µM free SecA (open circles),
in the presence of 2.0 µM refolding pRBP (filled
triangles) or 1.0 µM signal peptide (open
squares) was prepared and titrated with KI. F
The K
Figure 7:
Fluorescence emission spectra of ANS bound
to SecA. 1 µM SecA alone (solid line), with
refolding pRBP (dashed line), or with signal peptide (dotted line) was incubated for 1 h and then ANS was added.
Excitation wavelength was 370 nm, and all other conditions were the
same as described in Fig. 5.
Figure 8:
Kinetic progress curves of ANS binding to
SecA as measured by the fluorescence at 470 nm (at 30 °C). Unfolded
pRBP or signal peptide and ANS solution were mixed with SecA solution
before the measurement was started. Line a, SecA alone; line b, SecA + pRBP; line c, SecA + signal
peptide.
The kinetic progress profile of the
SecA binding to phenyl-Sepharose in the presence of refolding pRBP
qualitatively agrees with the above kinetics of ANS binding to SecA.
Upon dilution of unfolded pRBP with SecA solution, about 55% of SecA
was bound to phenyl-Sepharose initially within the dead time of mixing (Fig. 9), then SecA was released reaching to 37% after about 40
s. When only SecA was present, about 48% of the total SecA was bound to
phenyl-Sepharose which did not change with time. This result indicates
that the initial interaction of unfolded pRBP with SecA induced
significant increase in the exposure of hydrophobic sites on SecA. But
the hydrophobicity of SecA decreased with a time scale similar to that
of the release of ANS from SecA (Fig. 8, line b).
Figure 9:
Effect of unfolded pRBP on the binding of
SecA to phenyl-Sepharose.
Figure 10:
Thermal transition of SecA. Change in the
fluorescence intensity at 338 nm were recorded as a function of
temperature. First, unfolded pRBP was refolded by rapid dilution with
SecA solution and incubated for 30 min at 30 °C. The solution was
then cooled to 17 °C before starting the thermal transition
measurements. Excitation was made at 295 nm. The concentrations of SecA
and pRBP were 0.2 and 0.8 µM, respectively. T
Figure 11:
ATP binding of SecA. 2 µM SecA in 20 mM Tris-HCl (pH 7.3) was incubated with ATP
containing [
Figure 12:
Immunoblot analysis of protease K digests
of SecA. SecA was mixed with IMVs prepared from E. coli CP626.
pRBP unfolded in 1 M GdnHCl or signal peptide was added to
this solution and incubated for 20 min at 30 °C. Samples were
treated with proteinase K for 15 min at 30 °C and analyzed by
SDS-PAGE, followed by immunoblotting with anti-SecA antiserum. The
positions of molecular weight markers and SecA (98 kDa) are indicated
by arrows. The control is obtained by immunoblotting of the
SDS-PAGE gel of the IMV without the presence of added protein. The
total protein concentration (w/v) was adjusted to be the same in all
samples by adding the appropriate amount of carbonic
anhydrase.
When refolding pRBP and ATP
were added together, however, the digestion pattern was reversed as
shown in Fig. 13. When only SecA was present with IMV and
proteinase K was added, a faint band of 67 kDa plus the intact SecA
band could be seen. When these were mixed with unfolded pRBP and ATP,
the bands became more dense and three additional bands with smaller
molecular weight were also produced. Apparently, the combined effect of
pRBP and ATP brought about an increase in SecA penetration into IMV
membrane. It should also be noticed that all these bands disappeared
after a long period (more than 20 min) of digestion, demonstrating the
SecA penetration into IMV membrane is a reversible process.
Figure 13:
Proteinase K digestion of IMV-bound SecA
in the presence of ATP and refolding pRBP. ATP and unfolded pRBP was
added to the solution containing IMV-bound SecA and proteinase K
digestion was performed for the indicated time period at 30 °C. The
rest of the procedures were the same as described in Fig. 12.
The importance of interaction of SecA with anionic
phospholipids of E. coli for translocation of preprotein as
well as in vivo ATPase activities of SecA has already been
shown (Lill et al., 1990; de Vrije et al., 1988;
Kusters et al., 1991). The extent of in vivo SecA
binding to inner membrane of E. coli was found to be dependent
upon the presence of anionic phospholipids (Lill et al.,
1990). SecA interaction with model membrane of acidic phospholipid was
indirectly demonstrated by an increased SecA ATPase activity (Lill et al., 1990) and by the susceptibility of SecA to
staphylococcal protease V8 (Shinkai et al., 1991) when
liposome was added to SecA solution. Breukink et al. (1992)
also showed that the penetration of SecA into monolayer containing
negatively charged lipids is influenced by binding and hydrolysis of
ATP. The actual demonstration of SecA penetration into lipid bilayer
was observed by Ulbrandt et al.(1992) using fluorescence
quenching as a means of determining penetration depths. A deep
penetration at least the depth of outer monolayer was demonstrated, but
it was not possible to decide whether SecA traverses the bilayer. They
also observed a partial unfolding of SecA accompanying the penetration.
Our previous study established that SecA protein traverses the lipid
bilayer completely and that the penetration is inhibited by ATP while
ADP promotes it (Ahn and Kim, 1994). This conclusion was drawn from the
digestion of externally added SecA by the trypsin entrapped within the
phospholipid vesicles. The digestion products were analyzed by
SDS-PAGE, and the electrophoresis pattern was found to be dependent on
the kind of nucleotide present. The present investigation shows that
the isolated signal peptide enhances the digestion of SecA by
vesicle-entrapped trypsin while refolding pRBP has an opposite effect.
Refolding mRBP with or without the presence of isolated signal peptide
did not influence the SecA digestion. It was shown that SecA binds more
extensively to the vesicles in the presence of signal peptide and less
in the presence of refolding pRBP as compared to the binding when SecA
alone is present. The reason why the signal peptide promotes the
binding of SecA to the vesicles and subsequent penetration of bilayer,
while the refolding pRBP has an opposite effect, seems to be that
signal peptide converts the SecA into a more open form while refolding
pRBP makes it more compact. This conclusion was drawn from the
following experimental observations. First, more ANS bind to SecA in
the presence of SF peptide than in the absence of the peptide, while
refolding pRBP has an opposite effect (Fig. 7). It is clear
that, in the presence of SF signal peptide, more hydrophobic patches of
SecA were exposed to the surface, which increased ANS binding. Second,
the presence of refolding pRBP increased the thermal stability of SecA (Fig. 10), and this protein becomes more resistant to the
denaturation by GdnHCl. In view of the stabilizing effect of refolding
pRBP on SecA structure, it is somewhat surprising that refolding pRBP
decreased the intrinsic Trp fluorescence intensity, which indicates
that Trp residues are more extensively exposed to the solvent in the
presence of refolding pRBP. There is no easy explanation for this, but
this appears to be a common phenomenon judging from the reports for the
ATP-induced conformational change of the hsp90 and DnaK (Csermely et al., 1993; Palleros et al., 1992), among others.
Although this differential mode of interaction of signal peptide and
refolding pRBP with SecA explains the effect of these on the SecA
penetration of lipid bilayer, the reason for the difference is not
entirely clear. It is possible that the initial structural change
brought about by the signal peptide binding may facilitate the
subsequent binding of the mature domain, which converts the SecA into
more compact form. This was borne out by the kinetics of structural
change in SecA when this interacts with refolding pRBP ( Fig. 7and Fig. 8). Since the refolding mRBP by itself or
mRBP together with the isolated signal peptide had no effect, it is
clear that this occurs only when the mature domain is covalently linked
to the signal peptide. The essence of the results obtained here is
that, when refolding pRBP interacts with SecA, signal peptide is the
one first binds with SecA making it more lipid bilayer penetrable, and
this paves the way for the binding of the mature domain, which, in
turn, converts the SecA into membrane-inaccessible form. This is an
interesting observation by itself regardless of its relevance to the in vivo translocation of pRBP and merits further investigation
to elucidate the underlining mechanism. As to the relevance of the
observation we made here to the pRBP translocation, the results with
the IMV are of interest. It is noteworthy that ATP and refolding pRBP
separately affect the SecA penetration into IMV just as they did to
SecA insertion into lipid bilayer. However, when ATP and refolding pRBP
were mixed with IMV at the same time, an opposite effect of more
extensive SecA penetration can be seen. This confirms the results by
Kim et al.(1994), who used proOmpA at 0 °C using the
immunoblotting with anti-SecA antiserum as was here. Although the
results obtained by Economou and Wickner(1994) are somewhat different
from our results as well as from those by Kim et al.(1994),
possibly because of different band identification method, the general
conclusion that SecA penetrates the IMV membrane in the presence of
preprotein and ATP is the same. Another interesting aspect of our
result is that the SecA and its fragments disappear very rapidly (Fig. 13). The time-dependent change in the band density was not
investigated either by Kim et al.(1994) or by Economou and
Wickner(1994). But the latter authors demonstrated the reversible
nature of SecA penetration from exchange between inserted and
deinserted forms during translocation. This indicates the dynamic
nature of the topologies of intact SecA as well as its fragments. It is
possible that, under the condition given, SecA and its fragments
oscillate between insertion and deinsertion modes and they are
accessible to proteinase K during the deinsertion phase. Only an
extensive kinetic studies will be able to shed light on this. The
experimental results with IMV also underline the importance of the
integral protein members of the translocation machinery, especially the
SecY/E complex. It was proposed that the preprotein translocation is
carried out by protein-conducting channels, which are opened by signal
peptides (Simon and Blobel, 1992). From the cross-linking experiments,
it was also suggested that only SecA and SecY are in contact with
precursor proteins during translocation (Joly and Wickner, 1993),
suggesting a possibility that the channel forming proteins are SecY and
SecA. Any model of protein-conducting channel involving SecA must
accommodate the insertion-deinsertion cycle of this protein. It is
possible that equilibrium as well as kinetic studies of SecA
penetration into vesicles reconstituted with SecY/E in the presence of
refolding pRBP and ATP may be needed to clarify: 1) the difference in
this individual and combined effects of pRBP and ATP, and 2) how SecA
can be a part of a channel and still act as a ``shuttle.''
Volume 271,
Number 21,
Issue of May 24, 1996 pp. 12372-12379
©1996 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)are translocated with a targeting mechanism
very close to the one used by secretory proteins in mammalian cells.
The nascent precursor proteins of the latter subgroup initially bind to
the Ffh protein (Phillips and Silhavy, 1992), which has a high amino
acid sequence homology with the 54-kDa subunit of the signal
recognition particle in mammalian cells (Römisch et al., 1989). These two different targeting routes in E.
coli, however, converge at SecA with which the precursor proteins
bind. After this stage, the precursor proteins are translocated across
the membrane by SecY/E and some additional membrane proteins through an
as yet unknown mechanism.
-helix
formation, but its mutant peptide has a low -helix content (Yi et al., 1994; Chi et al., 1995).
Materials
Cibacron Blue Sepharose,
phenyl-Sepharose, and proteinase K were from Pharmacia Biotech Inc.
GdnHCl, trypsin, phosphatidylethanolamine (from bovine brain), dioleoyl
phosphatidylglycerol, and IODO-GEN were from Sigma. Stock solutions of
GdnHCl were prepared daily, and their concentrations were determined by
refractometry. Protein concentration was determined by the Bradford
method using bovine serum albumin as a standard (Bradford, 1976).
[
-
P]ATP (3000 Ci/mmol) and
Na
I (2 µl; 1 mCi) were from Amersham Corp.SecA Preparation
SecA protein was purified from a
SecA-overproducing strain (RR1/pMAN400) (Kawasaki et al.,
1989). The SecA-enriched fraction after 40-50% ammonium sulfate
precipitation was dialyzed against a buffer containing 25 mM Tris-HCl and 1 mM DTT (pH 7.5). This solution was loaded
onto a Cibacron Blue Sepharose column. The column was washed with a
buffer containing 25 mM Tris-HCl, 1 mM DTT, and 0.3 M KCl (pH 7.5) and then eluted with another buffer containing
25 mM Tris-HCl, 1 mM DTT, and 1.3 M KCl (pH
7.5). Fractions containing SecA were pooled and stored at -75
°C. The purity of this preparation as assayed by SDS-PAGE and
densitometry was close to 100%.Preparation of RBP
pRBP and mature RBP (mRBP) were
purified from the strains IQ87 (MC4100 secY
/pCI857,
pSP107) and SP114 (NR 69/pSP107), respectively, by ion exchange
chromatography as described in detail elsewhere (Teschke et
al., 1991) but after some minor modifications. The ammonium
sulfate precipitation following the sonication of cells was replaced
with CM-Sepharose column chromatography.Preparation of Signal Peptide
The revertant signal
peptide SF, in which the Leu at -17 position in the wild-type
peptide is replaced by a Pro and Ser at -15 position is replaced
by a Phe, was synthesized by a solid phase method on a MilliGen
(Burlington, MA) model 9060 automated peptide synthesizer. A second
revertant signal peptide, TI, where, in addition to the amino acid
replacement at -17 position, the Thr at -18 position was
substituted with an Ile, was also synthesized. The peptides were
purified by reverse-phase high performance liquid chromatography using
a Phenomenex W-porex C
column (15 cm 
1.0 cm),
elution being made with a water-acetonitrile linear gradient
(10-45% of acetonitrile) containing 0.1% trifluoroacetic acid.
The sequences of these peptides were confirmed by a MilliGen/Biosearch
6600 Prosequencer.Digestion of SecA by Vesicle-entrapped
Trypsin
Large unilamellar vesicles (LUV) were prepared with
phosphatidylethanolamine and dioleoyl phosphatidylglycerol (60:40, by
weight) by reverse phase evaporation method (Szoka et al.,
1978). The phospholipid concentration was determined using the method
of Vaskovsky et al.(1975). Trypsin was encapsulated within the
vesicles by the method used by Dumont and Richards(1984) and by
Rietveld et al.(1986). Digestion of SecA by vesicle-entrapped
trypsin was carried out by mixing solutions of SecA, vesicles, and
unfolded pRBP together and then incubating at 30 °C. These
components were dissolved in a buffer containing 1 mM DTT, 2
mM MgCl
, 20 mM HEPES (pH 7.5) and the
final concentrations were 3 µM for SecA, 0.6 mM for phospholipid (trypsin encapsulated LUV), and 10 µM for trypsin inhibitor. pRBP was initially unfolded in 1 M GdnHCl but the final GdnHCl concentration after mixing was low
enough for the pRBP to have the native structure as shown by circular
dichroism and fluorescence spectroscopy. For some experiments, 4.5
µM SF peptide, 6 µM native pRBP, or 6
µM refolding mRBP was replaced the refolding pRBP. Here as
well as in the subsequent experiments, the refolding RBP is
operationally defined as initially unfolded RBP in 1 M GdnHCl
being refolded after dilution with a solution containing SecA. These
experiments were performed with or without the presence of ATP or its
nonhydrolyzing analog, ATPS. The digestion reactions were
terminated by adding SDS sample buffer containing 3 mM phenylmethanesulfonyl fluoride and then placing the reaction
vessel in an ice bath. The samples were boiled for 5 min prior to the
electrophoresis on a denaturing 12% (w/v) sodium dodecyl sulfonyl
polyacrylamide gel (Laemmli, 1970).Quenching of SecA Trp Fluorescence by Vesicle-entrapped
Iodide
KI was entrapped within the vesicles as was described
before (Ahn and Kim, 1994). Solutions of SecA, iodide-entrapped
vesicles, and unfolded pRBP in 1 M GdnHCl, all in the same
buffer solution as in the SecA digestion (pH 7.5), were mixed together
and incubated at 30 °C for 30 min before fluorescence measurements.
Fluorescence intensity was measured using a Shimazu RF5000
spectrofluorometer in a thermostatted cuvette. The excitation
wavelength for the intrinsic tryptophan fluorescence was 295 nm.SecA Binding to Phospholipid Vesicles
Purified
SecA was iodinated using IODO-GEN (Markwell and Fox, 1978). Free iodine
was removed by Sephadex G25 chromatography. Binding experiments were
performed with 1 µMI-labeled SecA and 0.7
mM phospholipid (LUV) in the presence of either 4 µM refolding pRBP or 2 µM SF signal peptide in a buffer
containing 50 mM potassium phosphate, 100 mM KCl, and
1 mM DTT (pH 7.5). Protein-bound vesicles were centrifuged in
a Beckman TLA 100.2 rotor at 70,000 rpm for 1 h at 4 °C, and the
radioactivities of pellets and supernatant were determined using a
Beckman model 5500 counter.For the extractability experiments,
the SecA bound to phospholipid vesicles in the presence or absence of
SF signal peptide was treated with various concentrations of potassium
chloride and urea. Vesicles were sedimented, and the radioactivities of
pellets and supernatant were counted as described above.Conformational Change of SecA
Possible
conformational changes in SecA upon interaction with either refolding
pRBP or the SF signal peptide under various conditions were monitored
by fluorescence spectroscopy. Exposure of Trp of the SecA was tested by
mixing SecA with either SF peptide, TI peptide, refolding pRBP,
refolding mRBP, or combination of SF and refolding mRBP. All
experiments were performed at 30 °C with 0.4-8.0 µM SecA, 4-fold concentration of refolding pRBP, 4-fold concentration
of refolding mRBP, or 2-fold concentration of signal peptides in 50
mM potassium phosphate buffer, pH 7.5. pRBP and mRBP were
first unfolded in 1 M GdnHCl solution, and the final GdnHCl
concentration in the reaction samples after dilution was
0.02-0.04 M in all the experiments. The solutions of SF,
TI, and native pRBP also contained 0.02-0.04 M GdnHCl.
Fluorescence emission spectra were obtained as before.Effect of pRBP on the Stability of SecA
Thermal
unfolding of SecA was monitored by following the decrease in
fluorescence intensity at 338 nm (excitation 295 nm) with a 1
°C/min heating rate.Kinetic Studies of SecA-pRBP Interaction
In order
to check possible sequential interactions of the signal peptide and the
mature domain of pRBP with SecA, a solution containing ANS and pRBP
unfolded in 1 M GdnHCl was rapidly mixed with a 2 µM SecA solution and the ANS fluorescence intensity at 470 nm was
followed. The final concentration of ANS and pRBP was 200 and 8
µM, respectively, and the unfolded pRBP was diluted
50-fold. Control experiments were repeated either without pRBP or after
replacing of pRBP with SF signal peptide.I-SecA (
3 10
cpm/µg) was incubated with 0.2 µM unfolded pRBP
for different time periods at 30 °C, after which the
phenyl-Sepharose, suspended in 50 mM potassium phosphate (pH
7.5), was added to the reaction mixtures. After 1 min of incubation,
samples were centrifuged for 10 s. Radioactivities of the pellets and
supernatants were determined using a Beckman model 5500 counter.
From these the percentages of SecA bound to phenyl-Sepharose were
estimated.ATP Binding to SecA
2 µM SecA in 20
mM Tris-HCl (pH 7.3) was incubated with various concentrations
of ATP containing 1/1000 as much [-P]ATP.
Binding experiments were carried out in the presence of either the
signal peptide or the refolding pRBP. After 30 min of incubation at 30
°C, 0.1 volume of each sample was filtered through a spun
concentrator with molecular weight cutoff of 10,000. The radioactivity
of each filtrate was determined by a liquid scintillation counter
(Beckman LS6000LL) and compared with the control values without
proteins.Digestion of SecA Bound to the Outer Surface of
IMV
To compare with the digestion of SecA by the entrapped
trypsin within the phospholipid vesicles, we performed the proteinase K
digestion of SecA bound to E. coli inverted inner membrane
vesicle (IMV). IMVs were prepared from E. coli CP626 (MC4100 flhD
rbsB102::Tn10; Kim et
al., 1992) as described by Chang et al.(1978) and treated
with 6 M urea to inactivate SecA (Cunningham et al.,
1989). The concentrations of proteins and signal peptide were the same
as used in the digestion of SecA by vesicle-entrapped trypsin. Here
again, ATP or ATPS was also added for some experiments. 2 mg/ml
IMV and 5 µg/ml protease K were incubated at 30 °C for 15 min.
The reaction was stopped by the addition of SDS sample buffer
containing 3 mM phenylmethanesulfonyl fluoride as in the case
of trypsin digestion and followed by SDS-PAGE. Proteins separated on an
SDS-polyacrylamide gel were transferred to a nitrocellulose filter.
Rabbit anti-SecA serum was diluted 1000-fold and used in conjunction
with goat anti-rabbit immunoglobulin G-horseradish peroxidase conjugate
as a secondary antibody after 1000-fold dilution.
Effect of the pRBP and the Signal Peptide on the
Penetration of SecA into Phospholipid Vesicles
Fig. 1shows the SDS-PAGE of the fragments after 1 h of
digestion of external SecA by trypsin entrapped within phospholipid
vesicles. This is a sufficient time to reach the steady-state level of
digestion (Ahn and Kim, 1994). The digestion was performed in the
presence of initially unfolded pRBP, unfolded mRBP, or the SF signal
peptide. The effect of ATP or ATPS was also examined. When only
the SecA was present (lane a), fragments with molecular masses
of 67, 60, 31, and 20 kDa were obtained. The 98-kDa band is the intact
SecA and the band near the 20-kDa mark is from trypsin inhibitor. Two
minor bands of SecA fragments can also be seen between the intact SecA
and 67-kDa bands. There are some minor differences between this pattern
and the one shown previously (Ahn and Kim, 1994), which may arise from
the difference in phospholipid compositions. It was not possible to
determine the topology of the SecA from these multiple fragments.
-helix ()like RBP signal peptides
(Yi et al., 1994; Chi et al., 1995) had no effect.
Addition of either ATP (lane c) or its non-hydrolyzing analog,
ATPS (lane d) did not affect the intensity of intact
SecA, but there were some major changes in the fragments.S (lane g) in addition to refolding pRBP did not
affect the digestion pattern (lane e).
vesicle, phospholipid vesicle entrapped with iodide; vesicle, vesicles without entrapped
iodide.
Binding of SecA to Vesicles
Fig. 3shows
the binding of SecA to a model membrane with the same lipid composition
as in Fig. 1. Here, the SecA sedimented with the vesicles is
shown. More than 80% of the native SecA originally added to the
solution was bound to the vesicles. Although the presence of the SF
peptide appears to increase the SecA binding, this increase is within
the estimated experimental error. The addition of native pRBP also did
not affect the SecA binding appreciably. A tremendous reduction in the
SecA binding was observed when refolding pRBP was present.
I-labeled SecA was incubated with phospholipid
vesicles in the absence or presence of unfolded pRBP and signal
peptide. Vesicles with bound SecA was pelleted by ultracentrifugation,
and the radioactivities of pellets and supernatant were determined by
-ray counting.
I SecA with or without signal
peptide was treated with indicated concentration of KCl or urea and
sedimented by ultracentrifugation. The amount of membrane-bound SecA
was determined by -ray counting.
The Effect of pRBP and Signal Peptide on the Intrinsic
Tryptophan Fluorescence of SecA
The Trp emission spectra of SecA
obtained in the presence of refolding pRBP or signal peptide with the
excitation at 295 nm are shown in Fig. 5. Refolding pRBP caused
a decrease in fluorescence intensity by 12%, and revertant signal
peptides increased the intensity by about 30-40% as compared with
that of free SecA. However, there was no change in the
. We did not observe any change in fluorescence
intensity when either refolding mRBP or native pRBP was added. This
suggests that only refolding pRBP and signal peptides influence the
structure of SecA. The final concentration of GdnHCl in these
experiment was 0.04 M, and this concentration had no effect on
the intrinsic fluorescence intensity of SecA (data not shown). In the
presence of 0.04 M GdnHCl, pRBP has the same conformation as
that of native pRBP as assayed by tyrosine fluorescence and circular
dichroism at 222 nm (data not shown).
is the emission intensity in the
absence of iodide, F is the intensity in the presence of
iodide, K
is the Stern-Volmer quenching constant,
and [I] is the molar concentration of
iodide.
and F were determined as described under
``Experimental Procedures.''
value estimated from the slope was
2.55 M
for the native SecA. This value
increased to 3.18 M in the presence of
refolding pRBP and decreased to 1.37 M when
the signal peptide was added. It is clear that more SecA Trp residues
are exposed to the surface when refolding pRBP is present, but less Trp
residues are exposed to the solvent when the signal peptide is present. The Effect of pRBP and Signal Peptide on the ANS Binding
to SecA
Fig. 7shows that the native SecA binds ANS
significantly. This suggests that SecA has exposed hydrophobic regions
even in the native structure. When the signal peptide SF was present
together with SecA, the fluorescence intensity of ANS increased by
about 30%. TI peptide increased the ANS fluorescence intensity even
more. Since the signal peptides themselves showed no detectable binding
of ANS, it is clear that additional hydrophobic sites of SecA are
exposed when signal peptides are bound. However, upon addition of pRBP
unfolded in 1 M GdnHCl, the hydrophobic sites of SecA were
reduced as evidenced by a decrease in ANS fluorescence intensity by
about 29% when compared with that of free SecA. The final concentration
of GdnHCl after mixing was 0.02 M and under this condition
SecA appears to assume the native structure. This conclusion was drawn
from the observation that fluorescence intensity of SecA-bound ANS in
0.02 M GdnHCl was the same as the case of no GdnHCl (data not
shown). A control experiment showed no ANS binding to pRBP within the
GdnHCl concentration of up to 1.5 M. pRBP is fully unfolded in
0.8 M GdnHCl, and the transition midpoint concentration is
0.47-0.48 M when measured by tyrosine fluorescence and
circular dichroism (data not shown). This suggests that either
partially or fully unfolded pRBP has no hydrophobic clusters to which
ANS may bind. pRBP refolding in the presence of ANS also did not show
any fluorescence. The reduction in ANS fluorescence intensity by the
refolding pRBP suggests that the mature domain of the RBP converts the
SecA into more compact form by overcoming an opposing effect by the
signal peptide covalently linked to it. mRBP alone had no effect. When
equal mole amounts of signal peptide and refolding mRBP were added
together to SecA solution, the result was the same as the case when
signal peptide alone was added.
Sequential Interaction of the Signal Peptide and the
Mature Domain of Refolding pRBP with SecA
Fig. 8(line b) shows the time-course of ANS
binding to SecA when unfolded pRBP was diluted with SecA solution. This
may be compared with line a, which is a control experiment
without pRBP. There is an initial sharp increase in ANS binding above
the control reaching the level of another control experiment when the
SF peptide is present (line c). This is then followed by a
gradual decrease below the control value (line a). The time
course of ANS binding suggests that the signal peptide first binds with
the SecA, making its structure more open, and then the subsequent
binding of the mature domain makes the SecA structure even more compact
than the native SecA structure.
I-SecA was incubated with
unfolded pRBP for the indicated time period, and then phenyl-Sepharose
was added. Samples were centrifuged, and radioactivities of pellets and
supernatant were determined using a
-counter.
pRBP Increases the Stability of SecA
Fig. 10shows the thermal unfolding transition profile of SecA
measured by tryptophan fluorescence emission at 338 nm with excitation
at 295 nm. Native SecA showed transition temperature, T
, near 41 °C, which is similar to the result
obtained by Ulbrandt et al.(1992). However, the addition of
refolding pRBP resulted in a significant increase in the T to 45 °C. This result unequivocally
demonstrated that pRBP stabilizes SecA, which apparently is due to the
conversion of SecA into a more compact structure.
, thermal transition
temperature.
Effects of Signal Peptide and pRBP on ATP Binding to SecA
Fig. 11shows the effect of both refolding pRBP and
signal peptide on the ATP binding to SecA. When SecA alone is present,
the binding approaches the value of 1 mol of ATP/mol of SecA. This is
similar to the high affinity ADP binding to SecA (Mitchell and Oliver,
1993). When refolding pRBP is also present, the ATP binding is reduced
to about half of the value when SecA alone is present. The situation
becomes more complex when the SF signal peptide is present together
with SecA. The extent of binding at low ATP concentration is less than
the case of no SF peptide, but it becomes greater at higher ATP
concentration. Here again, the effect of pRBP on the ATP binding to
SecA is just opposite of the signal peptide. Although this should be
related to the opposite effects of these on the structure of SecA, the
precise nature of the effect is obscure at the moment. It should be
noted that the addition of refolding pRBP released ATP bound to SecA.
This result agrees with the earlier observation that proOmpA causes the
release of nucleotides bound to SecA in free solution (Shinkai et
al., 1991). However, these results appear to contradict with the
observation that the interaction of the preproteins with the
membrane-bound SecA stimulates the ATPase activity of SecA (Lill et
al., 1990). On the other hand, the synthetic signal peptides by
themselves competitively inhibited the enzyme activity and the
denatured mature proteins had no effect (Lill et al., 1990).
-P]ATP in the presence of
either 4 µM SF signal peptide or 6 µM unfolded pRBP for 30 min at 30 °C. Each sample was filtered
through a spun concentrator, and the bound ATP was determined by
counting the radioactivity of filtrate.
Digestion of the SecA Bound to IMV by Externally Added
Protease
This was performed with IMVs obtained from the strain
of E. coli CP626 (rbs102::Tn10). Digestion of the
bound SecA by external proteinase K at 30 °C produced major
proteolytic fragments with molecular mass of about 75 and 67 kDa (Fig. 12). Addition of either ATP or refolding pRBP greatly
increased the digestion, indicating less SecA penetration into IMV
membrane. On the other hand, the presence of SF signal peptide retarded
the digestion, suggesting that this peptide increased the SecA
penetration. These results are consistent with the results obtained
with the model membrane (Fig. 1).
)S, adenosine
5`-O-(thiotriphosphate).)
We thank Dr. Myeong-Jun Choi of Mogam Biotechnology
Research Institute for the synthesis of the signal peptide.
©1996 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:
|
|
 |
|
  H.-W. Wang, Y. Chen, H. Yang, X. Chen, M.-X. Duan, P. C. Tai, and S.-F. Sui Ring-like pore structures of SecA: Implication for bacterial protein-conducting channels PNAS, April 1, 2003; 100(7): 4221 - 4226. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ahn, D.-B. Oh, H. Kim, and C. Park The Phase Property of Membrane Phospholipids Is Affected by the Functionality of Signal Peptides from the Escherichia coli Ribose-binding Protein J. Biol. Chem., July 12, 2002; 277(29): 26157 - 26162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Baud, S. Karamanou, G. Sianidis, E. Vrontou, A. S. Politou, and A. Economou Allosteric Communication between Signal Peptides and the SecA Protein DEAD Motor ATPase Domain J. Biol. Chem., April 12, 2002; 277(16): 13724 - 13731. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ahn and H. Kim Effects of Nonlamellar-prone Lipids on the ATPase Activity of SecA Bound to Model Membranes J. Biol. Chem., August 21, 1998; 273(34): 21692 - 21698. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Schmidt, H. Ding, V. Ramamurthy, I. Mukerji, and D. Oliver Nucleotide Binding Activity of SecA Homodimer Is Conformationally Regulated by Temperature and Altered by prlD and azi Mutations J. Biol. Chem., May 12, 2000; 275(20): 15440 - 15448. [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 |