| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, August 11,
1994; and in revised form, November 4, 1994) From the
Mutations made at and near the methylation sites of the Escherichia coli aspartate receptor were found to affect the
methylation rates of the remaining methylation sites. The results
supported a model in which the methyltransferase enzyme contacts a
residue seven amino acids to the C terminus of a site being methylated.
The presence of a negatively charged residue at that position inhibits
methylation, whereas a neutral residue has no effect. Methylation sites
in the wild type receptor may also influence the methylation of other
sites which are 7 residues away through a physical contact with the
methyltransferase. Covalent modification is a common mechanism by which the
function of transmembrane receptors is regulated. In many instances,
these modifications are responsible for adaptation(1) , in
which the response to a constant stimulus is eventually diminished. In Escherichia coli, the chemotactic response to aspartate is
mediated by the aspartate receptor(2, 3) , identified
as the Tar protein(4) . Methylation of specific glutamate
residues in the cytoplasmic region of the aspartate receptor is
responsible for adaptation(5, 6) . The steady state
level of methylation changes in proportion to the extracellular
aspartate concentration, which counteracts the signal generated by
aspartate binding and prevents a sustained chemotactic response to a
constant stimulus. Methyl groups are added by a cytoplasmic
methyltransferase enzyme(7, 8) , and removed by a
cytoplasmic methylesterase
enzyme(9, 10, 11) . The methylation sites
of the aspartate receptor have been identified as glutamate residues
295, 302, 309, and 491, which are referred to as methylation sites one
through four, respectively(12, 13) . Because sites
one, two, and three are spaced in seven-amino acid intervals, they were
postulated to be on the same face of a single
Figure 1:
Amino acid sequences surrounding the
methylation sites. A stretch of amino acids containing methylation
sites one through three and a stretch containing site four are shown. Xxx indicates any amino acid. The boxes encapsulate
positions found in seven-amino acid intervals, with down being
equivalent to the C-terminal direction. Residue numbers are shown
underneath corresponding residues. The encapsulated glutamates are
methylation sites, which are designated on the right along
with their relative methylation rates. These rates were derived from Table 1and normalized so that the lowest rate equals
one.
Analysis of a series
of mutant aspartate receptors in which methylation site residues were
mutated from glutamates to aspartates (18) suggested that the
methyltransferase also contacts a residue seven amino acids to the
C-terminal side of a methylation site, extending the recognition
sequence to include residues on three helical turns. Given the spacing
between methylation sites one, two, and three, it was proposed that
site three is contacted by the methyltransferase when site two is
methylated and that site two is contacted when site one is methylated.
When an aspartate was present 7 residues to the C terminus of a
methylation site, no methylation occurred at that position, suggesting
that contact with an aspartate causes the methyltransferase to bind in
a nonproductive manner. Because a glutamine is found 7 residues to the
C terminus of site three and an alanine is found 7 residues to the C
terminus of site four (see Fig. 1), contact with these residues,
as with glutamate, apparently allows productive binding. To test the
ideas described above, a mutational analysis was extended to include
residues outside the set of methylation sites, specifically residues
seven amino acids to the C-terminal side of sites three and four.
Mutations were also made at methylation sites two and three, which are
7 residues to the C terminus of sites one and two, respectively. The
methylation rates of the mutant receptors were then determined. Mutants
will be described using the one-letter amino acid code for the residues
at methylation sites one through four, respectively. For instance, EEEE
represents a receptor in which all four sites are glutamates, and EQEE
represents a receptor in which site two has been converted to
glutamine. To describe receptors with mutations outside the set of
methylation sites, parenthesis will be placed to the C-terminal side of
the letter representing the methylation site closest to the mutation.
The parentheses will enclose the number of the mutated residue preceded
by the one-letter code for the wild type residue and followed by one
letter-code for the mutant residue. For example, EEE(Q316N)E represent
a receptor in which the glutamine 7 residues from site three (residue
316) has been changed to asparagine.
Figure 2:
Methylated peptides derived from receptor
mutants. Samples of extensively methylated receptors were subject to
proteolysis and HPLC. The amount of radioactivity eluting from the
column over time was monitored. Numbered peaks represent a methylated
peptide containing the indicated methylation site. These assignments
were based on retention times as described in the methods section. The
peak found at 5 min in each chromatograph was shown to be methanol (13) . The areas of the different peaks are very similar and do
not reflect the differing initial rates of methylation of the
individual sites (Table 1), suggesting that receptor methylation
was nearly
complete.
To determine which residues in the aspartate receptor
contribute to the recognition of sites by the methyltransferase,
mutations were made at methylation sites two and three, which are 7
residues from sites one and two, respectively, and also at Gln When
Ala To determine if site four in the
EEEE(A498E) and EEEE(A498D) receptors could be methylated at all, the
methylation patterns of receptors subject to extensive methylation
reactions were analyzed (Fig. 2). The reaction conditions were
designed to maximize the extent of methylation. No site four
methylation was detectable in the EEEE(A498D) receptor, while site four
in the EEEE(A489E) receptor was methylated to a similar extent as in
the EEEE and EEEE(A498Q) receptors. This suggests that the aspartate
substitution at Ala When the glutamine 7 residues
to the C terminus of site three (Gln Mutation of site
three to glutamine in the EEQE receptor caused an increase in the
methylation rates of all other positions (Table 1). The mutation
of site three to aspartate was previously found to enhance site one and
four methylation while nearly abolishing site two
methylation(18) . Thus, the two types of substitutions differ
primarily in their effects on site two. Mutation of site two to
glutamine in the EQEE receptor caused only minor changes in the
methylation rates of the remaining sites (Table 1). Mutation of
site two to either asparagine or alanine caused a roughly 4-fold
increase in the methylation rate of site one, while causing less than
2-fold changes at the other sites of methylation. It is informative to
compare these results to those of substituting an aspartate at site
two(18) . In every cases, changes of no more than 2-fold were
observed at sites three and four. However, while the aspartate
substitution at site two abolished site one methylation, each of the
neutral amino acid substitutions had a positive effect on the rate of
site one methylation. Two triple mutants were constructed in which
only site one remained a glutamate. When sites two, three, and four
were all aspartates, no site one methylation was detectable (Table 1). When site two was instead a glutamine and sites three
and four were aspartates, site one was methylated at a rate similar to
that of site one in the EEEE receptor. These results are consistent
those obtained from analysis of site one methylation in the EDEE (18) and EQEE receptors (Table 1). In essentially all the mutants analyzed here (Table 1),
changes occur in the methylation rates of every site. This effect of
methylation site substitutions was also observed in a study of the
related receptor for ribose and galactose in which alanine
substitutions were made(20) . It is likely that these changes
in methylation rate occur through several mechanisms, involving
alterations in both the structure of the receptor and in the
recognition of individual methylation sites by the methyltransferase.
The ability of alterations at and near the methylation sites to
influence other methylation sites appears to be a general property of
the receptor. In a previous analysis of aspartate substitutions at
the methylation sites of the aspartate receptor(18) , it was
found that the effects the mutations had on the methylation of
remaining sites could be divided into two categories, referred to as
``type A'' and ``type B.'' Both types of effects
are also apparent in the results described here (Table 1). Type B
effects are defined as an alteration in methylation rate that occurs at
all methylation sites. These effects, which are generally changes of
less than 3-fold, were postulated to be the result of perturbations in
receptor structure. Type A effects are defined as changes in
methylation rate occurring at a single methylation site that are of
greater magnitude or of opposite sign than effects occurring elsewhere.
Type A effects were postulated to result from a perturbation in the
recognition of an individual methylation site by the methyltransferase. An analysis of the results presented here (Table 1) provides
evidence that the methylation rate of a site is influenced by the
residue seven amino acids away from that site in the C-terminal
direction. For example, in the EEEE(A498D) receptor, there is a
complete lack of methylation at site four, which we classify as a type
A effect, accompanied by changes of roughly 2-fold in the methylation
rates of the other positions, which we classify as type B effects. The
results obtained with the EEEE(A498E) receptor show that the
introduction of a glutamate can also cause both type A and type B
effects. The residue seven amino acids away from a methylation site
can exert a positive as well as a negative influence on the methylation
of that site. This is best exemplified by the ENEE receptor, in which
the substitution of asparagine at site two caused a 4-fold increase in
the methylation rate at site one and slight decreases in the
methylation rates of sites three and four. Similar results were also
seen in the EAEE and EQEE mutants, indicating that the site two
substitutions enhance methylation at site one, positioned seven amino
acids to the N terminus of the mutations, through a mechanism distinct
from the one affecting the other positions. The appearance of type A
effects in many of the mutants described here, and only at sites seven
amino acids to the N terminus of a mutation, provides evidence that a
residue at this position can influence the binding of the
methyltransferase to a particular site. A model for the interaction
between the methyltransferase and the aspartate receptor (Fig. 3) can be derived from results presented here (Table 1) and previously(15, 18) . As shown in Fig. 3, when a glutamate or aspartate is present seven amino
acids to the C terminus of a methylation site, methylation is inhibited
because the geometry of the methyltransferase-receptor complex is
constrained. Perhaps this is due to an ionic interaction between a
positively charged residue on the methyltransferase and the negative
charge on the receptor. Contact with a glutamate inhibits but does not
completely prevent the methyltransferase from binding in a productive
manner, while contact with an aspartate prevents methylation entirely.
The negative influence of aspartate is demonstrated by the lack of
methylation at site one or two in the presence of an aspartate
substitution at site two or three, respectively(18) , and by
the lack of site four methylation in the EEEE(A498D) mutant (Fig. 2). The negative influence of glutamate is revealed by the
increase in site one methylation in the EQEE, ENEE, and EAEE mutants
relative to site one in EEEE (Table 1), and by the nearly
undetectable methylation of site four in the EEEE(A498E) mutant (Fig. 2). Additionally, the high rate of methylation at site
three in EEEE relative to sites one and two may be explained by the
presence of a glutamine rather than a glutamate 7 residues to its C
terminus (Fig. 1).
Figure 3:
The interaction between the
methyltransferase and the aspartate receptor. The cartoon shows a
proposed interface between a portion of the methyltransferase and a
portion of the aspartate receptor. The effect of the residue at site
two on methylation of site one is shown, but this is meant to
illustrate in general how a residue can influence a methylation site
positioned seven amino acids away in the N-terminal direction. The
methyltransferase contacts the receptor in three places; the active
site, indicated by the arrow, is positioned above a site to be
methylated; a second binding pocket contacts the rectangle between the methylation sites, which represents the residues found
one helical turn from each site; a positive charge in a third binding
pocket contacts a residue two helical turns, or seven amino acids, from
the site being methylated. When a glutamate is found at site two (A), the methyltransferase is attracted to it, but can still
methylate site one. When an aspartate is found at site two (B), the methyltransferase binds in a distorted manner,
preventing site one methylation. When an alanine is found at site two (C), the methyltransferase can bind in a manner allowing
optimal methylation of site one. The type of binding shown in C would also occur if site two was a glutamine, an asparagine, or
possibly a methylglutamate.
According to our model (Fig. 3),
when a glutamine, asparagine, or alanine is present seven amino acids
away from a methylation site, the methyltransferase may bind in a
productive manner. Differences in the effects of these 3 neutral
residues are not obvious, but there is some indication that the rate of
methylation is higher when an asparagine or alanine, rather that a
glutamine, is present. For instance, substitution of a glutamine for
the alanine normally found 7 residues to the C terminus of site four
caused a decrease in site four methylation through a type A effect (Table 1). Further, site one is methylated at a greater rate in
the EAEE and ENEE receptors than in the EQEE receptor. Perhaps
glutamine, which has a size similar to glutamate, has some interaction
with the methyltransferase, constraining it slightly, whereas the
smaller alanine and asparagine residues do not make contacts. An
alanine is present 7 residues to C-terminal side of site four in the
wild type receptor, which would be expected to allow an optimal
methylation rate (Fig. 1). Despite this, site four is methylated
at a relatively low rate, which is probably due to the deviation of the
surrounding residues from the consensus recognition sequence for
methylation(14, 15) . The near absence of methylation
at site four in the EEEE(A498E) mutant may be explained as a
combination of the negative influence of the substituted glutamate and
of the other residues surrounding site four. According to the model
presented in Fig. 3, the residue seven amino acids from a
methylation site lowers the rate of methylation at that position if it
is negatively charged, and has no influence on methylation if it is
neutral. Thus, the glutamate residues at methylation sites two and
three in the wild type receptor may have a negative influence on the
methylation rates of sites two and one, respectively. An intriguing
possibility is that the conversion of a methylation site to a neutral
methylglutamate by the methyltransferase may mimic the effect of
mutating a site to a neutral residue. Consequently, methylation of site
two or three would be expected to enhance the rate of methylation at
site one or two, respectively. Further, given that the results
presented here and elsewhere(18, 20) indicate that
virtually any substitution at a methylation site influences the
methylation rates of the other sites, it seems very likely that the
methylation of one site would influence the methylation of the others
through one or more mechanisms. Our model has some interesting
implications about the role of receptor methylation in chemotactic
behavior. As a bacterium swims up a gradient of aspartate, the number
of unmethylated sites on the aspartate receptor decrease as the
concentration of aspartate increases(21) . The proportion of
receptors with sites two and three available for methylation has been
found to decrease especially rapidly(14) . If methylation at
site two or three enhances the rate of methylation of site one or two,
respectively, it would help the overall rate of receptor methylation to
remain nearly constant despite a diminished number of unmethylated
sites. This in turn would help the adaptation rate to remain constant
across an aspartate gradient. Interestingly, it was found that bacteria
exposed to saturating concentrations of maltose, which signals through
the aspartate receptor, adapted to aspartate with kinetics similar to
those observed when maltose was absent(22) . The steady state
level of receptor methylation at the time of aspartate addition was
higher when maltose was added first, suggesting that the rate of
adaptation is not appreciably influenced by the absolute level of
receptor methylation. Our conclusions, combined with the recognition
sequence described previously(14, 15) , suggests that
the methyltransferase normally contacts at least three turns of helix
and 6 residues in binding to the receptor. A recognition sequence of
this length allows simultaneous recognition of two methylation sites.
Substrate recognition by casein kinase II and glycogen synthase
kinase-3 provide interesting parallels (23, 24) .
These enzymes phosphorylate serines and threonines found in close
proximity to negatively charged groups. In some cases, phosphorylation
at one site introduces the negative charge necessary for
phosphorylation of another. Thus, two phosphorylation sites may be
simultaneously contacted by the kinase, and the modification state of
one site influences the rate of modification at the other.
Volume 270,
Number 2,
Issue of January 13, 1995 pp. 751-755
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
helix(14, 15) . Site three is the most rapidly
methylated position, site two is methylated at roughly half the rate of
site three, and sites one and four are both methylated at less than a
tenth the rate of site three (14) (see Fig. 1). The
differences in rate may be related to the differences in the residues
surrounding each methylation site. Based on the pattern of residues
surrounding the methylation sites of the aspartate receptor and the
highly related receptor for serine(16, 17) , it was
first proposed that the methyltransferase optimally recognizes sites
within the ``consensus sequence''
Glu-Glu*-Xaa-Xaa-Ala-Ser/Thr, in which the starred glutamate is the
site of methylation and ``Xaa'' is any
residue(14, 15) . The 4 residues composing this
consensus sequence should all be on the same helical face. The
consensus sequence thus suggests that the methyltransferase contacts
residues on the same helical turn as a methylation site and residues on
the adjacent turn in the C-terminal direction.
Materials
Oligonucleotides were purchased from
Operon Technologies (Alameda, CA). Reagents for site-directed
mutagenesis were from Bio-Rad. Cycle sequencing reagents were from Life
Technologies, Inc. Tritiated AdoMet (15 Ci/mmol) was from Amersham
Corp. Scint A-XF scintillation fluid was from Packard. V8 protease was
from Pierce. Trypsin was from Worthington. Prestained protein molecular
weight markers (low range) were from Bio-Rad. A 30-cm Bondclone
C
reverse-phase column from Phenomenex was utilized. All
other materials were reagent grade.Strains and Plasmids
HCB721, a strain deficient in
all chemotaxis genes(19) , was provided by Dr. Howard Berg
(Harvard University). All plasmids were constructed by mutagenesis of
the pSK2 plasmid as described previously(18) . Codons changed
to glutamate, glutamine, aspartate, asparagine, or alanine were
specified by GAG, CAG, GAT, AAC, or GCC, respectively.Methylation Reactions
Cell membranes containing
various mutant receptors and the methyltransferase enzyme were prepared
and analyzed as described elsewhere(18) . All reactions were
performed by mixing equal volumes of a solution containing receptor in
30 mM sodium phosphate, pH 7.0, 1 mM phenylmethylsulfonyl fluoride, and a solution containing tritiated
AdoMet (100 µCi/ml) and methyltransferase in 30 mM sodium
phosphate, pH 7.0, 1 mM phenymethylsulfonyl fluoride, 2 mM 1,10-phenanthroline, and 5 mM EDTA. Initial rates of
methylation (see Table 1) were determined as described elsewhere (18) using 4 µM receptor and a 1 to 200 dilution
of methyltransferase extract. Samples intended for proteolysis were
allowed to react under these conditions for 2 min. Extensive
methylation reactions (see Fig. 2) contained 2 µM receptor and a 1 to 50 dilution of methyltransferase extract and
were allowed to proceed for 2 h at 37 °C.
Determination of Methylation Rates at Individual
Sites
Samples of receptors methylated under initial rate
conditions (see above) were subject to proteolysis followed by HPLC (
)as described previously(14, 18) .
Radioactive peaks were detected using a BetaRam flow-through detector
(INUS Systems). The resulting peptide maps were integrated using the
Kaleidagraph program (Abelbeck Software) in order to determine the
percent of the total methyl groups incorporated at each site. These
percentages were then multiplied by the overall initial rates of
methylation to determine the initial rates at each site (see Table 1). The average and standard deviation of the total rates
were determined from four or more assays and the percent incorporation
at each site was determined from the average of at least two assays.Analysis of Peptide Maps
The chromatographic
mobilities of methylation site-containing peptides derived from EEEE
were established previously(14, 18) . The mutation in
the EEE(Q316N)E receptor does not alter a residue in one of these
peptides. The mutations made at sites two and three alter the
composition of peptides containing the mutated sites only. The site
four-containing peptide is the 20-mer VTQQNASLVQE*SAAAAAALE (E* is site
four). The last alanine in this peptide is replaced by glutamine or
aspartate in the EEEE(A498Q) or EEEE(A498D) mutant, respectively. In
the EEEE(A498E) mutant, a new V8 cleavage site is introduced which
alters the site four-containing peptide to an 18 mer ending in
E*SAAAAAE. Each of these changes would be expected to make the site
four-containing peptide more polar and decrease its retention time on
the column, which is apparent for the EEEE(A498Q) (Fig. 2B) and EEEE(A498E) mutants (Fig. 2D). The apparent absence of a site
four-containing peptide in the EEEE(A498D) mutant is unlikely to be
caused by that peptide having the same chromatographic mobility as one
of the other methylation site-containing peptides, each of which is
only a 7 mer.
and Ala
, which are 7 residues to the C terminus of
sites three and four, respectively. The initial rates of methylation at
individual sites in the various mutant receptors are listed in Table 1. Methylation rates of receptors with aspartate
substitutions at sites one and two, as determined
previously(18) , are also listed for comparison.
was mutated to a glutamine, producing the EEEE(A498Q)
receptor, only minor changes in the various methylation rates were
apparent (Table 1). When Ala
was changed to either
an aspartate or a glutamate, no methylation at site four was detectable
under initial rate conditions, while changes of less than 3-fold
occurred at sites one, two, and three (Table 1). These results
are consistent with previous findings (18) that the presence of
an aspartate 7 residues to the C terminus of a methylation site
drastically reduces its methylation rate and suggests that a glutamate
may have a similar effect.
abolished methylation at site four,
while the glutamate substitution reduced the initial rate of site four
methylation to an undetectable level.
) was mutated to an
asparagine, the methylation rate at all four sites decreased (Table 1). Site three did not seem to be affected to a greater
extent than the others. Unfortunately, we were not able to evaluate the
effect of changing Gln
to either glutamate or aspartate
because receptors containing these substitutions were not stably
expressed, probably due to proteolytic degradation.
)
©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:
|
|
 |
|
  T. Emonet and P. Cluzel Relationship between cellular response and behavioral variability in bacterial chemotaxis PNAS, March 4, 2008; 105(9): 3304 - 3309. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. P. Alexander and I. B. Zhulin From the Cover: Evolutionary genomics reveals conserved structural determinants of signaling and adaptation in microbial chemoreceptors PNAS, February 20, 2007; 104(8): 2885 - 2890. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. G. Endres and N. S. Wingreen Precise adaptation in bacterial chemotaxis through "assistance neighborhoods" PNAS, August 29, 2006; 103(35): 13040 - 13044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. A. Mello and Y. Tu Perfect and Near-Perfect Adaptation in a Model of Bacterial Chemotaxis Biophys. J., May 1, 2003; 84(5): 2943 - 2956. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Djordjevic, P. N. Goudreau, Q. Xu, A. M. Stock, and A. H. West Structural basis for methylesterase CheB regulation by a phosphorylation-activated domain PNAS, February 17, 1998; 95(4): 1381 - 1386. [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 |
ABSTRACT
