| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, September 27,
1995; and in revised form, February 21, 1996) From the
During transcription initiation from galP2, one of the
two promoters of the Escherichia coli galactose operon with an
initially transcribed sequence of pppAUUUC, RNA polymerase (RNAP) is
known to engage nonproductive stuttering synthesis, which is sensitive
to the concentration of UTP. This study examines the effect of this
nonproductive synthesis on promoter clearance and determines other
parameters that might affect stuttering synthesis by analyzing a mutant
RNAP, RpoB3449, that has altered its function at this process at galP2. RpoB3449 has dramatically diminished stuttering
synthesis, and consequently, it has increased the rate of productive
initiation due to its enhanced rate of promoter clearance of galP2 compared with wild-type RNAP. Thus, a direct linkage between
promoter clearance and productive transcription is demonstrated. The
mechanism by which the mutant RNAP has altered the switch between
nonproductive stuttering synthesis and productive initiation during
promoter clearance is studied. Apparently, RpoB3449 has increased its
efficiency in incorporating CTP at the +5 position of the galP2 transcript leading to its reduced stuttering synthesis,
indicating that the rate of an RNAP incorporating the CTP after a
stretch of uridine residues is important for promoter clearance at galP2. Because RpoB3449 demonstrates ``wild-type''
stuttering synthesis at the mutant galP2 promoter, which
contains the 6 residue at the +5 position, it indicates that the
mutant RNAP has altered in binding CTP at this context. Further
experiments indicate that it is the +5 position per se of
the galP2 sequence rather than a particular nucleotide at that
position that is critical in determining the switch between the two
alternate pathways during transcription initiation. A checkpoint model
for the switch between nonproductive and productive initiations during
promoter clearance is discussed.
One of the potential regulatory steps in prokaryotic
transcription is promoter clearance, a transition step in transcription
initiation at which an RNA polymerase (RNAP) ( Nonproductive stuttering synthesis was reported only in a
mutant promoter in the past few years(6) , whereas abortive
synthesis is common at most promoters. Recently, evidence of biological
significance for these nonproductive syntheses in gene expression is
emerging. Nonproductive stuttering syntheses have been shown at the Escherichia coli native pyrBI and galP2 promoters(11, 12) , and the expression of these
two promoters is sensitive to changes in UTP
concentration(12, 13) . At these two promoters, RNAP
makes an abundant amount of nonproductive stuttered products and a
lesser amount of productive initiation products in vitro when
the concentration of UTP is high and vice versa. An E. coli mutant RNAP that overproduces abortive initiation products and
reduces productive initiation at the pyrBI promoter in
vitro has reduced expression from the promoter in
vivo(14) . CRP protein was shown to inhibit abortive
initiation leading to an enhanced productive initiation at the malT promoter(15) . lac repressor was suggested to
modify the initial transcribing complex so that promoter clearance was
inhibited(16) . In addition, DNA sequences flanking the
``classical'' promoter region have been implicated to
contribute to promoter function(17, 18) , suggesting
that initially transcribed sequences of promoter may affect
nonproductive syntheses and promoter clearance. Our knowledge of the
parameters that affect RNAP on abortive and stuttering synthesis as
well as of the mechanisms underlying the switch between nonproductive
and productive initiations during promoter clearance is very limited.
One approach to study these questions is to analyze mutant RNAPs that
have altered function in the switching between the two alternate
pathways during promoter clearance and to determine how these mutant
RNAPs are perturbed in the process. For this purpose, the mutant RNAPs
that confer rifampicin resistance (Rif In this
study, the effects of previously described Rif
Figure 1:
Transcription initiation from the gal and pyrBI promoters by wild-type and RpoB3449
RNAPs in vitro. RNA was labeled with
[
Figure 2:
Effects of different compositions of
nucleotides on stuttering synthesis at galP2. A,
transcription initiation at gal by wild-type and RpoB3449
RNAPs in vitro. The experiment was performed as described in Fig. 1, except that the compositions of NTPs were varied as
indicated. The concentrations of UTP in odd numbered lanes were high (H) 0.2 mM and in even numbered
lanes were low (L) 0.02 mM. The transcription
products were separated by electrophoresis on a 24% polyacrylamide, 7 M urea gel as in Fig. 1. B, the amounts of the
stuttered products (the sum of AU4 to AU30) produced from galP2 by the mutant RNAP (hatched columns) and wild-type enzyme (solid columns) at different conditions. Conditions for columns 1-8 are the same as that for lanes 1-8 (wild-type RNAP) or that for lanes 9-16 (RpoB3449)
in A.
The
effects of GTP and CTP on the stuttering synthesis by both RNAPs were
studied (Fig. 2). For RpoB3449, omission of GTP from a complete
transcription reaction leads to a large reduction in stuttering
synthesis compared with that of wild-type RNAP (compare lanes 15 and 16 with lanes 7 and 8 in Fig. 2A); RpoB3449 behaves as if all four nucleotides
were present (compare lanes 9 and 10 with lanes 1 and 2; see also Fig. 2B). However, when
CTP was omitted from a complete transcription reaction, both the mutant
and wild-type RNAPs made comparable amounts of stuttering synthesis
products (compare lanes 13 and 14 with lanes 5 and 6 in Fig. 2A; see also Fig. 2B). These results indicate that the presence of
CTP is responsible for the reduced stuttering synthesis by RpoB3449 at
the galP2 promoter. Because CTP is the next nucleotide to
be incorporated after the three continuous uridine residues of the
nascent RNA oligomer pppAUUU, it is hypothesized that RpoB3449 has
increased its efficiency in incorporating CTP at the fifth position of
the transcript. Therefore, this mutant RNAP accelerates the rate of
polymerization of pppAUUUC, resulting in reduced stuttering synthesis
and enhanced promoter clearance at galP2. If the proposed
model is correct, one would predict that stuttering synthesis at galP2 by both wild-type and the mutant RNAPs will be sensitive
to the concentration of CTP. Furthermore, at high enough concentrations
of CTP, wild-type RNAP will have reduced stuttering synthesis,
simulating the phenotype of RpoB3449. Conversely, when the
concentration of CTP is low enough to reduce the efficiency of the
mutant RNAP in incorporating the nucleotide, RpoB3449 should be able to
make significant amounts of stuttered products. To determine whether
the above predictions are true, transcription assays were performed at
a variety of CTP concentrations while keeping the other NTP
concentrations constant, and transcription products were analyzed. The
results are in complete agreement with the above hypothesis (Fig. 3). Overall, for both RNAPs there is an inverse
relationship between the amount of the stuttered products being made
and the concentration of CTP used in the assay (Fig. 3B). Furthermore, wild-type RNAP made few
stuttered products when the concentration of CTP was high (
Figure 3:
Effect of CTP concentration on stuttering
synthesis at galP2. A, transcription initiation at gal by wild-type and RpoB3449 RNAPs in vitro as a
function of CTP concentration. The experiment was performed as
described in Fig. 1except that the concentration of UTP was
kept at 0.2 mM in favor of stuttering synthesis, and CTP was
varied as indicated: 1 (lanes 1 and 8); 0.2 (lanes 2 and 9); 0.08 (lanes 3 and 10); 0.02 (lanes 4 and 11); 0.005 (lanes
5 and 12); 0.001 (lanes 6 and 13); and
0 mM (lanes 7 and 14). The transcription
products were separated by electrophoresis on a 24% polyacrylamide, 7 M urea gel as in Fig. 1. B, the amounts of the
stuttered products (the sum of AU4 to AU30) produced from galP2 by the mutant RNAP (hatched columns) and wild-type enzyme (solid columns) at different CTP
concentrations.
Figure 4:
Effect of GTP concentration on stuttering
synthesis at the new DNA template containing a C to G change at the
fifth position of the galP2 sequence. The experiment was
performed as described in the legend to Fig. 3except that the
concentration of GTP was varied as indicated: 1 (lanes 1 and 6); 0.2 (lanes 2 and 7); 0.02 (lanes 3 and 8); 0.002 (lanes 4 and 9); and 0
mM (lanes 5 and 10). The transcription
products were separated by electrophoresis on a 24% polyacrylamide, 7 M urea gel as in Fig. 1.
Because of its reduced stuttering synthesis, the mutant
RNAP is likely to have an increased rate of promoter clearance leading
to an increased productive initiation compared with wild-type RNAP. To
test this hypothesis, the accumulation of both stuttered and the
full-length galP2 transcripts (as the productive initiation
products) by the two RNAPs was determined as a function of time in a
single round transcription assay (i.e., in the presence of
heparin) (Fig. 5A). In these assays only transcription
from galP2 was measured because GalR repressor protein, which
only inhibits transcription from galP1 in vitro(29) and has no effect on stuttering synthesis at
galP2(12) , was included in the reactions. For wild-type RNAP,
stuttered products appeared as early as 20 s, indicating that the time
needed for RNAP to make oligomers of stuttered products is very short.
Full-length transcript P2 appeared at 40 s, the time needed for both
promoter clearance and the completion of the transcript elongation, and
its amount was increased gradually as time increased. The accumulation
of the full-length transcript was parallel with that of the stuttered
products as a function of time, indicating that at an any given time
only a fraction of the initially transcribing complexes has completed
the promoter clearance at galP2 to enter transcription
elongation and to finish a productive initiation cycle. Note that more
full-length products were made at 15 min than at 6 min, indicating that
even after 6 min there were still some wild-type RNAP molecules
remaining at the promoter (Fig. 5C). Therefore,
promoter clearance at galP2 is a rate-limiting step for
productive initiation due to stuttering synthesis for wild-type RNAP.
Figure 5:
Kinetics studies of the production of galP2 products by wild-type and RpoB3449 RNAPs. The experiment
was performed as described in Fig. 1, UTP was kept at 0.2 mM in favor of stuttering synthesis, and the reaction was stopped at
indicated time. GalR protein was added during preincubation period to
prevent transcription from galP1; therefore only transcription
from galP2 is studied. The transcripts from the same
experiment were separated either by electrophoresis on an 8%
polyacrylamide, 7 M urea gel to analyze the galP2 full-length transcript (P2,
However, in contrast to wild-type RNAP, the mutant RNAP made
increasing amounts of the full-length transcript rapidly and lesser
amount of the stuttered products (Fig. 5A).
Furthermore, the accumulation of the full-length transcript approached
the plateau within 3 min, when most RNAP molecules had finished a
productive initiation cycle (Fig. 5C). These results
indicate that at any given time, a larger fraction of the initially
transcribing RpoB3449 complexes has completed promoter clearance, and
consequently it takes less time for most of the mutant RNAP molecules
to finish a productive initiation cycle compared with wild-type RNAP.
Clearly, RpoB3449 has a faster rate of promoter clearance due to
reduced stuttering synthesis. The effect of the rate of promoter
clearance on the productive initiation was further manifested in a
multiple round transcription assay (i.e., in the absence of
heparin). As shown above, during stuttering synthesis wild-type RNAP
does not vacate the promoter, and it remains idling at the promoter for
a long duration. Thus, it follows that promoter turnover must be also a
rate-limiting step in productive initiation. One expected consequence
of this event is a reduced ability to productively initiate in a
multiple round transcription assay. In contrast, because RpoB3449 has
enhanced the rate of promoter clearance or promoter turnover, there is
a better chance for free mutant RNAP molecules to bind to vacated
promoters leading to a relative large difference in productive
initiation in a multiple round transcription assay compared with that
in a single round transcription assay. To test this predication, a
multiple round transcription assay was performed. The results are in
complete agreement with the above hypothesis (Fig. 5B).
For wild-type RNAP, the difference in the overall patterns of the
production of galP2 products, in particular for the
full-length transcript, was small between the multiple round
transcription (Fig. 5B) and the single round
transcription (Fig. 5A). Noticeable increase in the
production of the full-length galP2 products only occurred
after 6 min, and there was only about 40% increase in the production of
the productive initiation products in a multiple round transcription
compared with that in a single round transcription after 15 min (Fig. 5C). In contrast, for RpoB3449, there was a
marked difference in the amount of the full-length galP2 transcript between the multiple round transcription (Fig. 5B) and the single round transcription (Fig. 5A). Significant increases in the production of
the full-length galP2 products occurred at 3 min, and as time
increased the difference in the production of the full-length
transcript between the two assays becomes even larger, and there was an
over 3-fold increase in the production of the productive initiation
products in a multiple round transcription compared with that in a
single round transcription after 15 min (Fig. 5C). At
all time points, the mutant RNAP made more productive initiation
products than wild-type RNAP, and the effects became most prominent as
the time increases in the multiple round transcription assay (Fig. 5C). Clearly, the observed dramatic differences
in the productive initiation of galP2 between wild-type and
the mutant RNAPs at later time points reflect the amplified intrinsic
differences in the rates of promoter clearance between the two RNAPs in
a multiple round transcription assay. Stuttering synthesis has been reported both in prokaryotic
and eukaryotic organisms(33) . This work presents the first
analysis of a mutant RNAP that has altered nonproductive stuttering
synthesis during transcription initiation. The perturbation of
stuttering synthesis at galP2 by RpoB3449 has provided a
unique opportunity to study the effects of nonproductive synthesis on
promoter clearance and to study the mechanism (other than UTP
concentration) underlying the switch between nonproductive and
productive initiation during promoter clearance.
Figure 6:
A checkpoint model to explain the switch
between two alternate pathways. Nonproductive stuttering synthesis versus productive initiation during promoter clearance at the galP2 promoter. For more detail, see
text.
Furthermore, this study has demonstrated that the fifth position of
the galP2 initially transcribing sequence, rather than a
particular nucleotide at that position, is critical in determining
promoter clearance. This result indicates that this fifth position,
most likely because it lies immediately after the run of uridine
residues that are responsible for stuttering synthesis, imposes a high
energy barrier that an initially transcribing complex has to overcome.
However, it is not likely that in general the fifth position of the
initially transcribing sequence is critical in promoter clearance at
all promoters. Rather, it is the specific context at the galP2 sequence that makes the fifth position a critical one. It is known
that the stretch rU Why does
RNAP tend to make nonproductive initiation products during the
transition step between transcription initiation and elongation? It has
been speculated that in an initially transcribing complex, RNAP
establishes the RNA binding sites, which are postulated to be important
in maintaining the integrity of an elongation complex(36) . It
is conceivable that conversion from an initiation to an elongation
complex, accompanied by We are just beginning in understanding the
promoter clearance step in transcription initiation. This and other
work (14) indicate the importance of K
Volume 271,
Number 20,
Issue of May 17, 1996 pp. 11659-11667
©1996 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)switches from
an initial transcribing stage to an elongation stage (for reviews see (1) and (2) ). During promoter clearance, RNAP usually
also makes two forms of nonproductive initiation products in a
promoter-dependent manner: (i) abortive
synthesis(3, 4, 5) and (ii) stuttering
synthesis(6) , although productive stuttering synthesis at some
promoters also has been
reported(7, 8, 9, 10) . It is
conceivable that productive initiation could be modulated by turning
the switch between these alternate pathways (nonproductive versus productive synthesis) during the promoter clearance step at these
promoters.
) were chosen first,
because the antibiotic rifampicin inhibits transcription by preventing
an initial transcribing complex from entering an elongation
mode(3, 19) . Presumably, rifampicin plugs the pathway
leading a nascent RNA out of the active center(20) ; thus the
rifampicin-binding sites of RNAP could potentially interact with a
nascent RNA. It is reasoned that some of the Rif mutant
RNAPs, which have altered binding of rifampicin, are likely to be
altered in the interaction with nascent RNA leading to altered
nonproductive synthesis during promoter clearance. Most of the
Rif mutations are clustered in the middle of the rpoB gene encoding the 
subunit of RNAP(21, 22) ,
and their effects on transcription elongation,
termination/antitermination, and other functions have been
characterized(23, 24, 25, 26) .
Indeed, using this approach it was shown recently that some Rif RNAPs that overproduced abortive initiation products have reduced
productive initiation and that abortive synthesis is sensitive to the K
of an initially transcribing complex
for the nucleotide UTP during transcription initiation(14) .
Some lethal mutant RNAPs that block the initiation-to-elongation
transition were described, but the mechanisms underlying the blockage
have not been elucidated clearly(27, 28) . mutant RNAPs
on the nonproductive stuttering synthesis at galP2 are studied
to uncover other parameters (in addition to be sensitive to changes in
UTP concentration) that might affect promoter clearance at this
promoter. It is found that a Rif mutant RNAP, RpoB3449, has
diminished stuttering synthesis and enhanced productive initiation at
the galP2 promoter; thus it has altered the switch between the
two alternate pathways during promoter clearance. The effect of
RpoB3449 on the rate of productive initiation at galP2 is
analyzed, and the results indicate that promoter clearance is a
rate-limiting step for wild-type RNAP in the productive initiation of galP2. The mechanism by which RpoB3449 has altered promoter
clearance at galP2 is studied, and the results indicate that
the efficiency of incorporating CTP at the fifth position of galP2 transcript is critical in determining RNAP whether to engage in
nonproductive stuttering synthesis or in productive initiation.
Materials
Nucleotides were from Boehringer
Mannheim; salts of glutamate were from Fluka; and 
P-labeled nucleotides were from Amersham Corp. or ICN.
Plasmids were isolated using Qiagen column according to the
manufacturer's manual followed by phenol extraction, and DNA
fragments were purified by electroelution after isolation from a 1.5%
agarose gel followed by phenol extraction. The 348-bp EcoRI-BamHI fragment of pSA509 containing the gal promoter-regulatory region and the very strong Rho-independent
transcription terminator of the rpoC gene of E. coli was used as the DNA template for in vitro transcription(29) . Plasmid pBHM332 (a kind gift from Dr.
Charles Turnbough, Jr., University of Alabama at Birmingham), which was
constructed by replacing the 322-bp PvuII fragment of pUC19
with the 758-bp PvuII fragment containing the pyrBI promoter region(30) , was also used as the DNA template
for in vitro transcription. In pBHM332, the transcriptional
direction of pyrBI is opposite that of bla. RNAPs
were purified from E. coli K12 MG1655 background as
described(31) . Both the wild-type and mutant RNAP, RpoB3449,
were highly pure (>95%) and had comparable activities in
synthesizing the RNAI transcripts of pBR322. The gal repressor
(GalR protein, a kind gift of Dr. Yan Ning Zhou) was purified as
described(32) .Localized Mutagenesis, Cloning, and DNA
Sequencing
The C residue at the fifth position of the galP2 coding sequence in the pSA509 plasmid was replaced with G residue
as follows. First, two DNA fragments (A and B), which share
33-nucleotide sequences around the starting site (+1) of galP2 coding region, were each amplified by polymerase chain reaction
using pSA509 as a DNA template. The sequences of the two primers used
for amplifying DNA fragments A (
300 bp) were:
5`-TTCTAGACCTTCCCGTTTCGCTCAAGTTAG (DJJ121), which covers the sequence
upstream of the EcoRI site in pSA509, and
5`-TAGGCTTATGGTATCAAATAACCATAGCATAAC (DJJ122), which covers the
complementary sequences to the coding sequences from +19 to
-14 position and contains a G to C change at the +5
position. The sequences of the two primers used for amplifying DNA
fragments B (200 bp) were: 5`-GTTATGCTATGGTTATTTGATACCATAAGCCTA
(DJJ123), which covers the coding sequences from -14 to +19
position and contains a C to G change at the +5 position, and
5`-TCATAGAGTCTTGCAGACAAACTGCGCAAC (DJJ124), which covers the sequence
downstream of the BamHI site in pSA509. The amplified
fragments A and B were isolated from a 1.5% agarose gel and purified by
electroelution followed by phenol extraction. The purified DNA
fragments were then denatured and annealed at the overlapping sequences
followed by Taq polymerase extension in a limited polymerase
chain reaction (four cycles) to generate enough double strand DNA to be
used as DNA template. Another 20 cycles of polymerase chain reaction
were continued after adding two primers (DJJ121 and DJJ124) into the
above reactions to amplify the fragment C (500 bp), which contains EcoRI and BamHI sites at either end of the fragment
and a mutational change at the +5 position of galP2 coding sequence. The amplified DNA fragment C was digested with EcoRI and BamHI restriction enzymes followed by
purification of the resulting 350-bp EcoRI-BamHI
fragment as described above. This EcoRI-BamHI
fragment containing the mutation was ligated with the purified large EcoRI-BamHI-digested fragment of pSA509 resulting in
pDJJ51, followed by transformation. Plasmid pDJJ51 DNA was purified
from the Amp transformants, and the C to G change at the
+5 position of galP2 sequence was confirmed by automated
DNA sequencing carried out on a model 373A DNA sequencer (Applied
Biosystems Inc.) using the Taq polymerase dye terminator
sequencing protocol (Applied Biosystems Inc.) according to the
manufacturer's manual.In Vitro Transcription Assays
In vitro transcription assays were essentially as described(12) . A
complete transcription reaction mixture containing 20 mM Tris-glutamate (pH 8.0), 100 mM potassium glutamate, 10
mM magnesium glutamate, 5% glycerol, acetylated bovine serum
albumin (100 µg/ml), DNA at 5 nM, and RNA polymerase
at 20 nM for wild-type and 24 nM for RpoB3449, was
preincubated for 
10 min at 37 °C. When indicated, GalR protein
(87 nM as dimer) was also included during the preincubation
period. Unless otherwise mentioned, NTP concentrations were 0.2 mM for ATP, GTP, and UTP and 0.02 mM for CTP. The reaction
was started by the addition of NTPs including P-labeled
nucleotide as indicated and stopped after 10 min or at the indicated
times by the addition of a 0.2 volume of 0.2 M EDTA in 40%
glycerol with dyes and put on ice. For single round transcription
assays, heparin (final concentration, 100 µg/ml) was present in the
NTP mixture. After heating for 3 min in boiling water, samples were
loaded directly onto a polyacrylamide gel containing 7 M urea,
and transcripts were visualized by autoradiography. The nonproductive
initiation products were analyzed on a 24% gel, and the productive
initiation products were analyzed on a 8% gel. The transcription
products were quantified with an AMBIS Imaging System(TM) (San
Diego, CA), and the data were corrected for background.
RpoB3449 Has Reduced Stuttering Synthesis during
Transcription Initiation at galP2
RNAP makes nonproductive
stuttering synthesis products at the galP2 promoter, and the
stuttering synthesis is sensitive to the concentration of
UTP(12) . As shown in Fig. 1A, at high UTP (0.2
mM), wild-type RNAP makes a large amount of stuttered
products, whereas at low UTP (0.02 mM), the synthesis of
stuttered product is reduced. To study the effects of previously
described Rif mutant RNAPs on stuttering synthesis from the galP2 promoter, transcription initiation assays were
performed. Most Rif mutant RNAPs behaved like wild-type
RNAP in stuttering synthesis (data not shown). However, one mutant
RNAP, RpoB3449, which has a deletion of an alanine at the 532 position
of the subunit of RNAP, has greatly reduced nonproductive
synthesis at galP2 compared with wild-type RNAP (Fig. 1, A and C). Apparently, the effect of
the mutant RNAP on stuttering synthesis is promoter-specific, because
RpoB3449 behaves like wild-type RNAP at the pyrBI promoter:
both RNAPs make a large amount of stuttering synthesis products at 0.2
mM UTP (high) and a lesser amount at 0.02 mM UTP
(low) (Fig. 1, B and D). These results
indicate that the mutant RNAP is capable of engaging in stuttering
synthesis but has an altered function at the galP2 promoter
(when comparable amounts of the proteins were used).
-P]UTP in initiation assays containing
either high (H) 0.2 mM or low (L) 0.02
mM UTP as described under ``Experimental
Procedures.'' The transcription products were separated by
electrophoresis on a 24% polyacrylamide, 7 M urea gel. The
stuttered products from the galP2 promoter (of which AU4 is
the shortest with the others increasing in length by an increment of
one UMP) (A) and from the pyrBI promoter (of which
AAU4 is the shortest with the others increasing in length by an
increment of one UMP) (B) are indicated. The other less
abundant smaller sized transcripts are aborted products either from the gal promoters as described previously (39) or from the pyrBI(11) and other promoters presented in pBHM332. C, the amounts of the stuttered products (the sum of AU4 to
AU30) produced from galP2 by the mutant RNAP (hatched
columns) and wild-type enzyme (solid columns) at
different UTP concentrations. D, the amounts of the stuttered
products (the sum of AAU4 and AAU30) produced from pyrBI by
the mutant RNAP (hatched columns) and wild-type enzyme (solid columns) at different UTP
concentrations.
The Efficiency of an RNAP in Incorporating CTP at the
Fifth Position of the galP2 Nascent RNA Determines the Extent of
Stuttering Synthesis
To determine the mechanism by which
RpoB3449 has altered stuttering synthesis at galP2, the
effects of different conditions on this nonproductive initiation at galP2 were compared between wild-type and the mutant RNAPs (Fig. 2). It is known that wild-type RNAP stutters at galP2 when only the two nucleotides ATP and UTP are present (12) . Under this condition, RpoB3449 is capable of stuttering
synthesis at galP2 (Fig. 2A, lanes 11 and 12), just like wild-type RNAP (Fig. 2A, lanes 3 and 4; see also Fig. 2B). Because the patterns of stuttering syntheses
by the two RNAPs are similar and sensitive to the concentration of UTP,
it rules out the possibility that RpoB3449 had an altered apparent K
for UTP leading to reduced stuttering synthesis.
Because RpoB3449 has diminished stuttering synthesis in the presence of
four nucleotides, whereas it is proficient in stuttering synthesis in
the presence of only ATP and UTP, it suggests that the presence of
either GTP or CTP affects the ability of the mutant RNAP to make
stuttering synthesis products at the galP2 promoter.
0.2
mM, lanes 1 and 2 in Fig. 3A) but an abundant amount of the stuttered
products when the concentrations of CTP were low (
0.02 mM, lanes 4-6). When the concentration of CTP was very low
(0.001 mM, lane 6), wild-type RNAP made almost the
maximum amount of the stuttered products, comparable with that made in
the absence of CTP (lane 7). This result indicates that at 1
µM CTP most of the initially transcribing complexes are
trapped in the nonproductive synthesis mode. On the other hand,
RpoB3449 made only very few of the stuttered products when the
concentrations of CTP were >0.005 mM (lanes
8-12) but a significant amount of the stuttered products
when the concentration of CTP was very low at 0.001 mM (lane 13). In the absence of CTP in the reaction, the
mutant RNAP made the maximum amount of the stuttered products (lane
14). The effect is CTP-specific, because the stuttering synthesis
at galP2 by wild-type RNAP was not sensitive to the
concentration of GTP (data not shown). These results indicate that the
rate of incorporation of CTP at the fifth position of the galP2 transcript is important in determining the efficiency of
stuttering synthesis and promoter clearance.
The Fifth Position of the galP2 Transcript, Which Lies
Immediately after the Stretch of Uridine Residues, Is Critical for
Stuttering Synthesis
It is possible that it is not CTP per
se but rather the fifth position of the galP2 sequence (e.g., the base that follows the run of uridine residues) that
is critical in determining the efficiency of stuttering synthesis by
RNAP. If this was the case, one would predict that by changing CTP to
another nucleotide at that position, stuttering synthesis at galP2 would become sensitive to the concentration of the new nucleotide.
To determine whether in stuttering synthesis at galP2 the
effect is CTP- or position-specific, the C at the fifth position was
replaced with G by localized mutagenesis, and the effect of the
concentration of GTP on stuttering synthesis was analyzed. The results
showed that with the new DNA template, stuttering synthesis by both
RNAPs became sensitive to the changes in the concentration of GTP and
the amount of stuttered products made was an inverse function of the
concentration of GTP (Fig. 4). These results have demonstrated
that the effect is position-specific and not CTP-specific, indicating
that the fifth position of the galP2 sequence is a critical
point in determining stuttering synthesis and promoter clearance at galP2. Furthermore, in contrast to the stuttering synthesis
with the wild-type galP2 sequence (Fig. 3), the
patterns of stuttering synthesis between the wild-type and the mutant
RNAPs are basically the same with the new DNA template containing a C
to G change at the fifth position of the galP2 sequence (Fig. 4). The wild-type phenotype of stuttering synthesis by the
mutant RNAP with the new DNA template further argues for the notion
that the mutant RNAP has an enhanced efficiency in incorporating CTP at
the fifth position of the wild-type galP2 sequence, leading to
its reduced stuttering synthesis at the promoter. At the present, the
possibility that this mutant RNAP affects the K for CTP or other nucleoside triphosphates during transcription at
other contexts cannot be excluded.
Stuttering Synthesis Delays Promoter Clearance and
Reduces the Rate of Productive Initiation of an RNAP
-Because
an initially transcribing RNAP either completes promoter clearance to
enter elongation mode or engages in nonproductive stuttering synthesis
at galP2, one would predict that promoter clearance is a
rate-limiting step for productive initiation when stuttering synthesis
occurs. The alteration of RpoB3449 in stuttering synthesis at galP2 provides a unique opportunity to study the effects of the
nonproductive initiation on promoter clearance and productive
initiation.
125 nucleotides) or by
electrophoresis on a 24% polyacrylamide, 7 M urea gel to
analyze the galP2 nonproductive initiation products. A, kinetics studies of the production of galP2 products in a single round transcription assay (heparin was added
at the beginning of reaction). B, kinetics studies of the
production of galP2 products in a multiple round transcription
assay (no heparin was presented in the assay). At later time course
RpoB3449 appeared making small amount of the galP1 full-length
products (120 nucleotides), which migrated just below the galP2 full-length transcript, probably because some GalR
repressor was dissociated transiently from the operator sites to allow
reinitiating RNAP to initiate at galP1. C, the rates of
accumulation of the full-length galP2 transcripts by wild-type
and RpoB3449 RNAPs in a single round or multiple round transcription
assay. The full-length galP2 transcripts were scanned, and the
data were plotted as a function of time. Filled circle,
wild-type RNAP in a single round transcription assay; open
circle, wild-type RNAP in a multiple round transcription assay; filled triangle, RpoB3449 in a single round transcription
assay; open triangle, RpoB3449 in a multiple round
transcription assay.
Promoter Clearance Is a Rate-limiting Step in the Productive
Initiation Due to Stuttering Synthesis
During stuttering
synthesis, RNAP is able to engage in multiple cycles of initiation
without dissociation from the promoter. Furthermore, for each new cycle
of initiation, only a fraction of RNAP molecules is able to incorporate
the CTP at the fifth position of the galP2 transcript. Only
then an initially transcribing complex will be able to complete the
promoter clearance step to enter the elongation mode and finish a
productive initiation cycle. This is consistent with the observation
that the accumulation of the nonproductive initiation products is
parallel with that of the productive initiation products in the
presence of heparin over a long period of time. Such a prolonged delay
in promoter clearance will reduce the rate of the production of the
productive initiation products accompanied by an exaggerated
accumulation of nonproductive initiation products (both abortive and
stuttered products). The mutant RNAP has diminished nonproductive
synthesis; therefore it ``by-passes'' the rate-limiting step
leading to a faster rate in productive initiation at galP2 compared with wild-type RNAP. Therefore, the observed dramatic
differences between wild-type and the mutant RNAPs in the synthesis of
the galP2 products (Fig. 5) demonstrates a direct link
between nonproductive stuttering synthesis and productive initiation
during promoter clearance. This is an analogy to another case study
showing a direct link between abortive initiation and productive
initiation during promoter clearance(14) . These results
demonstrate that the strength of a promoter can be regulated beyond the
steps of RNAP binding to a promoter and isomerization, two parameters
that have been studied extensively in determine the strength of a
classical promoter (34) .A Checkpoint Model for the Switching between
Nonproductive Stuttering Synthesis and Promoter Clearance
By
analyzing the mechanism by which the mutant RNAP has altered stuttering
synthesis at galP2, a ``checkpoint'' model has
evolved for the switch between nonproductive stuttering synthesis and
productive initiation during promoter clearance (Fig. 6).
According to this model, the rate of incorporation of CTP at the fifth
position of galP2 transcript serves as a checkpoint in
determining the fate of an initially transcribing complex containing
the pppAUUU oligomer. Thus, if the effective concentration of CTP is
limiting relative to UTP, the probability of an initially transcribing
complex to produce pppAUUUC will be low, leading to enhanced stuttering
synthesis. Conversely, when the effective concentration of CTP is
abundant, the probability of an initially transcribing complex to
produce pppAUUUC will be high, leading to reduced stuttering synthesis
and enhanced promoter clearance. In essence, the competition between
the rate of incorporating an extra uridine and the rate of
incorporating the cytidine to the nascent RNAP oligomer pppAUUU will
determine the outcome of an initially transcribing complex into either
of the alternate pathways during transcription initiation. RpoB3449 is
in a sense a checkpoint mutant of RNAP that has altered the switch
between alternate pathways in transcription initiation of galP2, because it has increased efficiency in incorporating
CTP at the +5 position of galP2 transcript compared with
wild-type RNAP. The fact that RpoB3449 demonstrates wild-type
stuttering synthesis at the mutant galP2 promoter, which
contains the G residue at the +5 position, indicates that the
alteration of the mutant RNAP is probably in binding CTP.
dA sequences before the fifth position of the galP2 sequence in an initially transcribing complex are
intrinsically unstable(14, 35) . When 5`-Br-UTP was
used in the place of UTP, the stuttering synthesis at galP2 was reduced dramatically (data not shown); presumably the more
stable 5`-Br-UdA pairs prevented the short RNA oligomers from
dissociation from an initially transcribing complex. Therefore, the
competition between the rate of incorporation of an extra UTP, leading
to the release of a stuttered product, and the rate of incorporation of
CTP, leading to a stabilized initially transcribing complex at this
critical position of the galP2 sequence, will determine the
efficiency of stuttering synthesis and promoter clearance.
factor release is a stress-inducing
process. Such a stress state would likely sensitize RNAP to respond to
the availability of nucleotides for incorporation at putative critical
position(s) in the initially transcribed sequences at some promoters.
Because the mature RNA binding sites have not been established yet in
an initially transcribing complex, the association of oligomer to the
complex is likely to be weak and dissociation-prone, in particular, if
the initially transcribed sequences have a run of uridine residues. In
this regard, it is interesting to note that the termination efficiency
at intrinsic terminators also depends on the rate of RNAP incorporation
of a critical nucleotide that lies immediately after a run of uridine
residues at the end of terminator sites(37) . This similarity
between promoter clearance and termination suggests that there is a
common mechanism underlying the two processes. It would be interesting
to determine whether it is the particular nucleotide or the particular
position at those terminators that is important in determining
termination efficiency. of
an initially transcribing complex for critical nucleotide in promoter
clearance at some promoters. Other experiments show that at some other
promoters, promoter clearance is not sensitive to the changes in
nucleotide concentration but is stimulated by the transcript cleavage
factors GreA and GreB(38) . It is likely that the limiting
step(s) in promoter clearance at different promoters is different,
indicating that control of promoter clearance is a complex process and
appears to be promoter-specific (or context-dependent).
), rifampicin resistance; bp, base pair(s).
I am grateful to Drs. Sue Gargess, Jeff Roberts, Lucia
Rothman-Denes, and Yan Ning Zhou for comments on the manuscript, and I
thank Dr. Yan Ning Zhou for sequencing the mutated galP2 DNA
template.
©1996 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:
|
|
 |
|
  C. L. Turnbough Jr. and R. L. Switzer Regulation of Pyrimidine Biosynthetic Gene Expression in Bacteria: Repression without Repressors Microbiol. Mol. Biol. Rev., June 1, 2008; 72(2): 266 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Meng, C. L. Turnbough Jr., and R. L. Switzer Attenuation control of pyrG expression in Bacillus subtilis is mediated by CTP-sensitive reiterative transcription PNAS, July 27, 2004; 101(30): 10943 - 10948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. V. Sukhodolets, J. E. Cabrera, H. Zhi, and D. J. Jin RapA, a bacterial homolog of SWI2/SNF2, stimulates RNA polymerase recycling in transcription Genes & Dev., December 15, 2001; 15(24): 3330 - 3341. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Severinov T7 RNA polymerase transcription complex: What you see is not what you get PNAS, December 14, 2000; (2000) 21535298. [Full Text]
|
|
|
|
 |
|
 S. Bost, F. Silva, and D. Belin Transcriptional Activation of ydeA, Which Encodes a Member of the Major Facilitator Superfamily, Interferes with Arabinose Accumulation and Induction of the Escherichia coli Arabinose PBAD Promoter J. Bacteriol., April 1, 1999; 181(7): 2185 - 2191. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. Viswanathan and P. W. Doetsch Effects of Nonbulky DNA Base Damages on Escherichia coli RNA Polymerase-mediated Elongation and Promoter Clearance J. Biol. Chem., August 14, 1998; 273(33): 21276 - 21281. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. L. deHaseth, M. L. Zupancic, and M. T. Record Jr. J. Bacteriol., June 15, 1998; 180(12): 3019 - 3025. [Full Text]
|
|
|
|
|
|
M. V. Sukhodolets and D. J. Jin RapA, a Novel RNA Polymerase-Associated Protein, Is a Bacterial Homolog of SWI2/SNF2 J. Biol. Chem., March 20, 1998; 273(12): 7018 - 7023. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. N. Zhou and D. J. Jin The rpoB mutants destabilizing initiation complexes at stringently controlled promoters behave like "stringent" RNA polymerases in Escherichia coli PNAS, March 17, 1998; 95(6): 2908 - 2913. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Han and C. L. Turnbough Jr. Regulation of carAB Expression in Escherichia coli Occurs in Part through UTP-Sensitive Reiterative Transcription J. Bacteriol., February 1, 1998; 180(3): 705 - 713. [Abstract] [Full Text]
|
|
|
|
|
|
K. Severinov T7 RNA polymerase transcription complex: What you see is not what you get PNAS, January 2, 2001; 98(1): 5 - 7. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME |
