Originally published In Press as doi:10.1074/jbc.M112481200 on February 19, 2002
J. Biol. Chem., Vol. 277, Issue 18, 15407-15412, May 3, 2002
Generality of the Branched Pathway in Transcription Initiation by
Escherichia coli RNA Polymerase*
Motoki
Susa
§,
Ranjan
Sen¶, and
Nobuo
Shimamoto§
From the Structural Biology Center, National
Institute of Genetics, and the § Department of
Genetics, School of Life Science, The Graduate University for
Advanced Studies, 1111 Yata, Mishima, Japan, 411-8540
Received for publication, December 31, 2001, and in revised form, February 15, 2002
 |
ABSTRACT |
Transcription initiation has been assumed to be a
multi-step sequential process, although additional steps could exist.
Initiation from the T7A1 promoter, in particular, apparently behaves
in vitro in a manner that can be fully explained by the
sequential pathway. However, initiation from the
PRAL promoter has been shown to follow a
branched pathway from which a part of the enzyme-promoter complex is
arrested at the promoter raising the question as to which mechanism is
general. We found that a moribund complex, characteristic of the
arrested branch, is formed at the T7A1 promoter, especially in low salt
condition indicating that the initiation mechanism for this promoter is
also branched. The results of DNA footprinting suggested that
holoenzyme in the moribund complex is dislocated on DNA from the
position of productive complex. However, only a small fraction of the
binary complex becomes arrested at this promoter, and the
interconversion between subspecies of binary complex is apparently more
reversible than at the PRAL promoter, which
explains why the reaction pathway appears to be sequential. These
findings suggest a generality of the branched pathway mechanism, which
would resolve contradictory observations that have been reported for
various promoters.
| |
INTRODUCTION |
Transcription initiation in prokaryotes includes at least four
events: 1) the binding of holoenzyme to a promoter; 2) the isomerization of the resulting complex accompanied by strand opening; 3) the iterative synthesis and release of abortive transcripts; and 4)
the achievement of continuous elongation accompanied by the escape of
the enzyme from the promoter (for review see Ref. 1). Although these
events are required for transcription initiation in this order, they do
not necessarily represent the complete mechanism of transcription
initiation, because additional steps could exist. Nevertheless, it has
long been assumed that the events listed above is the complete
mechanism, mainly because of the lack of evidence for further
complications. This simplest mechanism may be called the sequential
pathway (Scheme 1A). In
vitro, the initiation from the bacteriophage T7A1 promoter in
particular shows the following two behaviors that are characteristic of
the sequential mechanism. Firstly, the promoter-RNA polymerase complex synthesizes a stoichiometric amount of full-length transcript in a
single-round transcription (2). Secondly, abortive synthesis does not
occur after the synthesis of full-length transcript (3), which is
consistent with the view that the transcription complex engaged in
abortive synthesis is a precursor of the complex synthesizing a long
RNA. However, these results do not prove that the sequential pathway is
applicable to initiation at all promoters or even at the T7A1 promoter
in all conditions.
View larger version (20K):
[in this window]
[in a new window]
|
Scheme 1.
Two models of the initiation pathway for
transcription. A, the sequential pathway composed of four
essential steps. B, the branched pathway that has been
applied to the initiation at the PRAL
promoter. The absence of abortive synthesis in the productive branch
has not been established. The black circle merely indicates
that a branching point exists at the level of the stage of binary
complex (within the broken rectangle) and could start at one
of the subspecies of indicated binary complex. The initiation at the
T7A1 promoter is considered to follow this model but with a rapid
conversion of moribund subspecies into productive subspecies
(broken arrows). See the Introduction for
details.
|
|
The above two behaviors characteristic of the sequential pathway are
not observed in single-round transcription from the
PRAL or LacUV5 promoter. The amount of
full-length transcript is much less than stoichiometric with the amount
of the binary complex (4), and abortive synthesis at these promoters
continues long after the completion of full-length transcription,
namely persistent abortive synthesis (5), suggesting the existence of
an initiation mechanism other than the sequential pathway. These
discrepancies are not attributable to heterogeneity in the preparation
of the RNA polymerase used for the following reasons. At the
PRAL promoter, RNA polymerase, which has been
isolated from the run-off elongation complex and re-used, displays the
same degree of abortive synthesis as the original enzyme, indicating
that a fraction of the previously productive enzyme becomes
nonproductive (5). Furthermore, the amount of full-length product in
single-round transcription from the PRAL
promoter increases to the same level of the binary complex if GreA,
GreB, and a high concentration of the initiation nucleotide are present
(6). This observation indicates that nonproductive enzyme can be
converted into a productive one. These observations imply the existence
of nonproductive pathway(s) that cause(s) the persistent abortive
synthesis and that the mechanism of initiation is branched at some
stage before abortive synthesis at these promoters. The most plausible
model for initiation from the PRAL promoter is shown in Scheme 1B (5). The moribund complex, which is
defined as the ability to synthesize only abortive transcripts, is
first generated in the nonproductive branch of initiation, the
promoter-arrested pathway. At the PRAL
promoter, the moribund complex is slowly converted into inactive
dead-end complex with a time constant of 10 min in the standard
condition (7). Because the moribund complex decays slower than the
productive complex, persistent abortive synthesis is observed. The
moribund complex is also identifiable in initiation at the
malT promoter (8). This finding indicates that the branched
pathway may arise in initiation at many promoters.
In a test of the generality of the branched pathway, one of the key
criteria would be whether or not the initiation at the T7A1 promoter,
whose mechanism appears the most sequential, actually follows a
branched pathway. Notably, a branched pathway would appear to be
sequential if the moribund complex converted into a productive complex
before converting into a dead-end complex so that most transcription
complexes would finally indulge in productive elongation. In this case,
a nearly stoichiometric amount of the full-length transcript should be
synthesized in single-round transcription and production of abortive
transcripts should cease before the full-length synthesis is completed.
Indeed, this situation occurs at the PRAL
promoter in the presence of the Gre factors and a high concentration of
initiating nucleotide (6).
Here we report that the promoter-arrested pathway can be detected in
initiation at the T7A1 promoter. We consistently observe that the
interconversion among subspecies of binary complex at the T7A1 promoter
is simply more reversible than at the PRAL promoter. These findings suggest a generality of the branched pathway
mechanism that can explain the seemingly contradictory characteristics
of various promoters.
| |
EXPERIMENTAL PROCEDURES |
The T7A1 template DNAs used in this study were prepared by PCR
using the plasmid pAR1435, which is a derivative of pBR322 with a
102-bp segment of T7 DNA containing the A1 promoter inserted at its
BamHI site (9). The other templates were described
previously by Sen et al. (3). All the transcription assays
were carried out as described previously (4, 5) in the standard
condition (50 mM Tris-HCl, pH 7.9, 10 mM
MgCl2, 0.1 M KCl) or the low salt condition (20 mM Tris-HCl, pH 7.9, 7 mM MgCl2).
The dissociation of the binary complex was measured by trapping the
dissociated enzyme with a 60-fold molar excess of the 190-bp DNA
fragment containing the PRAL promoter or with
40 µg/ml heparin. In the electrophoretic mobility shift assay, the
binary complex was formed at 37 °C for 10 min in the standard or low
salt condition in the presence of 13% glycerol. If necessary,
transcription was started by adding the substrate mixture containing 5 µM ATP, 100 µM GTP and CTP each UTP,
0.1 mg/ml heparin and then incubated for 20 min. The digestion with
HaeIII was carried out by incubation with 3.5 units of the
enzyme for an additional 30 min before electrophoresis in a 5%
polyacrylamide gel in 45 mM Tris borate, pH 8.0, buffer containing 1 mM EDTA. Exonuclease III-mediated DNA
footprinting was performed as described previously (7). In
Fe2+-induced site-specific radical cleavage, 1.2 pmol of
immobilized DNA template harboring the T7A1 or
PRAL promoter (10) and 1.0 pmol of RNA
polymerase were incubated at 37 °C for 10 min in the standard or low
salt condition for T7A1 and in the standard or standard +100
mM NaCl condition for PRAL. The
immobilized binary complex was washed with buffer lacking
MgCl2. A final concentration of 0.1 mM
Fe(NH4)2(SO4)2 was then
added, and the mixture was incubated for an additional 30 min.
Cleavage was terminated by the addition of phenol, and the reaction
mixture was analyzed in an 8% sequencing gel.
| |
RESULTS |
Persistent Abortive Synthesis at the T7A1 Promoter--
One
important line of evidence for the branched pathway at the
PR promoter is the occurrence of persistent
abortive synthesis. Because this phenomenon disappears in a high salt
condition (3), lower salt conditions might favor the formation of the
moribund complex at the T7A1 promoter. Although little sign of
persistent abortive synthesis at the T7A1 promoter has been detected in
the standard salt condition (3), we decided to test for its occurrence at low salt. Transcription from the T7A1 promoter was carried out using
a linear 154-bp DNA (Fig. 1A,
template I) and produced full-length transcripts
(74 bases or longer) together with abortive transcripts (shorter than
14 bases) in the standard salt condition (Fig. 1B). In the
low salt condition, similar amounts of long transcript were
synthesized, whereas the average length of abortive products increased
slightly. These moderate changes show that no anomalous reactions take
place at low salt.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 1.
Pulse-labeling experiments in the standard
and low salt conditions. A, the T7A1 template DNAs used
in this study. The arrows indicate transcription start
sites. The boxes indicate the region derived from T7 DNA
containing the A1 promoter and the transcription start site. Template I
was used in all assays except for the mobility shift experiments in
which template II was used. B, transcripts synthesized on
template I in single-round reactions with 30 nM DNA and RNA
polymerase in the standard or the low salt condition as indicated. The
numbers indicate the lengths of some transcripts.
C and D, the ratios of the amounts of 4-mer
(C) and 10-mer (D) to that of full-length product
plotted against the time of addition of [ -32P]ATP. The
filled and open squares represent the
ratios in the low and standard salt conditions, respectively.
|
|
The most sensitive method for the detection of persistent abortive
synthesis is a pulse-labeling assay (5) in which transcription is
started with unlabeled nucleotides and then the labeled initiating nucleotide [-32P]ATP is added at various time points.
In this assay, the amounts of the labeled full-length and abortive
transcripts indicate the residual proportion of binary complex that can
produce the corresponding transcripts at each time point. In the
sequential pathway, abortive synthesis should precede the full-length
synthesis, and thus the ratio of abortive to full-length transcripts
should not increase with the time elapsed prior to addition of labeled
ATP. In contrast, this ratio could increase in the branched pathway if
the moribund complex has a longer lifetime than the productive one. The
observed ratios for the abortive 4-mer or 10-mer transcripts were
plotted in Fig. 1, C and D, against the time of
addition of [-32P]ATP. In the standard salt condition,
the ratios increased by a maximum 4-fold for 60 s, and at low salt
they showed 5-20-fold increases. These results indicate that
persistent abortive synthesis occurs during initiation at the T7A1
promoter and that deviation from the sequential mechanism is more
extensive in the low salt condition.
Arrest of RNA Polymerase at the T7A1 Promoter--
Pulse-labeling
provides a catalytic assay and thus cannot determine what fraction of
binary complex becomes promoter-arrested. Therefore, this fraction was
measured by using an electrophoretic mobility shift assay.
Promoter-arrested complexes were formed on the 250-bp T7A1 DNA fragment
that harbors a HaeIII site at position +73 (Fig. 1A,
template II). To distinguish between promoter-arrested and run-off
elongation complexes, we removed the unlabeled downstream region of the
fragment, which would carry any of the latter complexes, by digestion
with HaeIII. The binary complex was formed in both salt
conditions (Fig. 2A, lanes 4 and 6). The observed amount of binary complex formed
decreased slightly in the low salt condition (Fig. 2B,
4NTP) despite the fact that DNA-protein complexes are generally stabilized at low salt. On the other hand, the
promoter-arrested complex was almost undetectable under the standard
condition (Fig. 2A, lane 7) but clearly existed in low salt
(lane 5), indicating that 2-3% of the binary complex was
arrested (Fig. 2B, right panel). These results suggest that
the fraction of the holoenzyme arrested at the T7A1 promoter is small
and that arrest is enhanced at the low salt condition. Therefore, the
observed persistent abortive synthesis is a sign of the formation of
moribund complex amplified by turnover of this complex. This result
explains why nearly stoichiometric amounts of full-length product are
obtained at the T7A1 promoter despite the occurrence of arrest at
this promoter.
View larger version (44K):
[in this window]
[in a new window]
|
Fig. 2.
Detection of promoter-arrested complexes by
gel shift assay. A, the 250-bp fragment (10 nM) with 32P-labeled nontemplate strand was
incubated with 25 nM RNA polymerase, and the indicated
reactions were carried out. The products were electrophoresed in 5%
polyacrylamide gel and visualized by autoradiography. B, the
amounts of RNA polymerase-promoter complexes observed as shifted band
(153 bp) + RNA polymerase (RNAP) were
quantified. Relative amounts of the complexes in the standard
(white bars) and low (gray bars) salt conditions
are indicated. The left and right
scales refer to the complexes before and after
transcription, respectively. The amount of binary complex in the
standard condition is taken as 100%.
|
|
An alternative method to detect the arrested complex is DNA
footprinting using exonuclease III (7), which can detect the limits of
the region of DNA protected by a bound protein. Thus, we examined the
footprints of polymerase bound to the DNA harboring the T7A1 promoter
in both conditions. The DNA was footprinted in the absence of
holoenzyme or after the formation of the binary complex or 20 min after
the addition of 4NTP (Fig. 3). Heparin was added with 4NTP to ensure single-round transcription. No difference between the footprints at the upstream boundary was observed between the two conditions (data not shown). In the standard condition, the
footprint on the nontemplate strand of the naked DNA (lane 2) was the same as that after transcription (lane 4).
This finding indicates that little RNA polymerase remains at the T7A1
promoter after 20 min, in agreement with the results of the mobility
shift assay. However, in the low salt condition, the footprint of the downstream boundary of the binary complex appears as enhanced bands at
positions +9 and +18 to +22 as well as reduced bands at +11 and +13
(lane 7). It should be noted that there is a similar footprint after the 20-min transcription in the low salt condition, although the bands are relatively faint (lane 8). This
observation indicates that a small fraction of the enzyme is still
sitting at the promoter, a conclusion again consistent with the finding from the mobility shift assay.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 3.
Detection of promoter-arrested complexes by
exonuclease III-DNA footprinting. The 5' end of the nontemplate
strand of the DNA harboring the T7A1 promoter was labeled with
32P and footprinted after the reactions indicated.
Positions on DNA are indicated relative to the transcription start as
+1. Digestion of the DNA was carried out in the absence (lanes
2 and 6) or presence (lanes 3 and
7) of RNA polymerase or after completion of transcription
for 20 min (lanes 4 and 8). The
reactions were performed in the standard (lanes 1-4) or the
low (lanes 5-8) salt condition. The open and
filled triangles, respectively, indicate the bands with
decreased or increased intensities upon the formation of binary complex
in the low salt condition.
|
|
There is a difference between the footprints of the binary complexes in
the two conditions. In low salt, the downstream edge of the enzyme
footprint is at position +22, whereas in the standard salt condition,
it is at +19 (lanes 3 and 7). This difference suggests that the RNA polymerase in the binary complex may be more
forward-tracked in the low salt condition than in the standard condition. Alternatively, exonuclease III may push or partially displace holoenzyme further upstream in the latter case so that the
observed differences reflect not the true boundaries but rather the
elasticity of the downstream edge. Whichever is the case, it is clear
that there is a physical difference between the downstream boundaries
in the two conditions. Here we will tentatively call the change forward
tracking. This possibility could result from a structural difference
between moribund and productive binary complexes if the earliest
branching point occurs at the stage of binary complex formation as in
the case of the PRAL promoter (6).
Mapping of the RNA Polymerase Catalytic Center onto DNA--
One
of the most important functional features of the enzyme complex is the
location of the catalytic center relative to DNA. Therefore, we
examined whether this location shifts in accord with the putative
forward tracking of RNA polymerase that was detected by exonuclease III
footprinting. According to Zaychikov et al. (11), a ferrous
ion (Fe2+) is allowed to replace the Mg2+ that
normally makes a chelate with the catalytic center. The resultant
Fe2+ chelate generates hydroxyl radicals that cleave the
template strand DNA nearby, allowing fine mapping of the catalytic
center onto the DNA in the two conditions. Fig.
4 shows the tracings of the
autoradiograms obtained by this mapping. The position of the catalytic
center of the RNA polymerase-T7A1 promoter complex in the standard
condition maps mainly at the position 2 with a secondary peak at 1
(Fig. 4A, gray line). In contrast, the mapping in the low
salt condition indicates significant forward shifting, such that the
two major peaks at 2 and 1 have similar strengths (black
line). The direction of the movement detected by the
Fe2+ cleavage is the same as was suggested by the
exonuclease III footprinting. Therefore, the binary complex with the
T7A1 promoter involves a subspecies that is positioned further forward
in the low salt condition. Because the productive subspecies should
maintain the same position of the catalytic center relative to the
template strand, the forward-tracked subspecies increasing at low salt presumably represents the moribund complex.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4.
The radical cleavage of promoter DNAs induced
by Fe2+ chelated at the catalytic center of RNA
polymerase. The profiles of band intensities of the template
strand are shown for the T7A1 promoter (A) and the
PRAL promoter (B). The intensities
have been normalized to give the same total density of cleaved bands in
both salt conditions. The broken lines and the gray
lines, respectively, indicate cleavage in the low and standard
salt conditions. Cleaved positions are indicated relative to the
transcription start site as +1.
|
|
This forward tracking of RNA polymerase at the T7A1 promoter in low
salt is in sharp contrast to that at the PRAL
promoter, where back-tracking of the binary complexes as a response to
reduced salt concentration is observed (Fig. 4B). Thus, at
the PRAL promoter, the direction of the
putative shift of moribund binary complex is the same as that of the
dead-end complex (7). Because the moribund complex is functionally
defined as a transcription complex that synthesizes only abortive
products, its structure could depend on the specific promoter involved,
giving a shift in either direction. The relative shift of the catalytic
center from the optimal position may decrease the catalytic activity of
the subspecies and thus explains why the moribund complex has a lower
affinity for the initiation nucleotide and a smaller elongating
activity (5).
Confirmation of Rapid Exchange among Subspecies of Binary Complex
Formed at the T7A1 Promoter--
We have presented two lines of
evidence that initiation at the T7A1 promoter follows a branched
pathway. The kinetic evidence is the existence of persistent abortive
synthesis shown by the pulse-labeling assay, and the biochemical
evidence is the existence of arrested complexes detected by mobility
shift and exonuclease III-footprinting assays. In both cases, the
deviation from the sequential pathway is more distinct in the low salt
condition, whereas the pathway in the standard condition appears to be
almost sequential. Despite the existence of the moribund complex,
almost no dead-end complex is formed in the standard salt condition. This result means that the moribund complex is converted into a
productive complex more rapidly than it is inactivated to form a
dead-end complex. Therefore, in the standard salt condition, the
productive and moribund subspecies formed at the T7A1 promoter are
expected to exchange rapidly, whereas those formed at the PRAL promoter exchange scarcely.
It is difficult to measure the conversion rates directly in order to
confirm this prediction. However, it is possible to measure the overall
rate of dissociation from promoters that involves a combination of the
exchange reaction and the breakdown of binary complex. Therefore, in
the standard salt condition, the overall rate at the T7A1 promoter is
expected to be more rapid than that at the
PRAL promoter. Consistent with this line of
reasoning, the reduction of arrest at the
PRAL promoter, which is produced either by
replacing 
70 with its region 3 mutant (3) or by the
addition of the Gre factors (6), is accompanied by acceleration of the
overall dissociation rate. In the case of the mutant ,
biphasic dissociation kinetics was actually observed, presumably one
phase because of exchange and the other phase because of breakdown. In
the case of the Gre factors, the dissociation kinetics is almost
monophasic, suggesting that one of the steps is too rapid to be measured.
We confirmed the existence of the expected rapid exchange at the T7A1
promoter in the standard salt condition by using the DNA fragment
harboring the PRAL promoter as a competitor
of the test promoter fragment. Holoenzyme was preincubated for 10 min with a 1.5-fold excess of the T7A1 promoter fragment to form binary complex, and then a large excess of the competitor was added. At each
time point examined, 3NTP and [-32P]ATP were added,
and reactions were continued for an additional 20 min. Because the
promoter on the competitor DNA encodes a G-start, only the initiation
from T7A1 was detected. Fig. 5 shows the
amounts of the full-length and abortive (6-mer) transcripts. The
kinetics are monophasic for both species, and the rate constants are
the same (0.2 min
1). As predicted, this value is much
larger than that for initiation at the PRAL
promoter (0.03 min1 for full-length and 9-mer species
(3)). This rapid and monophasic decay at the T7A1 promoter is
consistent with the expected rapid interconversion of the subspecies of
binary complex. A similar rapid decay at the T7A1 promoter was observed
in the low salt condition, indicating that the interconversion is still
rapid at low salt (data not shown).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 5.
Dissociation of the binary complexes at the
T7A1 promoter in standard salt condition. The amounts of the
labeled transcripts were plotted against the duration of incubation
with the competitors. The filled squares and
circles show the amounts of full-length and 6-mer abortive
transcript, respectively, when the PRAL DNA
was used as competitor. Open squares and circles
show the amounts of the full-length and the abortive transcript,
respectively, when heparin was used as competitor.
|
|
Metzger et al. (12) reported that the open complex formed at
the T7A1 promoter is rapidly inhibited by preincubation with heparin
and interpreted this as the result of rapid dissociation of the open
complex supposing a competitive role of heparin. Because an allosteric
role of heparin was postulated in a transcription study using whole T7
DNA (13), we examined whether heparin works as a competitor in our
experiment. We substituted heparin for the
PRAL DNA at a concentration sufficient to
destroy all of the activity on this template (0.04 mg/ml) when it had
been preincubated with holoenzyme. The data for heparin and the
PRAL fragment agreed within 15%, giving
almost the same decay curves (Fig. 5). Therefore, heparin is
essentially a competitive inhibitor. This finding is consistent with
the observation that for an artificial promoter with an extremely
strong affinity for RNA polymerase, a much higher concentration of
heparin (5 mg/ml) was needed for the inhibition of transcription with
no effects of heparin on transcription at concentrations as high as 1 mg/ml.1
| |
DISCUSSION |
Among many promoters for Escherichia coli RNA
polymerase, the A1 promoter of bacteriophage T7 (14) has been one of
the best studied in vitro, because it has a high affinity
for holoenzyme (2, 15) and is among the strongest (16). Using this
promoter, the order of binding of substrates (2) and the rate of
incorporation of single nucleotides were determined (2, 15) as well as the characteristics of abortive synthesis (17, 18). The concepts of
promoter clearance (16) and of initial transcribing complex/initial elongating complex (19) were established by experiments on promoters including T7A1, and the translocational movement of RNA polymerase away
from this promoter was systematically studied (20, 21). In addition to
these normal features of transcription, elongation arrest was first
discovered in the T7A1 transcription unit (22), and hydrolysis of
transcripts was also revealed (12, 23). Therefore, the T7A1 promoter is
the most representative promoter used in kinetic studies of
transcription by E. coli RNA polymerase. All of the kinetic
results obtained have been interpreted based on the assumption that the
mechanism of initiation at this promoter is sequential. However, the
new kinetic and structural evidence obtained in this study consistently
shows that initiation at the T7A1 promoter occurs by the same branched
pathway mechanism, which has recently been established for other
promoters such as the PRAL (5) and the
E. coli malT promoter (8).
This study also shows why initiation at the T7A1 promoter had appeared
to be sequential. If the moribund complex converts more rapidly into
productive complex than into dead-end complex, the branched pathway
becomes almost equivalent to the sequential pathway. The rapid and
monophasic dissociation of binary complex observed at the T7A1 promoter
suggests that there is indeed a rapid conversion between subspecies of
binary complex in standard salt conditions. A similar rapid and
monophasic dissociation was observed at the
PRAL promoter but only in the presence of the Gre factors, which it was concluded introduced reversibility between subspecies of the binary complex (6).
Abortive transcription at the T7A1 promoter was previously investigated
by a conventional kinetic assay, which detected persistent abortive
synthesis only for a misincorporation product (with A instead of G at
the fourth position) at 50 mM NaCl (12). The pulse-labeling
assay used in this study is significantly more sensitive and was able
to detect persistent synthesis of normal abortive transcripts and did
so even in the presence of 0.1 M KCl. Because persistent
abortive synthesis has also been observed at the
PRAL and lacUV5 promoters (5), we
conclude that there is no qualitative difference between initiation at
the T7A1 promoter and others. The difference is only quantitative. The
observed common features suggest that the branched pathway mechanism is general among the promoters for E. coli
70-holoenzyme.
In view of the many common features, we are inclined to hypothesize
that the earliest branching point of the reaction pathways exists at
the stage of binary complex for the T7A1 promoter as already
established for the PRAL promoter. This
hypothesis is substantiated by the results of DNA footprinting with
exonuclease III and for Fe2+-induced cleavage at the
catalytic center. In the low salt condition, a significant fraction of
binary complex is forward-tracked at the T7A1 promoter, and the
footprint of the binary complex agrees with that of the
promoter-arrested complex obtained after RNA synthesis. This agreement
suggests that the forward tracking of the footprint is because of the
formation of a significant amount of moribund binary complex. Because
the fraction of complex arrested after transcription is only 2-3%,
the major fraction of binary complex formed at an early stage is
converted into productive complex during RNA synthesis.
The footprints of catalytic center on binary complex dislocated in
opposite directions at the T7A1 and PRAL
promoters when salt concentration was reduced. Although this difference
is hard to be explained at present, speculations are possible. One
possible hypothesis on binary complex is that the moribund subspecies
is too stable in terms of translocation along DNA to escape from a
promoter, whereas the productive subspecies is in quasi-stable states
that have unfavorable positioning of DNA so that translocation is
possible. If the catalytic center locates near the 1 position in the
stable state independently of the promoter DNA and if the center in the
quasi-stable state is respectively close to 2 or +1 position at the
T7A1 or PRAL promoter, the observed opposite dislocations are merely a reflection of increased fraction of moribund
subspecies induced at low salt. However, other models are equally
possible. Irrespective of the models for the moribund complex, the
tracking should not be understood as a movement of RNA polymerase
molecule but rather as a distortion of the complex, because our
footprinting results show that at the T7A1 promoter the upstream
boundary of the enzyme does not move irrespective of the movement of
downstream boundary.
Mutations in region 3 of 70 are known to increase the
ratio of the amount of abortive transcripts to that of the full-length transcript (25). This alteration can be explained by an increase in
reversibility between the subspecies of binary complex (3). According
to protein footprinting of 70, this region is protected
in holoenzyme as well as in binary complex but exposed in the
promoter-arrested complex at the PRAL promoter (26), indicating that the structure of this region changes
upon the formation of the moribund complex. The region binds to a core
enzyme (26, 27) and lies in close proximity to promoter DNA between the
10 and the 35 boxes and at the transcription start site (+1) (28).
In fact, these features have been confirmed to exist in the crystal
structure of the complex of Thermus aquaticus holoenzyme with promoter DNA where the region 3.2 of the -subunit binds to the putative exit channel for
RNA.2 Therefore, all of these
results including those of the footprinting experiments in this study
consistently indicate that structural changes in this part of the
binary complex determine the fates of subspecies of the binary complex.
Despite its widespread use in kinetic studies, the T7A1 promoter is not
necessarily a typical one in terms of function. The binary complex at
this promoter dissociates rapidly in the presence of heparin (12),
whereas many other such complexes are insensitive to the reagent. The
heparin sensitivity of the binary complex at the T7A1 promoter was
shown in the present study to be the result of rapid dissociation of
the open binary complex and not to some allosteric effect of heparin
peculiar to this complex. The same mechanism of heparin sensitivity was
reported for the rrnBP1 promoter (24), which is one of the
strongest promoters in vivo but rather weak under standard
test conditions in vitro. Therefore, open binary complexes
at these promoters equilibrate reversibly with their free components.
These findings provide evidence against the widely held mistaken belief
that all open complexes are irreversibly formed and resistant to
heparin. The reversibility among subspecies of binary complex formed at
the T7A1 promoter is intrinsically high as compared with the complexes formed at other promoters that generate dead-end complexes.
Another atypical feature of this promoter is that it directs the
synthesis of full-length products far more efficiently than do other
promoters. It has been suggested that full-length transcript formation
is stoichiometric with the amount of preformed binary complexes at the
T7A1 promoter but not at other promoters (4). In other words, the high
efficiency is because of conversion of almost all binary complex into
the elongation complex, whereas at other promoters only a fraction of
binary complex achieve elongation, the rest being excluded. Indeed the
nearly full conversion was confirmed for the T7A1 promoter by the
results of DNA footprinting and mobility shift assays in this study. In
contrast, only a quarter to half of all binary complex clears the
PRAL promoter (4, 6).
Both the branched and sequential pathways could control the level of
transcription in the multiple-round transcription that occurs in
vivo. However, there are two distinct characteristics that are
specific to the branched pathway. The formation of dead-end complex
attenuates the level of transcription irrespective of the location of
the rate-limiting step in transcription initiation, whereas in the
sequential pathway, kinetic control is effective only at the
rate-limiting step. Therefore, initiation could be regulated by the
branched pathway, even if the rate-limiting step changes according to
the physiological environment. The second special characteristic of the
branched pathway is that it could amplify the effects of repressors and
activators. For example, if a repressor inactivates RNA
polymerase-promoter complex in the branched pathway, the inactivation
could be maintained long after dissociation of the repressor, and the
promoter could remain blocked until the arrest was relieved. Such
persistent repression would prevent the formation of a queue of RNA
polymerases in a transcription unit, which would be expected if
regulation were the result of an arrest in elongation.
| |
ACKNOWLEDGEMENTS |
We thank Dr. Richard S. Hayward for
critically reading the manuscript and Dr. Usha Padmanabhan for helpful comments.
| |
FOOTNOTES |
*
This work was supported in part by grants from Ministry of
Education (to N. S.) and an Itoh scholarship (to M. S.).The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
¶
Present address: Center for DNA Fingerprinting and
Diagnostics, ECIL Road, Nacharam, Hyderabad 500076, India.
To whom correspondence should be addressed. Tel.:
81-559-81-6843; Fax: 81-559-81-6844; E-mail:
nshima@LAB.nig.ac.jp.
Published, JBC Papers in Press, February 19, 2002, DOI 10.1074/jbc.M112481200
1
T. Gaal, R. L. Gourse, and N. Shimamoto,
unpublished result.
2
K. Murakami and S. A. Darst, unpublished result.
| |
REFERENCES |
| 1.
|
McClure, W. R.
(1985)
Annu. Rev. Biochem.
54,
171-204[CrossRef][Medline]
[Order article via Infotrieve]
|
| 2.
|
Shimamoto, N., Wu, F. Y.-H.,
and Wu, C.-H.
(1981)
Biochemistry
20,
4745-4755[CrossRef][Medline]
[Order article via Infotrieve]
|
| 3.
|
Sen, R.,
Nagai, H.,
Hernandez, V. J.,
and Shimamoto, N.
(1998)
J. Biol. Chem.
273,
9872-9877[Abstract/Free Full Text]
|
| 4.
|
Kubori, T.,
and Shimamoto, N.
(1997)
Nucleic Acids Res.
25,
2640-2647[Abstract/Free Full Text]
|
| 5.
|
Kubori, T.,
and Shimamoto, N.
(1996)
J. Mol. Biol.
256,
449-457[CrossRef][Medline]
[Order article via Infotrieve]
|
| 6.
|
Sen, R.,
Nagai, H.,
and Shimamoto, N.
(2001)
Genes Cells
6,
389-401[Abstract]
|
| 7.
|
Sen, R.,
Nagai, H.,
and Shimamoto, N.
(2000)
J. Biol. Chem.
275,
10899-10904[Abstract/Free Full Text]
|
| 8.
|
Tagami, H.,
and Aiba, H.
(1998)
EMBO J.
17,
1759-1767[CrossRef][Medline]
[Order article via Infotrieve]
|
| 9.
|
Dunn, J. J.,
and Studier, F. W.
(1983)
J. Biol. Chem.
166,
477-535
|
| 10.
|
Fujioka, M.,
Hirata, T.,
and Shimamoto, N.
(1991)
Biochemistry
30,
1801-1807[CrossRef][Medline]
[Order article via Infotrieve]
|
| 11.
|
Zaychikov, E.,
Martin, E.,
Denissova, L.,
Kozlov, M.,
Markovtsov, V.,
Kashlev, M.,
Heumann, H.,
Nikiforov, V.,
Goldfarb, A.,
and Mustaev, A.
(1996)
Science
273,
107-109[Abstract]
|
| 12.
|
Metzger, W.,
Schickor, P.,
Meier, T.,
Werel, W.,
and Heumann, H.
(1993)
J. Mol. Biol.
232,
35-49[CrossRef][Medline]
[Order article via Infotrieve]
|
| 13.
|
Pfeffer, S. R.,
Stahl, S. J.,
and Chamberlin, M. J.
(1977)
J. Biol. Chem.
252,
5403-5407[Abstract/Free Full Text]
|
| 14.
|
Stahl, S. J.,
and Chamberlin, M. J.
(1977)
J. Mol. Biol.
112,
577-601[CrossRef][Medline]
[Order article via Infotrieve]
|
| 15.
|
Nierman, W. C.,
and Chamberlin, M. J.
(1979)
J. Biol. Chem.
254,
7921-7926[Free Full Text]
|
| 16.
|
Kammerer, W.,
Deuschle, U.,
Gentz, R.,
and Bujard, H.
(1986)
EMBO J.
5,
2995-3000[Medline]
[Order article via Infotrieve]
|
| 17.
|
Smagowicz, W. J.,
and Scheit, K. H.
(1978)
Nucleic Acids Res.
5,
1919-1932[Abstract/Free Full Text]
|
| 18.
|
Oen, H., Wu, C.-W.,
Haas, R.,
and Cole, P. E.
(1979)
Biochemistry
18,
4148-4155[CrossRef][Medline]
[Order article via Infotrieve]
|
| 19.
|
Krummel, B.,
and Chamberlin, M. J.
(1989)
Biochemistry
28,
7829-7842[CrossRef][Medline]
[Order article via Infotrieve]
|
| 20.
|
Schickor, P.,
Metzger, W.,
Werel, W.,
Lederer, H.,
and Heumann, H.
(1990)
EMBO J.
9,
2215-2220[Medline]
[Order article via Infotrieve]
|
| 21.
|
Metzger, W.,
Schickor, P.,
Werel, W.,
Lederer, H.,
and Heumann, H.
(1989)
EMBO J.
8,
2745-2754[Medline]
[Order article via Infotrieve]
|
| 22.
|
Arndt, K. M.,
and Chamberlin, M. J.
(1990)
J. Biol. Chem.
213,
79-108
|
| 23.
|
Altmann, C. R.,
Solow-Cordero, D. E.,
and Chamberlin, M. J.
(1994)
Proc. Natl. Acad. Sci. U. S. A.
91,
3784-3788[Abstract/Free Full Text]
|
| 24.
|
Gaal, T.,
Bartlett, M. S.,
Ross, W.,
Turnbough, C. L., Jr.,
and Gourse, R. L.
(1997)
Science
278,
2092-2097[Abstract/Free Full Text]
|
| 25.
|
Hernandez, V. J.,
Hsu, L. M.,
and Cashel, M.
(1996)
J. Biol. Chem.
271,
18775-18779[Abstract/Free Full Text]
|
| 26.
|
Nagai, H.,
and Shimamoto, N.
(1997)
Genes Cells
2,
725-734[Abstract]
|
| 27.
|
Zhou, Y. N.,
Walter, W. A.,
and Gross, C. A.
(1992)
J. Bacteriol.
174,
5005-5012[Abstract/Free Full Text]
|
| 28.
|
Owens, J. T.,
Chmura, A. J.,
Murakami, K.,
Fujita, N.,
Ishihama, A.,
and Meares, C. F.
(1998)
Biochemistry
37,
7670-7675[CrossRef][Medline]
[Order article via Infotrieve]
|
Copyright © 2002 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:
|
|
|
|
 
Q. Wang, T. D. Tullius, and J. R. Levin
Effects of Discontinuities in the DNA Template on Abortive Initiation and Promoter Escape by Escherichia coli RNA Polymerase
J. Biol. Chem.,
September 14, 2007;
282(37):
26917 - 26927.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Hosoda, T. Matsuura, H. Kita, N. Ichihashi, K. Tsukada, and T. Yomo
Kinetic Analysis of the Entire RNA Amplification Process by Qbeta Replicase
J. Biol. Chem.,
May 25, 2007;
282(21):
15516 - 15527.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. V. Sukhodolets, S. Garges, and S. Adhya
Ribosomal protein S1 promotes transcriptional cycling
RNA,
August 1, 2006;
12(8):
1505 - 1513.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. A. Mooney and R. Landick
Tethering {sigma}70 to RNA polymerase reveals high in vivo activity of {sigma} factors and {sigma}70-dependent pausing at promoter-distal locations
Genes & Dev.,
November 15, 2003;
17(22):
2839 - 2851.
[Abstract]
[Full Text]
[PDF]
|
|
|
Copyright © 2002 by the American Society for Biochemistry and Molecular Biology.
TOP
ABSTRACT
INTRODUCTION
