Molecular Microbiology (1998) 29(5), 1179–1190
Preprotein transfer to the Escherichia coli translocase
requires the co-operative binding of SecB and the signal
sequence to SecA
Peter Fekkes,1 Janny G. de Wit,1 Jeroen P. W. van der
Wolk,1 Harvey H. Kimsey,2 Carol A. Kumamoto2 and
Arnold J. M. Driessen1*
1
Department of Microbiology, Groningen Biomolecular
Sciences and Biotechnology Institute, University of
Groningen, Kerklaan 30, 9751 NN Haren, The
Netherlands.
2
Department of Molecular Biology and Microbiology, Tufts
University School of Medicine, Boston, MA 02111, USA.
Summary
In Escherichia coli, precursor proteins are targeted
to the membrane-bound translocase by the cytosolic
chaperone SecB. SecB binds to the extreme carboxyterminus of the SecA ATPase translocase subunit,
and this interaction is promoted by preproteins. The
mutant SecB proteins, L75Q and E77K, which interfere with preprotein translocation in vivo, are unable
to stimulate in vitro translocation. Both mutants bind
proOmpA but fail to support the SecA-dependent membrane binding of proOmpA because of a marked reduction in their binding affinities for SecA. The stimulatory
effect of preproteins on the interaction between SecB
and SecA exclusively involves the signal sequence
domain of the preprotein, as it can be mimicked by a
synthetic signal peptide and is not observed with a
mutant preprotein (D8proOmpA) bearing a non-functional signal sequence. D8proOmpA is not translocated
across wild-type membranes, but the translocation
defect is suppressed in inner membrane vesicles
derived from a prlA4 strain. SecB reduces the translocation of D8proOmpA into these vesicles and almost
completely prevents translocation when, in addition,
the SecB binding site on SecA is removed. These data
demonstrate that efficient targeting of preproteins by
SecB requires both a functional signal sequence and
a SecB binding domain on SecA. It is concluded that
the SecB–SecA interaction is needed to dissociate
the mature preprotein domain from SecB and that
binding of the signal sequence domain to SecA is
Received 19 March, 1998; revised 1 June, 1998; accepted 2 June,
1998. *For correspondence. E-mail [email protected]; Tel.
(50) 363 2164; Fax (50) 363 2154.
Q 1998 Blackwell Science Ltd
required to ensure efficient transfer of the preprotein
to the translocase.
Introduction
The molecular chaperone SecB, a tetrameric protein with
a molecular mass of 68 kDa (Smith et al., 1996), targets a
subset of periplasmic and outer membrane proteins to the
translocase at the cytoplasmic membrane of Escherichia
coli (Kumamoto, 1991). SecB associates with the nascent
chains as they emerge from the ribosome (Kumamoto and
Francetic, 1993; Randall et al., 1997) and promotes both
their post- and co-translational translocation (Kumamoto
and Gannon, 1988). SecB interacts with the mature part
of the preprotein (Collier et al., 1988; Gannon et al., 1989;
Randall et al., 1990; De Cock et al., 1992; Fekkes et al.,
1995) and, by forming a stoichiometric complex, it maintains the preprotein in a translocation-competent state (Randall and Hardy, 1986; Lecker et al., 1989; 1990). It has been
suggested that SecB recognizes preproteins by binding
their signal sequence domain (Watanabe and Blobel,
1989; 1995; Altman et al., 1990a), but other studies show
that signal sequence recognition is not a prerequisite for
the binding reaction per se (Lecker et al., 1989; Liu et
al., 1989; Randall et al., 1990; Hardy and Randall, 1991).
In vitro, SecB interacts with a variety of proteins as long
as they are in an unfolded state (Hardy and Randall, 1991;
Fekkes et al., 1995). In vivo, however, SecB is very selective, and only a few proteins, mainly precursors of outer
membrane proteins, are recognized (Kumamoto and Francetic, 1993). Several attempts have been made to identify
a SecB binding motif in preproteins (Altman et al., 1990b;
Topping and Randall, 1994; Khisty et al., 1995; Smith et
al., 1996), but so far no consensus has been reached.
Residues in SecB involved in preprotein binding have
been identified by mutational studies (Gannon and Kumamoto, 1993; Kimsey et al., 1995). These mutations, which
result in a reduced translocation efficiency, have been
classified into three classes based on a colony colour
assay as a semi-quantitative measure of SecB activity,
and cluster in two regions (Kimsey et al., 1995). The first
class (class I) consists of mutations that reduce the ability
of SecB to form a stable complex with the precursor of maltose-binding protein (preMBP), but cause only a mild defect
in the rate of MBP export as judged from pulse–chase
1180 P. Fekkes et al.
experiments. The second class (class 2) is characterized
by a severe slow-down in the rate of MBP export without
any effect on the SecB–preMBP complex formation. The
third class (class 3) consists of SecB mutations that fail
to support the growth of E. coli on rich media, indicating
that there is no SecB activity at all.
In addition to the chaperone functions, SecB targets the
preprotein to SecA, the peripheral ATPase subunit of the
translocase (Hartl et al., 1990; Fekkes et al., 1997). SecA
recognizes both the signal sequence and the mature domain
of the preprotein (Cunningham and Wickner, 1989; Lill et al.,
1990; Kimura et al., 1991), but also binds SecB directly
(Hartl et al., 1990; Breukink et al., 1995). The latter interaction is confined to a short extreme carboxyl-terminal
polypeptide sequence of SecA, which bears a high instance
of positively charged amino acid residues (Fekkes et al.,
1997). The interaction of SecB with SecA bound at the
SecYEG complex is of high affinity (Hartl et al., 1990). Soluble SecB–SecA complexes have been observed (Hartl
et al., 1990; Hoffschulte et al., 1994), but such complexes
are formed with at least 10 times lower affinity than those
at the membrane (Den Blaauwen et al., 1997). Binding of
SecB to SecA at the membrane is stimulated by the preprotein resulting in the formation of a stable membranebound ternary complex (Hartl et al., 1990) in which the
preprotein is no longer associated with SecB (Fekkes et
al., 1997). The binding of ATP to SecA causes the dislocation of SecB from the translocase (Fekkes et al., 1997) and
the initiation of preprotein translocation.
To determine the catalytic mechanism of preprotein transfer and to identify the SecA binding domain on SecB, we
have analysed the SecB class 2 mutants L75Q and E77K.
These two mutants were thought to interfere with preprotein
translocation by binding the preprotein with higher affinity
compared with the wild-type SecB (Gannon and Kumamoto,
1993; Kimsey et al., 1995). Our data demonstrate that these
mutants are unable to support the SecB-dependent preprotein translocation because of a defect in the SecA binding.
The data show further that, for efficient delivery of the preprotein to SecA, both the SecA–SecB interaction and the
binding of the preprotein signal sequence domain to SecA
is necessary. The latter elicits a tightening of the SecB–
SecA interaction, causing the release of the mature preprotein domain from SecB for binding to SecA, thereby
providing specificity to the transfer reaction.
Results
SecBL75Q and SecBE77K fail to support in vitro
preprotein translocation
The class 2 mutants SecBL75Q and SecBE77K and wildtype SecB, all containing an amino-terminal hexahistidine
tag, were purified to homogeneity by nickel-NTA affinity
and anion-exchange chromatography. To study the effect
of the mutations in SecB on the efficiency of preprotein
translocation, in vitro translocation of [ 35S]-proOmpA into
urea-washed inverted inner membrane vesicles (IMVs)
derived from the wild-type cells was performed in the presence of wild-type SecA or SecAN880 (Fig. 1). SecAN880
is a functional SecA truncate that lacks the carboxy-terminal SecB binding site and is therefore unable to interact
with SecB (Fekkes et al., 1997). The presence of a His
tag promoted the activity of SecB, as the SecA-dependent
translocation of proOmpA was stimulated to a greater
extent by HisSecB (Fig. 1A, lane 3) than by wild-type
SecB (lane 2). In contrast, translocation was not stimulated by HisSecBE77K (lane 4) or HisSecBL75Q (lane
5). In the presence of SecAN880, hardly any stimulation
of translocation was observed with the SecB proteins
Fig. 1. SecBL75Q and E77K fail to stimulate
preprotein translocation. Translocation of
[ 35S]-proOmpA into IMVs of strain MC4100
was performed as described in Experimental
procedures before (A) or after (B) preincubation of proOmpA for 15 min at 378C. Translocation reactions were carried out in the
presence (lanes 1–5) or absence of SecA
(lane 6), or with SecAN880 (lanes 7–11), and
in the absence (lanes 1, 6 and 7) or presence
of SecB (lanes 2 and 8), His-tagged SecB
(lanes 3 and 9), His-tagged SecBE77K (lanes
4 and 10) or His-tagged SecBL75Q (lanes 5
and 11). The result of a typical experiment is
shown, which was repeated independently
three times. (A) and (B) represent different
exposures of the translocation reactions. The
proOmpA translocation efficiency of lane 2 in
(A) and (B) is about 23% and 8% respectively.
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Preprotein transfer from SecB to SecA 1181
(lanes 8–11), and the translocation was about as efficient
as with SecA in the absence of SecB (lanes 1 and 7). Only
the HisSecB showed a slight stimulation of translocation
(lane 9). When proOmpA was first diluted from urea into
a buffer in the absence of SecB, incubated for 15 min at
378C to render it translocation incompetent and then
added to IMVs, no translocation was observed (Fig. 1B,
lane 1). When the preincubation was performed in the presence of HisSecBE77K or HisSecBL75Q, it appeared that
the translocation-competent state of proOmpA was maintained (lanes 4 and 5), although the translocation was less
efficient than in the presence of wild-type SecB or HisSecB
(lanes 2 and 3). In the presence of SecAN880 (lanes 7–
11), wild-type and mutant SecB proteins were nearly equally
effective in maintaining the translocation-competent state
of proOmpA. These data indicate that the L75Q and E77K
mutations in SecB do not interfere with preprotein binding,
but suggest that they are defective in the targeting of
proOmpA to the translocation site at the membrane.
SecBL75Q and SecBE77K are defective in the targeting
of proOmpA to the membrane
To ascertain further that the absence of stimulation of
proOmpA translocation by HisSecBL75Q and HisSecBE77K is not caused by a deficient interaction with proOmpA, the HisSecB-dependent binding of urea-denatured
[ 35S]-proOmpA to Ni-NTA-coupled agarose beads was
determined (Fig. 2). In the presence of wild-type SecB,
little proOmpA was bound to the Ni-NTA-coupled agarose
beads, while with HisSecB, HisSecBL75Q and HisSecBE77K, substantial and nearly equivalent amounts of
proOmpA were bound by the resin. This observation, in
conjunction with the ability to maintain proOmpA in a translocation-competent state (Fig. 1B, lanes 8–11), demonstrates that the mutated SecB proteins still recognize
proOmpA. Subsequently, the effect of SecB on the membrane binding of proOmpA in the presence or absence of
SecA under non-translocating conditions (i.e. 08C and in
the absence of ATP) was measured using SecAN880 as
a control (Fig. 3). In the absence of SecA, only background
membrane binding of proOmpA was observed (white bars).
With both wild-type SecB and HisSecB, the addition of SecA
(grey bars) resulted in a large enhancement of the proOmpA
membrane binding. On the other hand, with HisSecBL75Q
and HisSecBE77K, the proOmpA binding with SecA (grey
bars) corresponded to the background levels observed
with SecAN880 (black bars). These data show that the
SecB mutants L75Q and E77K are disturbed in the targeting of proOmpA to membrane-bound SecA.
SecBL75Q and E77K are deficient in SecA binding
SecB binds with high affinity to the membrane-bound form
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Fig. 2. ProOmpA binds to SecBL75Q and E77K. [ 35S]-proOmpA
was preincubated for 30 min on ice with 2.5 mM (tetramer) SecB,
HisSecB, HisSecBL75Q or HisSecBE77K as described in
Experimental procedures. Subsequently, prewashed Ni-NTAcoupled agarose beads were added and, after further incubation
for 25 min on ice, the amount of [ 35S]-proOmpA bound to the beads
was determined.
of SecA (Hartl et al., 1990; Fekkes et al., 1997). In the
absence of proOmpA, SecB binds SecA with a dissociation constant (K d ) of 25 nM (Fekkes et al., 1997; see
also Fig. 4A). To determine whether the SecB mutants
are still able to associate with SecA, their ability to compete with [ 125I]-SecB for SecA binding was examined.
For this purpose, IMVs were used that were derived from
a SecYEG-overproducing strain that has previously been
shown to bind elevated amounts of SecB in a SecA-dependent fashion (Fekkes et al., 1997). In the assay, binding of
a fixed concentration of 125I-labelled wild-type SecB to IMVs
was measured in the presence of increasing concentrations
of unlabelled wild-type SecB, HisSecB, HisSecBL75Q or
HisSecBE77K protein. The inhibition curves for wild-type
SecB (Fig. 4A, W) and HisSecB (X) were nearly identical,
indicating that HisSecB has the same affinity for membranebound SecA as wild-type SecB. In contrast, the L75Q SecB
mutant (B) is a poor competitor, and HisSecBE77K (O) is
even worse. Only high concentrations of these SecB proteins could compete substantially with the [ 125I]-SecB for
SecA binding. This implies that these two mutants are
defective in the interaction with SecA. In the presence
of proOmpA, membrane-bound SecA binds SecB with an
enhanced affinity, i.e. a K d of 8 nM instead of 25 nM (Fekkes
et al., 1997; see also Fig. 4B). Even under these conditions,
inhibition curves of wild-type SecB (Fig. 4B, W) and HisSecB (X) were nearly identical, whereas the L75Q and
the E77K SecB mutants were equally poor competitors.
1182 P. Fekkes et al.
Taken together, these data demonstrate that the SecBL75Q
and E77K mutants are defective in their ability to interact
with the membrane-bound SecA.
To substantiate this conclusion, binding studies with
SecB were performed with the Schistosoma japonicum
glutathione-S -transferase (GST) protein that was fused
at its carboxy-terminus to the SecB binding domain of
SecA, i.e. GST218 (Fekkes et al., 1997). Both wild-type
SecB and HisSecB were bound efficiently by the GST218
fusion protein (Fig. 5, lanes 2 and 3), whereas no binding
was observed in the absence of the SecB binding domain,
i.e. with GST alone (lane 1, data not shown). GST218
failed to recognize the HisSecBE77K and L75Q (lanes 4
and 5), further demonstrating that these mutant SecB
proteins are disturbed in the interaction with SecA.
The signal sequence domain of precursor proteins
stimulates the SecA–SecB interaction
Fig. 3. SecBL75Q and E77K fail to target proOmpA to IMVs.
Urea-treated IMVs (250 mg ml¹1 ) derived from wild-type cells were
incubated on ice with [ 35S]-proOmpA (20 nM) and, as indicated,
100 nM SecB, HisSecB, HisSecBL75Q or HisSecBE77K. Incubations were performed in the absence of SecA (white bars) or in the
presence of 250 nM SecA (grey bars) or SecAN880 (black bars).
After 15 min, membranes were sedimented through a 100 ml 0.2 M
sucrose cushion, and the amount of proOmpA in the pellet and
supernatant fractions was quantified by counting in scintillation
liquid.
The interaction between SecA and SecB at the membrane
is stimulated by preproteins (Hartl et al., 1990; Fekkes et
al., 1997), and this phenomenon is retained in the SecA
binding mutant SecBE77K. To determine the mechanism
of stimulation and to establish which portion of the preprotein, i.e. the signal sequence and/or the mature domain, is
responsible for this phenomenon, the binding of a signal
Fig. 4. SecBL75Q and E77K are defective in SecA-dependent SecB binding to IMVs. Urea-treated IMVs (250 mg ml¹1 ) derived from SecYEGþ
cells were incubated on ice with 250 nM SecA, 10 nM [ 125I]-SecB and 13–4600 nM SecB (W), HisSecB (X), HisSecBL75Q (B) or HisSecBE77K
(O) in 100 ml of buffer B in the absence (A) or the presence (B) of proOmpA (300 nM). After 15 min, membranes were sedimented through a
100 ml 0.2 M sucrose cushion, and the amount of [ 125I]-SecB in the pellet and supernatant was quantified in a g-counter.
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Preprotein transfer from SecB to SecA 1183
Fig. 5. SecBL75Q and E77K fail to interact with the SecB binding
domain of SecA fused to glutathione-S -transferase. GST (lane 1)
and GST218 (lanes 2–5) were preincubated with 5 mg of SecB
(lanes 1 and 2), HisSecB (lane 3), HisSecBE77K (lane 4) and
HisSecBL75Q (lane 5). Subsequently, 10 ml of prewashed
glutathione-coupled agarose beads were added and, after two
washing steps with buffer HB, the GST proteins with bound SecB
protein were eluted with 20 mM reduced glutathione in buffer HB.
Samples were applied on a 15% SDS–PAGE gel and stained with
Coomassie brilliant blue. The positions of the SecB and HisSecB
are indicated. HisSecB had a higher apparent molecular mass
because of the presence of the hexaHis-tag.
sequence mutant of proOmpA, D8proOmpA, was examined. This mutant is defective in translocation as a result
of a deletion of the isoleucine at position 8 of the hydrophobic core of the signal sequence (Tanji et al., 1991;
Van der Wolk et al., 1998). Binding of SecB to the membrane-bound SecA was stimulated by proOmpA (Fig. 6A,
lane 2), but not by D8proOmpA (lane 3). OmpA (lane 4)
also failed to stimulate the binding of SecB to SecA.
These data suggest that the signal sequence is responsible for the stimulation, but do not rule out a possible role
for the mature preprotein domain. Therefore, the SecAdependent SecB binding to IMVs was measured in the presence and absence of a synthetic peptide corresponding
to the signal sequence of proOmpA (MKKTAIAIAVALAGFATVAQA; Movva et al., 1980; Lill et al., 1990). For better
resolution, these experiments were conducted with IMVs
obtained from cells overproducing SecYEG, but similar
results were obtained with wild-type IMVs. The presence
of the synthetic signal peptide (10 mM) had no effect on
the number of high-affinity SecB binding sites, but reduced
the K d value from 23 to 4 nM (Fig. 6B). The addition of a
non-related peptide (reduced bovine pancreatic trypsin
inhibitor) up to 250 mM had no effect on the SecB binding
to IMVs (data not shown). These data demonstrate that the
stimulation of the interaction between SecA and SecB at
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Fig. 6. The signal sequence domain of the preprotein stimulates
the interaction between SecA and SecB.
A. Preproteins with a defective or without a signal sequence fail to
enhance SecA-dependent SecB binding to IMVs. Binding of
[ 125I]-SecB (10 nM) to urea-treated membranes of wild-type cells
was measured in the absence (lane 1) or in the presence of
proOmpA (lane 2), D8proOmpA (lane 3) or OmpA (lane 4).
B. Scatchard plot analysis of SecA-dependent binding of
[ 125I]-SecB (4 nM to 5 mM) to urea-treated membranes derived from
SecYEGþ cells in the presence (W) or absence (X) of 10 mM
proOmpA signal peptide.
1184 P. Fekkes et al.
the membrane occurs primarily via the signal sequence
domain of the preprotein.
SecB and the signal sequence co-operate to transfer
the preprotein to SecA
In the stable membrane-bound complex of SecA, SecB
and preprotein, transfer of the preprotein from SecB to
SecA has already taken place. This is suggested by the
observation that proOmpA does not promote the binding
of SecB to SecAN880, a SecB binding-deficient SecA truncate (Fekkes et al., 1997), implying that SecB cannot bind
to SecA via the preprotein. To investigate the mechanism
of preprotein transfer further, binding of OmpA proteins
with different signal sequence domains to IMVs was measured in the presence of SecA or SecAN880 (Fig. 7).
Experiments were performed in the presence of SecB to
prevent non-specific interactions of the (pre)proteins with
the IMVs (Hartl et al., 1990). SecA supports the efficient
binding of proOmpA to the IMVs (lane 1, white bar), while
a reduced level of binding was observed in the presence
of SecAN880 (lane 1, black bar). SecA supports only low
levels of OmpA or D8proOmpA binding (lanes 2 and 3,
white bars), while the binding was almost completely abolished with SecAN880 (lanes 2 and 3, black bars). These
results demonstrate that both a functional SecB binding
site on SecA and a functional signal sequence on the preprotein are needed for efficient targeting and transfer of
preproteins from SecB to SecA.
To determine the functional impact of the co-operative
effect of SecB and the signal sequence on SecA targeting,
the in vitro translocation of proOmpA into IMVs with a wildtype or an altered signal sequence (D8proOmpA) was
analysed (Fig. 8). IMVs were derived from wild-type cells
(Fig. 8A) or from cells harbouring the prlA4 signal sequence suppressor mutation in secY (Fig. 8B), which enables
translocation of preproteins with a defective signal sequence
(Emr et al., 1981; Osborne and Silhavy, 1993; Van der Wolk
et al., 1998). In the presence of SecA, the translocation of
proOmpA into IMVs was stimulated by SecB (Fig. 8A,
lanes 3 and 4). This stimulation was largely absent when
SecAN880 was used (lanes 5 and 6; Fekkes et al., 1997;
see also Fig. 1). D8proOmpA was not translocated into
wild-type IMVs, regardless of the presence of SecB or
SecA (lanes 7–12). The translocation of proOmpA into
IMVs derived from the prlA4 strain in the presence of
SecA was far more efficient than in wild-type vesicles,
and even further stimulated by SecB (Fig. 8B, lanes 3
and 4). Under the conditions used, about 60% of the
added proOmpA was translocated into the IMVs. Again,
this stimulatory effect of SecB was largely abolished when
SecAN880 was used instead of SecA (lanes 5 and 6).
The prlA4 IMVs suppressed the D8proOmpA translocation defect in a SecA-dependent manner, although the
Fig. 7. SecB and the signal sequence domain co-operate in the
targeting of proOmpA to SecA. Urea-treated IMVs (250 mg ml¹1 )
derived from wild-type cells were incubated on ice with 100 nM
SecB and, as indicated [ 35S]-proOmpA, [ 35S]-OmpA or
[ 35S]-D8proOmpA (20 nM). Incubations were performed in the
presence of 250 nM SecA (white bars) or SecAN880 (black bars).
After 15 min, membranes were sedimented through a 100 ml 0.2 M
sucrose cushion, and the amount of radiolabelled protein in the
pellet and supernatant fractions was quantified by counting in
scintillation liquid.
preprotein was translocated with a lowered efficiency compared with proOmpA (compare lanes 3 and 9). The addition of SecB did not stimulate, but slightly reduced, the
translocation of D8proOmpA into these IMVs (lane 10).
This effect was exacerbated when the SecB binding
domain on SecA was also removed. SecB almost completely abolished the SecAN880-dependent translocation of
D8proOmpA into the prlA4 IMVs (lanes 11 and 12). Therefore, it appears that the SecA–SecB interaction is needed
for the release of the preprotein from SecB and that this
reaction is facilitated by the binding of the functional signal
sequence to SecA.
Discussion
The molecular chaperone SecB fulfils a dual function in
bacterial preprotein translocation. It maintains preproteins
in a translocation-competent state and targets them to the
SecA subunit of the translocase. The preprotein binding
site on SecB has been identified by mutational analysis
(Kimsey et al., 1995), and we have shown recently that
SecB recognizes the extreme carboxy-terminus of SecA
as its docking site (Fekkes et al., 1997). Here, we have
analysed a number of critical amino acid residues on
SecB that were thought to be involved in preprotein binding
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Preprotein transfer from SecB to SecA 1185
Fig. 8. Interaction of SecA with both SecB and
the preprotein signal sequence is essential for
optimal stimulation of translocation.
Translocation of [ 35S]-proOmpA (lanes 1–6)
and [ 35S]-D8proOmpA (lanes 7–12) into IMVs
of strain MC4100 (SecYþ ) (A) or SE6004
(PrlA4) (B) was performed as described in
Experimental procedures . Reactions were carried out in the absence (lanes 1, 2, 7 and 8) or
presence of SecA (lanes 3, 4, 9 and 10) or
SecAN880 (lanes 5, 6, 11 and 12), without
(lanes 1, 3, 5, 7, 9 and 11) or with SecB (lanes
2, 4, 6, 8, 10 and 12). The result of a typical
experiment is shown, which was repeated independently three times. (A) and (B) represent
identical exposures. [ 35S]-ProOmpA and [ 35S]D8proOmpA were synthesized by an in vitro
transcription/ translation reaction with a similar
yield of radioactivity. The translocation efficiency in lane 4 of (B) represents about 60% of
the input proOmpA.
(Gannon and Kumamoto, 1993; Kimsey et al., 1995), but
now prove to be part of the SecA binding site. Mutation
of leucine-75 into glutamine (L75Q) or glutamate-77 into
arginine (E77K) does not lead to loss of the ability of
SecB to bind the preprotein substrate proOmpA or to maintain proOmpA in a translocation-competent state. However,
the mutant SecB proteins fail to target the preprotein to the
membrane because of a defect in the interaction with SecA.
These data demonstrate that single point mutations can
separate the two functions of SecB in preprotein translocation, i.e. preprotein binding and targeting.
Mutations in SecB that alter the ability to stimulate preprotein translocation are clustered in two regions. Region 1
comprises amino acid residues 74–80, while region 2 starts
at position 20 and extends to amino acid 24 (Kimsey et al.,
1995). The residues of both regions are highly conserved
among the SecB proteins of Haemophilus influenzae
(Fleischmann et al., 1996) and Buchnera aphidicola (Lai
and Bauman, 1992) (Fig. 9). The unannotated genome
sequences of Actinobacillus actinomycetemcomitan, Neisseria meningitidis, Pseudomonas aeruginosa, Rhodobacter capsulatus and Vibrio cholerae also harbour SecB
homologues that exhibit a high degree of conservation
specifically in these regions. Class 1 mutations affecting
preprotein binding occur only in region 1 (Kimsey et al.,
1995). The L75Q and E77K mutations analysed in this
report are also located in region 1 but belong to class 2.
Other class 2 mutations have also been identified, and
these are located in region 2, i.e. aspartate-20 and glutamate-24 (Kimsey et al., 1995). Both residues are negatively charged and, like the residues characterized in this
paper, may well be involved in the SecA interaction. It is
therefore tempting to speculate that the interaction between
SecB and SecA is of an electrostatic nature. The strong
positively charged SecB binding site of SecA may bind to
a cluster of negatively charged residues present in the
SecA binding site of the tetrameric SecB.
To target preproteins efficiently, SecB has to interact in
a productive manner with SecA. When this interaction is
disturbed, either by the absence of the SecB binding site
on SecA (SecAN880) (Fekkes et al., 1997; this paper) or
by mutations in the SecA binding site on SecB, targeting
of preproteins to the membrane is no longer possible, and
translocation defects are apparent (Kimsey et al., 1995;
this paper). As these SecB mutants are still able to
maintain the translocation competence of preproteins, as
shown for proOmpA in this study, it appears that targeting
is an important aspect of the SecB function in vivo. The
translocation of a preprotein with a defective signal sequence in cells with the prlA4 mutation is impaired by
Fig. 9. SecA binding site on SecB. Alignment
of regions 1 and 2 of known bacterial SecB proteins. Eco, E. coli (Kumamoto and Nault, 1989),
GeneBank accession number M24489; Bap,
Buchnera aphidicola (Lai and Baumann, 1992),
GeneBank accession number M90644; Hin,
Haemophilus influenzae (Fleischmann et al.,
1995), GeneBank accession number U32758.
Mutants belonging to class 1 are defective in
preprotein binding, while class 2 mutants are
disturbed in SecA binding. The class 2 SecB
mutants, L75Q and E77K, are indicated. In E.
coli, region 1 stretches from amino acids 74 to
80 and region 2 from amino acids 20 to 24.
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
1186 P. Fekkes et al.
SecBL75Q but still occurs in the presence of wild-type
SecB (Francetic et al., 1993). With the current notion that
SecBL75Q is defective in SecA binding, it is evident that,
when both targeting factors are eliminated, i.e. a functional
signal sequence and the SecB–SecA interaction, translocation is abolished. A similar phenomenon is observed
in vitro with the SecAN880-mediated translocation of
D8proOmpA in prlA4 IMVs. However, in contrast to the
in vitro data, the translocation of signal sequence-defective
preproteins in vivo strictly requires SecB (Derman et al.,
1993; Francetic et al., 1993; Flower et al., 1994). This possibly reflects the dual role of SecB in protein translocation
and suggests that the targeting capacity of SecB in vivo is
even more important than in vitro. It is important to emphasize that in vitro, using urea-denatured preprotein, the ratedetermining steps in translocation may be different from in
vivo.
The stimulatory effect of the preprotein on the interaction
between SecB and SecA is presumably mediated through
the direct binding of the signal sequence to SecA. The
possibility that the signal sequence acts through direct
binding to SecB cannot be formally excluded. However,
this is less likely, as the signal sequence of the SecBbound preprotein seems to be exposed because it is easily
accessible for protease digestion (Topping and Randall,
1994; Khisty et al., 1995), and signal sequence binding
is not a prerequisite for the recognition of preproteins
bySecB (Lecker et al., 1989; Liu et al., 1989; Randall et
al., 1990; 1997; Hardy and Randall, 1991). Furthermore,
SecA has been shown to interact with the signal sequence
(Cunningham and Wickner, 1989; Lill et al., 1990). When
the signal sequence is mutated or deleted, preprotein recognition by SecA is impaired, and the SecA–SecB interaction
is no longer stimulated. Under these conditions, the targeting of the preprotein to the membrane is also defective,
even though the SecB binding domain on SecA is still functional. This suggests that SecB releases its cargo upon
binding to membrane-bound SecA, but that, because of
the lack of a proper signal sequence, the interaction of
the preprotein takes place only inefficiently. When the
SecB binding site on SecA is also removed, targeting is
completely abolished and, instead of stimulating translocation, SecB now even blocks the translocation of a preprotein with a defective signal sequence. This implies
strongly that the SecB–SecA interaction serves to dissociate the preprotein from SecB, consistent with previous
findings that, in the ternary SecA–SecB–preprotein complex, the preprotein is no longer bound to SecB (Fekkes et
al., 1997). The current study also demonstrates that the
SecB proteins defective in SecA interaction are unable to
support the high-affinity binding of proOmpA to SecA,
despite the fact that they have retained the ability to
form a complex with proOmpA. This further supports the
notion that the SecA–SecB interaction is not mediated
via binding of the preprotein to both SecA and SecB and
can be taken as a further indication that SecB is stimulated
by SecA to release the preprotein.
Because of the high content of b-structure in SecB and
in the outer membrane proteins, which form the main substrate for SecB, it has been suggested that preprotein
recognition is mediated through b–b interaction (MacIntyre et al., 1991; Breukink et al., 1992). This would mean
that the preprotein binding site on SecB consists of strands
with b-sheet secondary structure. Modelling of the polypeptide sequence of region 1 as a b-sheet reveals that
the amino acid residues involved in preprotein binding
(class 1 mutants) all face the same side of the b-sheet
yielding a hydrophobic patch (Kimsey et al., 1995). Similarly, the residues that are involved in SecA binding face
the opposite side and, in conjunction with region 2, may
form a negatively charged surface. Such close proximity
of the preprotein and SecA binding sites on SecB provides
a possible mechanism by which SecA can modulate the
preprotein binding activity of SecB. Binding of SecA to
the negatively charged face of region 1 may cause a conformational change in the preprotein binding site of SecB
on the opposite face of this b-sheet-like region, thereby
lowering the binding affinity of SecB for the mature portion
of the bound preprotein.
The data presented here provide a possible scenario for
the transfer of the preprotein from SecB to SecA (Fig. 10).
SecB interacts in the cytosol with newly synthesized preproteins by binding their mature domain. Within this binary
complex, the signal sequence is available to interact with
components such as the membrane-bound SecA. As a
result of this interaction, SecA changes its conformation
and binds SecB with higher affinity (step 1). Subsequently,
SecB undergoes a conformational change, and the mature
domain of the preprotein is unloaded. As the preprotein
is anchored to SecA via its signal sequence, it is automatically transferred from SecB to SecA, yielding a stable
ternary complex (step 2). Although SecB is now no longer
associated with the preprotein, it remains bound to SecA
and, in this state, it prevents newly synthesized preproteins
from entering the translocase. Once translocation is initiated
upon the binding of ATP, SecB is released into the cytosol
(step 3; Fekkes et al., 1997) to interact with another preprotein. The signal sequence fulfils a dual role in this catalytic preprotein transfer mechanism. First, it functions as
an anchor to hold on to SecA while the mature domain
of the preprotein is released from SecB. When this anchor
function is absent, either by removal or by mutation of the
signal sequence, the preprotein does not remain stably
bound at the membrane and is, to a large extent, released
into the cytosol. Secondly, by promoting the tight binding
of SecB to SecA, it ensures the efficient release of the
mature preprotein domain from its association with SecB.
It is concluded that SecB and the signal sequence of the
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Preprotein transfer from SecB to SecA 1187
Fig. 10. Model for the mechanism of preprotein transfer from SecB to SecA at the membrane. See Discussion for details.
preprotein co-operate in SecA targeting and that the preprotein is transferred from SecB to SecA by a mechanism
in which binding of the signal sequence to SecA promotes
the release of the mature preprotein domain from SecB for
binding to SecA.
[ 35S]-D8proOmpA (Van der Wolk et al., 1998) were synthesized by in vitro transcription/translation reaction, affinity purified (Crooke and Wickner, 1987) and stored frozen in 6 M
urea and 50 mM Tris-HCl, pH 7.8.
Construction of His-tagged wild-type and mutant SecB
Experimental procedures
Bacterial strains and plasmids
All strains were E. coli K-12 derivatives. Strains MC4100 (F¹
DlacU169 araD139 rps relA thiA motA ; Casadaban, 1976),
SE6004 (MC4100 alamBS60 prlA4 ; Emr et al., 1981) and
SF100 [F¹ DlacX74 galE galK thi rpsL (strA) DphoA(PvuII),
DompT ; Baneyx and Georgiou, 1990], bearing plasmid
pET340 (Van der Does et al., 1996) for overproduction of
the SecYEG complex, were used for the isolation of IMVs.
pET340 is a pTRC99A derivative with secYEG in tandem
and under the control of the trc promoter (Van der Does et
al., 1996). Overexpression of the hexaHis-tagged SecB proteins was carried out in strain MM152 (MC4100 secB ::Tn5
zhe ::Tn10 malTcon ; Kumamoto and Beckwith, 1985).
To create an Sph I site for cloning, the secB sequence was
changed from 58-CACATGTCA (encoding the first two residues of SecB, Met-Ser) to 58-CGCATGCCA (encoding MetPro) by oligonucleotide-directed mutagenesis as described
previously (Kimsey et al., 1995). The modified secB gene
was digested with Sph I and Msc I, and the resulting fragment
was cloned into pQE-32 (Qiagen). The DNA sequence of the
58 end of the resultant gene fusion was determined, and the predicted amino acid sequence of HisSecB was MRSGSH6GIRMP,
followed by the wild-type SecB sequence starting at amino
acid residue 3. Mutations in secB were introduced by replacing the Eco RV– Sac II secB fragment with fragments containing single base substitution mutations (Gannon and Kumamoto,
1993), yielding the plasmids pHKSB366, pHKSB387 and
pHKSB382 for HisSecB, HisSecBL75Q and HisSecBE77K
respectively.
Biochemicals
SecA (Cabelli et al., 1988), SecAN880 (Fekkes et al., 1997),
SecB (Weiss et al., 1988), proOmpA, D8proOmpA and OmpA
(Crooke et al., 1988) and GST and GST218 (Fekkes et al.,
1997) were purified as described. IMVs were prepared from
MC4100 or SE6004, or SF100 cells overproducing SecYEG,
as described previously (Chang et al., 1978) and treated with
6 M urea (Cunningham et al., 1989). [ 35S]-proOmpA and
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Overexpression and purification of hexaHis-tagged
SecB proteins
MM152 cells carrying F8 lacIQ lac þ and transformed with the
plasmids pHKSB366, pHKSB387 or pHKSB382 were grown
on LB at 378C to an OD660 of 1.0, induced with 1 mM IPTG
and grown for another 2 h. Cells were harvested by centrifugation (10 min, 10 000 × g ) and passed twice through a
1188 P. Fekkes et al.
French pressure cell (18 000 p.s.i.). After centrifugation
(15 min, 30 000 × g ), the supernatant was adjusted to 60 mM
imidazole, pH 8.0, and applied onto a nickel chelating column,
which was pretreated according to the manufacturer’s instructions (Pharmacia, Biotech). The column was washed with
several column volumes of 50 mM NaPi and 60 mM imidazole,
pH 8.0. His-tagged SecB proteins were eluted with a buffer
containing 50 mM NaPi , 500 mM imidazole and 300 mM
NaCl, pH 6.0. After overnight dialysis against 50 mM TrisHCl, pH 7.6, at 48C, the His-tagged proteins were applied to
a MonoQ FPLC anion-exchange column for further purification (Fekkes et al., 1997).
follows: purified GST fusion protein and SecB were preincubated at 208C for 10 min in PBS buffer (140 mM NaCl,
2.7 mM KCl, 10 mM Na2HPO4 , 1.8 mM K2HPO4 , pH 7.3). Subsequently, 10 ml of PBS buffer-washed, glutathione-conjugated
Sepharose 4B (Pharmacia) beads were added. After 10 min,
the mixture was transferred to a quick-spin column (Promega)
and centrifuged for 5 min at 5000 r.p.m. The column was
washed twice with 200 ml of PBS buffer, and GST was eluted
by a repeated wash with 50 ml of 20 mM reduced glutathione
in 50 mM Tris-HCl, pH 7.6. Eluted samples were pooled and
analysed by SDS–PAGE (Laemmli, 1970) and Coomassie
brilliant blue staining.
In vitro translocation of proOmpA
Other techniques
35
In vitro translocation of [ S]-proOmpA into E. coli IMVs was
assayed by its accessibility to added proteinase K (Cunningham and Wickner, 1989). Reaction mixtures (50 ml) contained
TB buffer [50 mM HEPES–KOH, pH 7.5, 30 mM KCl, 0.5 mg
ml¹1 BSA, 10 mM creatine phosphate, 5 mg ml¹1 creatine
kinase, 10 mM dithiothreitol (DTT) and 2 mM Mg(OAc)2 ],
1 mg of SecA, 1.6 mg of SecB, 1 ml of urea-denatured [ 35S]proOmpA and IMVs (15 mg of protein). This mixture was preincubated for 5 min at 378C, and the translocation reactions
were initiated by the addition of 2 mM ATP. After translocation, samples were treated with proteinase K (0.1 mg ml¹1 )
for 15 min on ice, precipitated with 7.5% (w/v) TCA, washed
with ice-cold acetone, solubilized in SDS sample buffer and
analysed by SDS–PAGE (Laemmli, 1970) and autoradiography. Translocation reactions were quantified by direct counting using the b-imager 2000 (Biospace Instruments).
Protein concentrations were determined according to the
method of Lowry et al. (1951) in the presence of SDS using
BSA as a standard. The SecB concentration was estimated
from quantitative total amino acid determination performed
by Eurosequence.
Binding studies
References
35
Binding of S-labelled proOmpA to His-tagged SecB was
measured as follows: SecB (9 mg) and 1 ml of [ 35S]-proOmpA
were incubated on ice in 50 ml of HB buffer (50 mM HEPES–
KOH, pH 7.6, 0.5 mg ml¹1 BSA). After 30 min, 50 ml of prewashed 50% Ni-NTA agarose slurry prewashed with HB
buffer was added and further incubated on ice for 25 min
with regularly vortexing. Subsequently, the mixture was centrifuged (5 min, 3000 r.p.m.) through a 200 ml sucrose cushion
(0.2 M sucrose in HB buffer). The pellet was washed with
150 ml of HB buffer, and the combined supernatant fractions
and the resulting pellet were counted in Optima Gold scintillation liquid (Packard Instrument).
Binding of [ 35S]-proOmpA, [ 35S]-D8proOmpA or [ 35S]-OmpA
to urea-treated IMVs derived from E. coli MC4100 cells was
measured as follows: SecB (100 nM), SecA (250 nM) and
[ 35S]-proOmpA, [ 35S]-D8proOmpA, or [ 35S]-OmpA (20 nM)
were incubated on ice in 100 ml of HB buffer. After 15 min,
the mixture was centrifuged (10 min, 30 000 p.s.i. in a Beckman airfuge at 48C) through a 100 ml sucrose cushion (0.2 M
sucrose in HB buffer), and the supernatant and the pellet
were counted in Optima Gold scintillation liquid (Packard).
Binding of 125I-labelled SecB to urea-treated IMVs derived
from E. coli MC4100 cells or SF100 cells overproducing
SecYEG was performed as described previously (Hartl et al.,
1990).
Binding of SecB to GST fusion protein was measured as
Acknowledgements
We would like to thank Tom Silhavy for the SE6004 strain,
and Bill Wickner for the synthetic proOmpA signal peptide.
These investigations were supported by a PIONIER grant of
the Netherlands Organization for Scientific Research (N.W.O.)
and the Netherlands Foundation for Chemical Research
(S.O.N.). H.J.K. and C.A.K. were supported by the National
Institute of Health (grant GM36415).
Altman, E., Emr, S.D., and Kumamoto, C.A. (1990a) The presence of both the signal sequence and a region of mature
LamB protein is required for the interaction of LamB with
the export factor SecB. J Biol Chem 265: 18154–18160.
Altman, E., Bankaitis, V.A., and Emr, S.D. (1990b) Characterization of a region in mature LamB protein that interacts
with a component of the export machinery of Escherichia
coli. J Biol Chem 265: 18148–18153.
Baneyx, F., and Georgiou, G. (1990) In vivo degradation of
secreted fusion proteins by the Escherichia coli outer
membrane protease OmpT. J Bacteriol 172: 491–494.
Breukink, E., Kusters, R., and De Kruijff, B. (1992) In vitro
studies on the folding characteristics of the Escherichia
coli precursor protein prePhoE. Eur J Biochem 208: 419–
425.
Breukink, E., Nouwen, N., Van Raalte, A., Mizushima, S.,
Tommassen, J., and De Kruijff, B. (1995) The C terminus
of SecA is involved in both lipid binding and SecB binding.
J Biol Chem 270: 7902–7907.
Cabelli, R.J., Chen, L., Tai, P.C., and Oliver, D.B. (1988)
SecA protein is required for secretory protein translocation
into E. coli membrane vesicles. Cell 55: 683–692.
Casadaban, M.J. (1976) Transposition and fusion of the lac
operon to selected promoters in Escherichia coli using
bacteriophage Lambda and Mu. J Mol Biol 104: 541–555.
Chang, C.N., Blobel, G., and Model, P. (1978) Detection of
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Preprotein transfer from SecB to SecA 1189
prokaryotic signal peptidase in an Escherichia coli
membrane fraction: endoproteolytic cleavage of
nascent f1 pre-coat protein. Proc Natl Acad Sci USA 75:
361–365.
Collier, D.N., Bankaitis, V.A., Weiss, J.B., and Bassford, Jr,
P.J. (1988) The antifolding activity of SecB promotes the
export of the E. coli maltose-binding protein. Cell 53: 273–
283.
Crooke, E., and Wicker, W. (1987) Trigger factor: a soluble
protein that folds proOmpA into a membrane assembly
competent form. Proc Natl Acad Sci USA 84: 5216–5220.
Crooke, E., Guthrie, B., Lecker, S., Lill, R., and Wickner, W.
(1988) ProOmpA is stabilized for membrane translocation
by either purified Escherichia coli trigger factor or canine
signal recognition particle. Cell 54: 1003–1011.
Cunningham, K., and Wickner, W. (1989) Specific recognition
of the leader region of precursor proteins is required for the
activation of translocation ATPase of Escherichia coli. Proc
Natl Acad Sci USA 86: 8630–8634.
Cunningham, K., Lill, R., Crooke, E., Rice, M., Moore, K.,
Wickner, W., et al. (1989) Isolation of SecA protein, a peripheral membrane protein of Escherichia coli plasma membrane that is essential for the functional binding and
translocation of proOmpA. EMBO J 8: 955–959.
De Cock, H., Overeem, W., and Tommassen, J. (1992) Biogenesis of outer membrane protein PhoE of Escherichia
coli : evidence for multiple SecB-binding sites in the mature
portion of PhoE. J Mol Biol 224: 369–379.
Den Blaauwen, T., Terpetschnig, E., Lakowicz, J.R., and
Driessen, A.J.M (1997) Interaction of SecB with soluble
SecA. FEBS Lett 416: 35–38.
Derman, A.A., Puziss, J.W., Bassford, Jr, P.J., and Beckwith,
J. (1993) A signal sequence is not required for protein
export in prlA mutants of Escherichia coli. EMBO J 12:
879–888.
Emr, S.D., Hanley-Way, S., and Silhavy, T.J. (1981) Suppressor mutations that restore export of a protein with a
defective signal sequence. Cell 23: 79–88.
Fekkes, P., Den Blaauwen, T., and Driessen, A.J.M. (1995)
Diffusion-limited interaction between unfolded polypeptides
and the Escherichia coli chaperone SecB. Biochemistry
34: 10078–10085.
Fekkes, P., Van der Does, C., and Driessen, A.J.M. (1997)
The molecular chaperone SecB is released from the carboxy-terminus of SecA during initiation of precursor protein
translocation. EMBO J 16: 6105–6113.
Fleischmann, R.D., Adams, M.D., White, O., Clayton, R.A.,
Kirkness, E.F., Kerlavage, A.R., et al. (1995) Whole-genome
random sequencing and assembly of Haemophilus influenzae Rd. Science 269: 496–512.
Flower, A.M., Doebele, R.C., and Silhavy, T.J. (1994) PrlA
and PrlG suppressors reduce the requirement for signal
sequence recognition. J Bacteriol 176: 5607–5614.
Francetic, O., Hanson, M.P., and Kumamoto, C.A. (1993)
prlA suppression of defective export of maltose-binding
protein in secB mutants of Escherichia coli. J Bacteriol
175: 4036–4044.
Gannon, P.M., and Kumamoto, C.A. (1993) Mutations of the
molecular chaperone protein SecB which alter the interaction between SecB and maltose-binding protein. J Biol
Chem 268: 1590–1595.
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190
Gannon, P.M., Li, P., and Kumamoto, C.A. (1989) The
mature portion of Escherichia coli maltose-binding protein
(MBP) determines the dependence of MBP on SecB for
export. J Bacteriol 171: 813–818.
Hardy, S.J.S., and Randall, L.L. (1991) A kinetic partitioning
model of selective binding of non-native proteins by the
bacterial chaperone SecB. Science 251: 439–443.
Hartl, F.-U., Lecker, S., Schiebel, E., Hendrick., J.P., and
Wickner, W. (1990) The binding cascade of SecB to SecA
to SecY/E mediates preprotein targeting to the E. coli
plasma membrane. Cell 63: 269–279.
Hoffschulte, H.K., Drees, B., and Müller, M. (1994) Identification of a soluble SecA/SecB complex by means of a subfractionated cell-free export system. J Biol Chem 269:
12833–12839.
Khisty, V.J., Munske, G.R., and Randall, L.L. (1995) Mapping
of the binding frame for the chaperone SecB within a natural
ligand, galactose-binding protein. J Biol Chem 270: 25920–
25927.
Kimsey, H.H., Dagarag, M.D., and Kumamoto, C.A. (1995)
Diverse effects of mutation on the effectivity of the Escherichia coli export chaperone SecB. J Biol Chem 270: 22831–
22835.
Kimura, E., Akita, M., Matsuyama, K., and Mizushima, S.
(1991) Determination of a region in SecA that interacts
with presecretory proteins in Escherichia coli. J Biol Chem
266: 6600–6606.
Kumamoto, C.A. (1991) Molecular chaperones and protein
translocation across the Escherichia coli inner membrane.
Mol Microbiol 5: 19–22.
Kumamoto, C.A., and Beckwith, J. (1985) Evidence for specificity at an early step in protein export in Escherichia coli.
J Bacteriol 163: 267–274.
Kumamoto, C.A., and Francetic, O. (1993) Highly selective
binding of nascent polypeptides by an Escherichia coli
chaperone protein in vivo. J Bacteriol 175: 2184–2188.
Kumamoto, C.A., and Gannon, P.M. (1988) Effects of Escherichia coli secB mutations on pre-maltose binding protein
conformation and export kinetics. J Biol Chem 263:
11554–11558.
Kumamoto, C.A., and Nault, A.K. (1989) Characterization of
the Escherichia coli protein- export gene secB. Gene 75:
167–175.
Laemmli, U.K. (1970) Cleavage of structural proteins during
the assembly of the head of bacteriophage T7. Nature
227: 680–685.
Lai, C.-Y., and Baumann, P. (1992) Sequence analysis of a
DNA fragment from Buchnera aphidicola (an endosymbiont of aphids) containing genes homologous to dnaG,
rpoD, cysE, and secB. Gene 119: 113–118.
Lecker, S.H., Lill, R., Ziegelhoffer, T., Georgopoulos, C.,
Bassford, Jr, P.J., Kumamoto, C.A., et al. (1989) Three
pure chaperone proteins of Escherichia coli, SecB, trigger
factor, and GroEL, form soluble complexes with precursor
proteins in vitro. EMBO J 8: 2703–2709.
Lecker, S.H., Driessen, A.J.M., and Wickner, W. (1990)
ProOmpA contains secondary and tertiary structure prior
to translocation and is shielded from aggregation by association with SecB protein. EMBO J 9: 2309–2314.
Lill, R., Dowan, W., and Wickner, W. (1990) The ATPase
activity of SecA is regulated by acidic phospholipids, SecY,
1190 P. Fekkes et al.
and the leader and mature domains of precursor proteins.
Cell 60: 271–280.
Liu, G., Topping, T.B., and Randall, L.L. (1989) Physiological
role during export for the retardation of folding by the
leader peptide of maltose-binding protein. Proc Natl Acad
Sci USA 86: 9213–9217.
Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J.
(1951) Protein measurement with Folin reagent. J Biol
Chem 193: 265–275.
MacIntyre, S., Mutschler, B., and Henning, U. (1991) Requirement of the SecB chaperone for export of a non-secretory
polypeptide in Escherichia coli. Mol Gen Genet 227: 224–
228.
Movva, N.R., Nakamura, K., and Inouye, M. (1980) Gene
structure of the OmpA protein, a major surface protein of
Escherichia coli required for cell–cell interaction. J Mol
Biol 143: 317–328.
Osborne, R.S., and Silhavy, T.J. (1993) PrlA suppressor
mutations cluster in regions corresponding to three distinct
topological domains. EMBO J 12: 3391–3398.
Randall, L.L., and Hardy, S.J.S (1986) Correlation of competence for export with lack of tertiary structure of the mature
species: a study in vitro of maltose-binding protein. Cell 46:
921–928.
Randall, L.L., Topping, T.B., and Hardy, S.J.S. (1990) No
specific recognition of leader peptide by SecB, a chaperone involved in protein export. Science 248: 860–863.
Randall, L.L., Topping, T.B., Hardy, S.J.S., Pavlov, M.Y.,
Freistroffer, D.V., and Ehrenberg, M. (1997) Binding of
SecB to ribosome-bound polypeptides has the same characteristics as binding to full-length, denatured proteins.
Proc Natl Acad Sci USA 94: 802–807.
Smith, V.F., Schwartz, L., Randall, L.L., and Smith, R.D.
(1996) Electrospray mass spectrometric investigation of
the chaperone SecB. Protein Sci 5: 488–494.
Tanji, Y., Gennity, J., Pollitt, S., and Inouye, M. (1991) Effect
of OmpA signal peptide mutations on OmpA secretion,
synthesis, and assembly. J Bacteriol 173: 1997–2005.
Topping, T.B., and Randall, L.L. (1994) Determination of the
binding frame within a physiological ligand for the chaperone SecB. Protein Sci 3: 730–736.
Van der Does, C., Den Blaauwen, T., De Wit, J.G., Manting,
E.H., Groot, N.A., Fekkes, P., et al. (1996) SecA is an
intrinsic subunit of the Escherichia coli preprotein translocase and exposes its carboxyl terminus to the periplasm.
Mol Microbiol 22: 619–629.
Van der Wolk, J.P.W., Fekkes, P., Boorsma, A., Huie, J.L.,
Silhavy, T.J., and Driessen, A.J.M. (1998) PrlA4 prevents
the rejection of signal sequence defective preproteins by
stabilizing the SecA–SecY interaction during the initiation
of translocation. EMBO J 17: 3631–3639.
Watanabe, M., and Blobel, G. (1989) Cytosolic factor purified
from Escherichia coli is necessary and sufficient for the
export of a preprotein and is a homotetramer of SecB.
Proc Natl Acad Sci USA 86: 2728–2732.
Watanabe, M., and Blobel, G. (1995) High-affinity binding of
Escherichia coli SecB to the signal sequence region of a
presecretory protein. Proc Natl Acad Sci USA 92: 10133–
10136.
Weiss, J.B., Ray, P.H., and Bassford, Jr, P.J. (1988) Purified SecB protein of Escherichia coli retards folding and
promotes membrane translocation of the maltose-binding
protein in vitro. Proc Natl Acad Sci USA 85: 8978–
8982.
Q 1998 Blackwell Science Ltd, Molecular Microbiology, 29, 1179–1190