Báo cáo Y học: Defective translocation of a signal sequence mutant in a prlA4 suppressor strain of Escherichia coli doc

Defective translocation of a signal sequence mutant in a
prlA4
suppressor strain of
Escherichia coli
Hendrik Adams
1
, Pier A. Scotti
2
*, Joen Luirink
2
and Jan Tommassen
1
1
Department of Molecular Microbiology and Institute of Biomembranes, Utrecht University, The Netherlands;
2
Department of
Microbiology, Institute of Molecular Biological Sciences, Biocentrum Amsterdam, The Netherlands
In the accompanying paper [Adams, H., Scotti, P.A., de
Cock, H., Luirink, J. & Tommassen, J. (2002) Eur. J. Bio-
chem. 269, 5564–5571.], we showed that the precursor of
outer-membrane protein PhoE of Escherichia coli with a Gly
to Leu substitution at position )10 in the signal sequence
(G-10L) is targeted to the SecYEG translocon via the signal-
recognition particle (SRP) route, instead of via the SecB
pathway. Here, we studied the fate of the mutant precursor
in a prlA4 mutant strain. prlA mutations, located in the secY
gene, have been isolated as suppressors that restore the
export of precursors with defective signal sequences.
Remarkably, the G-10L mutant precursor, which is nor-
mally exported in a wild-type strain, accumulated strongly in
a prlA4 mutant strain. In vitro cross-linking experiments

revealed that the precursor is correctly targeted to the prlA4
mutant translocon. However, translocation across the
cytoplasmic membrane was defective, as appeared from
proteinase K-accessibility experiments in pulse-labeled cells.
Furthermore, the mutant precursor was found to accumu-
latewhenexpressedinasecY40 mutant, which is defective in
the insertion of integral-membrane proteins but not in pro-
tein translocation. Together, these data suggest that SecB
and SRP substrates are differently processed at the SecYEG
translocon.
Keywords: inner membrane; prlA4; protein insertion; protein
translocation; Sec translocon.
Most proteins destined for the periplasm or the outer
membrane of Escherichia coli are transported across the
inner membrane by the membrane-embedded Sec system, a
complex consisting of the SecYEG translocon, the hetero-
trimer SecDFyajC and the peripheral ATPase SecA [1,2].
Targeting to the translocon is usually mediated by SecB,
which interacts with the mature portion of presecretory
proteins [3,4]. The SecB–preprotein complex is then targeted
to SecA, which in turn binds with high affinity to SecYEG
[5,6]. Upon initiation of translocation, SecB is released [7]
and the preprotein is translocated through the translocon by
an insertion–deinsertion cycle of SecA [8]. At the periplas-
mic side of the membrane, leader peptidase removes the
signal sequence from the precursor, and the mature protein
is released into the periplasm [9]. Targeting of inner-
membrane proteins is primarily controlled by the signal-
recognition particle (SRP) pathway [2,10,11]. The E. coli
SRP consists of a single protein, P48, and a 4.5S RNA, and

binds to particularly hydrophobic signal sequences when
they emerge from the ribosome [12,13]. The resulting SRP–
ribosome-nascent-chain (SRP–RNC) complex is then tar-
geted via FtsY to the inner membrane [14,15]. Upon release
of the SRP, the nascent chain inserts into the membrane
near the translocase components SecA, SecY, and SecG,
indicating that the SecB-targeting and SRP-targeting path-
ways converge at a common translocon [16].
The components of the Sec system were originally
identified by two different genetic approaches. One method
implicated the isolation of conditionally lethal mutants with
generalized secretion defects. The second approach eman-
ated from the idea that the signal sequence is recognized by
components of the export apparatus, and that specific
mutations in genes for Sec components would restore the
recognition of mutated signal sequences. Indeed, this
method resulted in the isolation of extragenic suppressor
mutations in prl (protein localization) genes directly
involved in protein translocation (prlA alleles of secY, prlD
alleles of secA and prlG alleles of secE) [17]. However, the
lack of allele specificity of prlA and prlG mutations with
respect to the suppression of signal sequence defects and the
observation that these prl mutants are able to translocate
even proteins without a signal sequence [18–20] argue
against the basic idea of the screening method. Studies with
a prlA4 suppressor strain revealed that the Sec translocon in
this strain facilitates translocation of preproteins with folded
domains [21], showed an increased affinity for SecA [22,23],
and is composed of subunits that are more loosely
associated than in the wild-type strain [24]. Therefore, it

has been proposed that prlA mutations cause a general
relaxation of the export apparatus [21,24] rather than a
specific change that results in bypassing a proofreading
mechanism of the Sec machinery [25].
The energy for precursor translocation is supplied by the
hydrolysis of ATP by SecA and by the proton-motive force
Correspondence to J. Tommassen, Department of Molecular
Microbiology, Utrecht University, Padualaan 8, 3584 CH Utrecht,
The Netherlands. Fax: + 31 30 2513655, Tel.: + 31 30 2532999,
E-mail:
Abbreviations: SRP, signal-recognition particle; RNC, ribosome-
nascent chain; pmf, proton-motive force; DSS, disuccinimidyl-
suberate; IMV, inverted inner-membrane vesicle; Sak, Staphylokinase;
TF, trigger factor; TMS, transmembrane segment.
*Present address:IECB-E
´
cole polytechnique ENSCPB,
Talence cedex, France.
(Received 29 May 2002, revised 10 September 2002,
accepted 16 September 2002)
Eur. J. Biochem. 269, 5572–5580 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03263.x
(pmf) [26]. Energy from ATP binding and hydrolysis is
probably used to confer conformational changes in the
SecA molecule, which lead to a cycle of insertion and
deinsertion of SecA into the membrane [8,27] and move-
ment of the precursor across the membrane [28]. The
mechanism by which the pmf stimulates the translocation
process is less clear. However, recent experiments indicate
that the insertion of the signal sequence in a transmembrane
orientation is stimulated by the pmf [29,30]. In our

laboratory, we use the SecB-dependent outer-membrane
protein PhoE [3,31] as a model protein to study protein
transport. In previous studies, we showed that a single
amino-acid substitution, G-10L (the residue at position )1
precedes the signal-peptidase cleavage site), in the hydro-
phobic core of the signal sequence of PhoE relieved the pmf
dependency of protein translocation [29] and shunts the
precursor via the SRP pathway to the Sec machinery [32].
Other experiments revealed that the prlA4 mutation in secY
reduced the pmf dependency of protein translocation of
wild-type precursors [21]. The initial goal of the present
study was to investigate whether the prlA4 mutation is able
to suppress the pmf dependency of the translocation of
(G-10L)prePhoE even further. Instead, we found a strong
accumulation of the (G-10L)PhoE precursor in the prlA4
mutant strain, and the step that was blocked in the
biogenesis pathway was identified.
MATERIALS AND METHODS
Reagents and biochemicals
Restriction enzymes were purchased from either Boehringer
Mannheim or Pharmacia. MEGAshortscript T7 transcrip-
tion kit was from Ambion, and [
35
S]methionine and
Tran
35
S-label were from Amersham International. The
cross-linker disuccinimidyl-suberate (DSS) and Super Signal
West Pico Chemiluminescent Substrate were from Pierce.
Bacterial strains and plasmids

The E. coli strains and plasmids used in this study are listed in
Table 1. To obtain a prlA4 derivative of strain CE1224, a
Tn10 marker was introduced close to the prlA4 allele in strain
NT1004 by P1 transduction using strain CAG12071 as the
donor and selection for tetracycline-resistant transductants.
Subsequently, the resulting strain, CE1510, was used as the
donor in a P1 transduction experiment with CE1224 as the
recipient. To confirm the prlA4 phenotype, tetracycline-
resistant transductants were transformed with plasmid
pNN6 and tested in pulse-labeling experiments for suppres-
sion of the secretion defect of (G-10R)prePhoE, which
contains a charged residue in the hydrophobic core of the
signal sequence. Tetracycline-resistant prlA
+
and prlA4
Table 1. Bacterial strains and plasmids used in this study. Cam
r
and Amp
r
indicate resistance to chloramphenicol and ampicillin, respectively.
Designation Relevant characteristics Description/reference
Strains
CE1224 F
–
, thr leu D(proA-proB-phoE-gpt) his thi argE lacY galK xyl rpsL supE ompR [57]
MC4100 F
–
, DlacU169 araD139 rpsL thi relA [58]
NT1060 MC4100 ptsF25 deoC1 lamBD60 T.J. Silhavy (pers. comm.)
NT1004 NT1060 prlA4 T.J. Silhavy (pers. comm.)

CAG12071 zhd-3082::Tn10 [59]
CE1510 NT1004 zhd-3082::Tn10 This study
CE1511 CE1224 zhd-3082::Tn10 This study
CE1512 CE1224 prlA4 zhd-3082::Tn10 This study
POP1730 ompR::Tn10 DmalB K. Bauer
AF111 MC4100 lamB14D [60]
ROA7 AF111 prlA7 T.J. Silhavy
ROA11 AF111 prlA11 T.J. Silhavy
ROA202 AF111 prlA202 T.J. Silhavy
CE1518 AF111 ompR::Tn10 This study
CE1519 ROA7 ompR::Tn10 This study
CE1520 ROA11 ompR::Tn10 This study
CE1521 ROA202 ompR::Tn10 This study
CU165 MC4100 secY40 zhd33::tet [44]
Plasmids
pLep-WT Amp
r
, lepB behind ara promoter [40]
pJP29 Cam
r
, wild-type phoE [33]
pNN6 pJP29 derivative encoding (G-10R)prePhoE [29]
pNN8 pJP29 derivative encoding (G-10L)prePhoE [29]
pNN100 Amp
r
, phoE gene behind tac promoter [61]
pHA106 pNN100 derivative encoding (G-10R)prePhoE This study
pHA108 pNN100 derivative encoding (G-10L)prePhoE This study
pC4Meth94PhoE Amp
r

, encodes truncated 94PhoE [12]
pC4Meth101FtsQ-WT Amp
r
, encodes truncated 101FtsQ [12]
pC4Meth(G-10L)94PhoE pC4Meth94PhoE derivative encoding (G-10L) mutant PhoE [32]
Ó FEBS 2002 Defective translocation in a prlA4 mutant (Eur. J. Biochem. 269) 5573
transductants were designated CE1511 and CE1512, respect-
ively. CE1518-1521 strains were obtained by P1 transduction
using strains AF111, ROA7, ROA11 and ROA202, respect-
ively, as recipients and strain POP1730 as the donor.
For the construction of plasmids pHA106 and pHA108,
encoding the G-10R and G-10L mutant PhoE, respectively,
under control of the tac promoter, a PstI/BamHI fragment
from pNN100 containing the signal-sequence-encoding part
of phoE was replaced by the corresponding fragment of
pNN6 and pNN8, respectively.
Pulse-labeling experiments
Strains carrying plasmid pJP29 or derivatives were grown
under phosphate limitation at 30 °C to induce PhoE
expression [33]. Strains, carrying plasmids with phoE under
tac promoter control, were grown for 4 h at 30 °Cin
synthetic minimal medium [34] supplemented with
100 lgÆmL
)1
ampicillin and 0.4% glucose. PhoE expression
was induced under these conditions with 1 m
M
isopropyl
b-
D

-thiogalactopyranoside for 1 h. Cells were pulse-labeled
for 45 s with Tran
35
S-label followed either by a chase period
with an excess of nonradioactive methionine/cysteine or by
chilling on ice. Subsequently, proteins were precipitated
with 5% (w/v) trichloroacetic acid, followed by immuno-
precipitation with a polyclonal PhoE-specific antiserum [35].
The precipitated proteins were separated by SDS/PAGE
[36] and visualized by autoradiography. Radiolabeled
proteins were quantified using the Imagequant software
(Molecular Dynamics) after scanning of the autoradiogram.
In vitro
transcription, translation, targeting
and cross-linking analysis
To generate truncated mRNA, plasmids encoding truncated
nascent chains of FtsQ or (G-10L)prePhoE (Table 1) were
linearized and transcribed as described [12]. The resulting
mRNAs, encoding (G-10L)94PhoE and 101FtsQ, were
translated in vitro in a lysate of strain MC4100 as described
[12,37] to produce RNC complexes. To allow SRP–RNC
complex formation, 350 n
M
reconstituted SRP was added
to the translation reaction [16]. After 5 min of incubation at
25 °C, samples were chilled on ice, and the resulting SRP–
RNC complexes were purified from the translation mixture
by centrifugation through a high-salt sucrose cushion [38]
and resuspended in RN buffer (100 m
M

potassium acetate,
5m
M
magnesium acetate, 2.5 m
M
Hepes/KOH, pH 7.9).
Inverted inner-membrane vesicles (IMVs) were isolated as
described [37]. Targeting reactions were performed as
described previously [16] by incubating purified SRP–
RNC complexes together with 1 m
M
FtsY, 50 l
M
GTP,
50 l
M
ATP, and IMVs (1.25 mgÆmL
)1
protein). Cross-
linking was induced with 1 m
M
DSSfor10minat25°C
and quenched at 0 °C by adding 0.1 vol. quenching buffer
(1
M
glycine/100 m
M
NaHCO
3
, pH 8.5). To separate

integral membrane from soluble and peripheral cross-linked
complexes, samples were treated with 0.18
M
Na
2
CO
3
(pH 11.3) for 15 min on ice. The membrane fractions
containing integral-membrane proteins were pelleted by
ultracentrifugation (10 min, 110 000 g) and resuspended in
RN buffer. Supernatant and pellet fractions were precipi-
tated with 10% (w/v) trichloroacetic acid, washed with cold
acetone, and resuspended in RN buffer. Samples were
immunoprecipitated as described [39] or mixed directly with
2 · SDS/PAGE gel loading buffer before electrophoresis.
Samples were analyzed on 12% or 4–15% gradient SDS/
polyacrylamide gels. Radiolabeled proteins were visual-
ized by phosphorimaging using a PhosphorImager 473
(Molecular Dynamics).
In vivo
membrane-targeting assay
Membrane targeting of leader peptidase was studied in vivo
essentially as described [40]. Briefly, cells containing pLep-
WT, encoding leader peptidase, were induced with 0.2%
arabinose and labeled in the mid-exponential phase with
Tran
35
S-label. After spheroplasting, the cells were treated
with proteinase K to degrade translocated proteins, fol-
lowed by immunoprecipitation [35] with polyclonal anti-

bodies directed against leader peptidase and analysis by
SDS/PAGE and autoradiography.
Proteinase K-accessibility experiments
Cells of prlA4 mutant strain NT1004, carrying plasmid
pNN8, were grown under phosphate limitation to induce
PhoE expression [33] before pulse-labeling with Tran
35
S-
label for 60 s. Directly after the pulse, an excess of
nonradioactive methionine and cysteine was added, and
the cells were collected by centrifugation (2 min, 16 000 g).
For spheroplasting, cells were resuspended in ice-cold buffer
A [40% (w/v) sucrose, 1.5 m
M
EDTA, 33 m
M
Tris/HCl
(pH 8.0)] and incubated with lysozyme (final concentration
5 lgÆmL
)1
). After 10 min incubation on ice, incubation was
continued at 37 °C for 10 min followed by addition of
10 m
M
MgCl
2
. Aliquots of the spheroplast suspension were
incubated on ice for 30 min in the presence or absence of
proteinase K (final concentration 50 lgÆmL
)1

). Subse-
quently, 2 m
M
phenylmethanesulfonyl fluoride was added
to the cell suspension, and incubation was continued for
5 min on ice. Proteins were precipitated with 5% (w/v)
trichloroacetic acid and analyzed by SDS/PAGE and
autoradiography.
Western immunoblot analysis
Total cellular proteins were solubilized in sample buffer for
10 min at 100 °C, followed by separation by SDS/PAGE
on 15% polyacrylamide gels. After transfer of proteins to
nitrocellulose filters (0.45 lm; Schleicher and Schuell) using
a Mini Trans-Blot Cell (Bio-Rad Laboratories), the blots
were incubated with antibodies directed against SecB [41]
and developed by chemiluminescence according to the
manufacturer’s (Pierce) recommendations.
RESULTS
Accumulation of (G-10L)prePhoE in a
prlA4
mutant strain
We have previously shown that the prlA4 mutation in secY
reduced the pmf dependency of protein translocation [21].
Similarly, a single amino-acid substitution, replacing the
helix-breaking glycine at position )10 in the hydrophobic
core of the PhoE signal sequence by leucine (G-10L),
relieved the pmf dependency of PhoE protein translocation
5574 H. Adams et al.(Eur. J. Biochem. 269) Ó FEBS 2002
[29]. To investigate whether the reported effects are additive,
we studied the in vivo translocation kinetics of wild-type and

mutant PhoE in prlA4 mutant strain CE1512, using
processing as a criterion for translocation. The prlA4
phenotype of the strain was confirmed by studying the
processing kinetics of (G-10R)prePhoE. The translocation
of this mutant precursor, which contains a charged residue
in the hydrophobic core of the signal sequence, was severely
hampered in the wild-type strain and considerably improved
in the prlA4 mutant (Fig. 1A). Directly after the pulse, the
mature form of the G-10R mutant was barely detectable in
the wild-type strain, whereas the majority of the protein was
already processed in the prlA4 mutant. Processing was
completed in the prlA4 cells after a 2-min chase period,
whereas in the wild-type cells, at this time point, still about
50% of the synthesized (G-10R)prePhoE was unprocessed.
Interestingly, whereas the (G-10L) mutant precursor was
efficiently processed in wild-type cells, it accumulated
abundantly in the prlA4 suppressor strain, indicating that
its translocation across the inner membrane was affected by
the prlA4 mutation. Thus, whereas the prlA4 mutation has
been isolated as a strong suppressor of signal sequence
defects, it seemed to have an adverse effect on the
translocation of the G-10L PhoE precursor.
To test whether the (G-10L)prePhoE accumulation is
specific for the prlA4 allele, we determined the effect of
several other prlA suppressor mutations with the amino-
acid substitutions being located in distinct topological
domains of SecY [25]. Again, suppression of the transloca-
tion defect of (G-10R)prePhoE was used as a control for the
suppressor phenotype (Fig. 1B). Quantification of the
results revealed that 39% of the total amount of radiolabe-

led G-10R mutant PhoE was processed directly after the
pulse in the wild-type strain. In the prlA7 and prlA202
suppressor strains, processing was improved and the
amount of mature PhoE increased to 50% and 60%,
respectively, of the total amount of PhoE synthesized during
the pulse. The prlA11 suppressor increased the amount of
PhoE only by 2% compared with wild-type cells. Although
the prlA7 and prlA202 alleles tested improved the pro-
cessing of (G-10R)prePhoE, precursor accumulation of
(G-10L)prePhoE was not observed in these prlA suppres-
sors. In conclusion, these data suggest that (G-10L)prePhoE
accumulation is specific for the prlA4 allele.
Targeting of G-10L nascent PhoE to the PrlA4 Sec
translocon
Whereas wild-type PhoE is targeted to the Sec translocon by
SecB [3,31], targeting of (G-10L)prePhoE is mediated by
SRP [32]. Therefore, the accumulation of (G-10L)prePhoE
in the prlA4 mutant may result from a defect in SRP-
mediated targeting to the mutant translocon. To study this
possibility, we examined the targeting of (G-10L)prePhoE
nascent chains to SecY in vitro in cross-linking experiments.
After translation, RNCs of (G-10L)94PhoE polypeptides
were saturated with reconstituted SRP. The SRP–RNC
complexes were purified and incubated with IMVs, derived
from either a wild-type or a prlA4 mutant strain to allow
targeting. After cross-linking with the bifunctional cross-
linking reagent DSS, peripheral and soluble cross-linked
complexes were separated from integral-membrane cross-
linked complexes by Na
2

CO
3
extraction, and the complexes
were analyzed by SDS/PAGE (Fig. 2A). In the Na
2
CO
3
pellet, at least two (G-10L)94PhoE cross-linked complexes,
one at  110 kDa and one at  46 kDa, could be detected
with both wild-type and prlA4 IMVs (Fig. 2A, lanes 4 and
5). In both cases, the 110-kDa complex could be immuno-
precipitated with antiserum directed against SecA, indica-
ting that it is a complex of the radiolabeled (G-10L)94PhoE
Fig. 1. Processing kinetics of prePhoE and mutant derivatives in wild-type and prlA mutant strains. (A) Cells of prlA
+
strain CE1511 (prlA
+
)orprlA4
mutant CE1512, carrying plasmids encoding wild-type PhoE (WT) or mutant derivatives with G-10R or G-10L substititutions in the signal
sequence, were grown under phosphate limitation to express PhoE. The cells were pulse-labeled at 30 °Cfor45swithTran
35
S-label and chased
with unlabeled methionine/cysteine for the indicated periods. The proteins were analyzed, after immunoprecipitation with antibodies against PhoE,
by SDS/PAGE and autoradiography. The precursor and mature forms of PhoE are indicated by p and m, respectively. (B) Cells of strains CE1518
(prlA
+
), CE1519 (prlA7), CE1520 (prlA11) and CE1521 (prlA202), carrying plasmids encoding wild-type PhoE (WT) or mutant derivatives under
control of the tac promoter, were grown in synthetic minimal medium at 30 °C until the mid-exponential phase. After 1 h induction with isopropyl
b-
D

-thiogalactoside, aliquots of each culture were pulse-labeled with Tran
35
S-label for 45 s, followed by the addition of an equal volume of ice-cold
10% trichloroacetic acid. Radiolabeled proteins were subsequently analyzed as described under (A). G, R and L indicate PhoE proteins with a Gly,
Arg and Leu at position )10 in the signal sequence () 10), respectively.
Ó FEBS 2002 Defective translocation in a prlA4 mutant (Eur. J. Biochem. 269) 5575
and SecA (Fig. 2B, lanes 1 and 3). In addition, cross-linking
adducts of  220 kDa and  40 kDa were also immuno-
precipitated from the Na
2
CO
3
pellet with SecA antiserum.
We assume that the  220-kDa product corresponds to
cross-linked complexes between (G-10L)94PhoE and the
dimeric form of SecA. The  40-kDa product in the
Na
2
CO
3
pellet probably contains proteolytic fragments of
the SecA cross-linked products, which is in agreement with
earlier reports [42]. The fuzzy  46-kDa product (Fig. 2A,
lanes 4 and 5) was immunoprecipitated with anti-SecY
serum (Fig. 2B, lanes 2 and 4), demonstrating that the
(G-10L)94PhoE nascent chains are targeted to the Sec
translocon in the prlA4 IMVs as well as in wild-type IMVs.
The small difference in the electrophoretic mobilities of the
PrlA4-(G-10L)94PhoE adduct and SecY-(G-10L)94PhoE
adduct (Fig. 2A, lanes 4 and 5) probably results from the

amino-acid substitutions in the mutant PrlA4 protein.
In the Na
2
CO
3
supernatant, at least three major cross-
linking adducts, of apparent molecular mass  110,  65
and  55 kDa, could be detected, both when wild-type and
when prlA4 IMVs were present (Fig. 2A, lanes 8 and 9).
Immunoprecipitation revealed that these adducts represent
cross-linking to SecA (data not shown), trigger factor (TF)
and P48 (Fig. 2C), respectively, demonstrating that (G-
10L)94PhoEinteractswithP48andTFinbothcases.As
described in the accompanying paper [32], several additional
low-molecular-mass cross-linking adducts (< 30 kDa)
were revealed (Fig. 2A, lanes 8 and 9), but the identity of
these complexes is unknown. To investigate whether the
cross-linking adducts are similar to those of other substrates
of the SRP pathway, FtsQ was used as a model substrate.
This class II membrane protein, with a short N-terminal
cytoplasmic tail [43], was synthesized as a slightly longer
nascent chain (101 residues) than (G-10L)94PhoE to expose
properly its signal-anchor domain. Indeed, 101FtsQ inter-
actedproperlywithbothSecYandSecA(Fig.2A,lanes13
and 14) as demonstrated by immunoprecipitations with
anti-SecY and anti-SecA serum, respectively (Fig. 2B, lanes
5–8). In conclusion, the results show that the targeting of the
(G-10L)94PhoE nascent chains and other SRP substrates to
the SecY protein in the translocon is not defective in the
prlA4 mutant.

(G-10L)prePhoE is inefficiently translocated
in the
prlA4
mutant
Whereas the targeting of the (G-10L)prePhoE and other
SRP substrates to the translocon is apparently unaffected in
the prlA4 mutant, their subsequent insertion into the mutant
translocon might be impaired. To test this possibility,
protease-accessibility experiments were conducted after
pulse-labeling of cells expressing the G-10L mutant protein
(Fig. 3). Indeed, the precursor of G-10L PhoE that
accumulated in the prlA4 mutant was not sensitive to
proteinase K after spheroplasting of the cells (Fig. 3, lane
2), whereas mature PhoE, which is translocated, was
degraded. These results show that the precursor of
(G-10L)PhoE, although correctly targeted to the translocon,
Fig. 2. Targeting of SRP–RNCs to the prlA4 mutant Sec translocon.
[
35
S]Methionine-labeled (G-10L)94PhoE or 101FtsQ was incubated
with 350 n
M
reconstituted SRP. The SRP–RNCs were subsequently
purified and incubated with NT1060 (prlA
+
) and NT1004 (prlA4)
IMVs. The cross-linker DSS was used to analyze SRP–RNC interac-
tions. After quenching, peripherally bound and soluble proteins were
separated from the inner membranes by carbonate extraction. Samples
were either directly (A) or after immunoprecipitation with the indica-

tedantisera(BandC)subjectedtoSDS/PAGE,andcross-linked
complexes were visualized with a PhosphorImager. The positions of
molecularmassmarkerproteins(MW)areindicatedontheright.
Arrowheads indicate relevant cross-linked complexes.
5576 H. Adams et al.(Eur. J. Biochem. 269) Ó FEBS 2002
is inefficiently translocated across the inner membrane in a
prlA4 mutant.
To investigate whether membrane insertion of SRP
substrates is also defective in a prlA4 mutant, we examined
the insertion kinetics of leader peptidase [40]. Directly after
pulse-labeling, leader peptidase already appeared to be
accessible to proteinase K in spheroplasts of both the wild-
type and the prlA4 mutant strain (Fig. 4A). The intactness
of the spheroplasts was confirmed by the inaccessibility to
proteinase K of SecB, which was only degraded after
solubilization of the spheroplasts with Triton X-100
(Fig. 4B). These results indicate that insertion of the SRP
substrate leader peptidase into the Sec translocon is not
affected in the prlA4 mutant.
Processing of (G-10L)prePhoE is impaired in secY40
mutant cells
The results presented so far indicate that the prlA4 mutation
has distinct effects on the translocation kinetics of a SecB-
targetedandanSRP-targetedprecursor,i.e.wild-typeand
G-10L mutant PhoE, respectively. A possible explanation is
that these substrates are targeted to different domains of
SecYEG. A cold-sensitive secY40 mutant has previously
been described that is impaired in inner-membrane protein
insertion, whereas protein export is unaffected [44]. In
addition, this result suggested that different regions of SecY

may be involved in protein export and membrane protein
insertion. To test whether the secY40 mutation affects the
translocation of wild-type and (G-10L)prePhoE differently,
pulse–chase experiments were conducted. Indeed, even at
the permissive temperature, the (G-10L)prePhoE was only
slowly processed in the secY40 mutant, whereas wild-type
prePhoE was completely processed within 30 s after the
pulse (Fig. 5). These results are consistent with the hypo-
thesis that wild-type and G-10L mutant prePhoE are
targeted to different domains of the SecYEG translocon.
DISCUSSION
Signal sequences often contain an a-helix breaker in the
hydrophobic core [45]. We have previously shown that
substitution of the helix breaker in the signal sequence of
Fig. 4. In vivo membrane insertion of leader peptidase in wild-type and
prlA4 mutant cells. (A) Cells of strain NT1060 (prlA
+
)oritsprlA4
derivative NT1004, carrying plasmid pLep-WT, were grown in syn-
thetic minimal medium to mid-exponential phase at 30 °C. Expression
of leader peptidase was induced for 5 min by the addition of 0.2%
arabinose. Subsequently, cells were pulse-labeled for 15 s with Tran
35
S-
label and chased with unlabeled methionine/cysteine for the indicated
periods. Insertion of leader peptidase was assessed by analyzing its
protease accessibility after spheroplasting of the cells. The proteins
were analyzed after immunoprecipitation with anti-Lep antibodies by
SDS/PAGE and autoradiography. (B) As a control for membrane
intactness, the spheroplasts were treated with proteinase K (PK),

without or with solubilization by Triton X-100 (TX-100) and analyzed
by immunoblotting using antiserum directed against SecB.
Fig. 3. Proteinase K accessibility of (G-10L)prePhoE in prlA4 mutant
cells. NT1004 (prlA4) cells, expressing (G-10L)prePhoE from plasmid
pNN8, were pulse-labeled for 60 s with Tran
35
S-label, followed by the
addition of an excess of methionine/cysteine. After spheroplasting, the
cells were incubated in the absence or presence of proteinase K (PK).
Subsequently, the undigested proteins were precipitated with
trichloroacetic acid and analyzed by SDS/PAGE and autoradio-
graphy. The precursor and mature forms of PhoE are indicated by p
and m, respectively.
Fig. 5. Processing of (G-10L)prePhoE is impaired in secY40 cells. Cells
of strain CU165 (secY40) carrying plasmid pJP29 encoding wild-type
PhoE (WT) or pNN8 encoding (G-10L)prePhoE, were grown under
phosphate limitation to express PhoE. The cells were pulse-labeled
at 37 °Cfor45swithTran
35
S-label and chased with unlabeled
methionine/cysteine for the indicated periods. Radiolabeled proteins
were analyzed by SDS/PAGE and autoradiography. The precursor
and mature forms of PhoE are indicated by p and m, respectively.
Ó FEBS 2002 Defective translocation in a prlA4 mutant (Eur. J. Biochem. 269) 5577
PhoE by a helix-promoting residue (e.g. G-10L) relieved the
pmf dependency of the translocation of this precursor [29].
Furthermore, we showed that such substitutions resulted in
the re-routing of this SecB substrate to the SRP-targeting
pathway [32]. In the present study, we show that (G-
10L)prePhoE accumulates in its precursor form when

expressedinaprlA4 mutant strain and that its translocation
across the inner membrane is impaired.
The results described here for (G-10L)prePhoE are
reminiscent of those reported previously for staphylokinase
(Sak), a secreted protein from the Gram-positive bacterium
Staphylococcus aureus, which was efficiently processed and
exported to the periplasm in wild-type E. coli cells, but
accumulated in its precursor form when expressed in a prlA4
mutant of E. coli [46]. Sequence examination suggests the
presence of an unusually long a-helix in the core region of
the Sak signal sequence as is the case in the (G-10L)prePhoE
signal sequence. The export defect of Sak in the prlA4
mutant was suppressed by a small four-amino-acid deletion
or by amino-acid substitutions introducing a strong helix
breaker, such as glycine, into the a-helical core region of the
signal sequence [46]. Similarly, streptokinase, an extracellu-
lar protein of streptococcal strains, is also blocked from
being secreted in E. coli prlA4 mutant cells [47], presumably
because of the long a-helix in its signal sequence.
The prlA4 allele actually contains two missense mutations
in the secY gene, resulting in the amino-acid substitutions
F286Y and I408N in transmembrane segments (TMS) 7
and 10, respectively [48]. The mutation in TMS 10 is
responsible for the suppressor phenotype and enables the
translocation of preproteins with a defective or completely
missing signal sequence [20,25,49]. Furthermore, this muta-
tion was reported to result in a looser association between
the subunits of the SecYEG translocon [24] and in increased
affinity of SecA for the SecY protein [23]. The mutation in
TMS 7 was probably acquired as a secondary mutation

which restored the stability while conserving the flexibility of
the system. Further, it was shown that the mutation in
TMS 7 (F286Y) is responsible for the observed export
defect of Sak [46,48,50], suggesting that this defect is not
related to the suppressor phenotype. Consistently, we did
not observe a processing defect when (G-10L)prePhoE was
expressedinotherprlA suppressor strains, i.e. prlA7 (A277E
in TMS 7 and L407R in TMS 10), prlA11 (V407R in
TMS 10 and V411G in TMS 10), and prlA202 (I287S in
TMS 7). Therefore, we speculate that the export defect of
(G-10L)prePhoE, like that of Sak, is caused by the
secondary mutation in TMS 7 of SecY.
Recent studies have indicated that the SecYEG complex
facilitates the insertion of inner-membrane proteins in
addition to catalyzing protein translocation [51,52]. Mem-
brane protein insertion and protein translocation are two
distinct processes that are likely to impose different func-
tional requirements on the translocase. The mechanism by
which the Sec complex mediates both of these functions is still
unclear. Available evidence indicates that the translocon
forms an aqueous channel that permits the translocation of
polypeptides into the periplasm [53,54]. However, it can also
open laterally to allow the exit of transmembrane regions
into the lipid bilayer [55]. Possibly, the passage of exported
proteins and the insertion of integral membrane proteins are
facilitated by distinct regions of the translocon. Indeed,
studies in eukaryotes indicated that signal sequences of
secreted proteins and signal-anchor domains of membrane
proteins are positioned differently in the Sec61p translocon
[56]. Recently, a secY40 mutant of E. coli (carrying an A363S

substitution in cytoplasmic domain 5 of SecY) was shown to
be defective in inner-membrane protein insertion, whereas
protein export was unaffected by the mutation [44]. Our data
show that the translocation of the SRP substrate (G-
10L)prePhoE is affected by the secY40 mutation as well,
whereas the SecB substrate wild-type PhoE is not, indica-
ting that these highly related precursors are differently
processed at the SecYEG translocon. Furthermore, the (G-
10L)prePhoE accumulation in prlA4 cells is caused by
inefficient translocation, whereas wild-type prePhoE is
correctly translocated. However, the membrane insertion
of another SRP substrate, leader peptidase, was not affected
by the prlA4 mutation. To explain our results, we propose
that SRP and SecB substrates are targeted to different
domains of the translocon. The cytoplasmic domain 5, where
the secY40 mutation is located, is involved in the docking of
SRP substrates. Also after docking at the translocon, SRP
substrates and SecB substrates are differently processed by
the translocon. In the case of SRP substrates, the translocon
opens laterally to allow the insertion of integral-membrane
proteins. However, when the hydrophobic a-helix is too
short to span the inner membrane, the SRP substrate may be
transferred to the translocation pathway, which is normally
used by SecB substrates. This transfer appears to be defective
in the prlA4 mutant.
ACKNOWLEDGEMENTS
We would like to thank Malene Urbanus for her efforts with the
cross-linking experiments. Furthermore, we would like to thank
Elaine Eppens and Margot Koster for helpful discussions and
interest in the work, William Wickner, Annemieke van Dalen and

Arnold Driessen for providing antisera, and Katharina Bauer, Ann
Flower, Chris Harris and Tom Silhavy for their gifts of strains. This
work was supported by EU grant HPRN-CT-2000-00075 from the
European Community.
REFERENCES
1. Driessen, A.J., Fekkes, P. & van der Wolk, J.P. (1998) The Sec
system. Curr. Opin. Microbiol. 1, 216–222.
2. Mu
¨
ller, M., Koch, H.G., Beck, K. & Scha
¨
fer, U. (2001) Protein
traffic in bacteria: multiple routes from the ribosome to and across
the membrane. Prog. Nucleic Acid Res. Mol. Biol. 66, 107–157.
3. de Cock, H., Overeem, W. & Tommassen, J. (1992) Biogenesis of
outer-membrane protein PhoE of Escherichia coli. Evidence for
multiple SecB-binding sites in the mature portion of the PhoE
protein. J. Mol. Biol. 224, 369–379.
4. Knoblauch, N.T., Rudiger, S., Schonfeld, H.J., Driessen, A.J.,
Schneider-Mergener, J. & Bukau, B. (1999) Substrate specificity of
the SecB chaperone. J. Biol. Chem. 274, 34219–34225.
5. Manting, E.H., van der Does, C. & Driessen, A.J. (1997) In vivo
cross-linking of the SecA and SecY subunits of the Escherichia coli
preprotein translocase. J. Bacteriol. 179, 5699–5704.
6. van der Does, C., Manting, E.H., Kaufmann, A., Lutz, M. &
Driessen, A.J. (1998) Interaction between SecA and SecYEG in
micellar solution and formation of the membrane-inserted state.
Biochemistry 37, 201–210.
7. Fekkes, P., van der Does, C. & Driessen, A.J. (1997) The mole-
cular chaperone SecB is released from the carboxy-terminus of

SecA during initiation of precursor protein translocation.
EMBO J. 16, 6105–6113.
5578 H. Adams et al.(Eur. J. Biochem. 269) Ó FEBS 2002
8. Economou, A., Pogliano, J.A., Beckwith, J., Oliver, D.B. &
Wickner, W. (1995) SecA membrane cycling at SecYEG is driven
by distinct ATP binding and hydrolysis events and is regulated by
SecD and SecF. Cell 83, 1171–1181.
9. Dalbey, R.E. & Wickner, W. (1985) Leader peptidase catalyzes the
release of exported proteins from the outer surface of the
Escherichia coli plasma membrane. J. Biol. Chem. 260,
15925–15931.
10. de Gier, J.W., Valent, Q.A., von Heijne, G. & Luirink, J. (1997)
The E.coliSRP: preferences of a targeting factor. FEBS Lett. 408,
1–4.
11. Herskovits, A.A., Bochkareva, E.S. & Bibi, E. (2000) New
prospects in studying the bacterial signal-recognition particle
pathway. Mol. Microbiol. 38, 927–939.
12. Valent, Q.A., de Gier, J.W., von Heijne, G., Kendall, D.A., ten
Hagen-Jongman, C.M., Oudega, B. & Luirink, J. (1997) Nascent
membrane and presecretory proteins synthesized in Escherichia
coli associate with signal-recognition particle and trigger factor.
Mol. Microbiol. 25, 53–64.
13. Lee, H.C. & Bernstein, H.D. (2001) The targeting pathway of
Escherichia coli presecretory and integral-membrane proteins is
specified by the hydrophobicity of the targeting signal. Proc. Natl.
Acad.Sci.USA98, 3471–3476.
14. Luirink, J., ten Hagen-Jongman, C.M., van der Weijden, C.C.,
Oudega, B., High, S., Dobberstein, B. & Kusters, R. (1994) An
alternative protein-targeting pathway in Escherichia coli:studies
on the rol of FtsY. EMBO J. 13, 2289–2296.

15. de Leeuw, E., Poland, D., Mol, O., Sinning, I., ten Hagen-
Jongman, C.M., Oudega, B. & Luirink, J. (1997) Membrane
association of FtsY, the E.coliSRP receptor. FEBS Lett. 416,
225–229.
16. Valent, Q.A., Scotti, P.A., High, S., de Gier, J.W., von Heijne, G.,
Lentzen, G., Wintermeyer, W., Oudega, B. & Luirink, J. (1998)
The Escherichia coli SRP- and SecB-targeting pathways converge
at the translocon. EMBO J. 17, 2504–2512.
17. Bieker, K.L., Phillips, G.J. & Silhavy, T.J. (1990) The sec and prl
genes of Escherichia coli. J. Bioenerg. Biomembr. 22, 291–310.
18. Derman, A.I., Puziss, J.W., Bassford, P.J. Jr & Beckwith, J. (1993)
A signal sequence is not required for protein export in prlA
mutants of Escherichia coli. EMBO J. 12, 879–888.
19. Flower, A.M., Doebele, R.C. & Silhavy, T.J. (1994) PrlA and
PrlG suppressors reduce the requirement for signal sequence
recognition. J. Bacteriol. 176, 5607–5614.
20. Prinz, W.A., Spiess, C., Ehrmann, M., Schierle, C. & Beckwith, J.
(1996) Targeting of signal sequenceless proteins for export in
Escherichia coli with altered protein translocase. EMBO J. 15,
5209–5217.
21. Nouwen, N., de Kruijff, B. & Tommassen, J. (1996) prlA sup-
pressors in Escherichia coli relieve the proton electrochemical
gradient dependency of translocation of wild-type precursors.
Proc. Natl. Acad. Sci. USA 93, 5953–5957.
22. van der Wolk, J.P., Fekkes, P., Boorsma, A., Hui, J.L., Silhavy,
T.J. & Driessen, A.J. (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.
23.deKeyzer,J.,vanderDoes,C.,Swaving,J.&Driessen,A.J.

(2002) The F286Y mutation of PrlA4 tempers the signal sequence
suppressor phenotype by reducing the SecA binding affinity.
FEBS Lett. 510, 17–21.
24. Duong, F. & Wickner, W. (1999) The PrlA and PrlG phenotypes
are caused by a loosened association among the translocase
SecYEG subunits. EMBO J. 18, 3263–3270.
25. Osborne, R.S. & Silhavy, T.J. (1993) PrlA suppressor mutations
cluster in regions corresponding to three distinct topological
domains. EMBO J. 12, 3391–3398.
26. Schiebel, E., Driessen, A.J., Hartl, F.U. & Wickner, W. (1991)
DH
+
and ATP function at different steps of the catalytic cycle of
preprotein translocase. Cell 64, 927–939.
27. Breukink, E., Demel, R.A., de Korte-Kool, G. & de Kruijff, B.
(1992) SecA insertion into phospholipids is stimulated by nega-
tively charged lipids and inhibited by ATP: a monolayer study.
Biochemistry 31, 1119–1124.
28. Manting, E.H. & Driessen, A.J. (2000) Escherichia coli translo-
case: the unravelling of a molecular machine. Mol. Microbiol. 37,
226–238.
29.Nouwen,N.,deKruijff,B.&Tommassen,J.(1996)DH
+
dependency of in vitro protein translocation into Escherichia coli
inner-membrane vesicles varies with the signal sequence core
region composition. Mol. Microbiol. 19, 1205–1214.
30. van Dalen, A., Killian, A. & de Kruijff, B. (1999) DW stimulates
membrane translocation of the C-terminal part of a signal
sequence. J. Biol. Chem. 274, 19913–19918.
31. de Cock, H. & Tommassen, J. (1992) SecB-binding does not

maintain the translocation-competent state of prePhoE. Mol.
Microbiol. 6, 599–604.
32. Adams, H., Scotti, P.A., de Cock, H., Luirink, J. & Tommassen, J.
(2002) The presence of a helix-breaker in the hydrophobic core of
signal sequences of secretory proteins prevents recognition by the
signal-recognition particle in Escherichia coli. Eur.J.Biochem.,
269, 5564–5571.
33. Bosch, D., Leunissen, J., Verbakel, J., de Jong, M., van Erp, H. &
Tommassen, J. (1986) Periplasmic accumulation of truncated
forms of outer-membrane PhoE protein of Escherichia coli K-12.
J. Mol. Biol. 189, 449–455.
34. Lugtenberg, B., Peters, R., Bernheimer, H. & Berendsen, W.
(1976) Influence of cultural conditions and mutations on the
composition of the outer membrane proteins of Escherichia coli.
Mol. Gen. Genet. 147, 251–262.
35. Bosch, D., de Boer, P., Bitter, W. & Tommassen, J. (1989) The role
of the positively charged N-terminus of the signal sequence of
E.coliouter-membrane protein PhoE in export. Biochim. Biophys.
Acta 979, 69–76.
36. Lugtenberg, B., Meijers, J., Peters, R., van der Hoek, P. & van
Alphen, L. (1975) Electrophoretic resolution of the Ômajor outer
membrane proteinÕ of Escherichia coli K-12 into four bands. FEBS
Lett. 58, 254–258.
37.deVrije,T.,Tommassen,J.&deKruijff,B.(1987)Optimal
posttranslational translocation of the precursor of PhoE protein
across Escherichia coli membrane vesicles requires both ATP and
the proton-motive force. Biochim. Biophys. Acta 900, 63–72.
38. High, S., Flint, N. & Dobberstein, B. (1991) Requirements for the
membrane insertion of signal-anchor type proteins. J. Cell Biol.
113, 25–34.

39.Luirink,J.,High,S.,Wood,H.,Giner,A.,Tollervey,D.&
Dobberstein, B. (1992) Signal-sequence recognition by an
Escherichia coli ribonucleoprotein complex. Nature (London) 359,
741–743.
40. de Gier, J.W., Mansournia, P., Valent, Q.A., Phillips, G.J.,
Luirink, J. & von Heijne, G. (1996) Assembly of a cytoplasmic
membrane protein in Escherichia coli is dependent on the signal-
recognition particle. FEBS Lett. 399, 307–309.
41. de Cock, H. & Tommassen, J. (1991) Conservation of components
of the Escherichia coli export machinery in prokaryotes. FEMS
Microbiol. Lett. 64, 195–199.
42. Scotti, P.A., Valent, Q.A., Manting, E.H., Urbanus, M.L.,
Driessen, A.J., Oudega, B. & Luirink, J. (1999) SecA is not
required for signal-recognition particle-mediated targeting and
initial membrane insertion of a nascent inner membrane protein.
J. Biol. Chem. 274, 29883–29888.
43. Carson, M.J., Barondess, J. & Beckwith, J. (1991) The FtsQ
protein of Escherichia coli: membrane topology, abundance, and
Ó FEBS 2002 Defective translocation in a prlA4 mutant (Eur. J. Biochem. 269) 5579
cell division phenotypes due to overproduction and insertion
mutations. J. Bacteriol. 173, 2187–2195.
44. Newitt, J.A. & Bernstein, H.D. (1998) A mutation in the Escher-
ichia coli secY gene that produces distinct effects on inner-mem-
brane protein insertion and protein export. J. Biol. Chem. 273,
12451–12456.
45. Shinde, U.P., Guru Row, T.N. & Mawal, Y.R. (1989) Export of
proteins across membranes: the helix reversion hypothesis. Biosci.
Rep. 9, 737–745.
46. Iino, T. & Sako, T. (1988) Inhibition and resumption of processing
of the staphylokinase in some Escherichia coli prlA suppressor

mutants. J. Biol. Chem. 263, 19077–19082.
47. Mu
¨
ller, J., Reinert, H. & Malke, H. (1989) Streptokinase muta-
tions relieving Escherichia coli K-12 (prlA4) of detriments caused
by the wild-type skc gene. J. Bacteriol. 171, 2202–2208.
48. Sako, T. & Iino, T. (1988) Distinct mutation sites in prlA sup-
pressor mutant strains of Escherichia coli respond either to sup-
pression of signal peptide mutations or to blockage of
staphylokinase processing. J. Bacteriol. 170, 5389–5391.
49. Emr, S.D., Hanley-Way, S. & Silhavy, T.J. (1981) Suppressor
mutations that restore export of a protein with a defective signal
sequence. Cell 23, 79–88.
50. Sako, T. (1991) Novel prlA alleles defective in supporting
staphylokinase processing in Escherichia coli. J. Bacteriol. 173,
2289–2296.
51. Urbanus, M.L., Scotti, P.A., Froderberg, L., Saaf, A., de Gier,
J.W., Brunner, J., Samuelson, J.C., Dalbey, R.E., Oudega, B. &
Luirink, J. (2001) Sec-dependent membrane protein insertion:
sequential interaction of nascent FtsQ with SecY and YidC.
EMBO Report 2, 524–529.
52. van der Laan, M., Houben, E.N., Nouwen, N., Luirink, J. &
Driessen, A.J. (2001) Reconstitution of Sec-dependent membrane
protein insertion: nascent FtsQ interacts with YidC in a SecYEG-
dependent manner. EMBO Report 2, 519–523.
53. Manting, E.H., van der Does, C., Remigy, H., Engel, A. &
Driessen, A.J. (2000) SecYEG assembles into a tetramer to form
the active protein translocation channel. EMBO J. 19, 852–861.
54. Bessonneau, P., Besson, V., Collinson, I. & Duong, F. (2002) The
SecYEG preprotein translocation channel is a conformationally

dynamic and dimeric structure. EMBO J. 21, 995–1003.
55. Duong, F. & Wickner, W. (1998) Sec-dependent membrane pro-
tein biogenesis: SecYEG, preprotein hydrophobicity and translo-
cation kinetics control the stop-transfer function. EMBO J. 17,
696–705.
56. Nilsson, I., Whitley, P. & von Heijne, G. (1994) The COOH-
terminal ends of internal signal and signal-anchor sequences are
positioned differently in the ER translocase. J. Cell Biol. 126,
1127–1132.
57. Tommassen, J., van Tol, H. & Lugtenberg, B. (1983) The ultimate
localization of an outer membrane protein of Escherichia coli K-12
is not determined by the signal sequence. EMBO J. 2, 1275–1279.
58. Casadaban, M.J. (1976) Transposition and fusion of the lac genes
to selected promoters in Escherichia coli using bacteriophage k and
l. J. Mol. Biol. 104, 541–555.
59. Nichols, B.P., Shafiq, O. & Meiners, V. (1998) Sequence analysis
of Tn10 insertion sites in a collection of Escherichia coli strains
used for genetic mapping and strain construction. J. Bacteriol. 180,
6408–6411.
60. Flower, A.M., Osborne, R.S. & Silhavy, T.J. (1995) The allele-
specific synthetic lethality of prlA-prlG double mutants predicts
interactive domains of SecY and SecE. EMBO J. 14, 884–893.
61. Nouwen, N., Tommassen, J. & de Kruijff, B. (1994) Requirement
for conformational flexibility in the signal sequence of precursor
protein. J. Biol. Chem. 269, 16029–16033.
5580 H. Adams et al.(Eur. J. Biochem. 269) Ó FEBS 2002