|
|
|
|||||||
|
|||||||||
(Received for publication, May 30,
1995; and in revised form, August 21, 1995) From the
The expression of the kdpFABC operon, coding for the
K
K In all cases
examined to date, the signal transduction of the sensor kinase/response
regulator systems occurs via a phosphorylation cascade. The
phosphorylation of the KdpD protein in vitro, most probably at
the conserved residue His-673, and the subsequent transfer of the
phosphoryl group to the response regulator KdpE was shown recently
(Nakashima et al., 1992; Voelkner et al., 1993).
Furthermore, the interaction of KdpE Compared to other
sensor kinases, the transmitter domain (C-terminal part) of KdpD is
highly conserved, whereas the input domain shows no homology to any
other protein sequenced so far. In the case of KdpD the input domain,
which in general is responsible for sensing the signal, is rather
large, comprising the N-terminal region of 400 amino acids, the
membrane-spanning segments, and approximately 200 amino acids of the
C-terminal region. In order to understand the mechanism of signal
perception, a topological analysis of KdpD in the membrane has been
performed using protein fusions to PhoA and LacZ and protease
susceptibility experiments.
Figure 1:
Construction
of kdpD deletions. Part of the kdpD gene, which was
used to construct the desired plasmids, is shown in more detail. The
position of the primers D3, D4, D5, D6, and D7 within the kdpD sequence are indicated in the upper part. Black arrows represent identical (D3) or complementary (D4, D5, D6, D7)
sequences to kdpD, whereas the dotted part of the arrow represents newly introduced sequences comprising BamHI (B) and EcoRV (EV)
recognition sites. The orientation of the arrows indicates the
direction of polymerization during the PCR. The numbered open bars (1-4) represent the sequence of the predicted
membrane-spanning helices (Walderhaug et al., 1992). Numbered arrows at the bottom of the pPV2 sequence indicate
the positions where the complementary sequences within the primers
D4-D7 stops. The positions (in bp) of EcoRV sites used for the
construction are displayed (EV 4699 and EV 5908). The
numbering of the nucleotides is given in bp according to the kdpDE sequence published by Walderhaug et al.(1992). The four
sequences in the lower part of the figure show the constructed deletion
plasmids pPV2-D4 to pPV2-D7. The hatched lines indicate the
fusion of the newly introduced EcoRV site to the EcoRV site at bp 5908 (for the cloning strategy, see
``Experimental Procedures''). BamHI sites used for
the construction of the lacZ and phoA fusions are
indicated by arrows designated B. The truncated KdpD
proteins and the amino acids lacking therein are given on the right.
The first
fusion point was the EcoRV site at bp 4699 at the beginning of
the kdpD gene (Fig. 1). Plasmid pPV14 (kdp`D-lacZ) and pPV15 (kdp`D-phoA-kdpD`) were
constructed by replacing the 1209-bp EcoRV fragment (bp
4699-5908, see Fig. 1) with the lacZ and phoA fragment, respectively. Prior to that the BamHI ends of
the fragment DNA had been filled in using Klenow fragment. For the
other fusion points, the newly introduced BamHI site in the
deleted kdpD genes of plasmids pPV2-D4, pPV2-D5, pPV2-D6, and
pPV2-D7 were used (Fig. 1). For that purpose the BamHI-restricted lacZ and phoA fragment,
respectively, were ligated with the BamHI vector part of
plasmid pPV2-D4 to pPV2-D7. The resulting plasmids were named pPV2-D4L
and pPV2-D4P to pPV2-D7L and pPV2-D7P, respectively. In this cloning
step the kdpD gene downstream of the BamHI site and
most of the kdpE gene in plasmids pPV2-D4 to pPV2-D7 were
deleted due to a second BamHI site (bp 7614) at the end of the kdpE gene. In the case of lacZ fusions, strain
TKV2209 (lacZ
The determination of AP activity was performed with
cells of strain DHB4 harboring plasmids pPV2, pPV15, and pPV2-D4P to
pPV2-D7P, respectively, grown to logarithmic phase in KML complex
medium with chloramphenicol and 1 mM
isopropyl-
It has already been shown that the KdpD protein is located in
the cytoplasmic membrane of E. coli (Walderhaug et
al., 1992; Voelkner et al., 1993). The hydrophobicity
plot according to the algorithm of Kyte and Doolittle(1982) revealed
that KdpD is predominantly of hydrophilic character except for four
narrow-spaced, membrane-spanning segments in the middle of the primary
structure (Fig. 2). However, experimental evidence in support of
this topological model was still lacking.
Figure 2:
Hydropathy profile of the KdpD protein.
The average local hydrophobicity at each residue according to the
algorithm of Kyte and Doolittle(1982), using a window of 15 amino
acids, is plotted on the vertical axis versus the residue
number on the horizontal axis.
Figure 3:
Detection of truncated Kdp proteins.
Everted membrane vesicles from strain TKV2209 carrying plasmid pPV2-D4,
pPV2-D5, pPV2-D6, and pPV2-D7, respectively, were prepared as described
under ``Experimental Procedures.'' Proteins (30 µg/lane)
were separated on 7.5% SDS-PAGE, and a polyclonal anti-KdpD antiserum
was used for immunoblotting (dilution 1:100,000). The protein standard
and a KdpD reference were applied to the first
lane.
Figure 4:
Detection of KdpD-LacZ fusion proteins.
Exponentially growing cells of strain TKV2209 harboring plasmids pPV2
(KdpD), pPV14 (Kdp`D-LacZ), and pPV2-D4L (Kdp`D4-LacZ) to pPV2-D7L
(Kdp`D7-LacZ), respectively, were harvested by centrifugation and
resuspended in gel electrophoresis sample buffer. ``Whole
cell'' proteins (40 µg/lane) were separated on 7.5% SDS-PAGE,
and a monoclonal anti-LacZ antiserum (Boehringer) was used for
immunoblotting.
Figure 7:
Topology of the KdpD protein. The
illustrated model was derived from both hydropathy analysis and studies
with LacZ/PhoA fusion proteins as described in the text. Transmembrane
helices are displayed as white boxes, and the first and last
residue of the predicted hydrophobic region are indicated. The
positions of the LacZ/PhoA fusion point are marked by arrows,
and the corresponding amino acid residue is indicated. The net charge
of the hydrophilic regions before, between or after the hydrophobic
domains, taking up to 10 amino acids into account, is circled.
Arg, Lys, and His are considered positive, Glu and Asp are considered
negative.
The detection of the PhoA fusion
proteins, Kdp`D4-PhoA to Kdp`D7-PhoA, with predicted sizes ranging from
95 to 105 kDa, turned out to be more difficult. Only the Kdp`D5-PhoA
fusion protein could clearly be detected with an anti-PhoA antiserum
(data not shown). This phenomenon could be explained by the instability
of these fusion proteins or by the inability of the anti-PhoA
antibodies as well as the anti-KdpD antibodies to detect their antigen
within the fusion protein.
It has already been
well established that the alkaline phosphatase is inactive in the
cytoplasm (Manoil et al., 1990), whereas the
Judging from the immunoblot (Fig. 4), the concentration of the KdpD-LacZ fusion proteins in
the cell lysates are almost identical for Kdp`D5-LacZ to Kdp`D7-LacZ.
In the case of Kdp`D4-LacZ, the protein concentration is 3-4
times higher, whereas in the case of Kdp`D-LacZ the protein
concentration is roughly 2 times lower compared to the other three
fusion proteins. Taking these different protein levels into account,
the
Figure 5:
Protease treatment of KdpD.
KdpD-containing spheroplasts and spheroplasts permeabilized with Triton
X-100 from strain TKR2000/pPV5 were prepared as described under
``Experimental Procedures.'' Samples (0.5 mg/ml) with
(+) or without(-) the protease kallikrein (100 µg/ml)
were incubated for 1 h at 37 °C. Proteins (20 µg/lane) were
separated on 9% SDS-PAGE and a polyclonal anti-KdpD antiserum was used
for immunoblotting.
The sensor kinase KdpD is related to a group of sensor
kinases, like PhoR (Makino et al., 1986) and EnvZ (Comeau et al., 1985), which exhibit a modular design of an
N-terminal, mostly periplasmic input domain located between two
membrane-spanning segments. Sequence homology of KdpD to other sensor
kinases begins at approximately residue 660 with the C-terminal
transmitter domain. This domain carries the five characteristic
sequence motifs (blocks H, N, G1, F, and G2; Parkinson and
Kofoid(1992)), and possesses autokinase and phosphotransferase activity
(Nakashima et al., 1992; Voelkner et al., 1993;
Altendorf et al., 1994). Therefore, the C-terminal domain of
KdpD, which must interact with ATP and the cytoplasmic protein KdpE, is
surely located in the cytoplasm. That also applies to the short
hydrophobic region from residue 840 through 852, which appears as a
prominent peak in the hydrophobicity plot (Fig. 2). KdpD
possesses an unusually extended N-terminal input domain, which is
mostly hydrophilic except for the central membrane-spanning helices (Fig. 2). This region does not show homology to any other
protein sequenced so far. Our working model (Fig. 6A)
suggests that this domain of about 400 amino acid residues is also
cytoplasmic and that the central hydrophobic region consists of four
membrane-spanning segments (Walderhaug et al., 1992). However,
the N-terminal region possesses a stretch of 17 hydrophobic amino acids
from residues Val-27 to Ala-43, which corresponds to the first peak in
the hydrophobicity plot (Fig. 2). As predicted by the algorithm
of Rao and Argos(1986), this region could also be a membrane-spanning
helix. Therefore, a second membrane model of KdpD, in which the
N-terminal input domain is mostly periplasmic is possible (Fig. 6B). As a consequence of the cytoplasmic
localization of the C terminus, in this model only three
membrane-spanning segments could exist in the middle of the protein, as
predicted by the algorithm of Rao and Argos (1986; data not shown).
Figure 6:
Models of KdpD. Two possible models (A and B) for the topology of the KdpD protein in the
cytoplasmic membrane are shown. The postulated phosphorylation site
His-673 is indicated.
To be able to distinguish between these two models for KdpD, PhoA
and LacZ fusion proteins have been used, which are only active in the
periplasm or in the cytoplasm (for an overview, see Traxler et
al.(1993)). The five fusion points were chosen in such a way that
the number of the central membrane-spanning helices, and in particular,
the location of the N-terminal domain could be clearly established. We
provide experimental evidence for a cytoplasmic localization of the
N-terminal domain of KdpD leading to the topology model of KdpD as
given in Fig. 7. In detail, the high As
shown with the ProW protein (Whitley et al., 1994) long
periplasmic N-terminal tails fused to reporter proteins could
artificially remain cytoplasmic, if the translocation signals are
localized in the following transmembrane domain. Therefore, the
topological model of KdpD presented in Fig. 7was confirmed not
only by the LacZ and PhoA fusion studies, but also by the protease
treatment of spheroplasts with and without Triton X-100. Using
spheroplasts in the absence of Triton X-100, KdpD was completely
resistant to kallikrein digestion. On the other hand, in permeabilized
spheroplasts KdpD was no longer protected against kallikrein, because
the cytoplasmic regions of the protein carrying potential cleavage
sites are now accessible to the protease. An analysis of the amino
acid distribution within the polypeptide chain is also in accord with
the model proposed in Fig. 7. The hydrophobic stretch from
residues Val-27 to Ala-43 is interrupted by a lysine and is closely
followed by two arginines. Since signal sequences are extremely
sensitive to the presence of positively charged residues at their
C-terminal end (Andersson and von Heijne, 1993), this segment is
unlikely to function as a translocation signal. Furthermore, the
overall amino acid composition of the N-terminal region (residue
1-400) is more similar to that of a typical cytoplasmic than a
periplasmic region (Nakashima and Nishikawa, 1992) as checked with the
TOP-PRED program. A final argument in favor of the
model presented stems from the analysis of a mutant strain, in which
the four membrane-spanning segments in the middle of KdpD have been
deleted. The truncated protein is still associated with, but can be
removed from the membrane by treatment with 2 M urea. This
observation lends support to the notion that the N- and C-terminal
regions are both cytoplasmic. Induction of the Kdp system occurs
when the K
Volume 270,
Number 47,
Issue of November 24, 1995 pp. 28282-28288
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-translocating Kdp-ATPase, is under the control of
the two regulatory proteins KdpD and KdpE, which belong to the group of
sensor kinase/response regulator systems. The topology of the KdpD
protein in the cytoplasmic membrane was investigated using LacZ and
PhoA fusions at different sites within the polypeptide chain and by
treating spheroplasts in the presence or absence of Triton X-100 with
the protease kallikrein. The results revealed that KdpD has four
membrane-spanning segments in the middle of the polypeptide chain,
whereas N and C terminus are both cytoplasmic.
ions play a crucial role in maintaining
turgor in bacterial cells (Epstein, 1986). Therefore, bacteria have
established several types of K
influx and efflux
systems as well as secondary porters and stretch-activated channels to
adjust the intracellular K
concentration in response
to changes in the osmolarity of the medium (Bakker, 1993). In the case
of Escherichia coli the Kdp-ATPase, an inducible, high
affinity K
uptake system (K
for transport: 2 µM; Epstein et al.,
1978) is expressed as an emergency system under growth conditions,
where the three constitutive K
uptake systems TrkH,
TrkG, and Kup (Bossemeyer et al., 1989; Dosch et al.,
1991) are unable to maintain the required cellular pool of
K
. The Kdp-ATPase is composed of the three
membrane-bound subunits KdpA, KdpB, and KdpC (Laimins et al.,
1978). The structural genes are organized in the kdpFABC operon (Hesse et al., 1984), whereby the function of the
open reading frame kdpF is still unknown (Altendorf et
al., 1992). The adjacent kdpDE operon codes for two
proteins that regulate the expression of the kdpFABC operon
(Polarek et al., 1992). KdpD is a membrane-bound protein (98.7
kDa), whereas KdpE is a cytoplasmic protein (25.2 kDa) (Walderhaug et al., 1992). Both proteins belong to the class of sensor
kinase/response regulator systems, which are characterized by a
homologous signal-transduction pathway for the adaptive response of
environmental stimuli (Parkinson and Kofoid, 1992).
P with the kdpFABC promoter, resulting in a 10-fold increase of the transcription of
the kdpFABC operon, was demonstrated (Nakashima et
al., 1993). For KdpD the stimulus seems to be a decrease in turgor
(Laimins et al., 1981; Epstein, 1992). However, further
observations lend support to the notion that turgor is apparently not
the sole regulatory signal for the expression of kdp (Csonka
and Hanson, 1991; Asha and Gowrishankar, 1993).
Materials
All chemicals used were of analytical
grade. Nitrocellulose membranes (0.45 µm) were obtained from
Schleicher & Schüll (Dassel, Germany (FRG)). T4
DNA-ligase, alkaline phosphatase, Klenow fragment of DNA polymerase I,
T4 polynucleotide kinase, restriction endonucleases, and dNTPs were
purchased from Boehringer (Mannheim, FRG), Life Technologies, Inc.
(Eggenstein, FRG), Biolabs (Schwalbach, FRG), or Pharmacia (Freiburg,
FRG). Agarose was obtained from Roth (Karlsruhe, FRG). Goat anti-rabbit
IgG alkaline phosphatase conjugate and 5-bromo-4-chloro-3-indolyl
phosphate were purchased from Biomol (Hamburg, FRG),
isopropyl-
-D-thiogalactoside from Serva (Heidelberg,
FRG), and leupeptin,
5-bromo-4-chloro-3-indolyl--D-galactopyranoside, o-nitrophenyl--D-galactoside, and p-nitrophenyl phosphate were obtained from Sigma (Munich,
FRG). Tfl thermostable DNA polymerase was purchased from
BIOzym (Hameln, FRG), while the T7 sequencing kit was obtained from
Pharmacia. The protease kallikrein and anti--galactosidase
antibody were obtained from Boehringer (Mannheim, FRG).
Sheep-anti-(mouse Ig) horseradish peroxidase-linked whole antibody and
the ECL-chemiluminescence Western blotting detection reagents were
purchased from Amersham Buchler (Braunschweig, FRG). Oligonucleotides
for sequencing and PCR
were synthesized by Dr. H. Lill
(University of Osnabrück).Bacterial Strains, Plasmids, and Media
E. coli strains and plasmids used are listed in Table 1. All strains
are derivatives of E. coli K12. Strain TKV2209 carries a
deletion from EcoRV (bp 4699; see Fig. 1) in kdpD to BamHI (bp 7614) in kdpE. The deletion was
constructed in vitro on a plasmid, and the transfer to the
chromosome was performed using the polA technique as described
by Gutterson and Koshland(1983). KML complex medium was prepared as
described (Epstein and Davies, 1970), and antibiotics were used in the
following concentrations: 25 µg/ml chloramphenicol, 50 µg/ml
ampicillin, and 100 µg/ml carbenicillin.
Recombinant DNA Techniques and PCR
For preparation
and handling of recombinant DNA and for transformation of E. coli cells, standard procedures were used (Sambrook et al.,
1989; Ausubel et al., 1987). Treatment of DNA with enzymes
(restriction enzymes, alkaline phosphatase, Klenow fragment of DNA
polymerase I, polynucleotide kinase, ligase) was carried out under
conditions recommended by the suppliers. Prior to ligations vector DNA
fragments were treated with alkaline phosphatase. DNA fragments were
recovered from agarose gels using the Gene-clean kit from Dianova
(Hamburg, FRG). Sequencing of double-stranded DNA was performed by the
dideoxynucleotide chain termination method (Sanger et al.,
1977) using a Pharmacia T7 sequencing kit. Polymerase chain reactions
were carried out with the Tfl thermostable DNA polymerase, and
PCR products were purified using the Magic-Prep kit from Promega
(Madison, WI).Construction of Different kdpD Deletions
Four kdpD deletions were generated by using the PCR method followed
by two additional subcloning steps. Primers used for PCR and their
positions within the kdpD gene are shown in Fig. 1.
Primer D3 (5`-CCCGTCATCCCAAACGCTG-3`) corresponds to bp 4673-4692
just upstream of the first EcoRV site in the kdpD gene. The sequence of primer D4
(3`-GTTCACCGCACATGTCCTAGGGCTATAG-5`) is in the 3` part complementary to
bp 5497-5511. In the 5` part (underlined), a new sequence
containing a BamHI and an EcoRV restriction site was
created. These new restriction sites were used to construct ``in
frame'' KdpD deletion (EcoRV) or fusion (BamHI)
proteins. The sequences of primer D5
(3`-CTACCGCAAACTACGCCTAGGGCTATAG-5`), primer D6
(3`-CGATAAAATACCTGCCCTAGGGCTATAG-5`), and primer D7
(3`-GCGGGGTGCGCCGTGCCTAGGGCTATAG-5`) are in the 3` part complementary
to the sequences of bp 5578-5592 (D5), bp 5638-5652 (D6),
and bp 5710-5724 (D7), respectively. The 5` parts of the primer
D5, D6, and D7 (underlined) carry the same new BamHI-EcoRV recognition sequence as primer D4. PCR
amplification was performed using plasmid pPV2 as template with the
primer pairs D3 and D4, D3 and D5, D3 and D6, as well as D3 and D7. The
resulting PCR products were purified and treated with Klenow fragment
and T4 polynucleotide kinase. For more convenient handling, the PCR
products were first cloned in the SmaI site of vector pUC19.
The resulting plasmids were named pUCD4, pUCD5, pUCD6, and pUCD7. In a
second step the four different kdpD deletion plasmids were
constructed. For this purpose the kdpD` containing the 1209-bp EcoRV fragment of plasmid pPV2 was replaced by the EcoRV-digested PCR fragments of plasmids pUCD4, pUCD5, pUCD6,
and pUCD7, respectively, resulting in plasmids pPV2-D4, pPV2-D5,
pPV2-D6, and pPV2-D7 (see Fig. 1). Verification of the four kdpD deletions was achieved by restriction analysis and
plasmid sequencing.Construction of kdpD-lacZ and kdpD-phoA
Fusions
The fusions of the kdpD gene to the lacZ or the phoA gene were generated at five different sites
within the kdpD gene. For this purpose the 3068-bp BamHI fragment of plasmid pZ19 (Haardt, 1993) containing the lacZ reporter gene was used, which is deleted for the first 19
nucleotides of the gene, but includes the stop codon. The phoA reporter gene was taken from plasmid pSWFII (Ehrmann et
al., 1990). The 1300-bp BamHI fragment of pSWFII is
deleted for the first 78 nucleotides of phoA coding for the
signal sequence and lacks a stop codon at the 3` end.) was transformed with the
ligation mixture and recombinant plasmids were selected on KML agar
plates containing chloramphenicol and the chromogenic substrate
5-bromo-4-chloro-3-indolyl-
-D-galactopyranoside (40
mg/liter). In contrast, strain DHB4 (phoA)
was used for the phoA fusion plasmids and transformants were
selected on KML agar plates with chloramphenicol,
isopropyl-
-D-thiogalactoside (1 mM), and the
chromogenic substrate 5-bromo-4-chloro-3-indolyl phosphate (40
mg/liter). The constructed plasmids were tested by restriction analysis
and by immunodetection of the synthesized fusion proteins with
antibodies raised against KdpD, LacZ, and PhoA.Assay of
For the -Galactosidase and Alkaline Phosphatase
Activity-galactosidase assay, strain TKV2209
harboring plasmids pPV2, pPV14, and pPV2-D4L to pPV2-D7L, respectively,
was grown in KML complex medium with chloramphenicol to logarithmic
phase and the -galactosidase activity was measured as the rate of o-nitrophenyl--D-galactoside hydrolysis in
permeabilized cells according to Miller(1972). Therefore, 1 ml of cells
were washed in 50 mM phosphate buffer, pH 7.2. After addition
of 20 µl toluol, the suspension was incubated at 37 °C for 5
min. The reaction was started by addition of 50 µl of o-nitrophenyl--D-galactoside (20 mM),
and after development of sufficient yellow color the reaction was
stopped with 1.5 ml of Na
CO
(200 mM).
Before measuring the absorbance at 420 nm, cell debris were removed by
centrifugation for 10 min at 13,000 
g in a
microcentrifuge.-D-thiogalactoside. The rate of p-nitrophenyl phosphate hydrolysis in permeabilized cells was
measured according to Michaelis et al.(1983). In this case 0.5
ml of cells were diluted with 0.5 ml of buffer A (10 mM Tris/HCl, pH 8.0, 150 mM NaCl) and washed twice with the
same buffer. Subsequently, 0.1 ml of the sample were added to 0.9 ml of
buffer B (1 mM Tris/HCl, pH 8.0, 1 mM
ZnCl). Cells were permeabilized by the addition of 25
µl of SDS (0.1%) and 25 µl of chloroform and incubated at room
temperature for 5 min. The enzymatic reaction was started by the
addition of 0.1 ml of p-nitrophenyl phosphate (0.4% in 1 M Tris/HCl, pH 8.0) and incubated at 37 °C, until sufficient
yellow color was observed. The reaction was stopped by the addition of
120 µl of 2.5 M KHPO
, and cell
debris were removed by centrifugation for 10 min at 13,000 g before measuring the absorbance at 420 nm. In both cases
units of enzyme activity are defined as OD
1000/t v OD (Miller,
1972; Brickmann and Beckwith, 1975), where OD
represents
the reaction mixture, OD
the cell density just before the
assay, t the reaction time in minutes, and v the
volume of cell culture used for the assay in ml.
Preparation of Everted Membrane Vesicles
Cells
were grown in KML complex medium supplemented with chloramphenicol to
an optical density (OD) of approximately 1.0. Membranes
were prepared according to Siebers and Altendorf(1988), but instead of
using a Ribi Press fractionator cells were disrupted by sonification
for 3 min at 4 °C on ice (duty cycle, 50%; Branson sonifier).
Furthermore, the washing step in low ionic strength buffer was omitted
and buffers were changed from Hepes/Tris to Tris/HCl.
Preparation of Spheroplasts
Exponentially grown
cells (OD = 0.6) were washed once in 33 mM Tris/HCl, pH 8.0, 100 mM KCl, and 5 mM MgSO
. Subsequently, cells were resuspended in the same
buffer plus 40% (w/v) sucrose. Spheroplasts were prepared by the
addition of EDTA (20 mM final concentration) and lysozyme (100
µg/ml) and incubated at 4 °C for 15 min. The spheroplasts were
used immediately without further washing.Proteolysis with Kallikrein
Spheroplasts alone or
in the presence of 0.5% Triton X-100 were used in the corresponding
buffer at a protein concentration of 0.5 mg/ml. Proteolysis was carried
out with kallikrein (100 µg/ml) for 1 h at 37 °C. The reaction
was stopped by the protease inhibitor leupeptin (1 µM),
followed by the precipitation of the protein with 10% trichloroacetic
acid. The precipitated proteins were washed once with acetone (4
°C), dried under a stream of nitrogen, and further used for
SDS-PAGE.Analytical Procedures
Protein concentrations were
determined according to Hartree(1972). SDS-PAGE was performed as
described by Lugtenberg et al.(1975) using 7,5% or 9%
polyacrylamide gels. The electrophoretic transfer of proteins from
SDS-polyacrylamide gels to nitrocellulose membranes and immunodetection
with anti-KdpD antiserum (dilution 1:100,000, Voelkner et al.,
1993) and alkaline phosphatase-conjugated secondary antibody was
carried out as described by Voelkner et al.(1993), except that
the blocking buffer was changed from 3% bovine serum albumin to 5%
(w/v) powdered skim milk. Furthermore, in the case of anti-LacZ as
primary antibody (Boehringer), the peroxidase-conjugated secondary
antibody was detected with the detection reagents of the ECL-Western
blotting system as recommended by the supplier.
Construction of kdpD Deletion Plasmids
In order to fuse
PhoA or LacZ to the proposed periplasmic or cytoplasmic loops of KdpD,
various deletions comprising the central membrane-spanning helices were
generated, together with the introduction of a convenient BamHI site as described under ``Experimental
Procedures.'' The PCR products used for the cloning steps and the
resulting plasmids carrying the deleted kdpD genes are
displayed in Fig. 1. Plasmid pPV2-D4 carries the 822-bp EcoRV fragment of the PCR product obtained with primer pair D3
and D4 instead of the original 1209-bp EcoRV fragment of pPV2.
In analogy the original EcoRV fragment was replaced by a
903-bp EcoRV fragment (primers D3 and D5) in pPV2-D5, by a
963-bp EcoRV fragment (Primers D3 and D6) in pPV2-D6, and by a
1035-bp EcoRV fragment (primers D3 and D7) in pPV2-D7,
respectively. DNA sequencing of the deletions in plasmids pPV2-D4 to
pPV2-D7 was performed to verify the correct positioning of the newly
created BamHI sites and to proof that all constructions are
``in frame'' deletions.Detection of the Truncated KdpD Proteins
Everted
membrane vesicles of strain TKV2209 harboring plasmid pPV2-D4, pPV2-D5,
pPV2-D6, and pPV2-D7, respectively, were prepared as described under
``Experimental Procedures.'' The synthesis of the four
proteins KdpD4, KdpD5, KdpD6, and KdpD7 was confirmed by immunoblotting
using polyclonal antibodies raised against wild type KdpD protein (Fig. 3). Their predicted sizes, ranging from 85 kDa to 93 kDa,
are in good agreement with those calculated from the immunoblot. The
results also demonstrated that all proteins carry ``in
frame'' deletions. Furthermore, the truncated proteins could all
be detected in the membrane fraction. As already shown by Puppe et
al., KdpD proteins lacking all four membrane-spanning
helices, like the KdpD4 protein, are still associated with, but not
integrated into the membrane, probably due to an interaction of a
stretch of hydrophobic amino acids in the N-terminal part of KdpD with
the membrane. However, the amount of the four truncated proteins
synthesized is different (Fig. 3), although their genes are
expressed from the same plasmid background. The protein level is the
highest for KdpD4, followed by KdpD6, and decreased from KdpD7 to
KdpD5. This phenomenon could possibly be explained by the different
stability of the truncated forms in the cytoplasmic membrane.
Construction of KdpD-LacZ and KdpD-PhoA Fusion
Proteins
To investigate the membrane topology of the KdpD
protein in more detail, gene fusions of kdpD ``in
frame'' to lacZ and phoA were constructed as
described under ``Experimental Procedures.'' The five
different fusion points are shown in Table 2and Fig. 1.
For the first fusion at residue Asp-127, the EcoRV site at bp
4699 was used. The resulting fusion protein Kdp`D-LacZ contains only
the kdpD sequence upstream of the EcoRV site, whereas
the fusion protein Kdp`D-PhoA-KdpD` also contains the kdpD sequence proximal to the second EcoRV site (residue
Ile-531 to Met-894), because the stop codon of the phoA gene
is lacking. The 130-kDa Kdp`D-LacZ protein could be detected with the
anti-LacZ antiserum (Fig. 4). However, the fusion protein seems
to be unstable, because the second detectable, faster moving band has
the mobility of the -galactosidase. The 105-kDa Kdp`D-PhoA-KdpD`
protein could be detected with the anti-KdpD antiserum, since the
antigenic C-terminal region of KdpD is still present (data not shown).
The four other fusions at residues Gln-398, Ala-425, Arg-445, and
Thr-469, respectively, were constructed by using the newly created BamHI site in the deleted kdpD genes of plasmids
pPV2-D4, pPV2-D5, pPV2-D6, and pPV2-D7. As shown in Fig. 1and Fig. 7, these fusion points are lying immediately before or
between the predicted membrane-spanning helices. The corresponding
fusion proteins Kdp`D4-LacZ, Kdp`D5-LacZ, Kdp`D6-LacZ, and Kdp`D7-LacZ,
with predicted sizes between 155 and 165 kDa, could all be detected
with the anti-LacZ antiserum (Fig. 4). It can clearly be seen
that the level of the Kdp`D4-LacZ protein is elevated in comparison to
the other fusion proteins.
Enzyme Activities of the Fusion Proteins
In order
to determine the location of the fusion points, leading to the overall
topology of the KdpD protein, the AP and -galactosidase activities
were measured as described under ``Experimental Procedures'';
the results are shown in Table 2. In this context, it should be
noted that the activities are normalized by the amount of permeabilized
cells and not by the amount of fusion protein.-galactosidase remains inactive in the periplasm (Lee et
al., 1989). The fusion proteins Kdp`D-LacZ/Kdp`D-PhoA-KdpD`,
Kdp`D4-LacZ/Kdp`D4-PhoA, and Kdp`D6-LacZ/Kdp`D6-PhoA exhibited high
levels of -galactosidase, but very low AP activity. For that
reason we concluded that the fusion residues Asp-127, Gln-398, and
Arg-445 are located in cytoplasmic regions of the KdpD protein. In
contrast, the fusion proteins Kdp`D5-LacZ/Kdp`D5-PhoA showed high AP
activity, but low -galactosidase activity, consistent with a
periplasmic exposure of the fusion residue Ala-425. The only fusion
without a definite result is that at position Thr-469. The activities
of both fusion proteins Kdp`D7-LacZ/Kdp`D7-PhoA are only slightly
elevated compared to the basal level.-galactosidase activity of Kdp`D-LacZ would be higher, that of
Kdp`D4-LacZ would be 3-4 times lower and that of the other three
would not change at all. Based on this kind of normalization of enzyme
activity, the same conclusions can still be drawn for the topology of
KdpD as was the case for the normalization based on the amount of
permeabilized cells (Table 2).Protease Treatment of KdpD
Further evidence that
the N-terminal part of KdpD is located in the cytoplasm stems from
protease susceptibility studies. In both the N- and the C-terminal part
of KdpD, but not in the central hydrophobic region, seven potential
cleavage sites (before Phe-Arg or Leu-Arg) for kallikrein can be found.
Treatment of spheroplasts permeabilized by Triton X-100 with kallikrein
exhibits an complete degradation of KdpD, whereas in the case of
spheroplasts without Triton X-100 the sensor kinase is not touched at
all (Fig. 5). In contrast, trypsin causes degradation of KdpD
even in spheroplasts (data not shown), since potential cleavage sites
for this protease reside in the postulated periplasmic loops of the
central hydrophobic region.
-galactosidase and
low AP activities of the KdpD-LacZ and KdpD-PhoA fusions at residues
Asp-127, Gln-398, or Arg-445 suggest a cytoplasmic location of these
fusion points. On the other hand, the low -galactosidase activity
of the Kdp`D5-LacZ fusion at residue Ala-425 or of Kdp`D7-LacZ at
residue Thr-469 indicates a location of the fusion points at the
periplasm. In case of the PhoA fusions, high AP activity could clearly
be demonstrated for the Kdp`D5-PhoA fusion, indicating again a
periplasmic location of that fusion point. However, Kdp`D7-PhoA
exhibited low instead of high AP activity. This latter observation can
possibly be explained by the inability of the Kdp`D7 fusions to export
the reporter proteins due to ineffective or deleted topogenic signals
or due to premature splitting of the reporter proteins. These arguments
also apply to the slightly elevated -galactosidase activity of
Kdp`D7-LacZ. These kind of problems are already known from topological
studies of other membrane proteins. In the case of the LacY protein
(Calamia and Manoil, 1990), an alkaline phosphatase fusion to a known
periplasmic domain has low AP activity presumably because it follows a
poor export signal. Detailed studies revealed that a charged residue in
the membrane-spanning segment N-terminal to this periplasmic domain
limit export of the reporter protein (Calamia and Manoil, 1992). In the
case of KdpD only the third membrane-spanning segment contains a
charged residue. Therefore, in analogy, the low AP activity of
Kdp`D7-PhoA may also be explained by an inefficient export of PhoA
fused to the periplasmic domain C-terminal to the third
membrane-spanning segment, since the interaction of the third
membrane-spanning segment with the fourth transmembrane helix of KdpD
might be necessary for a correct membrane insertion. When sequences
located C-terminal of the fusion point are required for a proper
topological assembly, as demonstrated for the MalK protein (McGovern et al., 1991), sandwich gene fusions (Ehrmann et al.,
1990) are sometimes a better tool to determine the topology of a given
protein than fusions in which the C-terminal part of the protein is
replaced by the reporter protein. Furthermore, problematic fusions
could be counteracted by shifting the fusion point by some amino acids
as demonstrated for the LacY protein (Calamia and Manoil, 1990). In addition, the distribution of charged
residues may function as a topogenic determinant (Boyd and Beckwith,
1989) and the ``positive-inside'' rule (von Heijne, 1992) is
in line with the model of KdpD presented here. All cytoplasmic regions
which are at the border of hydrophobic domains possess a net positive
charge, whereas the two periplasmic loops possess a net negative or no
net charge at all (Fig. 7). concentration of the medium becomes
growth-limiting or if the turgor pressure decreases (Altendorf et
al., 1992). Based on the topological model presented, it is
tempting to speculate that the four membrane-spanning helices of KdpD
could sense ``turgor'' as changes in the stretch of the plane
of the membrane, whereas the cytoplasmic N-terminal domain could be a
sensor of the ionic conditions in the cytoplasm, which may modulate the
primary signal. Experiments are under way to test this intriguing
hypothesis.
)))
We thank Dr. G. von Heijne (Stockholm University) for
interpretations of the data obtained with the TOP-PRED program for
KdpD, Dr. M. Ehrmann (University of Konstanz) for suggesting the use of
kallikrein and for other useful information, Dr. G. Deckers-Hebestreit
for critically reading the manuscript, H. Gerdes for technical
assistance, and J. Petzold for typing the manuscript.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Hamann, P. Zimmann, and K. Altendorf Reduction of Turgor Is Not the Stimulus for the Sensor Kinase KdpD of Escherichia coli J. Bacteriol., April 1, 2008; 190(7): 2360 - 2367. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Zimmann, A. Steinbrugge, M. Schniederberend, K. Jung, and K. Altendorf The Extension of the Fourth Transmembrane Helix of the Sensor Kinase KdpD of Escherichia coli Is Involved in Sensing J. Bacteriol., October 15, 2007; 189(20): 7326 - 7334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Jung, T. Odenbach, and M. Timmen The Quorum-Sensing Hybrid Histidine Kinase LuxN of Vibrio harveyi Contains a Periplasmically Located N Terminus J. Bacteriol., April 1, 2007; 189(7): 2945 - 2948. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ballal, M. Bramkamp, H. Rajaram, P. Zimmann, S. K. Apte, and K. Altendorf An Atypical KdpD Homologue from the Cyanobacterium Anabaena sp. Strain L-31: Cloning, In Vivo Expression, and Interaction with Escherichia coli KdpD-CTD J. Bacteriol., July 15, 2005; 187(14): 4921 - 4927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Heermann, K. Altendorf, and K. Jung The N-terminal Input Domain of the Sensor Kinase KdpD of Escherichia coli Stabilizes the Interaction between the Cognate Response Regulator KdpE and the Corresponding DNA-binding Site J. Biol. Chem., December 19, 2003; 278(51): 51277 - 51284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. T. Verhamme, J. C. Arents, P. W. Postma, W. Crielaard, and K. J. Hellingwerf Glucose-6-phosphate-dependent phosphoryl flow through the Uhp two-component regulatory system Microbiology, December 1, 2001; 147(12): 3345 - 3352. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ouchane and S. Kaplan Topological Analysis of the Membrane-localized Redox-responsive Sensor Kinase PrrB from Rhodobacter sphaeroides 2.4.1 J. Biol. Chem., June 11, 1999; 274(24): 17290 - 17296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Wood Osmosensing by Bacteria: Signals and Membrane-Based Sensors Microbiol. Mol. Biol. Rev., March 1, 1999; 63(1): 230 - 262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Peter, B. Weil, A. Burkovski, R. Krämer, and S. Morbach Corynebacterium glutamicum Is Equipped with Four Secondary Carriers for Compatible Solutes: Identification, Sequencing, and Characterization of the Proline/Ectoine Uptake System, ProP, and the Ectoine/Proline/Glycine Betaine Carrier, EctP J. Bacteriol., November 15, 1998; 180(22): 6005 - 6012. [Abstract] [Full Text]
|
|
|
|
|
|
K. Jung and K. Altendorf Individual Substitutions of Clustered Arginine Residues of the Sensor Kinase KdpD of Escherichia coli Modulate the Ratio of Kinase to Phosphatase Activity J. Biol. Chem., October 9, 1998; 273(41): 26415 - 26420. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Malli and W. Epstein Expression of the Kdp ATPase Is Consistent with Regulation by Turgor Pressure J. Bacteriol., October 1, 1998; 180(19): 5102 - 5108. [Abstract] [Full Text]
|
|
|
|
 |
|
 N. Uozumi, T. Nakamura, J. I. Schroeder, and S. Muto Determination of transmembrane topology of an inward-rectifying potassium channel from Arabidopsis thaliana based on functional expression in Escherichia coli PNAS, August 18, 1998; 95(17): 9773 - 9778. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Jung and K. Altendorf Truncation of Amino Acids 12-128 Causes Deregulation of the Phosphatase Activity of the Sensor Kinase KdpD of Escherichia coli J. Biol. Chem., July 10, 1998; 273(28): 17406 - 17410. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Peter, A. Burkovski, and R. Kramer Osmo-sensing by N- and C-terminal Extensions of the Glycine Betaine Uptake System BetP of Corynebacterium glutamicum J. Biol. Chem., January 30, 1998; 273(5): 2567 - 2574. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Jung, B. Tjaden, and K. Altendorf Purification, Reconstitution, and Characterization of KdpD, the Turgor Sensor of Escherichia coli J. Biol. Chem., April 18, 1997; 272(16): 10847 - 10852. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Puppe, P. Zimmann, K. Jung, M. Lucassen, and K. Altendorf Characterization of Truncated Forms of the KdpD Protein, the Sensor Kinase of the K+-translocating Kdp System of Escherichia coli J. Biol. Chem., October 4, 1996; 271(40): 25027 - 25034. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Heermann, K. Altendorf, and K. Jung The Hydrophilic N-terminal Domain Complements the Membrane-anchored C-terminal Domain of the Sensor Kinase KdpD of Escherichia coli J. Biol. Chem., May 26, 2000; 275(22): 17080 - 17085. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Jung, M. Veen, and K. Altendorf K+ and Ionic Strength Directly Influence the Autophosphorylation Activity of the Putative Turgor Sensor KdpD of Escherichia coli J. Biol. Chem., December 15, 2000; 275(51): 40142 - 40147. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |