|
|
|
|||||||
|
|||||||||
(Received for publication, December 12, 1994; and in revised form, February 7, 1995) From the
The Escherichia coli aidB gene is part of the adaptive
response to DNA methylation damage. Genes belonging to the adaptive
response are positively regulated by the ada gene; the Ada
protein acts as a transcriptional activator when methylated in one of
its cysteine residues at position 69. Through DNaseI protection assays,
we show that methylated Ada ( In Escherichia coli, the adaptive response to DNA
methylation damage is activated by sublethal concentrations of
methylating agents such as methyl methanesulfonate or N-methyl-N`-nitro-N-nitrosoguanidine
(MNNG)( For both the ada/alkB and
the alkA promoters, the Ada-binding site has been determined
by DNaseI protection experiments; two short sequences common to both ada and alkA promoter regions (AAANNAA and AAAGCGCA)
have been proposed as the Ada-binding sequence (Ada box) by different
papers ( (10) and (11) , respectively). Mutations in
the AAAGCGCA sequence have been shown to abolish transcriptional
activation of the ada gene by Counterparts of the ada genes have been found in many other bacteria, as well as
in eukaryotic and mammalian cells(13, 14) . In Salmonella typhimurium and Bacillus subtilis,
Ada-binding sites have been identified and
sequenced(15, 16) ; in both of these bacterial
species, no precise consensus sequence has been assigned. However, it
has been shown that E. coli Ada protein is able to activate
transcription from the S. typhimurium ada gene(15) . The aidB gene is ada regulated(17) , but
unlike other adaptive response genes, aidB can also be induced
by oxygen-limiting conditions, or by the addition of sodium acetate to
the growth medium, at a final pH of
6.0-6.5(18, 19) . This second mechanism of
induction is ada independent and requires a functional rpoS gene. rpoS encodes
Figure 1:
aidB promoter and regulatory
region. Sequence numbering is relative to the first transcriptional
start site (+1). The nucleotides protected from DNaseI by
Figure 2:
DNaseI protection of aidB promoter region. A, coding strand. Lane1, no Ada protein; lane2, 12 pmol of
unmethylated Ada; lane3, 12 pmol of methylated Ada. B, template strand. Lanenumbering is the
same as A. The sequencing ladder of a known DNA fragment
(GATC) is shown as a molecular weight standard. The solidlines indicate protection by
Although the borders of the protected region
appear to be less clear on the template strand, the pattern emerging
from DNaseI protection experiments is very similar; the bands
corresponding to positions -62 to -38 are protected by Fig. 3compares
the aidB Ada-binding site to those from the E. coli ada and alkA genes(3) , from the S. typhimurium
ada gene(15) , and from the B. subtilis alkA gene(16) . The Ada-binding site from aidB displays a higher similarity with the Ada boxes from the B.
subtilis alkA gene (14 identical nucleotides out of 18, 78%
identity) and for the S. typhimurium ada gene (67% identity)
than for both the ada and alkA genes from E. coli (56% identity).
Figure 3:
Comparison between Ada-binding sites in
the regulatory regions of different genes. The regions with higher
similarity are indicated by largeletters. The
nucleotides conserved in each sequence are underlined. Numbers give the position of the different Ada-binding regions
relative to the transcriptional start sites. Abbreviations are as
follows: st, S. typhimurium; bs, B.
subtilis; ec, E.
coli.
Figure 4:
Gel retardation experiments. A, aidB regulatory region (BamHI-StyI fragment,
106 bp). Lanes1 and 6, free DNA; lanes2-5, 0.3, 1, 3.3, and 10 pmol of unmethylated Ada
protein; lanes7-10, same concentrations of
Figure 5:
In vitro transcription
experiments. DNA fragments used were HindII-EcoRI
(176 bp) for ada and BamHI-MluI (246 bp) for aidB. A lacUV5 promoter fragment (205 bp) was always
added as an internal standard. The expected RNA transcripts were
92-98 nucleotides for ada, 118-120 nucleotides for aidB, and 72 nucleotides for lacUV5. The sequencing
ladder of a known DNA fragment was used to confirm the size of the
transcripts. A: lane1, p lacUV5
alone; lane2, p lac and p ada, no
Ada protein; lane3, 10 pmol of unmethylated Ada
protein added; lane4, 10 pmol of methylated Ada
added. B: lane1, p lacUV5 alone; lane2, p lac and p aidB no Ada
protein; lane3, 10 pmol of unmethylated Ada protein
added; lane4, 10 pmol of methylated Ada
added.
The aidB gene is ada regulated in
vivo(17) , and the Ada protein, in its methylated form ( The location of the Ada-binding site in aidB may provide information on the -35 and -10
promoter elements and define the areas to target in future
site-specific mutagenesis analysis. The aidB promoter, like
many other positively controlled promoters, lacks -10 and
-35 consensus sequences; based on the position of the
transcriptional start site, the most likely candidate for a -10
sequence is CATACT (Fig. 1). This hexamer maintains the sixth
(T) and the second (A) nucleotide residues that are strongly conserved
in positively controlled promoters (22) and is similar to
CAGCCT, which is the -10 sequence for the ada promoter (11) . Assuming that CATACT is the -10 for aidB,
a possible -35 box might be CTGTCA, which is located 18 base
pairs upstream and is similar to TTGACA, the consensus sequence for the
-35 region. In Fig. 3, we have compared the Ada-binding
site of aidB with the E. coli ada and alkA genes, as well as to the ada gene from S. typhimurium(15) and the alkA gene from B.
subtilis(16) . It is noteworthy that when the ada gene from S. typhimurium is cloned into E. coli,
it functions as part of the adaptive response, i.e. its
Ada-binding site is recognized by the E. coli methylated Ada
protein(15) . Common elements among these binding sites are the
presence of an AGCAA-like sequence (AGCAA in aidB and in alkA from both E. coli and B. subtilis,
ACGCAA in ada from S. typhimurium, and AGCGCAA in the ada gene from E. coli) preceded by an A/T-rich
region. Another common sequence is a short AAT that is always present
five nucleotides upstream of the AGCAA-like element. These observations
suggest that a possible recognition sequence for The affinity of The lower affinity of
Volume 270,
Number 14,
Issue of April 7, 1995 pp. 8285-8289
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
EVIDENCE FOR A NEW CONSENSUS SEQUENCE FOR Ada-BINDING SITES (*)
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
Ada) is able to bind a DNA
sequence between 40 and 60 base pairs upstream of the aidB transcriptional startpoint. Binding of Ada is
necessary to activate transcription of the adaptive response genes;
accordingly, in vitro transcription of aidB is
dependent on the presence of Ada. Unmethylated Ada protein
shows no protection against DNaseI digestion in the aidB promoter region nor does it promote aidBin vitro transcription. The aidB Ada-binding site shows only weak
homology to the proposed consensus sequences for Ada-binding sites in E. coli (AAANNAA and AAAGCGCA) but shares a higher degree of
similarity with the Ada-binding regions from other bacterial species,
such as Salmonella typhimurium and Bacillus subtilis.
Based on the comparison of five different Ada-dependent promoter
regions, we suggest that a possible recognition sequence for Ada might be AATnnnnnnGCAA. Higher concentrations of Ada
are required for the binding of aidB than for the ada promoter, suggesting lower affinity of the protein for the aidB Ada-binding site. Common features in the Ada-binding
regions of ada and aidB are a high A/T content, the
presence of an inverted repeat structure, and their position relative
to the transcriptional start site. We propose that these elements, in
addition to the proposed recognition sequence, are important for
binding of the Ada protein.
)(1, 2) . Upon activation, cells are
protected from the toxic and mutagenic effects of these agents. The Ada
protein is the key enzyme of the adaptive response; this protein is a
methyltransferase able to remove methyl groups from damaged DNA,
transferring them to two of its own cysteine residues. Ada protein
methylated at Cys-69 (Ada) can bind to specific DNA
sequences in the promoter region of several genes; binding of Ada makes the promoters accessible to RNA polymerase and
therefore activates their transcription(3, 4) .
Moreover, a direct protein-protein interaction takes place, since RNA
polymerase mutants resulting from deletions in the C-terminal domain of
the 
subunit are unable to support transcription from the ada promoter, even when Ada is present(5) .
Unmethylated Ada protein, on the contrary, seems to be involved in
negative regulation of the ada promoter(6) . Methylated Ada
activates transcription of its own gene, ada, and of at least
two other genes, alkA and aidB. The ada promoter is also responsible for the transcription of another
gene, alkB, whose function is unknown(7) . The product
of the alkA gene is a glycosylase that recognizes and excises
methylated DNA bases(8) . The aidB gene encodes a
protein homologous to mammalian isovaleryl-coenzyme A dehydrogenase.
Overexpression of the AidB protein is able to counteract MNNG-mediated
mutagenesis in vivo(9) , and this effect seems to be
specific for MNNG. (
)It seems possible that the AidB protein
is involved in a detoxification pathway for MNNG, although it is still
unclear if this function is related to its isovaleryl-coenzyme A
dehydrogenase activity.Ada(11) ,
although the same is true for mutations outside of this proposed
consensus sequence(11, 12) .
, an
alternative factor for RNA polymerase that is required for
expression of stationary phase-specific genes(20) . In this
report, we will consider only the mechanism of the ada-dependent activation of aidB. It has already been
shown that methylated Ada protein is able to bind to the aidB promoter region in vitro(9) , although aidB lacks both the AAANNAA and the AAAGCGCA sequences common to ada and alkA. This suggests that both of these
sequences might be non-essential features of the Ada-binding site. In
this paper, we describe the aidB Ada-binding site and discuss
its importance as a general model for gene regulation by the Ada
protein.
Chemicals
Restriction endonucleases
were purchased from Boehringer Mannheim.
[-
P]dATP,
[-P]dCTP, and
[-P]CTP were from Amersham Corp. E.
coli RNA polymerase and RNase-free DNaseI were from Pharmacia
Biotech Inc.DNA
For gel retardation assays, DNA
fragments were labeled with [-P]dATP by
end-filling with DNA polymerase I (Klenow fragment). The following
fragments were used: for ada, a 176-bp HindIII-EcoRI fragment from plasmid
pYN3066(3) ; for aidB, a 106-bp BamHI-StyI fragment from pMV120 ( (9) and Fig. 1). For DNaseI protection assays, a 246-bp BamHI-MluI aidB fragment from plasmid pMV120
was labeled with [-P]dATP at the BamHI site (template strand); for the coding strand, a BamHI-BstXI fragment (Fig. 1) was subcloned
into the BamHI-MluI sites of pSport1 (Life
Technologies, Inc.) using a BstXI-MluI linker
oligonucleotide (Operon Technologies, Alameda, CA) and producing
plasmid pMV442. The EcoRI-MluI fragment from pMV442
(165 bp) was labeled at the MluI site with
[-P]dCTP by the Klenow end-filling
reaction. For in vitro transcription, the HindIII-EcoRI fragment from ada, the BamHI-MluI fragment from aidB, and the lacUV5 promoter (a 205-bp EcoRI fragment from plasmid
pYN3072(3) ) were used. All DNA fragments were isolated by
electrophoresis and purified from agarose using a Geneclean kit (Bio
101, La Jolla, CA).
Ada are shown in boldfacetype. The twoarrows running in opposite directions indicate an
imperfect palindromic sequence. The possible -35 and -10
promoter elements are boxed. The two transcriptional start
sites identified in vivo(8) are numbered+1 and +3, and the smallarrowabove indicates direction of
transcription. The two possible translation starts for the AidB protein
are outlined, and the dottedunderline shows the
Shine-Dalgarno sequence. Restriction sites used in the in vitro experiments are underlined, and the different enzymes are
indicated.
Gel Retardation Assays
Methylation of the
Ada protein was produced by incubating the protein (120 pmol) with 1
µg of methylated calf thymus DNA for 15 min at 37 °C in 20
µl of 70 mM Na-Hepes, pH 7.8, 1 mM EDTA, and 1
mM dithiothreitol. For unmethylated Ada, the protein was
incubated in the same conditions with 1 µg of unmethylated DNA.
10,000 cpm of labeled DNA, corresponding to approximately 10 fmol, were
used in each reaction. The binding reaction was performed at 22 °C
for 45 min in 10 mM Tris-HCl, pH 7.6, 50 mM NaCl, 5%
glycerol, 0.05% Nonidet P-40, 0.1 µg of salmon sperm DNA, and 10
µg/ml bovine serum albumin in a final volume of 20 µl. The
samples were loaded on a 6% native polyacrylamide gel and run in 0.25
TBE at 12 V/cm.DNaseI Protection Assays
Samples were
treated as described for the gel retardation assay. After 45 min at 22
°C, 6 mM CaCl and 200 ng of DNaseI were added,
and samples were incubated for 30 s at 22 °C. DNaseI digestion was
stopped with an equal volume (25 µl) of DNaseI stop solution (600
mM NaCl, 100 mM EDTA, 30 µg/ml of salmon sperm
DNA), and samples were precipitated with ethanol and resuspended in
loading buffer. Samples were heated for 3 min at 100 °C and loaded
onto a denaturing polyacrylamide/urea gel.In Vitro Transcription
0.5 pmol of the
promoter regions from lac, ada, and aidB were pre-incubated as described for the gel retardation
experiment, with the addition of 4 pmol of RNA polymerase.
Transcription reactions were started by the addition of 6 mM MgCl, 0.2 mM ATP, 0.2 mM UTP, 0.2
mM GTP, 120 µg/ml heparin, and 1 µCi of
[-P]CTP. Incubation was for 6 min at 30
°C. After addition of 1000-fold unlabeled CTP, the reaction was
continued for 5 min and blocked with an equal volume of DNaseI stop
solution (see above). Samples were processed as described for DNaseI
protection experiments and analyzed on a 6% polyacrylamide/urea gel.
DNaseI Protection Experiments
To
identify the Ada-binding site in the aidB promoter region,
DNaseI protection experiments were performed. Results are shown in Fig. 2A for the coding strand and in Fig. 2B for the template strand. For the coding strand, clear protection
by 12 pmol of Ada can be observed between nucleotides
-62 to -38 relative to the transcriptional start site. This
area is marked by the solidline in Fig. 2A. Two DNaseI-hypersensitive sites, at positions
-35 and -28, could also be detected. The region between
these two hypersensitive sites also shows a weak protection by Ada.
Ada; arrows show DNaseI-hypersensitive sites.
Ada (marked by a solidline in Fig. 2B). Two hypersensitive sites are also present at
nucleotides -68 and -75 from the transcription start. It is
noteworthy that the position of the hypersensitive sites is downstream
of the protected region in the coding strand and upstream of it in the
template strand. This might suggest that binding of Ada
induces a symmetrical distortion of the DNA helix in the bordering
regions. No DNaseI-protected or -hypersensitive site was detectable
with 12 pmol of unmethylated Ada in either of the two strands (Fig. 2, A and B).
Gel Retardation Assays
Since aidB does not share a very high homology with the Ada-binding site from
the ada promoter, it was possible that the methylated Ada
protein could display a different affinity for the two sites. To answer
this question, we performed a gel retardation assay with both the ada and the aidB Ada-binding regions in the presence
of different concentrations of unmethylated and methylated Ada protein.
100% retardation is obtained at a concentration of Ada as
low as 0.3 pmol for ada (Fig. 4B), while it
takes an approximately 10-fold higher concentration for aidB (Fig. 4A). Binding was also observed with
unmethylated Ada, although 100% retardation was not obtained for aidB at the concentrations tested and was obtained for ada only at 10 pmol. At high concentrations of Ada, new
bands with a slower electrophoretic mobility began to appear,
particularly for the ada promoter. Although we cannot rule out
that more than one molecule of Ada might be able to bind
specifically at the same time, it seems likely that the slower bands
result from nonspecific interaction between DNA and Ada
and/or residual unmethylated protein.
Ada. B, ada promoter region (HindIII-EcoRI, 176 bp). Lanes1 and 6, free DNA; lanes2-5: 0.3,
1, 3.3, and 10 pmol of unmethylated Ada protein; lanes7-10, same concentrations of Ada.
In Vitro Transcription
In the reaction
conditions described by Teo et al.(4) , in which Ada is able to stimulate transcription from the ada promoter, we could not detect any transcription from the aidB promoter region. However, in a cell-free transcription/translation
system, Ada was required for transcription of the aidB gene from the supercoiled plasmid pMV426(9) , suggesting
that binding of Ada is necessary for aidB transcription. Fig. 5shows in vitro transcription
assays with purified RNA polymerase performed using the reaction buffer
from the footprint experiments instead of the transcription buffer
previously used(4) . In these conditions, transcription from
the aidB promoter produces an mRNA of approximately 120
nucleotides in size, determined using a known DNA sequencing ladder as
a molecular weight marker (Fig. 5B, lanes2-4). This value is in good agreement with the
position of the transcriptional start sites identified in vivo ( (9) and Fig. 1). This transcript was the only one
to be affected by the presence of Ada (Fig. 5B, lane4). It is likely that
the other transcripts originate from nonspecific transcription from
either promoter-like elements or the ends of the DNA fragment. Fig. 5A shows transcription from the ada promoter using the same experimental conditions; the size of the
transcripts (92-98 nucleotides) and the presence of multiple
start sites are in agreement with the results of Teo et
al.(4) .
Ada), is able to interact with the aidB promoter
region and stimulate its transcription(9) . However, neither of
the two consensus sequences so far proposed for Ada-binding sites (10, 11) has been found in aidB. With DNaseI
protection experiments, it has been possible to identify the specific
binding site for Ada; this area spans approximately 24
nucleotides, slightly more than the number of nucleotides protected in
both the ada and the alkA promoter
regions(4) . Ada-binding sites are centered at different
positions relative to the promoter elements and the transcriptional
start site in ada and alkA, but they are always
located in the proximity of the -35 region; in the ada gene, the protein binds immediately upstream, protecting the
residues from -40 to -60, while in alkA the
binding site overlaps the -35 region, spanning from -32 to
-52(4, 21) . The location of the Ada-binding
site of aidB is the same as that of ada; Ada protects from DNaseI digestion an area extending from
-38 to -62.Ada might
be AATnnnnnnGCAA. However, it is likely that elements other than the
simple nucleotide sequence play a role in the recognition of the
binding site by Ada. The A/T-rich sequence is likely to
determine local DNA structural modification such as
bending(23) . Another common feature is the location of the
Ada-binding site, immediately upstream of the -35 box. This
position is important for protein-protein interaction between Ada and RNA polymerase, interaction known to be necessary
for Ada-mediated transcriptional activation(5) . In the aidB and ada genes from E. coli, the AGCAA (aidB) and the AGCGCAA (ada) sequences are at the
center of an imperfect palindrome ( Fig. 1and (4) );
although this structure might be involved in the regulation of the two
genes, it does not appear to be a general feature for Ada-binding
sites.Ada for its binding site in ada is higher than for aidB, as determined using gel
retardation assays. Fig. 4shows that a 100% bandshift is
observed at a concentration of 0.3 pmol of Ada for the ada promoter, while the same effect is obtained at
concentrations of Ada at least 10-fold higher for the aidB regulatory region. This suggests that additional
determinants, such as local modification of DNA structure, may be
necessary for optimal interaction between Ada and its
binding site. This view is supported by the fact that, in the ada promoter, mutations affecting local DNA bending have been shown to
abolish transcriptional activation by Ada (12) ,
although they fall outside any consensus sequence so far proposed for
the Ada-binding site.Ada for aidB is likely to play a role in the lower transcriptional
levels displayed by aidB both in vitro and in
vivo. In the in vitro experiments, the levels of induced
transcription for aidB are much lower than for the ada promoter (Fig. 5). In experiments performed in vivo using aidB::lacZ chromosomal fusions, aidB transcription requires higher concentrations of methylating agents
for induction compared with the ada or the alkA promoters(24) . aidB induction, however, is
neither indirect nor accidental, since aidB has been shown to
be effective in counteracting MNNG-mediated mutagenesis(9) . It
is likely that induction of adaptive response genes may take place at
different levels of methylation damage, and this ``fine
tuning'' may be determined by different affinities of Ada for different Ada-binding sites. According to this
hypothesis, aidB might be regarded as an ``emergency
gene,'' activated when ada and alkA induction
are no longer enough to counteract high levels of methylation damage.
)Ada, methylated Ada.)
We thank Graham Walker, Bradford Saget, and Robert
Pohlman for the gift of purified Ada protein and for useful discussion
and Laurel Hajec for critical reading of 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:
|
|
 |
|
  M. S. Rohankhedkar, S. B. Mulrooney, W. J. Wedemeyer, and R. P. Hausinger The AidB Component of the Escherichia coli Adaptive Response to Alkylating Agents Is a Flavin-Containing, DNA-Binding Protein J. Bacteriol., January 1, 2006; 188(1): 223 - 230. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Lacour, A. Kolb, and P. Landini Nucleotides from -16 to -12 Determine Specific Promoter Recognition by Bacterial {sigma}S-RNA Polymerase J. Biol. Chem., September 26, 2003; 278(39): 37160 - 37168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Bhende and S. M. Egan Genetic Evidence that Transcription Activation by RhaS Involves Specific Amino Acid Contacts with Sigma 70 J. Bacteriol., September 1, 2000; 182(17): 4959 - 4969. [Abstract] [Full Text]
|
|
|
|
|
|
P. Landini and S. J. W. Busby Expression of the Escherichia coli Ada Regulon in Stationary Phase: Evidence for rpoS-Dependent Negative Regulation of alkA Transcription J. Bacteriol., November 1, 1999; 181(21): 6836 - 6839. [Abstract] [Full Text]
|
|
|
|
|
|
P. Landini and S. J. W. Busby The Escherichia coli Ada Protein Can Interact with Two Distinct Determinants in the sigma 70 Subunit of RNA Polymerase According to Promoter Architecture: Identification of the Target of Ada Activation at the alkA Promoter J. Bacteriol., March 1, 1999; 181(5): 1524 - 1529. [Abstract] [Full Text]
|
|
|
|
|
|
P. Landini, J. A. Bown, M. R. Volkert, and S. J. W. Busby Ada Protein-RNA Polymerase sigma Subunit Interaction and alpha Subunit-Promoter DNA Interaction Are Necessary at Different Steps in Transcription Initiation at the Escherichia coli ada and aidB Promoters J. Biol. Chem., May 22, 1998; 273(21): 13307 - 13312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Landini, T. Gaal, W. Ross, and M. R. Volkert The RNA Polymerase alpha Subunit Carboxyl-terminal Domain Is Required for Both Basal and Activated Transcription from the alkA Promoter J. Biol. Chem., June 20, 1997; 272(25): 15914 - 15919. [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 |