|
|
|
|||||||
|
|||||||||
(Received for publication, July 11, 1995) From the
To determine the role in transmembrane signaling of the
N-terminal peptide of the first transmembrane region of the aspartate
receptor, it was subjected to extensive mutagenesis. Drastic changes
did not alter the chemotactic ability of the receptor to aspartate
significantly. Thus the cytoplasmic N terminus of the first
transmembrane region does not play an essential role in transmembrane
signaling, and the entire signal that is transmitted to the cytoplasmic
domain must be sent through the second transmembrane region. This
eliminates the models requiring an interaction of this N-terminal
peptide with the remaining cytoplasmic portion of the receptor.
Receptor-mediated transmembrane signaling is important in all
living organisms from bacteria to human beings. Most receptors function
as either homodimers or heterodimers (for recent reviews see Refs. 1
and 2). The dimerization process is sometimes induced by ligand binding
and sometimes totally independent of it. The human growth hormone
receptors belong to the first category(3) . Others, such as
human insulin receptor (4) and the bacterial chemotactic
receptors(5) , do not change their oligomeric states upon
ligand binding. These receptor families share similar transmembrane
topology (one or two single transmembrane segments connecting an
extracellular ligand binding domain and an intracellular signal domain
in each subunit). They are believed to transmit signals by a common
mechanism, because chimeric receptors containing the ligand binding
domain of one receptor and the signaling portion of another are still
capable of transmembrane signaling(6, 7) . Due to the
detailed understanding of the biochemical pathways of the bacterial
chemotaxis system and the relative ease in manipulating bacterial
proteins, the aspartate receptor provides us with an excellent system
to study the general principles underlining the signal transduction
mechanism. Each aspartate receptor subunit has a periplasmic ligand
binding domain, two transmembrane segments (transmembrane (TM) ( When the alanine residue (at position 19) in the
middle of TM 1 was replaced by a lysine residue (A19K), the mutant
receptor was nonfunctional, even though the binding affinity for
aspartate remained similar to that of the wild type(11) . The
pseudorevertants of this A19K mutant were largely found within a
40-residue region (residues 264-303) of the cytoplasmic domain.
Thus there might be some interactions between the cytoplasmic domain
and the TM 1, probably through the cytoplasmic extension of TM 1, a
six-residue tail at the N-terminal end of the receptor sequence, and
this interaction might be important for transmembrane signaling. Short or long C-terminal tails exist in a number of receptors, but
their role is still uncertain at the moment (for recent reviews see (12, 13, 14) ). The N-terminal tail of the
aspartate receptor is highly conserved in the chemotactic receptor
family (Table 1). In order to clarify its role in the signal
transduction process, extensive mutations have been introduced into
that region, and the cells with the modified receptors have been
examined.
The minimal medium contains Vogel-Bonner citrate
salt(16) , 1% glycerol, 100 µg/ml ampicillin, and 500
mg/liter of each of the following: L-histidine, L-methionine, L-leucine, L-threonine, and
thiamine. The Luria broth contains 1% tryptone, 0.5% yeast extract, and
1% NaCl. The minimal plates have 0.3% agar in minimal medium. The
aspartate plates are the same as the minimal ones except that aspartate
was added to 100 µM. The tryptone plates have 1.3%
tryptone, 0.6% NaCl, 0.3% agar, and 100 µg/ml ampicillin.
where B and F are the concentrations of bound
and free aspartate, respectively, B
For the
methylation of purified receptors in a reconstituted system, partially
purified receptor sample was added into the reconstitution buffer (50
mM NaPi, pH 7.0, 40% glycerol, 1/10 volume of null membrane,
0.5% OG, and 1 mM PMSF) and incubated at room temperature for
45 min and then the procedure above was followed.
Several mutants within the N-terminal cytoplasmic tail region
of the Salmonella typhimurium aspartate receptor were made (Table 2). Three of them were deletion mutants: Arg In order to optimize the
overexpression of the aspartate receptor, an EcoRI restriction
site was placed immediately upstream of the tar We performed in vivo swarm assays
to test the effects of the mutations on chemotaxis. As shown in Fig. 1, cells harboring each of the mutant receptor genes showed
normal responses toward aspartate in the medium compared with those
with wild type receptor genes, and the cells containing no receptor
gene showed the same swarm rates in the presence and absence of
aspartate. Fig. 2shows the swarm rates of the mutants on
tryptone plates. The results agree with those in Fig. 1. Thus it
seems that a wide variety of mutations at the N-terminal tail region do
not impair the chemotactic ability of the cells significantly.
Figure 1:
Swarm rates of the
N-terminal mutants on minimal and aspartate plates. Swarm assays were
performed at 30 °C on the minimal medium plate with or without
attractant aspartate. Swarm rates (mm/h) were reported as the slopes of
the linear fits of the time-course data points. For each receptor, the column on the left is the swarm rate in the absence
of aspartate and the column on the right is that in
the presence of aspartate. Top, class A mutants (see
``Results'' for details); middle, class B mutants; bottom, class C mutants. wt, wild
type.
Figure 2:
Swarm
rates of the N-terminal mutants on tryptone plates. Swarm assays were
performed at 30 °C on the tryptone plates. Swarm rates (mm/h) were
reported as the slopes of the linear fits of the time-course data
points. Top, class A mutants (see ``Results'' for
details); middle, class B mutants; bottom, class C
mutants. wt, wild type.
The
exceptions were the R4G mutant of class B ( Fig. 1and Fig. 2) and the R4C mutant of class A ( Fig. 1and Fig. 2). The R4G mutant in class B was very poorly expressed.
However, when it was subcloned into pBTac2, the expression level was
comparable with other mutants, and it swarmed normally ( Fig. 1and Fig. 2). The R4C mutant showed lower swarm
rates than most of the others on both aspartate and tryptone plates,
but its rates were still higher than those of the negative controls.
For each swarm assay, we did immunoblotting analysis to monitor the
expression level of the receptor. The differences in swarm rates that
did occur on minimal and aspartate plates of different classes could be
traced to the differences in expression level (data not shown). Because swarm assay is only a semiquantitative assay to measure
chemotaxis and because several factors could affect swarm rates, we
also did some in vitro biochemical studies on some of the
mutants. First we did immunoblotting experiments on the membrane
samples and found that the mutant receptors still associate with the
membrane after high salt washes (data not shown), indicating that the
mutated N-terminal tail region could still function as a leader
peptide. To determine whether the mutations caused any significant
change in the tertiary and quartenary structures of the receptor, the
aspartate binding affinities of the mutants were tested. Table 3listed the binding parameters of the wild type and some of
the mutants. The dissociation constants (K
We then used methylation assays as
a test of signal transduction in vitro. As shown in Fig. 3and 4 for the wild type and each of the mutants tested,
aspartate increased the methylation rates in a similar manner to wild
type. In the reconstituted system (Fig. 4), the addition of
aspartate increased the methylation rate of the wild type by 1.8-fold.
For the
Figure 3:
Methylation rate ratios of the N-terminal
mutants in a membrane vesicle system. Methylation assays were performed
at 37 °C in the presence and the absence of aspartate. The
methylation rates were the slopes of the linear fits of the time-course
data points. The ratio of each receptor was the value of the
methylation rate in the presence of aspartate divided by that in the
absence of aspartate. wt, wild
type.
Figure 4:
Methylation rates of the N-terminal
mutants in a reconstituted system. Partially purified receptor samples
were incubated with the null membrane at room temperature for at least
45 min prior to the methylation assays, which were performed at 37
°C in the presence and the absence of aspartate. The methylation
rates were the slopes of the linear fits of the time-course data
points. The rates were normalized to the methylation rate of the wild
type receptor in the absence of aspartate. For each receptor, the column on the left is the rate in the presence of
aspartate, and the column on the right is that in the
absence of aspartate.
To probe the possible motion of TM 1 within each subunit
(either a piston-like motion or a rotation), we compared the rates of
disulfide bond formation of two cysteine mutants in the presence and
the absence of aspartate. They looked virtually identical for each
mutant, as shown in Fig. 5. Based on a previous cross-linking
study(24) , residues 4 and 4` are at the dimer interface, with
the closest distance among the N-terminal tail residues. The inserted
cysteine residue of the ICys2 mutant is located at the position for
methionine residue in the wild type receptor. The side chain of this
residue faces the second transmembrane segment within the same subunit.
Figure 5:
Cross-linking time-courses of the ICys2
mutant. Cross-linking reactions were performed in the presence of 1
mM copper-phenanthroline at 37 °C. Aliquots were removed
at various times, and the reaction was quenched to yield the reaction
coordinate, the fraction of the receptor that was cross-linked (cpm of
the dimeric receptor divided by the sum of cpm of the dimeric and
monomeric receptor). The filled circles were the points in the
presence of aspartate, and the open circles were those in the
absence of aspartate. The time-courses of the R4C mutant looked similar
to those of the ICys2 mutant.
The results reported herein indicate that major changes can
be made in the N-terminal peptide projecting into cytoplasm with very
minor effects on either transmembrane signaling or aspartate binding
ability. The mutants did not interfere with insertion into the membrane
or the folding of the cytoplasmic domain. Previous studies showed
that a TM 1-less receptor (first 30 amino acid residues deleted) was
not functional as determined by both swarm assays and methylation
assays(25) . However, we found that deletion of four residues
at the N-terminal end did not affect chemotaxis significantly. Thus the
TM 1 transmembrane region is important, but the peptide extending from
the TM 1 region into the cytoplasm is not. This suggests that the
transmembrane helices (TM 1 and TM 1`) are important in maintaining the
structural integrity of the whole receptor and probably the
transmembrane signaling. A distortion of these interactions would lead
to abolished function of the receptor, as shown by the A19K mutant in
TM 1 (11) and the 204 mutants in TM 2(26) . It has
also been found that when TM 1 and 1` were cross-linked by a
Cys If the interaction
of the N-terminal hexapeptide with the cytoplasmic signaling domain is
eliminated as a source of the indicated transmembrane conformational
change, then the entire transmission of the signal that is delivered to
the cytoplasmic domain must go through TM 2. Clearly that severely
limits the mechanism for such transmembrane signaling. A rotation
model, as suggested by Maruyama et al.(29) , seems
unlikely as a sole contributor to the transmembrane signaling. Rotation
of the two cytoplasmic subunits relative to each other would be
excluded as a transmembrane signaling option by the results of Milligan
and Koshland(10) , in which the cytoplasmic portion of one
subunit could be eliminated with only a minor effect on signaling. Thus
a piston model(30) , in which the cytoplasmic domain moves
relative to the membrane, or a model involving relative motion of
transmembrane segments TM 2 and TM 2` seems indicated. Such a model
could also explain the transmembrane signaling of the epidermal growth
hormone and low density lipoprotein receptors, which have one
transmembrane region per receptor subunit.
Volume 270,
Number 41,
Issue of October 13, 1995 pp. 24038-24042
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
)1 and TM 2), and one cytoplasmic domain. The native
receptor has two subunits that do not associate or dissociate during
signaling(5) . The aspartate binding sites, as seen in the
x-ray crystal structure, lie at the interface between the two
subunits(8) . Cross-linking experiments with introduced
cysteine residues at various locations of the receptor showed that the
signal is a conformational change of the receptor(9) . It can
be transmitted through one subunit of the functional
dimer(10) .
Bacterial Strains and
Plasmids
Escherichia coli strain CJ236 (dut
ung) was from BioRAD. XL1-Blue and XL2-Blue were from Stratagene.
RP3808 (
(cheA-cheZ)2209 tsr-1) and RP8611 (tsr7028 (tar-tap)5201 zbd::Tn5 (trg)100) were
gifts from Dr. J. S. Parkinson at the University of Utah. HCB721 (15) was from Dr. H. C. Berg at Harvard University. Plasmids
pEMBL18+ and pBTac2 were from Boehringer Mannheim.Chemicals and
Media
N-Ethylmaleimide, n-octyl-
-D-glucopyranoside (OG), S-adenosyl-L-methionine, -aminooctyl-agarose
gel, thiamine, and the amino acids L-histidine, L-methionine, L-leucine, and L-threonine
were from Sigma. L-Aspartic acid, Zwittergent 3-12, and
isopropylthio-
-D-galactoside were from Calbiochem. S-adenosyl-L-[
H]methionine, L-[2,3]-[H]aspartic acid, and S-protein A were purchased from Amersham Corp.
Dithiothreitol was from Boehringer Mannheim. Mutagenic oligonucleotides
were either from Operon Technologies (Alameda, CA) or from the DNA
Synthesis Facility of the University of California at Berkeley. The
Centriprep-30 ultrafiltration tubes were from Amicon. The T4 Mutagene
kit, the protein assay kit, and the Affi-Protein A MAPS II kit were
from Bio-Rad. The fmol
DNA sequencing system was
from Promega.
Mutant Construction
Site-directed
mutagenesis was carried out according to Kunkel(17) . For
technical reasons, one set of the oligonucleotides contain an EcoRI restriction site right upstream of the start codon of
the tar
gene, in addition to the desired
mutations. A second set has only the individual mutations in the
oligonucleotides. The pXC09 plasmid contains the wild type tarS gene, as an insert in the pEMBL18 vector. Single-stranded pXC09
DNA template was prepared from CJ236. The mutagenesis reactions were
performed using a Mutagene kit according to the user's manual.
The mutagenic products were used to transform XL1-Blue or XL2-Blue
cells. Plasmids were prepared by standard alkaline lysis method, and
the mutants were confirmed by double-stranded DNA sequencing. The class
A plasmids are pEMBL18-based. The receptor gene in this class was
expressed under the control of its natural promoter. Class B plasmids
were constructed by ligating the small EcoRI-HindIII restriction fragments of class A
plasmids into the pBTac2 vector cleaved with the same restriction
enzymes. The receptor genes in this class were expressed under the
control of the tac promoter.Swarm Assay
The plasmids listed in Table 2were used to transform E. coli strain RP8611,
which has none of the four chemotactic receptor genes. Cells were grown
from single colonies in a 1-ml Luria broth with 100 µg/ml
ampicillin (LB+Amp) at 30 °C overnight. Then they were
inoculated onto the center of the minimal, aspartate, or tryptone
plates, using a sealed Pasteur pipet tip. The plates were incubated at
30 °C overnight with the soft surface facing up. Swarm diameters
were measured five times over an 8-10-h period starting
approximately 14 h after inoculation. Swarm rates were extrapolated
from the diameter versus time plots as the slopes of the
linear fits of the points.
Overexpression of the Aspartate
Receptor
Overexpression of the aspartate receptor was
carried out by transforming RP3808 or HCB721 with various plasmids.
Cells containing the class B plasmids were grown in 50 ml of
LB+Amp over night at 37 °C. Then overnight culture was
transferred into 1 liter of minimal medium and grown at 30 °C.
Cells were harvested when A reached
1.0-1.6. Cells containing the class A plasmids were grown in 50
ml of LB+Amp at 30 °C overnight. 23 ml of the overnight
culture were transferred into 1 liter fresh LB+Amp and allowed to
continue growing at 30 °C for 8-10 h before harvesting. Cells
containing the class C plasmids were grown in 10 ml of LB+Amp
overnight at 37 °C. The overnight culture was transferred into 1
liter of LB+Amp and grown at 30 °C.
Isopropylthio-
-D-galactoside was added to 0.5 mM when A reached 0.5-1.0. Cells were
harvested 3 h later. The harvested cells were frozen immediately in
liquid nitrogen and stored at -80 °C until the membrane
preparation.
Membrane Preparation
All steps were
carried out at 4 °C. Frozen cells were thawed on ice and low salt
buffer (100 mM NaPi, pH 7.0, 10% glycerol, 5 mM EDTA,
and 1 mM PMSF) was added to 6-12 ml/liter culture. Cells
were broken by sonication. The lysed cells were centrifuged at 5,000
g for 15 min until there was no visible pellet formed.
The supernatant portion was then spun at 38, 000 rpm (Ti 45 rotor) for
25 min. The pellet was washed twice in high salt buffer (20 mM NaPi, pH 7.0, 2 M KCl, 10% glycerol, 5 mM EDTA,
and 1 mM PMSF) and once in final buffer (50 mM Tris-Cl, pH 7.5, 10% glycerol, 5 mM EDTA, 1 mM PMSF, and 5 mM 1,10-phenanthroline). Finally the membrane
was resuspended in 1 ml/liter culture of final buffer, mixed well,
frozen in liquid nitrogen, and stored at -80 °C until use.
For the membrane preparations of the cysteine mutants, 5 mM dithiothreitol was added in all buffers. The membrane prepared
from RP3808 with pEMBL18 was called null membrane, because it had no tar gene present. The amounts of the receptor in the membrane
samples were quantified by quantitative immunoblotting analyses, using
purified receptor as standard and S-labeled protein A. The
anti-Tar antibody was purified from anti-serum (
-Tar
#9206 prepared by H.-P. Biemann) using the Affi-Protein A MAPS II
kit.Receptor Purification
All steps were
carried out at 4 °C. The membrane sample was thawed on ice. The
detergent OG was added to 1.4% (w/v), and the sample was left on ice
for 10 min. The solubilized membrane proteins were separated from
membrane vesicles by ultracentrifugation at 95,000 rpm for 10 min (TLA
100.3 rotor). Ammonium sulfate was added to 20% (w/v) to the
supernatant fraction. The precipitated protein was pelleted by
centrifugation at 5,000 rpm for 10 min (HS-4 rotor) and was resuspended
in Zwittergent buffer (20 mM Tris-Cl, pH 7.5, 10% glycerol, 1
mM EDTA, 0.1 mM PMSF, and 0.4% Zwittergent
3-12). Then the proteins were loaded onto an
-aminooctyl-agarose column pre-equilibrated with Zwittergent
buffer. After washing the column with three volumes of the Zwittergent
buffer, the proteins were eluted with a linear gradient of 0-750
mM NaCl in 20-bed volume OG buffer (same as the Zwittergent
buffer except that it contained 1.2% OG instead of 0.4% Zwittergent
3-12). The receptor peak was identified by SDS-polyacrylamide gel
electrophoresis, and the fractions were pooled and dialyzed against OG
buffer overnight. The receptor sample was concentrated to 0.3-0.6
mg/ml of total protein, using Centriprep-30 ultrafiltration tubes. An
equal volume of 100% glycerol (w/v) was added to the protein solution,
and it was stored at -20 °C. Protein concentration was
determined by Bradford protein assay (18) using the Bio-Rad
protein assay kit.
Aspartate Binding Assay
The binding
assays were done as published ( (19) and (20) and
references therein) with minor modifications as follows. Membrane
samples were mixed with various amounts of aspartate in a final volume
of 200 µl. After incubating on ice for more than 10 min, 90 µl
of the mixture were transferred into a TLA100 ultracentrifuge tube
containing 1 µl of either water or 100 mM cold aspartate.
Membrane was pelleted by a 10-min centrifugation at 95,000 rpm.
Triplicates of 20 µl of the supernatant fraction were added to a
scintillation vial containing 150 µl of water. Then 2 ml of
scintillation mixture was added, and radioactivity was counted. Binding
data were analyzed (using the Kaleidagraph 3.0 program, Abelbeck
Software) by fitting the curve of bound aspartate concentration versus free aspartate concentration to the Hill equation,
is the
maximal bound aspartate concentration, K
is the
dissociation constant, and n
is the Hill
coefficient.Methylation Assays
Methylation assays
were performed either in the membrane vesicle system or in the
reconstituted system. The methyltransferase was prepared according to
Shapiro and Koshland(21) . For membrane methylation assays,
receptor-containing membranes were mixed with an equal volume of the
methylation mixture (100 µCi of S-adenosyl-L-[H]methionine, 150
mM NaPi, pH 7.0, 30 mM EDTA, 1 mM PMSF, and
4-10 µl of methyl-transferase/ml of mixture) with or without
2 mM aspartate. At various times, part of the reaction mixture
was removed and dotted on a piece of 3 M filter paper, and the
paper immediately dropped in 10% trichloroacetic acid with stirring.
After the final time point was taken, the filter papers were left in
10% trichloroacetic acid for 10 min and then washed twice with 10%
trichloroacetic acid and twice with methanol for 10 min each. Finally
the filter papers were air-dried and added to a vial containing 3 ml of
scintillation mixture. The radioactivity was counted. The counts/minute
values were plotted against time. The methylation rate was reported as
the slope of the linear fitting of the time points.Cross-linking of the Cysteine-containing Mutant
Receptor
The membrane samples were spun down to remove an
excess amount of dithiothreitol and resuspended in the final buffer.
After pre-incubation at 37 °C for 5 min, the catalyst
copper-(phenanthroline)(9) was then added to a
final concentration of 1 mM to start the reaction. At various
time points, an aliquot of the reaction mixture was removed, added to a
tube containing 3 SDS sample buffer with 10 mM EDTA
(to chelate Cu) and 10 mMN-ethylmaleimide (to quench the free thiol groups), and
frozen immediately in liquid nitrogen. The samples were then subjected
to SDS-polyacrylamide gel electrophoresis, and the regions containing
the dimeric and monomeric form of the receptor were excised. The
content of radioactive methyl groups in the gel slice was then
determined by a previously described methanol diffusion
assay(22) .
deleted (R4), four residues deleted from Phe
to
Ile
(2-5), and Ile and Arg
deleted (5-6). Four of them were point mutations that
converted Arg to Gly (R4G), Glu (R4E), Lys (R4K), and Cys
(R4C), respectively. The last one (ICys2) was an insertion of a
cysteine residue between Met and Phe. Two
cysteine-containing mutants, R4C and ICys2, allowed us to cross-link
the receptor molecules with disulfides. start codon so that the tar gene could be
subcloned into another vector that has a strong promoter. It was
designed in the same oligonucleotide carrying each of the several
N-terminal mutations (class B mutants). This approach also made the
screening easier. Unfortunately, this EcoRI site altered the
putative ribosome binding site for the receptor synthesis(23) .
It indeed lowered the expression level of the receptor (data not
shown). These mutants were eventually subcloned into the plasmid
pBTac2, and the expression of these class C receptor mutants were much
better (data not shown). The oligonucleotides for the rest of the
mutants (class A) only contained the desired mutations, and tar gene is under the control of its natural
promoter (Table 2).
) were
within a 2-fold range of that of the wild type. The mutations at the
N-terminal cytoplasmic region did not alter the binding affinity for
aspartate of the receptor significantly. The index for cooperatively
(the Hill coefficient (n)) of each mutant tested
(except for R4K) was within the previously published range,
0.6-0.8(20) , indicating that the binding behavior was
similar to that of the wild type.
R4 mutant and 2-5 mutant, the increase was 1.9-
and 1.7-fold, respectively. The reactions with no purified receptor
showed minimal levels of methylation in the presence and the absence of
aspartate.
-Cys(24) ,
Cys
-Cys(27) , or
Cys
-Cys
(9) disulfide bond,
the receptor could still signal as determined by methylation. This
observation leads to the conclusion that TM 1 and TM 1` do not change
position relative to each other during transmembrane signaling, and it
is confirmed by similar reactions of a
Cys
-Cys/C
-Cys
double cross-linked receptor(28) .
)-D-glucopyranoside; LB+Amp,
Luria broth with 100 µg/ml ampicillin; PMSF, phenylmethylsulfonyl
fluoride.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
ABSTRACT
This article has been cited by other articles:
|
|
 |
|
  X.-N. Zhang, J. Zhu, and J. L. Spudich The specificity of interaction of archaeal transducers with their cognate sensory rhodopsins is determined by their transmembrane helices PNAS, February 2, 1999; 96(3): 857 - 862. [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 |