Previous Article | Next Article 
Journal of Bacteriology, December 1998, p. 6719-6728, Vol. 180, No. 24
0021-9193/98/$04.00+0
Copyright © 1998, American Society for Microbiology. All rights reserved.
Coupled Changes in Translation and Transcription during
Cobalamin-Dependent Regulation of btuB Expression in
Escherichia coli
Xiangwu
Nou and
Robert J.
Kadner*
Department of Microbiology, University of
Virginia School of Medicine, Charlottesville, Virginia 22908
Received 20 August 1998/Accepted 2 October 1998
 |
ABSTRACT |
The level of the vitamin B12 transport protein BtuB in
the outer membrane of Escherichia coli is strongly reduced
by growth in the presence of cobalamins. Previous analyses of
regulatory mutants and of btuB-lacZ fusions indicated that
the primary site of btuB gene regulation was at the
translational level, and this required sequences throughout the
240-nucleotide (nt) leader region. Cobalamin-dependent regulation of
transcriptional fusions was of a lesser magnitude but required, in
addition to the leader, sequences within the first 100 nt of the coding
sequence, termed the translated regulatory region (TRR). To analyze the
process of transcription-level regulation of btuB in
E. coli, the levels and metabolism of btuB RNA
were analyzed by S1 nuclease protection assays, and mutations that
alter the coupling of translational and transcriptional control were
analyzed. Expression of transcriptional fusions was found to correlate
with changes in the level of intact btuB RNA and was
related to changes in the metabolic stability of the normally
long-lived RNA. Mutational analysis showed that the btuB
start codon and a hairpin structure that can sequester the
Shine-Dalgarno sequence are necessary for cobalamin-dependent regulation and that translation of the TRR is necessary for extended RNA stability and for expression of the transcriptional fusion. The
absence of regulation at the stage of transcription initiation was
confirmed by the findings that several truncated btuB RNA fragments were expressed in a constitutive manner and
that the normal regulatory response occurred even when the
btuB promoter and upstream sequences were replaced by the
heterologous bla and lac promoters.
Transcription driven by phage T7 RNA polymerase was not regulated by
cobalamins, although some regulation at the translational level was
retained. Cobalamin-dependent changes in RNA structure were suggested
from the RNase III-dependent production of a transcript fragment that
is made only in the presence of cobalamin and is independent of the
regulatory outcome. These results indicate that the primary control of
btuB expression by cobalamin occurs at the level of
translation initiation, which directly affects the level and stability
of btuB RNA in a process that requires the presence of the
intact translated regulatory region.
| |
INTRODUCTION |
Expression of the btuB
genes of Escherichia coli and Salmonella enterica
serovar Typhimurium, encoding the outer membrane cobalamin (Cbl)
transporter BtuB, and of the cob operon of S. typhimurium, encoding the biosynthetic pathway for Cbls, is
markedly depressed during growth in the presence of Cbls, such as
vitamin B12 (CN-Cbl) (6, 9, 25). The properties
of this Cbl-dependent regulatory system deduced from the expression of
btuB-lacZ fusions suggest the operation of a novel mode of
gene regulation. The regulatory system responds to the intracellular
level of 5'-deoxyadenosyl-Cbl (Ado-Cbl), which is formed from CN-Cbl by
the action of the btuR or cobA gene product
(13, 20). Unlike control systems that depend on repressor
action, Cbl-mediated regulation is independent of gene copy number and
affects expression of btuB-lacZ fusions at both the
transcriptional and translational levels. The modulation of
translational fusions by Cbl is at least 25-fold, while that of
transcriptional fusions is modest: about 5-fold in E. coli (14) and even less in Salmonella typhimurium
(17). The transcripts of the btuB gene or the
cbiA gene, which is the first gene of the large
cob operon, contain a 241- or 468-nucleotide (nt)
leader before the start of the respective coding region.
Regulatory mutations selected for increased expression under
repressive conditions occur at numerous sites throughout these leader
regions (14, 18, 19). No candidate regulatory proteins have
been identified, since all unlinked mutations yet obtained that confer
altered regulation affect either entry of CN-Cbl into the cell or its conversion to Ado-Cbl (6a, 20).
Analysis of transcriptional and translational fusions carrying various
portions of the respective regulatory regions has suggested the
involvement of separate elements in different stages of Cbl-dependent control. The primary site of regulation appears to be at translation initiation and requires the integrity of most or all of the leader region. Cbl-dependent control of transcriptional lac fusions
requires the integrity of both the leader region and the initial part
of the btuB coding region out to around residue +350
(7, 14, 17). This translated regulatory region (TRR),
between nt 241 and 350, includes a putative Rho-independent terminator,
a G+C-rich stem and loop followed by a series of U residues, whose
integrity is essential for transcriptional control (7).
Unlike most attenuators, candidate terminators for btuB and
cbiA lie past the site of translation initiation, in the
early portion of the protein-coding sequences (11). Control
of translational fusions occurs in the absence of the translated
btuB sequences and appears to be moderated by formation of
an RNA hairpin that includes or is near the Shine-Dalgarno sequence
(17, 19). Expression of btuB requires the
presence of the B12 box, a 17-nt element that is conserved in the
leaders of the btuB and cbiA genes
(18). Precise deletion of the B12 box sequences resulted in
a very low and unregulated level of btuB expression
(7).
Although the extent of regulation by Cbl of btuB-lacZ
transcriptional fusions in S. typhimurium is slight
(17), an appreciable level of control of btuB in
E. coli and of cob in S. typhimurium occurs (14, 18). This control depends on
the presence of each respective TRR, which has been suggested to
function as an attenuator, although no evidence for this proposal was
obtained. To investigate the complex regulation seen with gene fusions,
we analyzed the changes in btuB RNA levels detected by S1
nuclease protection assays under a variety of conditions that affect
the frequency of btuB translation or transcription
initiation or RNA stability. Prompted by the finding that alteration of
translation dramatically affected expression of transcriptional fusions
and RNA levels and stability, we concluded that the TRR is specifically
responsible for changes in btuB RNA levels, rather than that
this apparent regulation is a nonspecific reflection of the lability of
untranslated RNA. We propose that the negative action of the TRR is
prevented by the passage of ribosomes through it, which is modulated by the primary site of regulation at the stage of ribosome binding.
The btuB regulatory region, with the location of the DNA
hybridization probes and the sites of reporter fusions used in this study, is schematically shown in Fig. 1.
View larger version (18K):
[in this window]
[in a new window]
|
FIG. 1.
Schematic representation of btuB regulatory
region. The DNA regions carried in the plasmids used in this study are
shown under the coordinate numbers relative to the start of
transcription. The locations of the BamHI sites used as
junctions of fusion to lacZ in the three fusions are
indicated. The locations of the B12 box, conserved in Cbl-dependent
promoters, the potential stem-loop structures in the transcript that
comprise the Shine-Dalgarno sequence and its complement, and the
attenuator are marked. The vertical arrows indicate the locations of
the 3' termini of the partial-length transcript fragments. On the
bottom are indicated the extent of the three probes used in this study,
aligned with the 5' end on the right; the position of the label is
indicated by the symbols.
|
|
| |
MATERIALS AND METHODS |
Bacterial strains and plasmids.
The bacterial strains,
plasmids, and oligonucleotide primers used in this study are listed in
Tables 1 and
2. In most experiments, the host strain
was RK5173, a metE derivative of MC4100. Strain JM109 was
used as the host of plasmids pXN26 and pXN27, which were very unstable
in RK5173. For plasmids carrying btuB-lacZ fusions
transcribed from the T7 late promoter, the host strain was RK3515, a
btuB+ transductant of BL21(DE3), in which
production of T7 RNA polymerase is induced by addition of
isopropyl-
-D-thiogalactopyranoside (IPTG). Cells were
routinely grown in Luria-Bertani medium. Ampicillin (100 µg/ml) was
added to maintain selection for plasmids. When indicated, Ado-Cbl was
added to a concentration of 1 µM to establish repressive conditions.
Unless indicated, cells were grown at 37°C with vigorous aeration.
All plasmids used in this study are derived from pRS414 or pRS415,
which allow formation of translation or transcription fusions
to
lacZ, respectively (
22). Inserts were introduced
as
EcoRI-
BamHI
fragments. The insert's
EcoRI site occurs at
60 in the wild-type
btuB
sequence, and the
BamHI sites were introduced as linker
insertions
or substitutions, as previously described (
7).
The
BamHI site
at each fusion junction occurs in the same
translational reading
frame. In the translational fusions, the
btuB coding sequences
are coupled to the 9th codon of
lacZ (
22). In the transcriptional
fusions, the
lacZ gene is separated from the
btuB junction by
about 100 bp of
trp-lac DNA and possesses its own
ribosome-binding
site. Construction of the plasmids carrying
btuB-lacZ fusions
was performed as described previously
(
7).
Construction of btuB mutants and fusions.
All
nucleotide coordinates are numbered relative to the btuB
transcription start site. Plasmid DNAs were purified and introduced into host cells by standard protocols (21). All restriction endonucleases, DNA polymerases, and DNA ligase were used according to
the recommendations of the manufacturer.
Mutations were introduced into the
btuB leader region by a
one-step or two-step PCR process. The first step used mutagenic
oligonucleotides (XNO-26, XNO-27, or XNO-47) and oligonucleotide
XNO-4,
which anneals downstream of the
BamHI site in plasmid
pRS415,
as primers and plasmid pEB450X DNA as a template. This product
was purified by gel electrophoresis and used as an adapter in
the
second PCR step. In this step, oligonucleotide XNO-15, which
anneals to
plasmid pAG1 DNA at vector sequences upstream of the
EcoRI
site, and oligonucleotide XNO-4 were used as primers, and
plasmid pAG1
was used as a template, in the presence of the product
of the first
PCR. Thus, the product of the first PCR step, which
contains the
desired mutation, was further extended and amplified,
resulting in a
DNA fragment containing
EcoRI and
BamHI sites and
the desired mutation. The
EcoRI-
BamHI fragment
was generated,
purified, and ligated into the corresponding sites of
pRS414 and
pRS415, and the presence of the expected mutation was
verified
by sequencing. Mutants M43, M44, M45, and M57 were constructed
by one-step PCR mutagenesis and cloned into pEB450 plasmids as
EcoRI-
HindIII or
HindIII-
BamHI
fragments.
The two-step PCR scheme was also used to construct
btuB-lacZ
fusions under control of heterologous promoters. Oligonucleotides
XNO-51 for
lac and XNO-52 for
bla are composed of
the 14 nt upstream
of the site of transcription initiation of the
appropriate promoter
followed by the first 14 nt of
btuB
transcribed sequences. These
primers were used with oligonucleotide
XNO-4 as the primer and
plasmid pEB450X DNA as the template in the
first-step PCR. In
the second step, oligonucleotides XNO-49 for
lac and XNO-50 for
bla, which anneal upstream of
the appropriate promoter region
and contain an
EcoRI site,
were used along with oligonucleotide
XNO-4 as primers and pGEM-3Z DNA
(Promega) as a template, in the
presence of the first PCR product as an
adapter. The resulting
EcoRI-
BamHI fragment was
subcloned as described above. To construct
the
btuB-lacZ
fusions under control of the T7 promoter, oligonucleotide
T7-btu, which
contains an
EcoRI site and 16 nt of the T7 consensus
sequence fused to the first 22 nt of
btuB-transcribed
sequences,
was used in PCR with XNO-4 as the primer and pEB450X DNA as
the
template, and the resulting
EcoRI-
BamHI
fragment was subcloned
as described
above.
RNA extraction and S1 nuclease protection analysis.
Total
cell RNA was prepared by the hot phenol extraction method described by
Emory and Belasco (5), as modified. About 109
logarithmic-phase cells were collected by centrifugation for 20 s
at 4°C, and the cell pellet was immediately frozen until all samples
for the experiment were collected. The cells were suspended in 450 µl
of lysis buffer (150 mM sucrose, 10 mM sodium acetate [pH 4.5]) on
ice and mixed with 50 µl of 10% sodium dodecyl sulfate (SDS).
Samples were heated at 70°C for 3 min and then extracted three times
with phenol at 70°C. Nucleic acids were precipitated by addition of
1/10 volume of sodium acetate and 2 volumes of ethanol. The sample was
air dried and suspended in 100 µl of DNase buffer (20 mM sodium
acetate [pH 4.5], 20 mM NaCl, 10 mM MgCl2) and treated
with 1 to 2 U of DNase I for 30 min at 25°C. This digestion was
followed by phenol extraction and ethanol precipitation, and the
product was suspended in 15 µl of water and stored at 20°C.
Probe
A was synthesized by PCR with primers XNO-15 and
XNO-16 and with plasmid pAG1 DNA as the template. This product extends
from plasmid vector sequences upstream of the site of insertion
of the
btuB-lacZ fusion to position +470 of the wild-type
btuB sequence. Probe
A was internally labeled by
asymmetric PCR with
primer XNO-16, in a labeling reaction mixture
containing 3 µM
[
-
32P]dCTP, 25 µM unlabeled dCTP,
and 125 µM other deoxynucleoside
triphosphates (dNTPs).
Probe
B is a restriction fragment generated
by digestion of
plasmid pXN5 with
EagI and
PvuII. In plasmid
pXN5,
an
EagI restriction site (C'GGCCG) replaced
the sequence at the
start of
btuB transcription
(TTGCCG) by PCR mutagenesis. A fill-in
reaction catalyzed by
the Klenow fragment of DNA polymerase I
with
[
-
32P]dCTP and dGTP resulted in labeling of the 3' end
of the DNA
strand complementary to
btuB RNA, and the
position of the labeled
nucleotide hybridized to the transcript
position at +1. Probe
C was synthesized by PCR with primers XNO-15 and
5'-labeled XNO-2
with plasmid pAG1 DNA as template. This product
extends from upstream
vector sequences to position +269. The PCR
primers were labeled
at their 5' ends with [
-
32P]ATP
and polynucleotide kinase. The
bla-specific fragment was
synthesized and labeled by PCR with XNO-14 and 5'-labeled XNO-9
as
primers and plasmid pAG1 DNA as a
template.
For the S1 nuclease protection assay, modified from that described by
Emory and Belasco (
5), 1 µl of extracted RNA was
mixed
with 16 µl of formamide and 2 µl of hybridization buffer
(400 mM
PIPES [pH 6.4], 400 mM NaCl, 10 mM EDTA) along with the
indicated
32P-labeled DNA probes in a volume of 1 µl. This mixture
was heated
to 80°C for 4 min and then held at 50°C for 4 to 16 h. To the
sample was added 300 µl of S1 buffer (280 mM NaCl, 50 mM
sodium
acetate [pH 4.5], 4.5 mM ZnSO
4) containing 5 µg
of denatured herring
sperm DNA and 40 U of S1 nuclease. Following
incubation for 90
min at 37°C, the reaction was stopped by addition
of 80 µl of
stop buffer (4 M ammonium acetate, 20 mM EDTA), and the
DNA was
precipitated with 5 µg of yeast tRNA and 1 ml of ethanol. The
precipitate was suspended in sequencing gel loading buffer, resolved
on
a 6% polyacrylamide sequencing gel, and analyzed with a Molecular
Dynamics
PhosphorImager.
-Galactosidase assays.
For assays of -galactosidase
expression, cells were grown in minimal salts medium A with 0.5%
glycerol and 0.5% casein hydrolysate. The level of -galactosidase
was determined as previously described (7), by measuring the
rate of hydrolysis of 2 mM
o-nitrophenyl--D-galactopyranoside in cells
permeabilized with SDS-CHCl3. All assays were performed in
triplicate, and most were repeated.
| |
RESULTS |
Hairpin structure at the ribosome-binding site affects
regulation.
Previous work concluded that translational control
requires the presence of sequences up to the start of translation,
whereas transcriptional control requires, in addition, 60 to 100 nt of the TRR. We examined the involvement in btuB
expression of a potential hairpin structure formed by the
Shine-Dalgarno sequence and a complementary sequence 12 bp
upstream (Fig. 2). Mutations were introduced that alter each stem of this hairpin. In mutant M26, the
complementary sequence was changed from GCATCC
to CGATCC (an additional G-to-T
substitution at +208 created an NruI site to facilitate
mutant identification). Mutant M27 changed the
Shine-Dalgarno sequence GGATGCT to
GGATCGA, which has little effect on the
match to the consensus. The double mutant M26/27 combined both sets of changes and restored the capacity for pairing of the Shine-Dalgarno sequence and its upstream complement. The effect of these changes was
measured in EB450 transcriptional and translational
btuB-lacZ fusions carrying the btuB promoter
followed by 450 nt of transcribed sequences before fusion to the
lacZ reporters (Fig. 1).
View larger version (23K):
[in this window]
[in a new window]
|
FIG. 2.
Hairpin structure comprising the Shine-Dalgarno sequence
and its upstream complement and the mutations that affect either the
stem or the initiation codon. The translation initiation codon is
indicated in the oval. The base changes introduced by the mutations are
indicated. The units of -galactosidase activities of transcriptional
and translational EB450 fusions are presented. WT, wild type.
|
|
Translational fusions expressing the M26 or M27 substitutions showed a
slight increase in expression but almost complete loss
of Cbl-dependent
regulation (Fig.
2). The 1.4- to 1.7-fold increase
in expression
may reflect increased access of the ribosome-binding
site because of
its decreased sequestration by the upstream complement.
The combination
of both substitutions in the compensatory M26/27
mutant resulted in
restoration of the wild-type degree of regulation,
indicating that the
pairing of these two sequences must occur
as part of the process of
translational control, as had been previously
proposed from similar
studies with the
S. typhimurium btuB and
cbiA genes (
17,
19).
Strikingly, the M26 and M27 substitutions resulted in almost complete
loss of Cbl-dependent regulation of the transcriptional
fusion,
although they had no effect on the level of expression
(Fig.
2). The
compensatory M26/27 combination restored wild-type
regulation. Using
the S1 nuclease protection assay described below,
it was seen that the
relative levels of
btuB RNA paralleled the
changes in
reporter activity. Thus, the ability to form the hairpin
that includes
the ribosome-binding site is crucial for both translational
and
transcriptional regulation. This apparent coordination of
translational
and transcriptional activity could reflect the participation
of this
hairpin in formation of an RNA conformation that leads
to RNA
attenuation or cleavage, to the increased lability of untranslated
RNA,
or to a specific effect of translation on the activity of
the
TRR.
The extent of translated sequences affects transcription.
To
test whether btuB translation affects transcription, the AUG
start codon was converted to UCG in the M43 substitution, and assayed
for effect on expression in EB450 transcriptional and translational
fusions. As expected, this substitution resulted in complete loss
of expression of the translational fusion. Surprisingly, there was a
drastic reduction in expression of the transcriptional fusion to about
2% of the wild-type level (Fig. 3). The
amount of full-length btuB RNA was also greatly reduced but
was decreased further by Ado-Cbl (data not shown).
View larger version (19K):
[in this window]
[in a new window]
|
FIG. 3.
Effect of mutations that affect btuB
translation on regulation of btuB-lacZ fusions. The initial
portion of the btuB coding region is shown at the top, along
with nucleotide coordinates. The paired arrows show the location of the
putative transcription terminator. Symbols indicate the consequences of
the mutations, as follows: X, translation termination or blockage;
vertical bar, frameshift mutation; solid line, wild-type BtuB amino
acid sequence; dashed line, frameshifted amino acid sequence. The
expression of units of -galactosidase activity from EB450
transcriptional and translational fusions in cells grown in the absence
or presence of Ado-Cbl is presented for each mutant.
|
|
To test whether the decrease in transcription required the presence of
the TRR, the M43 mutation was placed in the context
of the EB270
transcriptional fusion. This fusion contains the
btuB
promoter and transcribed region to +270. It is deleted for
most of the
TRR and exhibits high-level constitutive expression,
but almost fully
repressible behavior as a translational fusion
(repression ratio of 13)
(
7). Expression from this EB270-M43
fusion was comparable to
that of the wild-type EB270 fusion and
was increased by the presence of
Ado-Cbl (1,700 U without Ado-Cbl
and 2,900 U with Ado-Cbl). These
results showed that the decreased
transcription activity that
occurred in the EB450 fusion upon
blockage of translation depends
on the integrity of the translated
regulatory region, suggesting that
translation of the TRR might
overcome its negative effect on
transcription.
The ability of translating or stalled ribosomes to affect attenuator
function (reviewed in reference
11) or
endonucleolytic
digestion of RNA (
3,
16) is well known. To
test whether the
extent of translation or the presence of specific
polypeptide
sequences in the translated product affected
transcriptional activity,
several translation termination mutations
were introduced early
in the
btuB coding sequence and
analyzed in EB450 transcriptional
and translational fusions (Fig.
3).
The potential attenuator element
in the
btuB TRR extends
from codon 7 to codon 17 (i.e., nt +260
to +290). Mutation M44
introduced a chain-terminating UAG sequence
at codon 6. It eliminated
expression of the translational fusion
and strongly reduced both
expression and regulation of the transcriptional
fusion. Thus, ribosome
attachment and initiation of translation
are not sufficient to
alleviate the inhibition of
transcription.
Mutation M45 introduced a +1 frameshift at codon 3, resulting in a
change in the distal reading frame until termination occurs
at codon
33. This mutant showed reduced transcriptional activity,
but a
substantial degree of Cbl-mediated regulation (repression
ratio of
around 3) (Fig.
3). Mutation M47 introduced a
1 frameshift
at codon
30, resulting in a change in the amino acid sequence
until termination
occurs at codon 49. This mutant showed almost
wild-type levels of
transcriptional expression and regulation,
even though the
translational fusion was negative. Thus, loss
of ribosome movement past
codon 49 of
btuB does not affect transcriptional
activity.
The M45/47 combination allows translation of the entire
btuB
gene, although the amino acid sequence from codon 3 to codon
30 is
changed by the frameshift. This combination gave wild-type
levels of
expression and regulation of the transcriptional fusion
and nearly
wild-type expression of the translational fusion, showing
that the
sequence of the translated product is not a factor in
determining the
level and regulation of transcription. Similar
results were obtained
with a
1 frameshift at codon 7, which terminates
translation at codon
8. Its low and unregulated transcriptional
expression was fully
restored by combination with M45 (+1 shift
at codon 3). These results
show that transcriptional activity
is controlled by translation through
the TRR between positions
240 and 330, rather than through formation of
transcript secondary
structures or the general lability of untranslated
RNA.
Differential regulation of btuB RNA fragments.
The
transcripts produced from Cbl-regulated genes have not been previously
described. S1 nuclease protection was used to detect changes in
btuB RNA levels, length, or stability. Three hybridization probes were used to identify the origin of RNA transcribed from three btuB-lacZ fusions, designated EB450, EB315, and EB270
(Fig. 1). These fusions contain the btuB promoter
region and 450, 315, or 270 nt, respectively, of
btuB-transcribed sequence joined at an introduced
BamHI site as a transcriptional fusion to
trp-lacZYA in the plasmid vector pRS415 (7).
These inserts exhibit comparable degrees of Cbl-dependent regulation
when present as translational fusions, with repression ratios of 19, 21, and 13, respectively. In transcriptional fusions, they exhibit
repressible, partially repressible, or constitutive behavior,
respectively, attributed to the progressive deletion of portions of the
TRR in the two shorter inserts. Examination of these nested fusions
with different regulatory behaviors allowed the simplest test of
whether changes in the patterns of transcripts are associated with
regulation. Expression from plasmid-borne btuB-lacZ fusions
was used to increase the amount of btuB RNA. Elevated gene
copy number does not appear to affect btuB regulation by
Cbls (1, 25). The transcript of the chromosomal
btuB allele was detectable and showed changes similar to
those of the plasmid-borne EB450 fusion. RNA was extracted from cells
grown in the absence and presence of 1 µM Ado-Cbl, hybridized to one
of the three btuB probes and a bla DNA probe, subjected to S1 nuclease digestion, resolved by electrophoresis, and
detected by autoradiography or phosphorimaging. The bla
transcript served as a control for RNA loading.
Hybridization to probe
A, extending from upstream vector
sequences to +470 of
btuB sequence and internally labeled by
asymmetric
PCR synthesis, allowed detection of all
btuB-derived RNA species.
Cells carrying the vector plasmid,
pRS415, yielded the 470-nt
product derived from the chromosomal
btuB allele; its levels were
strongly depressed by growth
with Ado-Cbl (Fig.
4, lanes 1 and
2). Each of the
btuB-lacZ fusion templates gave rise
to multiple
RNA species with different regulatory responses (lanes 3 to
8).
The major transcripts were the expected full-length species of
450, 315, and 270 nt, respectively. The amount of the 450-nt species
expressed from EB450 was strongly reduced during growth with Ado-Cbl,
whereas the 315-nt species from EB315 was less strongly decreased,
and
the 270-nt species from EB270 showed only a slight change.
Quantification of band intensities showed that the repression
ratios
for the full-length species (amount in absence of Ado-Cbl/amount
in
presence of Ado-Cbl; normalized to the amount of the 185-nt
bla transcript fragment) from EB450, EB315, and EB270 were
4.6,
1.7, and 1.3, respectively. These values agreed with the
repression
ratios for
-galactosidase expressed from these
transcriptional
fusions: 4.9, 2.0, and 1.1, respectively
(
7). This correspondence
indicates that the expression from
btuB-lacZ transcriptional fusions
is a valid measure of
btuB RNA levels.
View larger version (36K):
[in this window]
[in a new window]
|
FIG. 4.
S1 nuclease protection analysis of transcripts expressed
from three btuB-lacZ fusions. RNA was extracted from cells
of strain RK5173 carrying the plasmids with the indicated
btuB-lacZ fusions. Cells were grown as
indicated in the absence or presence of 1 µM Ado-Cbl and extracted
RNA was hybridized with 32P-labeled probe A. In all assays,
a constant amount of labeled probe complementary to the bla
gene transcript was included. The sizes of the protected DNA fragments
are indicated.
|
|
The other major
btuB RNA species were a group of fragments
ca. 165 nt long. High-resolution electrophoresis of these fragments
next to a sequencing ladder showed that these fragments were 162
to 167 nt in length (data not shown). The amount of these fragments
did not
change significantly under repressive conditions, indicating
that the
btuB promoter is expressed in a constitutive manner.
Several
RNA species were present in smaller amounts. Products
migrating between
280 and 300 nt and at 215 nt were constitutively
produced. Fragments of
240 to 250 nt were depressed by Ado-Cbl,
and a 210-nt product was
present only in repressed cells. The
formation of the 210-nt fragment
in the presence of Ado-Cbl occurred
with all three
btuB-lacZ
fusions, indicating that its production
is not directly correlated with
transcriptional activity. Products
smaller than 150 nt were not
detected.
The origin of the sequences represented by these fragments was
determined by hybridization to two other probes (Fig.
1). Probe
B was labeled on its 3' end at the residue complementary to
+1
of the
btuB transcript, and thus its label is protected
only by
btuB transcripts that contain the authentic 5' end.
All RNA species
detected by probe
B were identical to those
detected with the
internally labeled probe
A (data not
shown). Probe
C was labeled
on its 5' end at the position
complementary to nt +269 and can
detect
btuB RNA species
that extend beyond +269, including any
that might be generated by
endonucleolytic cleavage. The only
significant protected
btuB species was the 269-nt fragment representing
the
full-length product of the three templates (data not shown).
These
results showed that all of the RNA fragments detected with
probe
A are sense-strand
btuB products and start at
nucleotide
+1. The truncated fragments could result from transcription
termination
or endonucleolytic cleavage, but if formed by nuclease
action,
the distal fragment must be much more labile than the
promoter-proximal
fragment.
Kinetics of repression of btuB RNA levels.
The S1
nuclease protection assay was used to determine the rate of change in
the levels of the btuB transcripts following addition of 1 µM Ado-Cbl. The ratio of the full-length btuB transcript to the bla transcript was calculated by ImageQuant analysis
(Fig. 5). The level of intact
btuB RNA from the repressible EB450 fusion began to decline
within minutes to a steady-state level of 22% of the derepressed
level. The lag probably reflects the time for accumulation of Ado-Cbl
to effective levels in the cell. The 470-nt RNA from the chromosomal
allele also showed a similar decline. It took about 20 min before the
new steady-state level of btuB RNA was achieved, which was
longer than expected if this RNA had a lifetime typical of mRNA in
E. coli.
View larger version (24K):
[in this window]
[in a new window]
|
FIG. 5.
Time course of decrease of btuB RNA levels
following addition of Ado-Cbl. RNA was extracted from cells of strain
RK5173 carrying the indicated fusion plasmids. Cells were grown in
minimal medium A to mid-log phase. Ado-Cbl was added to 1 µM, and
samples were removed at the indicated times for the S1 protection assay
by using internally labeled probe A. The amount of
full-length btuB RNA, normalized to the intensity of the
bla transcript, is plotted as a function of the time after
addition of Ado-Cbl.
|
|
The full-length
btuB RNA level in the partially repressible
EB315 strain fell to 60% of the initial level with kinetics similar
to
those in the EB450 strain. The RNA level in the poorly repressed
EB270
strain fell slightly, to 80% of the initial level. In all
three
strains, the levels of the unregulated 290- and 165-nt species
were not
substantially affected, and the 210-nt species increased
under
repressive conditions, but only after a lag of 5 to 10 min,
i.e., with
slower kinetics than the repression of the full-length
species.
Decreased stability of btuB RNA during repression.
The metabolic stability of btuB RNA was estimated from the
rate of its decay following the addition of rifampin to cells growing in the absence (Fig. 6A) or upon
simultaneous addition of Ado-Cbl (Fig. 6B). The amounts of the
btuB RNA species were determined by S1 nuclease protection
assays and quantified with the ImageQuant program (Fig. 6C). The
full-length btuB RNA expressed from all three
btuB-lacZ fusions in the absence of Ado-Cbl was long-lived, with half-lives (t1/2s) in the range of 9.4 to
13.4 min. All of the partial-length btuB transcripts were
also long-lived, suggesting that their stability is determined within
the first 160 nt. The bla RNA exhibited the short
t1/2 of 2 to 4 min, typical of most bacterial
mRNAs (8).
View larger version (78K):
[in this window]
[in a new window]
|
FIG. 6.
Metabolic stability of btuB RNA in the
presence or absence of Cbl following the addition of rifampin. RNA was
extracted from cells of RK5173 carrying plasmid pRS415 with the
indicated btuB::lacZ fusion-encoding
inserts at the indicated times after the addition of rifampin and was
subjected to S1 nuclease protection of internally labeled probe
A. (A) Cells in minimal medium. (B) Cells that received
Ado-Cbl at the same time as rifampin. (C) Amount of full-length
btuB transcript, as a fraction of the initial value, derived
from the three templates, for cells in the absence (open symbols) and
presence (solid symbols) of Ado-Cbl.
|
|
In the presence of Ado-Cbl, the stability of the full-length products
of EB450 and EB315 and of the chromosome-encoded
btuB transcript was markedly reduced, with
t1/2s in
the range of 2
to 4 min (Fig.
6B and C). The
t1/2 for the EB270 transcript decreased
slightly
to around 8.3 min. The stability of the constitutively
expressed 165-nt
fragments was not appreciably affected by Ado-Cbl.
Thus, the decreased
amount of intact
btuB RNA under repressive
conditions
is associated with its decreased stability. This is
a seemingly
paradoxical situation, since addition of Ado-Cbl results
in increased
RNA lability in the presence of rifampin but a slow
decrease in RNA
levels when transcription occurs. This behavior
can be explained
by the constitutive synthesis of a long-lived
RNA whose level and
stability is affected by Cbl only after it
has reached a length
beyond nt
300.
Effect of RNases on regulation and stability of btuB
RNA.
The formation of the several truncated species of
btuB RNA and the increase in its lability in the presence of
Ado-Cbl could result from transcription attenuation or endonucleolytic
cleavage. The effect of mutations that block expression of various
RNases involved in RNA processing on the distribution and regulation of
the transcripts from the EB450 fusion was analyzed by S1 nuclease protection assay. Two rnc mutants defective in RNase III
activity displayed a marked decrease in the 165-nt fragments and an
increase in the Cbl-induced 210-nt fragment (Fig.
7), but they had no significant effect on
the regulation of the full-length transcript or on lacZ expression. These results suggest that an RNA secondary structure is
formed that is cleaved by RNase III at position 165 immediately downstream of the B12 box. Since RNase E is essential for viability, its involvement was tested following thermal inactivation of alleles that confer a temperature-sensitive growth phenotype. We found there
was no substantial effect on the distribution of the RNA species or on
the regulation of full-length RNA levels or lacZ expression
as a result of individual inactivation of RNase E, II, D, PH, BN, or D
or of polynucleotide phosphorylase (data not shown). Many of these
RNases have redundant activities, and defects in processing are
observed only when most of them have been inactivated (10,
12).
View larger version (86K):
[in this window]
[in a new window]
|
FIG. 7.
Effect of mutations affecting RNase III and Hfq-1
activities on the pattern of btuB-specific transcripts. RNA
was extracted from the indicated parental and mutant cells carrying the
EB450 fusion and grown in the absence or presence of Ado-Cbl. The RNA
was used for S1 nuclease protection of internally labeled probe
A.
|
|
The
hfq gene encodes an RNA-binding protein, host factor-1,
that can disrupt RNA secondary structure and allow increased
translation
or replication of
rpoS mRNA or phage Q
RNA,
respectively (
15,
24). An
hfq null mutant
carrying the EB450 fusion displayed
normal regulation of
-galactosidase levels and of full-length
btuB RNA (Fig.
7), indicating that Hfq function is not needed
for transcription
elongation or BtuB regulation. However, the
amount of the Cbl-induced
210-nt fragment was strongly reduced,
indicating a possible role for
changes in RNA secondary structure
in the generation of this
fragment.
Regulation of transcription driven by heterologous promoters.
To test whether promoter sequences and the nature of the transcription
complex play any role in regulation, the btuB promoter in
the EB450 transcriptional and translational fusions was replaced with
the heterologous lac, bla, and phage T7 RNA
polymerase-dependent 
10 late promoters. The wild-type transcription
start site was retained in all cases (data not shown). The levels of
-galactosidase differed in response to promoter strengths and the
degree of induction with IPTG (Table 3).
Two different regulatory outcomes were seen. Expression of
transcriptional and translational fusions driven by the lac
and bla promoters was reduced by Ado-Cbl to a degree similar
to that with the btuB promoter. S1 nuclease protection assays revealed no qualitative difference in the distributions or
relative amounts of the btuB transcript fragments (data not shown). Thus, the promoter and upstream sequences play no detectable role in Cbl-dependent regulation.
Different behavior was seen when transcription was driven by T7 RNA
polymerase, which is widely used for gene expression,
because of its
high rate of elongation, single-subunit composition,
and diminished
response to transcriptional pause and termination
signals (
4,
23). Plasmids pXN17 and pXN18 contain the upstream
region of the
T7
10 promoter joined at the +1 position to the
EB450
transcriptional and translational fusions, respectively.
Although the
T7 promoter extends into the transcribed region,
T7 RNA polymerase
started transcription at the same position as
that used by the
wild-type promoter. Expression of
-galactosidase
in strain RK5173
was very low, indicating the dependence on T7
RNA polymerase (Table
3).
Expression in strain RK3513, a
btuB+
transductant of the T7 RNA polymerase-producing strain, BL21(DE3),
was
increased more than 10-fold by induction of T7 RNA polymerase
synthesis
with 0.5 mM IPTG for 1.5 h. The transcriptional fusion
was not
appreciably affected by Ado-Cbl, with repression ratios
of <1.3.
Although the levels of the full-length transcript were
not decreased
following growth with Ado-Cbl, the 210-nt fragment
was still induced
(data not shown). Many of the other
btuB fragments
were
still present following T7 polymerase-driven transcription,
but those
around 280 and 295 nt were absent. Cbl-dependent regulation
of the
translational fusion still occurred, but to a diminished
degree, with
repression ratios of >2.5. These results suggest
that transcriptional
regulation depends on the nature and properties
of the transcription
complex, such as its response to termination
signals, rather than a
feature of the RNA transcript, such as
a secondary structure that is
cleaved by an
RNase.
| |
DISCUSSION |
Previous studies using reporter fusions to the E. coli
btuB gene (7, 14) and to the S. typhimurium
btuB and cbiA genes (17-19) revealed a
similar and unexpected pattern of dependence on the length of the
transcribed sequences for Cbl-dependent regulation at both the
transcriptional and translational levels. In all three Cbl-regulated
genes, translational control appears to be the major site of control
over gene expression. The results presented here agreed with
studies of the other Cbl-regulated genes that the hairpin formed over
or near the ribosome-binding site is a major component in translational
control. This paper provides for the first time information relevant to
the process of transcriptional control and changes in the transcript
during Cbl-dependent regulation. Ravnum and Andersson (17)
concluded that expression of btuBSt was not
subject to a substantial degree of transcriptional control, based on
the behavior of lac fusions. However, the other
Cbl-regulated genes exhibit an appreciable degree of apparent
transcriptional control, and the results presented here indicate that
this is a specific feature of btuBEc expression.
This conclusion is based on our demonstration of changes in
btuB RNA levels. The fact that btuB-lacZ fusions
of different lengths exhibit different regulatory behaviors allowed
analysis of the role of the TRR and correlations of changes in the
transcripts with changes in the regulatory behavior of the fusions.
There was concern that the apparent regulation of transcriptional
fusions might be an artifactual consequence of readthrough into the
lacZ gene from ribosomes that initiated translation in the
btuB sequence. Several lines of evidence eliminated
this concern. First, introduction of a translational stop at codon 49 completely blocked btuB translation, but had no effect on
the level or regulation of the btuB-lacZ transcriptional
fusion or of btuB RNA. Second, the changes in the level of
full-length btuB RNA matched well the behavior of
transcriptional fusions. Third, Cbl-dependent regulation is associated
with changes in the stability of full-length btuB RNA.
Finally, two different btuB-lacZ transcriptional fusions were constructed with an RNase III cleavage site upstream of the lacZ gene to dissect the translation of lacZ from
that of btuB. The repression ratios in the two constructs
were 4.9 and 6.0, which are very similar to those in the pRS415 fusion
vector. Thus, the possibility that transcriptional regulation is a
consequence of translational readthrough from btuB sequences
is untenable.
Cbl-dependent regulation of transcriptional fusions and of
btuB RNA levels could reflect the general or nonspecific
lability of untranslated RNA, such that when btuB
translation is prevented, the distal sequences are subject to
endonucleolytic degradation. We found that transcriptional control of
btuB requires the integrity of the TRR. The EB270
fusion, which showed greatly reduced transcriptional regulation, still
possesses 30 nt of btuB sequence before the fusion and
100 nt of trp-lac sequences before the start of the lacZ gene, which should provide a suitable target for
RNA turnover. In addition, transcriptional fusions to sequences
upstream of the start of translation show the same activity of
-galactosidase as fusions after the start of translation, suggesting
that translation does not provide an inherent increase in stability
(7). However, it is difficult to eliminate the impact of the
lability of untranslated RNA.
If the TRR confers a specific mechanism involved in btuB RNA
turnover under repressive conditions, this region might function either
as a transcriptional attenuator or as a site for endonuclease cleavage.
The existence of an attenuator was suggested by Lundrigan et al.
(14) and Ravnum and Andersson (17), based on the
presence of a potential G+C-rich stem and loop followed by 4 U residues in the transcript. Deletions to position +270 or further upstream, which affect this structure, completely eliminated transcriptional control (7). Further deletions between +285 to +303 resulted in reduced regulation, whereas deletions ending at +345 of distal showed normal regulation. Transcript fragments ending at +280 to
+300 should result from termination at the attenuator, but their
amounts increased only modestly under repressive conditions, indicating
either that attenuation is not operative or that the attenuated
products are labile. Base changes in the putative attenuator element
partially reduced transcriptional regulation (unpublished data). Thus,
we can conclude that the putative attenuator sequence is important for
transcriptional control, but downstream sequences out to between +315
and +345 are also involved to some degree. Note that deletions that
result in complete loss of transcriptional control retain
translation-level regulation (7). We found no evidence for
the action of RNase E or III in regulation, although RNase III cleavage
of the btuB transcript was observed. The involvement of a
Cbl-regulated turnover mechanism in addition to attenuator action is
indicated from the kinetics of RNA turnover. The full-length RNA is
long-lived, but becomes labile in the presence of Ado-Cbl. This result
cannot be explained by the operation of an attenuation event as the
sole regulatory factor affecting btuB RNA, since the
lability is imposed upon completed RNA molecules or even nascent transcripts past the site of attenuation.
There is no control process operative at the level of transcription
initiation. Normal regulation was seen when transcription was
driven by heterologous sigma 70-dependent promoters, as long as the
proper start site was retained. In addition, some btuB transcripts are expressed in a constitutive manner, notably the 165-nt
fragments. Dissociation of transcriptional and translational regulation
occurred with T7 RNA polymerase, which is less responsive to
transcriptional pausing and termination signals than is E. coli RNA polymerase (4). This result suggests that
transcriptional regulation depends on an activity of the transcription
complex, such as polymerase pausing or termination, rather than on the formation of an RNA structure whose cleavage by an RNase triggers RNA degradation. The retention of translational regulation
indicates that this level of control is independent of the properties
of the transcription complex and may be determined by RNA secondary structure, perhaps involving access to the ribosome-binding site.
Specific transcript fragments were produced, whose 3' ends occur
at potentially relevant sites. The 165-nt fragments end at the
downstream side of the B12 box and were greatly decreased in RNase
III-deficient mutants. The 3' ends of the constitutively expressed 235- to 250-nt fragments occur just past the stem-loop structure
formed by the Shine-Dalgarno hairpin, and the 3' end of the Cbl-induced
210-nt fragment occurs just before this hairpin. However, in no case
was there an invariant correlation between the level of any fragment
and the level of btuB expression. The intact transcript and
most of its fragments were long-lived under nonrepressing conditions.
The inability to observe fragments shorter than 165 nt suggests that
the first 165 nt of the transcript, which includes the B12 box, are
needed for the enhanced stability. As discussed above, the presence of
the TRR sequences between +270 and +315 is needed for the decrease in
RNA stability. This decrease in stability appears to be unaffected by
the absence of RNase E or RNase III. Stabilizing elements, such as
double-stranded structures at the 5' ends of RNA, have been documented
(8).
Thus, all of the results obtained here indicate that translating
ribosomes prevent the decreases in btuB RNA levels and
stability that are brought about by the presence of the TRR. This
behavior appears to represent a novel and specific mechanism of gene
regulation. Attenuators and RNase-cleavage sites that are affected by
translation are well known (3, 11), but there are few cases
known in which regulated translation of a coding sequence is directly
coupled to these determinants of RNA stability. Most attenuators act at a site before the regulated structural gene. The placement of the
transcriptional regulatory region distal to the start of translation allows the translational regulatory process to simultaneously affect
the transcriptional process.
The mechanism of Cbl-dependent translation regulation remains a subject
worthy of further study. It is known from previous studies to require
the integrity of most of the leader region, including the proper
transcription start site (unpublished data and reference
7) and the potential for formation of the hairpin around the ribosome-binding site. However, no candidate regulatory proteins that might influence the structure of the RNA in a
Cbl-dependent manner have surfaced yet. Changes in the conformation of
the transcript can be deduced from the appearance of the
Cbl-inducible 210-nt fragment. We are attempting to define changes in
btuB RNA secondary structure in the presence of
Ado-Cbl and to test for specific binding of Ado-Cbl to the RNA.
We conclude that the translated region is essential for transcriptional
control and the Cbl-dependent decrease in RNA stability. The activity
of this region is directly regulated by the Cbl-inhibited translation
of the btuB gene, such that translation through this region
must occur to prevent termination and RNA turnover. The presence of
Ado-Cbl may cause changes in RNA secondary structures that affect
accessibility of the ribosome-binding site. Questions under study
address the precise role of the attenuator and distal sequences, the
relationship of translation to attenuator action, and, most
importantly, how Ado-Cbl is sensed and affects RNA structure or
translation initiation.
| |
ACKNOWLEDGMENTS |
We thank Sidney Kushner, Malcolm Winkler, and Murray Deutscher
for provision of strains and Joanna Goldberg and Igor Olekhnovich for
helpful comments.
This work was supported by grant GM19078 from the National Institutes
of Health.
| |
FOOTNOTES |
*
Corresponding author. Mailing address: Department of
Microbiology HSC#441, School of Medicine, University of Virginia,
Charlottesville, VA 22908. Phone: (804) 924-2532. Fax: (804) 982-1071. E-mail: rjk{at}virginia.edu.
| |
REFERENCES |
| 1.
|
Aufrère, R.,
M. Tempête, and J.-P. Bohin.
1986.
Regulation of expression of the gene for vitamin B12 receptor cloned on a multicopy plasmid in Escherichia coli.
Mol. Gen. Genet.
205:358-365[Medline].
|
| 2.
|
Babitzke, P.,
L. Granger,
J. Olszewski, and S. R. Kushner.
1993.
Analysis of mRNA decay and rRNA processing in Escherichia coli multiple mutants carrying a deletion in RNase III.
J. Bacteriol.
175:229-239[Abstract/Free Full Text].
|
| 3.
|
Braun, F.,
J. Le Derout, and P. Régnier.
1998.
Ribosomes inhibit an RNase E cleavage which induces the decay of rpsO mRNA of Escherichia coli.
EMBO J.
17:4790-4797[Medline].
|
| 4.
|
Chevrier-Miller, M.,
N. Jacques,
O. Raibaud, and M. Dreyfus.
1990.
Transcription of single-copy hybrid lacZ genes by T7 RNA polymerase in Escherichia coli: mRNA synthesis and degradation can be uncoupled from translation.
Nucleic Acids Res.
18:5787-5792[Abstract/Free Full Text].
|
| 5.
|
Emory, S. A., and J. G. Belasco.
1990.
The ompA 5' untranslated RNA segment functions in Escherichia coli as a growth-rate-regulated mRNA stabilizer whose activity is unrelated to translational efficiency.
J. Bacteriol.
172:4472-4481[Abstract/Free Full Text].
|
| 6.
|
Escalante-Semerena, J. C., and J. R. Roth.
1987.
Regulation of cobalamin biosynthetic operons in Salmonella typhimurium.
J. Bacteriol.
169:2251-2258[Abstract/Free Full Text].
|
| 6a.
| Franklund, C. V. Unpublished observations.
|
| 7.
|
Franklund, C. V., and R. J. Kadner.
1997.
Multiple transcribed elements control expression of the Escherichia coli btuB gene.
J. Bacteriol.
179:4039-4042[Abstract/Free Full Text].
|
| 8.
|
Hansen, M. J.,
L.-H. Chen,
M. L. S. Fejzo, and J. G. Belasco.
1994.
The ompA 5' untranslated region impedes a major pathway for mRNA degradation in Escherichia coli.
Mol. Microbiol.
12:707-716[Medline].
|
| 9.
|
Kadner, R. J.
1978.
Repression of synthesis of the vitamin B12 receptor in Escherichia coli.
J. Bacteriol.
136:1050-1057[Abstract/Free Full Text].
|
| 10.
|
Kelly, K. O., and M. P. Deutscher.
1992.
The presence of only one of five exoribonucleases is sufficient to support the growth of Escherichia coli.
J. Bacteriol.
174:6682-6684[Abstract/Free Full Text].
|
| 11.
|
Landick, R.,
C. L. Turnbough, Jr., and C. Yanofsky.
1996.
Transcription attenuation, p. 1263-1286.
In
F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella. Cellular and molecular biology, 2nd ed., vol. 1. ASM Press, Washington, D.C.
|
| 12.
|
Li, Z., and M. P. Deutscher.
1996.
Maturation pathways for E. coli tRNA precursors: a random multienzyme process in vivo.
Cell
86:503-512[Medline].
|
| 13.
|
Lundrigan, M. D., and R. J. Kadner.
1989.
Altered cobalamin metabolism in Escherichia coli btuR mutants affects btuB regulation.
J. Bacteriol.
171:154-161[Abstract/Free Full Text].
|
| 14.
|
Lundrigan, M. D.,
W. Köster, and R. J. Kadner.
1991.
Transcribed sequences of the Escherichia coli btuB gene control its expression and regulation by vitamin B12.
Proc. Natl. Acad. Sci. USA
88:1479-1483[Abstract/Free Full Text].
|
| 15.
|
Muffler, A.,
D. Fischer, and R. Hengge-Aronis.
1996.
The RNA-binding protein, HF-1, known as a host factor for phage Q RNA replication, is essential for rpoS translation in Escherichia coli.
Genes Dev.
10:1143-1151[Abstract/Free Full Text].
|
| 16.
|
Petersen, C.
1993.
Translation and mRNA stability in bacteria: a complex relationship, p. 117-145.
In
J. G. Belasco, and G. Brawerman (ed.), Control of messenger RNA stability. Academic Press, San Diego, Calif.
|
| 17.
|
Ravnum, S., and D. I. Andersson.
1997.
Vitamin B12 repression of the btuB gene in Salmonella typhimurium is mediated via a translational control which requires leader and coding sequences.
Mol. Microbiol.
23:35-42[Medline].
|
| 18.
|
Richter-Dahlfors, A. A., and D. I. Andersson.
1992.
Cobalamin (vitamin B12) repression of the Cob operon in Salmonella typhimurium requires sequences within the leader and the first translated open reading frame.
Mol. Microbiol.
6:743-749[Medline].
|
| 19.
|
Richter-Dahlfors, A. A.,
S. Ravnum, and D. I. Andersson.
1994.
Vitamin B12 repression of the cob operon in Salmonella typhimurium: translational control of the cbiA gene.
Mol. Microbiol.
13:541-553[Medline].
|
| 20.
|
Roth, J. R.,
J. G. Lawrence, and T. A. Bobik.
1996.
Cobalamin (coenzyme B12): synthesis and biological significance.
Annu. Rev. Microbiol.
50:137-181[Medline].
|
| 21.
|
Sambrook, J.,
E. F. Fritsch, and T. Maniatis.
1989.
Molecular cloning: a laboratory manual, 2nd ed.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
|
| 22.
|
Simons, R. W.,
F. Houman, and N. Kleckner.
1987.
Improved single and multicopy lac-based cloning vectors for protein and operon fusions.
Gene
53:85-96[Medline].
|
| 23.
|
Studier, F. W.,
A. H. Rosenberg,
J. J. Dunn, and J. W. Dubendorff.
1990.
Use of T7 RNA polymerase to direct expression of cloned genes.
Methods Enzymol.
185:60-89[Medline].
|
| 24.
|
Tsui, H.-C.,
H.-C. E. Leung, and M. E. Winkler.
1994.
Characterization of broadly pleiotropic phenotypes caused by an hfq insertion mutation in Escherichia coli.
Mol. Microbiol.
13:35-49[Medline].
|
| 25.
|
Wei, B.-Y.,
C. Bradbeer, and R. J. Kadner.
1992.
Conserved structural and regulatory regions in the Salmonella typhimurium btuB gene for the outer membrane vitamin B12 transport protein.
Res. Microbiol.
143:459-466[Medline].
|
Journal of Bacteriology, December 1998, p. 6719-6728, Vol. 180, No. 24
0021-9193/98/$04.00+0
Copyright © 1998, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Borovok, I., Gorovitz, B., Schreiber, R., Aharonowitz, Y., Cohen, G.
(2006). Coenzyme B12 Controls Transcription of the Streptomyces Class Ia Ribonucleotide Reductase nrdABS Operon via a Riboswitch Mechanism.. J. Bacteriol.
188: 2512-2520
[Abstract]
[Full Text]
-
Chowdhury, S., Ragaz, C., Kreuger, E., Narberhaus, F.
(2003). Temperature-controlled Structural Alterations of an RNA Thermometer. J. Biol. Chem.
278: 47915-47921
[Abstract]
[Full Text]
-
Rodionov, D. A., Vitreschak, A. G., Mironov, A. A., Gelfand, M. S.
(2003). Comparative Genomics of the Vitamin B12 Metabolism and Regulation in Prokaryotes. J. Biol. Chem.
278: 41148-41159
[Abstract]
[Full Text]
-
VITRESCHAK, A. G., RODIONOV, D. A., MIRONOV, A. A., GELFAND, M. S.
(2003). Regulation of the vitamin B12 metabolism and transport in bacteria by a conserved RNA structural element. RNA
9: 1084-1097
[Abstract]
[Full Text]
-
SILVERMAN, S. K.
(2003). Rube Goldberg goes (ribo)nuclear? Molecular switches and sensors made from RNA. RNA
9: 377-383
[Abstract]
[Full Text]
-
Cadieux, N., Bradbeer, C., Reeger-Schneider, E., Koster, W., Mohanty, A. K., Wiener, M. C., Kadner, R. J.
(2002). Identification of the Periplasmic Cobalamin-Binding Protein BtuF of Escherichia coli. J. Bacteriol.
184: 706-717
[Abstract]
[Full Text]
-
Stormo, G. D., Ji, Y.
(2001). Do mRNAs act as direct sensors of small molecules to control their expression?. Proc. Natl. Acad. Sci. USA
98: 9465-9467
[Full Text]
-
Miranda-Rios, J., Navarro, M., Soberon, M.
(2001). A conserved RNA structure (thi box) is involved in regulation of thiamin biosynthetic gene expression in bacteria. Proc. Natl. Acad. Sci. USA
10.1073/pnas.161168098v1
[Abstract]
[Full Text]
-
Nou, X., Kadner, R. J.
(2000). Adenosylcobalamin inhibits ribosome binding to btuB RNA. Proc. Natl. Acad. Sci. USA
10.1073/pnas.130013897v1
[Abstract]
[Full Text]
-
Nou, X., Kadner, R. J.
(2000). Adenosylcobalamin inhibits ribosome binding to btuB RNA. Proc. Natl. Acad. Sci. USA
97: 7190-7195
[Abstract]
[Full Text]
-
Miranda-Rios, J., Navarro, M., Soberon, M.
(2001). From the Cover: A conserved RNA structure (thi box) is involved in regulation of thiamin biosynthetic gene expression in bacteria. Proc. Natl. Acad. Sci. USA
98: 9736-9741
[Abstract]
[Full Text]
Top
Abstract
Introduction
