JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M509376200 on October 6, 2005

J. Biol. Chem., Vol. 280, Issue 48, 39860-39869, December 2, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/48/39860    most recent
M509376200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Weaver, J. R.
Right arrow Articles by Goodrich, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Weaver, J. R.
Right arrow Articles by Goodrich, J. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

The Sequence at Specific Positions in the Early Transcribed Region Sets the Rate of Transcript Synthesis by RNA Polymerase II in Vitro*

Jessica R. Weaver1, Jennifer F. Kugel2, and James A. Goodrich3

From the Department of Chemistry and Biochemistry, University of Colorado, Boulder, Colorado 80309-0215

Received for publication, August 24, 2005 , and in revised form, October 3, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
To further understand the mechanism of promoter escape by RNA polymerase II, we have systematically investigated the effect of core promoter sequence on the rate of transcript synthesis in vitro. Chimeric and mutant promoters were made by swapping sequences between the human interleukin-2 promoter and the adenovirus major late promoter, which exhibit different rates of transcript synthesis. Kinetic studies at these promoters revealed that sequences downstream of the start sites set the rate of transcript synthesis. Specifically, the sequences at +2 and +7/+8 are critical for determining the rate; when either +2 is a C (nontemplate strand) or +7/+8 is a TT (nontemplate strand), transcript synthesis is slow. At +7/+8, the thermodynamic stability of the RNA:DNA hybrid controls the overall rate of transcript synthesis. Our data support a model in which the rate-limiting step during transcript synthesis by RNA polymerase II in vitro occurs at the point in the reaction at which early ternary complexes transform into elongation complexes.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Eukaryotic mRNA transcription is a multistep reaction involving numerous protein factors and DNA elements. RNA polymerase II (Pol II)4 catalyzes mRNA synthesis; however, promoter-specific initiation of transcription requires additional proteins known as the general transcription factors, including TFIIA, TFIIB, TFIID, TFIIE, TFIIF, and TFIIH (1). In addition, a multitude of other factors, including regulators, co-regulators, chromatin, and chromatin modifying factors, work in concert with Pol II and the general transcription factors to regulate levels of transcription (2–4). The DNA elements that control mRNA transcription are contained in the promoters of genes, which include regulatory elements that bind activators and repressors and core promoter elements that bind Pol II and the general transcription factors.

During the early stages of transcript synthesis unstable initiated complexes that synthesize 2- and 3-nt RNAs (5–9) transform into stable elongation complexes that complete synthesis of the transcript (9, 10). This transition occurs during promoter escape, and involves a complex series of molecular transformations including changes in the melted region of DNA, the release of general transcription factors, and formation of the full-length RNA:DNA hybrid (6, 9, 11–15). The primary transformations in the DNA involve the formation and initial movement of the transcription bubble. Permanganate footprinting studies showed that the DNA melts from –9 to +2 upon assembly of the preinitiation complex (6). Early transcript synthesis then results in continuous extension of the downstream edge of the melted region. By synthesis of an 11-nt RNA the upstream edge of the transcription bubble re-anneals through at least –2 (6). Recent studies indicate that this re-annealing cannot occur until the bubble is 18 residues in length and a 7-nt RNA is synthesized (15). In addition to the DNA, the protein components of the preinitiation complex change en route to formation of elongation complexes. For example, TFIIB and TFIIE release from the ternary complex by synthesis of a 10-nt RNA (16). This is consistent with structural studies of yeast Pol II and TFIIB, which revealed that the presence of the N terminus of TFIIB in the active center of the polymerase will compete with transcript RNAs longer than 10 nt (17). The RNA also influences structural transformations in early transcription. The RNA forms a 8-bp RNA:DNA hybrid in the elongation complex, which is thought to be a primary stability determinant for the elongation complex (18, 19). Both abortive RNA synthesis and transcript slippage, which are most prevalent prior to formation of a stable elongation complex, are influenced by the RNA:DNA hybrid stability (20–22). Clearly numerous complex events must occur during promoter escape as preinitiation complexes transform into elongation complexes.

We previously found promoter escape to be the slowest of the transcript synthesis steps in vitro at both the adenovirus major late promoter (AdMLP) (9, 13) and the interleukin-2 promoter (IL-2P) (23). There was, however, a significant difference in the rate constants for transcript synthesis at the two promoters. This suggested that characteristics unique to each of the promoters led to different rates of transcript synthesis. Whereas the role of the core promoter in recruiting the transcription machinery is well established, its influence on the synthesis steps of the transcription reaction has not been investigated in detail. Studies to address this question could provide fundamental insight into the mechanism of the early steps of transcript synthesis, which are important for establishing the elongation complex.

To further understand the mechanism of promoter escape we have investigated the effect of the core promoter sequence on the rate of transcription. Taking advantage of the difference in the rates of transcript synthesis at the IL-2P and AdMLP, we made chimeric and mutant promoters that swapped IL-2P and AdMLP sequences to determine which sequences set the rate of transcript synthesis. We found that the sequence downstream of the transcription start site is the primary determinant of the rate of transcript synthesis, with the sequences at +2 and +7/+8 being critical. Investigating the mechanism by which sequence affects the rate of transcription revealed that the thermodynamic stability of the RNA:DNA hybrid at +7/+8 could control the rate. These studies support a model in which the rate-limiting step during transcript synthesis in vitro occurs at the point at which the elongation complex first forms.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Preparation of Promoters and Transcription Reagents—Plasmid pBS-IL-2-Luc, which has been described previously (5), contains the region of the IL-2 promoter from –326 to +46. Plasmid pBS-IL-2(MLP(–40/+46))-Luc was generated using PCR to replace the region from –41 to +45 of pBS-IL-2-Luc with the region from –40 to +46 of pBSMLP-G-less (consists of AdMLP sequence from –40 to +10 and a G-less cassette from +11 to +42) (5). To make each of the other chimeric plasmids listed in TABLE ONE, PCR was used to generate two DNA fragments, both of which contained the designated mutations in overlapping regions. The two DNA fragments were then used together as a PCR template to generate a third DNA fragment. This final DNA fragment was digested using XhoI and HindIII and ligated into the pBS-IL-2-Luc plasmid.


View this table:
[in this window]
[in a new window]
  		TABLE ONE
Promoters used (in the order presented in the text)

 
Recombinant human TFIIB and TFIIF and native human Pol II were prepared as previously described (13). Human TBP was expressed in Escherichia coli by induction with 1 mM isopropyl {beta}-D-thiogalactopyranoside for 30 min at 37 °C. TBP was purified by ion-exchange chromatography in buffer H (20 mM Hepes (pH 7.9), 10% glycerol, 0.05 mM EDTA, 2 mM MgCl2, 1 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, and 0.1 mM sodium metabisulfite) containing varying concentrations of KCl. First, cellular lysate was passed over a DE52 column and washed with buffer H containing 0.2 M KCl. The flow-through containing TBP was loaded on a P11 column and eluted with buffer H containing 1 M KCl. The eluate was ammonium sulfate precipitated (25% saturation). The pellet was resuspended in buffer H and dialyzed in buffer H containing 0.1 M KCl. TBP was further purified on a Mono S column with a linear gradient from 0.1 to 0.5 M KCl in buffer H. Fractions containing TBP were pooled and dialyzed against buffer DB100 (20 mM Tris·HCl (pH 7.9), 20% glycerol, 1 mM dithiothreitol, 0.1 mM EDTA, 100 mM KCl).

In Vitro Transcription Reactions—Transcription reactions were performed in buffer A containing 10 mM Tris·HCl (pH 7.9), 10 mM Hepes (pH 8.0), 10% glycerol, 1 mM dithiothreitol, 4 mM MgCl2, 50 mM KCl, 50 µg/ml bovine serum albumin, and 15 units/ml RNA Guard (Amersham Biosciences). Reactions contained the following concentrations of transcription factors and template DNA: 3 nM TBP, 14 nM TFIIB, 2 nM TFIIF, 3 nM Pol II, and 1–1.2 nM DNA template. Unless otherwise indicated, nucleotides were added to the assay at final concentrations of 625 µM ATP, 625 µM UTP, 25 µM [{alpha}-32P]CTP (5 µCi per reaction), and 100 µM 3'-O-methylguanosine triphosphate (3'-Me-GTP, Amersham Biosciences). When used, dinucleotides (UpA, CpU, and ApC (Dharmacon)) were added at final concentrations of 0.3–1.0 mM. In all cases, control reactions omitting the dinucleotide were performed and run on gels adjacent to reactions containing dinucleotide. Transcripts that initiated with dinucleotide migrated at a different position in the gel from those that initiated with an NTP, thereby enabling us to quantitate only the transcripts that initiated with dinucleotide. When used, 625 µM rTTP (TriLink Biotechnologies) or 5-bromo-UTP (Ambion) replaced the UTP in transcription reactions.

In general, experiments to measure rates of transcript synthesis were performed as follows, with differences indicated in the figures and figure legends. Transcription factors and Pol II were incubated in buffer A (200 µl) at 37 °C for 2 min. Promoter DNA in buffer A (200 µl) was then added to this mixture and the incubation was continued for 16 min at 37 °C to allow preinitiation complexes to form. At this point, competitor DNA (40 µl of pBSMLP(+1G), in which the +1 position of the AdMLP is changed from an A to a G (5)), was added to a final concentration of 9 nM. After 4 min, a solution of nucleotides (40 µl) was added to initiate transcription, and the reaction was incubated at 37 °C. At variable time points, 24-µl portions of the reaction were transferred to tubes containing 100 µl of stop solution (3.1 M ammonium acetate, 10 µg of carrier yeast RNA, 15 µg of proteinase K, and 20 mM EDTA). RNA was ethanol precipitated and resolved by 14% denaturing PAGE.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 1.
The core regions of AdMLP and IL-2P set different rates of transcript synthesis. A, schematic showing the method used to measure rates of RNA synthesis in vitro. B, the rate of transcript synthesis at the AdMLP is 5-fold slower than at the IL-2P. Representative data for transcript synthesis at the IL-2P (squares, solid curve) and the AdMLP (circles, dashed curve) are shown.

 
Rate Constant Calculations—Full-length RNA product was quantitated using an Amersham Biosciences PhosphorImager. After subtracting background, the PhosphorImager units (PI) were plotted versus time in seconds and fit to the equation: PI = PImax(1-e–kt) using Kaleidagraph® or Prism® to solve for PImax and k (the rate constant). The plots in Figs. 1B, 4, 5B, and 6 were generated by dividing the PI units at each time point by the PImax for the data set.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
The Early Transcribed Regions of the IL-2P and AdMLP Set Different Rates of Transcript Synthesis—We previously found that transcript synthesis (i.e. all steps that occur after the addition of nucleotides to preinitiation complexes) was significantly slower at the AdMLP than at the IL-2P in a transcription system reconstituted from TFIIA, TFIID, TFIIB, TFIIF, Pol II, TFIIE, and TFIIH (23). Our goals were to understand what properties of the IL-2P and AdMLP cause different rates of transcript synthesis and to further decipher the mechanism of promoter escape. To begin we constructed two plasmids that are identical in sequence except for an 85-bp region containing either the core promoter region of IL-2P or the core promoter region of AdMLP-G-less (hereafter referred to as AdMLP). To study the most basic mechanism of promoter escape, we chose to work in a human reconstituted system containing the minimum set of factors required to obtain promoter-specific transcription from a negatively supercoiled plasmid: TBP, TFIIB, TFIIF, and core Pol II (9, 10, 13).

A schematic depicting the method used to measure the rate of a single round of transcript synthesis is shown in Fig. 1A. Pol II, TBP, TFIIB, and TFIIF were incubated for 2 min after which promoter DNA was added. Preinitiation complexes were allowed to form, and then a competitor DNA was added to sequester unbound transcription factors and Pol II, thereby preventing them from initiating additional rounds of transcription beyond the first. A nucleotide mixture including ATP, CTP, UTP, and 3'-Me-GTP was added to allow transcription through the first C in the template strand. Under these conditions, transcription from the IL-2P and AdMLP produced 27- and 49-nt products, respectively (transcribed sequences are shown in Fig. 2A). The rate constants for transcript synthesis at the IL-2P and AdMLP were 11.5 x 10–3 s–1 and 2.31 x 10–3 s–1, respectively (Fig. 1B and TABLE ONE). Therefore, transcript synthesis at the AdMLP is 5-fold slower than at the IL-2P in a minimal in vitro transcription system. This difference is not because of the different transcript lengths produced from the two promoters (data not shown). Given that the two plasmids differ only in the characteristics of their core promoters, we conclude that the core regions of IL-2P and AdMLP set different rates of transcript synthesis in vitro.



View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 2.
The early transcribed regions set the rates of transcript synthesis at the IL-2P and AdMLP. A, alignment of the nontemplate strand sequences of the core regions of the human IL-2P and the AdMLP. Locations of TATA box, Inr, and the TFIIB-recognition element (BRE) (for AdMLP) are indicated, with bases having identity to the consensus sequences shown in bold. The spacing between the TATA box and start site is indicated for each promoter. B, the downstream regions, as opposed to the upstream regions or the spacing between the TATA boxes and start sites, set the rate of transcript synthesis. Shown on the left are schematics of the chimeric promoters. Rate constants are the average of at least three experiments, and the errors represent 1 S.D. C, a strong correlation exists between the rate constants for transcript synthesis and the origin of the downstream region in the chimeric promoters. The rate constants from B are plotted versus the origins of the upstream region, downstream region, and promoter spacing. Filled circles indicate IL-2P origin and open circles indicate AdMLP origin. D, the +1 to +10 regions of the core promoters set the rate of transcript synthesis. Schematic of the A(+1/+10) promoter. Representative data were plotted and fit to a single exponential.

 
Fig. 2A shows an alignment of the core regions of the IL-2P and AdMLP sequences (nontemplate strand). There is a difference in the spacing between the TATA box and the start site at the two promoters: 24 bp at IL-2P and 23 bp at AdMLP. Therefore, the AdMLP sequence shown in Fig. 2A has a space inserted between –5 and –6, which results in an alignment having the maximum amount of sequence identity between the two promoters. Even so, outside of the TATA boxes, little sequence similarity exists between IL-2P and AdMLP.

Two possible causes of the difference in the rates of transcript synthesis between the IL-2P and AdMLP are: 1) differences in the sequences of the two core promoters, and 2) the difference in the spacing between the TATA box and transcription start site in the two promoters. To begin to test the effects of spacing and sequence on the rate of transcript synthesis we constructed four chimeric promoters (Fig. 2B) that differed in the upstream core promoter region, downstream core promoter region, and the spacing between the TATA box and the transcription start site. The name of each promoter contains three terms (in order): 1) the origin of the region upstream of the start site (A for AdMLP or I for IL-2P), 2) the origin of the region downstream of the start site (A or I), and 3) the spacing between the TATA box and the start site in the chimera (23 or 24 bp). To change the spacing between the TATA box and the start site, the junctions between the upstream and downstream regions were varied. The rate constants for transcript synthesis at the four chimeric promoters are plotted in Fig. 2B, along with the rate constants for IL-2P and AdMLP from Fig. 1. The rate constants at the AI23 and AI24 promoters approximated that of wild-type IL-2P. The rate constant at the IA23 promoter was similar to wild-type AdMLP, and the rate constant at the IA24 promoter was intermediate, although most similar to AdMLP. Considering the origins of the upstream region, downstream region, and spacing, only the origin of the downstream region was predictive of the rate of transcript synthesis measured on each chimeric promoter. When the downstream region was from AdMLP the rate was slow; when the downstream region was from IL-2P the rate was fast.

This conclusion is more easily visualized in the plot in Fig. 2C. Here, the six rate constants are plotted in each of three columns: upstream region, downstream region, and spacing. The shading of each point depicts the origin of the upstream region, downstream region, or spacing, with filled circles for IL-2P and open circles for AdMLP. In each column, the points cluster into two groups: those with larger (IL-2P-like) rate constants and those with smaller (AdMLP-like) rate constants. It is only in the downstream region column that the open circles and filled circles segregate into two separate clusters, as highlighted by the gray ovals. Therefore, the downstream region is the primary determinant of the rate of transcript synthesis in this system. As an additional test of the effect of spacing, we measured the rate of transcript synthesis on the wild-type IL-2P in the presence of UpA, which forces transcription to initiate at –1. Under these conditions, the rate constant for transcript synthesis was 8.80 x 10–3 s–1 (TABLE ONE), confirming that altering the spacing from 24 to 23 bp does not substantially affect the rate of transcript synthesis at the IL-2P.



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 3.
Sequences at +2 and +8 control the rate of transcript synthesis. A, alignment of the +1 to +10 transcribed sequences of three promoters. B, the sequences at +2 and +8 in the A(+1/+10) promoter set a slow rate of transcript synthesis. Transcribed sequences (nontemplate strand) are shown for the A(+1/+10) promoter, the IL-2P initiated with UpA, and the mutant promoters. Dashes indicate positions in the mutant promoters that are identical to the A(+1/+10) promoter. The bar plot shows the rate constants for transcript synthesis measured on the four mutant promoters (y axis is logarithmic). For reference, the upper dashed line indicates the rate constant previously measured at IL-2P initiated with UpA, whereas the lower dashed line indicates the rate constant previously measured at the A(+1/+10) promoter. Rate constants are the average of at least three experiments and the error bars represent 1 S.D. C, the sequences at +2 and +8 together set the rate of transcript synthesis. The bar plot shows the rate constants for transcript synthesis measured on the three mutant promoters (y axis is logarithmic).

 
We previously found that the slow step of transcript synthesis in our system occurs after synthesis of a 4-nt RNA at both promoters and is complete by synthesis of a 15-nt RNA at the AdMLP (9, 13, 23). We therefore hypothesized that the early transcribed regions of the AdMLP and IL-2P are responsible for setting the two different rates of transcript synthesis observed at these promoters. To test this we created a new construct, A(+1/+10), that is based on the IA23 promoter, which had the slowest rate of transcript synthesis of the four chimeric promoters tested in Fig. 2B. A(+1/+10) differs from IA23 in that the sequence downstream of +10 is from IL-2P instead of AdMLP (see Fig. 2, B and D). In effect, the A(+1/+10) promoter consists of the IL-2P promoter with the region from –1 to +9 replaced with the +1 to +10 region of AdMLP. The rate constant for transcript synthesis measured at the A(+1/+10) promoter was 1.63 x 10–3 s–1 (Fig. 2D). Therefore, replacing the early transcribed region of IL-2P with that of AdMLP decreases the rate of transcript synthesis to that observed at AdMLP. To further test the role of the early transcribed region in setting the rate of transcript synthesis, we created a promoter in which the +1 to +10 region of the AdMLP is replaced with the IL-2P sequence from –1 to +9 (I(–1/+9)). The rate constant for transcript synthesis at this promoter was 3.5-fold greater than at the AdMLP (TABLE ONE). Lastly, control experiments showed that the slow rate of transcript synthesis at A(+1/+10) was independent of the nucleotide used for labeling. Together these experiments show that the early transcribed regions of AdMLP and IL-2P set different rates of transcript synthesis.

The Sequences at Positions +2 and +7/+8 Set the Rate of Transcript Synthesis—We next wanted to determine whether sequences at specific positions in the early transcribed region set different rates of transcript synthesis. To devise a strategy for making mutations in the +1 to +10 region we aligned the sequences of the 9 promoters tested thus far. To initially limit the number of possibilities, we assumed that the sequence at each position (+1 through +10) acted independent of neighboring sequences. This analysis showed that at two positions, +2 and +5, the sequences between promoters with fast and slow rates were always different. The alignment of three transcribed sequences in Fig. 3A depicts this observation. We therefore hypothesized that the sequences at +2 and/or +5 would contribute to setting the different rates of transcript synthesis. Because the alignment of the +1 to +10 regions of AdMLP and IL-2P initiated with UpA differed at only five positions (+1, +2, +5, +8, +9) we made a series of four mutant promoters that progressively changed the A(+1/+10) promoter at each of these positions to the sequence of IL-2P initiated with UpA as shown in Fig. 3B. Comparing rate constants for transcript synthesis at these mutant promoters would reveal the contributions of the sequences at the five positions, including +2 and +5, to setting the rate of transcript synthesis.

The bar plot in Fig. 3B displays the rate constants for transcript synthesis at the four mutant promoters (also see TABLE ONE). For comparison, the rate constants previously measured for A(+1/+10) and IL-2P initiated with UpA are displayed as the lower and upper dashed lines, respectively. As the sequence is progressively changed from AdMLP toward IL-2P (from left to right) the rate constant for transcript synthesis increases. The progression jumps in two distinct places. Both of these jumps occur with a sequence change at a single position in the early transcribed region: 1) changing +8 from T to C (nontemplate strand) increases the rate of transcript synthesis 2-fold, and 2) changing +2 from C to A increases the rate of transcript synthesis an additional 2.5-fold. Changes in the sequence at +1, +5, and +9 did not substantially affect the rate of transcript synthesis. Hence, as hypothesized, the sequence at +2 contributes to setting the rate of transcript synthesis; all promoters with slow rates of transcript synthesis have a C at +2. A strong correlation does not exist between the sequence at +8 and the rate of transcript synthesis. This suggested that although the sequence at +8 contributes to setting the rate of transcript synthesis, it does not function independently, but is likely influenced by the sequence at one or more other positions.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 4.
The sequences at +2 and +7/+8 in IL-2P set the rate of transcript synthesis. A, changing the sequence at +2 in IL-2P from a T to a C (nontemplate strand) slows the rate of transcript synthesis. Reactions in panels A and C were initiated with the dinucleotide ApC. The IL-2P curve from Fig. 1B is shown for reference (dashed curve) in all panels. B, changing the sequence at +7 in IL-2P from a C to a T (nontemplate strand) slows the rate of transcript synthesis. C, mutating both +2 and +7 in IL-2P slows the rate of transcript synthesis, but the effects of the mutations are not additive.

 
Given the way the series of mutants in Fig. 3B was made, it was possible that the sequence changes at +2 and +8 did not act alone, but required changes at other positions. To test this we changed the sequences at +2 and +8 in the A(+1/+10) promoter together and individually to the corresponding sequences of IL-2P initiated with UpA as shown in Fig. 3C. Changing both +2 and +8 caused an 8-fold increase in the rate constant for transcript synthesis. Thus, changes at +2 and +8 dramatically altered the rate of transcription, and changes at other positions (e.g. +5 and +9) were not required to observe this effect. When +2 and +8 were individually mutated, the rate constants for transcript synthesis increased to 5.06 x 10–3 s–1 and 2.41 x 10–3 s–1, respectively. Neither individual mutation had the same level of effect as the double mutant, therefore the sequences at +2 and +8 both contribute to setting the rate of transcript synthesis.

We next asked whether the sequences at +2 and +8 of the IL-2P were important for setting a fast rate of transcript synthesis when transcription was initiated from the natural start site. To test the sequence at +2 we made a single point mutation in IL-2P that replaced the T at +2 with a C (I-T2C), which we predicted would slow transcript synthesis. To alleviate the possibility that transcript synthesis would be slow because of the reduced CTP concentration used for labeling the transcript, reactions were initiated with the dinucleotide ApC. Under these conditions, introducing a Cat +2 decreased the rate constant for transcript synthesis 2.7-fold (Fig. 4A, solid curve) compared with IL-2P initiated at the natural start site (dashed curve). Thus, the sequence at +2 of IL-2P is important for setting the rate of transcript synthesis.

Testing the effect of the sequence at +8 of IL-2P was not as straight forward for two reasons. First, it appeared that the effect of the sequence at +8 was influenced by sequences at another yet unidentified position(s). Second, when the natural start sites of IL-2P and AdMLP are aligned, both promoters contain a T (nontemplate strand) at +8, yet the rates of transcript synthesis were different (Fig. 3A). We hypothesized that sequences neighboring +8 influence the rate of transcript synthesis (i.e. sequence at +7 or +9). When the AdMLP and IL-2P are aligned relative to their natural start sites, the sequences at +9 are identical, whereas those at +7 differ (refer to Fig. 3A). We therefore changed the IL-2P sequence at +7/+8 from CT to TT, which is the +7/+8 sequence in the AdMLP. The single point mutation in the IL-2P (I-C7T) decreased the rate constant for transcript synthesis 3.8-fold (Fig. 4B, compare the dashed and solid curves). Taken together with earlier results, this indicates that the sequences at +7 and +8 cooperate to set the rate of transcript synthesis. Specifically, when the sequence at +7/+8 is TT then the transcript synthesis is relatively slow.

To determine whether the respective decreases in the rates of transcript synthesis caused by placing a C at +2 and TT at +7/+8 were additive, we made a construct containing two mutations: I-T2C/C7T. The rate of transcript synthesis at this promoter was 3.99 x 10–3 s–1 (Fig. 4C), which is similar to that measured at the I-T2C promoter and greater than that at the I-C7T promoter. Therefore, placing a C at +2 and TT at +7/+8 of IL-2P individually decreased the rate of transcript synthesis, but together did not have an additive effect.

The Effects of the C and TT Sequences on the Rate of Transcript Synthesis Are Position Specific—Fig. 5A depicts a two-dimensional matrix with the three +2 sequences in the tested promoters listed along the bottom and the three +7/+8 sequences in the tested promoters listed along the left side. The average rate constants (with standard deviations if obtained) for the seven different combinations of +2 and +7/+8 sequences tested are shown. Among these promoters, transcript synthesis was always relatively slow when either the sequence at +2 was a C or the sequence at +7/+8 was TT. By contrast, an A or a T at +2 permitted a fast rate of transcript synthesis (as long as the sequence at +7/+8 was not TT) and TC or CT at +7/+8 permitted a fast rate of transcription (as long as the sequence at +2 was not a C).

In examining the sequences of promoters tested thus far (TABLE ONE), it is apparent that C and TT are found at many positions other than +2 and +7/+8 in promoters with fast rates. For example, the IL-2P naturally contains C residues in its nontemplate strand at +3, +5, +7, and +9. Moreover, when transcription was initiated at the IL-2P using UpA, the C residues were moved to +4, +6, +8, and +10 relative to the new transcription start site, and the rate of transcript synthesis remained fast. Hence, having a C at any position from +3 through +10 did not cause a slow rate of transcript synthesis. Similarly, in considering the sequence TT, the IL-2P naturally has the sequence TTT at +10/+11/+12 and the I-C7T has a TT at +6/+7. At both of these promoters, the rate of transcript synthesis is fast, hence TT at +6/+7, +10/+11, and +11/+12 does not cause a slow rate of transcript synthesis. This strongly suggests that the effects of the C and TT are unique to +2 and +7/+8, respectively.

To further test the effect of C and TT at other positions in the early transcribed region, we used the dinucleotide CpU to shift the transcriptional start site upstream two positions in the I-C7T promoter. This moves TT out of +7/+8 and places a C at +1 and TT at both +8/+9 and +9/+10 relative to the new start site (sequences are shown in Fig. 5B). In doing so, the rate constant for transcript synthesis increased almost 3-fold (Fig. 5B, triangles). Hence, neither a C at +1 nor TT at +8/+9 and +9/+10 caused a slow rate of transcript synthesis. As a control, UpA was used to shift the start site upstream by only one position, which maintains a TT at +7/+8 in the I-C7T promoter. Under these conditions, the rate of transcript synthesis remained slow (Fig. 5B, squares). Together our data show that a C causes a slow rate of transcript synthesis when present at +2, but not at +1 nor at any position from +3 through +10. In addition, TT causes a slow rate of transcript synthesis when present at +7/+8, but not at +6/+7, +8/+9, +9/+10, +10/+11, and +11/+12.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 5.
ACat +2 and TT at +7/+8 are unique in their ability to set a slow rate of transcript synthesis. A, having either a +2 C or +7/+8 TT causes a slow rate of transcript synthesis. See text for explanation of the matrix. B, shifting the start site of transcription, thereby changing the sequence at +7/+8, can change the rate of transcript synthesis. At the I-C7T promoter, initiating transcription at –2 using CpU moves TT out of +7/+8 and increases the rate of transcript synthesis (triangles). Initiating transcription at –1 using UpA retains a TT at +7/+8 and the rate of transcript synthesis remains slow (squares). The I-C7T sequence (nontemplate strand) and the sequences of the RNA transcripts are shown. Positions +7 and +8 are underlined.

 
The Slow Rate of Transcript Synthesis Set by the +7/+8 TT Is Because of Low Thermodynamic Stability of the RNA:DNA Hybrid at This Position—We were intrigued by the observation that TT (nontemplate strand) at +7/+8 causes a slow rate of transcript synthesis because previous measurements of the thermodynamic stability of RNA:DNA hybrids have shown that rUrU:dAdA is by far the least stable of all of the 16 different 2-bp RNA:DNA hybrids (24). By contrast, the DNA:DNA hybrid dTdT:dAdA has moderate stability compared with other sequences (25). Moreover, experiments have shown that a (rU:dA)5 stretch was at least 200-fold less stable than the corresponding (rA:dT)5 stretch, leading the authors to propose that these unstable RNA:DNA hybrids may have biological functions (26). We hypothesized that the slow rate of transcript synthesis at promoters containing a +7/+8TT was because of the low thermodynamic stability of the resulting rUrU: dAdA hybrid at these positions. To begin to test this we made two additional point mutations. First, we asked whether changing +8 from a T to a C could suppress the effect of the I-C7T mutation, which resulted in a 4-fold decrease in the rate of transcript synthesis (Fig. 4B). We made a double mutant promoter (I-C7T/T8C) that has TC instead of TT at +7/+8; therefore, the RNA:DNA hybrid at +7/+8 would be rUrC: dAdG instead of rUrU:dAdA. This would change the {Delta}G0 of the RNA: DNA hybrid at +7/+8 from –0.2 to –1.5 kcal/mol, making it significantly more stable (24). Introducing the second mutation increased the rate of transcript synthesis 4.4-fold from 3.02 to 13.6 x 10–3 s–1 (Fig. 6A), thereby reversing the effect of the I-C7T mutation. This confirms the importance of TT at +7/+8 in setting a slow rate of transcription. Second, we mutated +7 in IL-2P to an A, which changed the sequence at +7/+8 from CT to AT. Notably, this mutation would not change the {Delta}G0 for the RNA:DNA hybrid at +7/+8, however, it would significantly decrease the {Delta}G0 for the DNA:DNA hybrid at +7/+8 (from –1.16 to –0.73 kcal/mol) (25). The rate constant for transcript synthesis at this promoter (I-C7A) was 24.1 x 10–3 s–1 (TABLE ONE). Thus, decreasing the thermodynamic stability of the DNA:DNA hybrid at +7/+8 does not slow the rate of transcript synthesis.



View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 6.
The slow rate of transcript synthesis at promoters containing +7/+8 TT is because of the low thermodynamic stability of the rUrU:dAdA RNA:DNA hybrid. A, changing the sequence at +7/+8 from TT to TC increases the rate of transcript synthesis. Data obtained with the I-C7T/T8C promoter were fit to a single exponential (solid curve). The I-C7T curve from Fig. 4B is shown for reference (dashed curve). The sequences (nontemplate strand) of the I-C7T and I-C7T/T8C promoters are shown. Positions +7 and +8 are underlined. B, increasing the stability of the RNA:DNA hybrid at +7/+8 with the use of rTTP increases the rate of transcript synthesis at the I-C7T promoter. The I-C7T curve obtained with UTP (from Fig. 4B) is shown for reference (dashed curve).

 
To directly test whether the effect of TT at +7/+8 on the rate of transcript synthesis was because of the low stability of the RNA:DNA hybrid at these positions, we asked whether stabilizing the RNA:DNA hybrid at +7/+8 using the modified nucleotides 5-methyl-UTP (rTTP) and 5-bromo-UTP (rBr-UTP) in place of UTP would increase the rate of transcript synthesis. Both poly(rT:rA) and poly(rBrU:rA) have significantly higher Tm values than does poly(rU:rA) (27). In addition, rBr-UTP has previously been used to stabilize RNA:DNA hybrids during transcription (20, 28). We replaced UTP with either rTTP or rBr-UTP in reactions containing the I-C7T promoter, which has TT at +7/+8. Both modified nucleotides increased the rate of transcript synthesis: rTTP increased the rate 3-fold (Fig. 6B) and rBr-UTP increased the rate 2.5-fold (TABLE ONE). Hence, these hybrid-stabilizing nucleotides reversed the majority of the 3.8-fold decrease caused by the C7T mutation. To ensure that the increased rate of transcript synthesis in the presence of the modified nucleotides was because of the TT at +7/+8, we measured the rate constants for transcript synthesis at the wild-type IL-2P with rTTP or rBr-UTP in place of UTP. Neither modified nucleotide affected the rate constant for transcript synthesis at the IL-2P (TABLE ONE). We conclude that the low thermodynamic stability of the rUrU:dAdA hybrid that results from a TT at +7/+8 causes a slow rate of transcript synthesis.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
To better understand the mechanism of early transcription we used chimeric and mutant promoters to investigate how core promoter sequence influences the rate of transcription in vitro. We found that the region of the core promoter downstream of the start site sets a fast or slow rate of transcript synthesis at the IL-2P or AdMLP, respectively, with sequences at +2 and +7/+8 being critical determinants of the rate. A slow rate of transcript synthesis occurs when +2 is C or when +7/+8 is TT. We also found that the rate of transcript synthesis depends on the thermodynamic stability of the RNA:DNA hybrid at +7/+8. These studies establish that the core promoter sequence can set the rate of RNA synthesis in vitro.

The Sequence at +7/+8 Influences the Rate of Transcript Synthesis—The mechanism by which the sequence at +7/+8 sets the rate of transcript synthesis involves the thermodynamic stability of the RNA: DNA hybrid. Transcript synthesis is slow when the RNA:DNA hybrid at +7/+8 is rUrU:dAdA, which has a significantly lower thermodynamic stability than all of the 15 other 2-bp RNA:DNA hybrid sequences (24). Pal and Luse (21, 22) previously found that RNA:DNA hybrid stability in the early transcribed region contributes to transcript slippage, and the amount of transcript slippage decreased at +8/+9. We have observed transcript slippage at some of the promoters studied here, however, the amount of slippage does not correlate with the rate of transcript synthesis.

We propose that the transformation between an early ternary complex and a functional elongation complex is rate-limiting in our in vitro transcription system, and that the stability of the RNA:DNA hybrid at +7/+8 can control the rate of this transformation. A less stable hybrid at +7/+8 decreases the rate at which a functional elongation complex can form. Our data indicate that the eighth nucleotide must be added to the growing RNA transcript and the RNA:DNA hybrid at +7/+8 must form before the rate-limiting step can occur. Therefore, the first point in the reaction at which the rate-limiting step could occur is just prior to or during translocation of the Pol II active site after synthesis of an 8-nt RNA. Notably, synthesis of an 8-nt RNA is also required to form the 8-bp RNA:DNA hybrid found in the elongation complex, which is predicted to be a primary determinant of elongation complex stability (18, 19). Hence, translocation after synthesis of an 8-nt RNA is the first point in the transcription reaction at which: 1) the rate-limiting step can occur, 2) a full-length RNA:DNA hybrid can be established, and 3) a stable elongation complex can form. It is possible that the rate-limiting step occurs later in the reaction, between +9 and +15. For this to occur, the +7/+8 positions of the RNA:DNA hybrid would have to be uniquely situated such that the rate of a critical downstream transformation in the ternary complex is affected. This is because a TT (and hence low thermodynamic stability in the RNA:DNA hybrid) at positions other than +7/+8 did not result in a slow rate of transcript synthesis.

Luse and colleagues (15) recently monitored bubble collapse (the abrupt reannealing of the upstream portion of the melted DNA) during early transcription. They propose that bubble collapse defines the transition between initiation complexes and stable elongation complexes, and show that bubble collapse can occur only after the transcript is minimally 7 nt in length (15). The kinetic studies presented here are consistent with these findings.

The Sequence at +2 Influences the Rate of Transcript Synthesis—The mechanism by which the sequence at +2 can set the rate of transcript synthesis is less clear than that at +7/+8. Position +2, unlike +7/+8, is not in the region in which the rate-limiting step occurs (between synthesis of a 4- and 15-nt RNA (9, 13)). The sequence at +2 has not previously been implicated in affecting transcription. Position +2 is contained within the Inr, but the sequence at this position does not affect Inr function; any of the four possible base pairs is allowed at +2 according to the Inr consensus sequence (29–31). We speculate that this enables the sequence at +2 to be utilized to control the rate of transcript synthesis. It has been shown that the TAF subunits of TFIID make contacts within this region of DNA (32–34). It is possible that specific contacts between position +2 and a TAF(s) could be important for regulating the rate of transcription. In this scenario, the Inr and the sequence at +2 could independently control both preinitiation complex assembly and the rate of transcript synthesis, respectively.

We considered the possibility that the effect of the +2 sequence was because of the RNA:DNA hybrid stability at either +1/+2 or +2/+3. Using a variety of di- and trinucleotides to alter the RNA:DNA hybrid stability at +1/+2 and +2/+3, we did not find a correlation with the rate of transcript synthesis (data not shown). It is possible that the sequence at +2 affects the rate of transcript synthesis via direct contact of the RNA transcript or the RNA:DNA hybrid with the polymerase. It is unlikely that the +2 sequence affects the rate of transcript synthesis via the stability of the DNA:DNA hybrid because the DNA at position +2 is melted in ternary complexes containing 4–10 nt RNAs (6).

We propose that the sequence at +2 of either the RNA or the DNA controls the initial separation of the 5'-end of the RNA from the DNA during the formation of an elongation complex, and that this event can limit the overall rate of transcript synthesis. The RNA:DNA hybrid length in the elongation complex is likely to oscillate between 7 and 8 bp during each cycle of catalysis and translocation. This is evident from the high resolution crystal structure of a yeast Pol II elongation complex paused by omission of UTP, which was needed for addition of the 11th nucleotide to the growing RNA transcript (19). In this complex, the RNA was 10 nt in length, the active site of the polymerase had translocated to register 11, and the 3 nucleotides at the 5'-end of the RNA were separated from the DNA, leaving a 7-bp RNA:DNA hybrid. This structure can be extrapolated to the initial formation of the elongation complex. During translocation of the polymerase active site after addition of the 8th nucleotide, the 5'-end of the RNA would separate from the DNA, and a 7-bp RNA:DNA hybrid would transiently exist until the 9th nucleotide is added to the growing RNA. We propose that the sequence at +2 can influence the initial separation of the 5' nucleotide of the RNA from the DNA, thereby affecting the overall rate of transcript synthesis.

It is also possible that the sequence at +2 affects the rate at which TFIIB releases from the transcribing complex. The yeast Pol II/TFIIB co-crystal structure shows that the N terminus of TFIIB projects into the active center of the polymerase where it occupies space that will ultimately contain the growing RNA transcript; it is predicted that TFIIB must vacate this space before the RNA is 9–10 nt in length (17). Interactions between the RNA at position +2 and either Pol II or TFIIB could influence the rate at which TFIIB releases. Luse and colleagues (15) predict that the displacement of TFIIB from the active center of the polymerase is coupled with bubble collapse and formation of an elongation complex. It is possible that TFIIB release is coupled to the slow step in transcript synthesis, and changing the rate of release could influence the overall rate of transcript synthesis.

When the effects of the +2 and +7/+8 sequences are considered together, we arrive at a concerted model for the rate-limiting step of transcript synthesis in a minimal system. We propose that a single step limits the rate of transcript synthesis in vitro and that this step occurs during translocation of the polymerase active site after addition of the 8th nucleotide, which is the first point at which an elongation complex can form. It is likely that bubble collapse and the release of TFIIB occur at this point in the reaction as well (15). The rate of this transformation depends on: 1) the stability of the RNA:DNA hybrid at +7/+8, which anchors the 3' end of the RNA to the DNA, and 2) the sequence of the RNA or DNA at +2, which could affect the initial separation of the 5'-end of the RNA from the DNA. This concerted model explains our observation that sequences at two distinct and separate positions can affect the rate of a single step in the transcription reaction.

It will be interesting to determine how general this principle is for determining the rate of transcription at other eukaryotic promoters. Once the rates of transcript synthesis have been determined at a number of promoters, a comparative analysis of the sequences at +2 and +7/+8 should reveal whether a C or TT, respectively, could generally affect the rate of transcript synthesis. Whereas the use of a minimal transcription system enabled us to determine that the sequences at +2 and +7/+8 are important for setting the rate of transcript synthesis, future experiments will investigate the influence of these sequences on transcription in cells.


   FOOTNOTES
 
* This work was supported in part by United States Public Health Service Grant GM55235 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 Supported in part by National Institutes of Health Predoctoral Training Grant T32 GM07135. Back

2 To whom correspondence may be addressed. Tel.: 303-735-0955; Fax: 303-492-5894; E-mail: jennifer.kugel{at}colorado.edu. 3 To whom correspondence may be addressed. Tel.: 303-492-3273; Fax: 303-492-5894; E-mail: james.goodrich{at}colorado.edu.

4 The abbreviations used are: Pol II, RNA polymerase II; TFII, transcription factor of RNA polymerase II; nt, nucleotides; AdMLP, adenovirus major late promoter; IL-2P, interleukin-2 promoter; TBP, TATA-binding protein; 3'-Me-GTP, 3'-O-methylguanosine triphosphate; PI, PhosphorImager units; Inr, initiator; Br-UTP, 5-bromo-UTP. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Orphanides, G., Lagrange, T., and Reinberg, D. (1996) Genes Dev. 10, 2657–2683[Free Full Text]
  2. Naar, A. M., Lemon, B. D., and Tjian, R. (2001) Annu. Rev. Biochem. 70, 475–501[CrossRef][Medline] [Order article via Infotrieve]
  3. Workman, J. L., and Kingston, R. E. (1998) Annu. Rev. Biochem. 67, 545–579[CrossRef][Medline] [Order article via Infotrieve]
  4. Malik, S., and Roeder, R. G. (2000) Trends Biochem. Sci. 25, 277–283[CrossRef][Medline] [Order article via Infotrieve]
  5. Goodrich, J. A., and Tjian, R. (1994) Cell 77, 145–156[CrossRef][Medline] [Order article via Infotrieve]
  6. Holstege, F. C. P., Fiedler, U., and Timmers, H. T. M. (1997) EMBO J. 16, 7468–7480[CrossRef][Medline] [Order article via Infotrieve]
  7. Luse, D. S., Kochel, T., Kuempel, E. D., Copploa, J. A., and Cai, H. (1987) J. Biol. Chem. 262, 289–297[Abstract/Free Full Text]
  8. Cai, H., and Luse, D. S. (1987) J. Biol. Chem. 262, 298–304[Abstract/Free Full Text]
  9. Kugel, J. F., and Goodrich, J. A. (2000) J. Biol. Chem. 275, 40483–40491[Abstract/Free Full Text]
  10. Kugel, J. F., and Goodrich, J. A. (2002) Mol. Cell. Biol. 22, 762–773[Abstract/Free Full Text]
  11. Samkurashvili, I., and Luse, D. S. (1998) Mol. Cell. Biol. 18, 5343–5354[Abstract/Free Full Text]
  12. Dvir, A. (2002) Biochim. Biophys. Acta 1577, 208–223[Medline] [Order article via Infotrieve]
  13. Kugel, J. F., and Goodrich, J. A. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 9232–9237[Abstract/Free Full Text]
  14. Dvir, A., Tan, S., Conaway, J., and Conaway, R. (1997) J. Biol. Chem. 272, 28175–28178[Abstract/Free Full Text]
  15. Pal, M., Ponticelli, A. S., and Luse, D. S. (2005) Mol. Cell. 19, 101–110[CrossRef][Medline] [Order article via Infotrieve]
  16. Zawel, L., Kumar, K. P., and Reinberg, D. (1995) Genes Dev. 9, 1479–1490[Abstract/Free Full Text]
  17. Bushnell, D. A., Westover, K. D., Davis, R. E., and Kornberg, R. D. (2004) Science 303, 983–988[Abstract/Free Full Text]
  18. Kireeva, M. L., Komissarova, N., Waugh, D. S., and Kashlev, M. (2000) J. Biol. Chem. 275, 6530–6536[Abstract/Free Full Text]
  19. Westover, K. D., Bushnell, D. A., and Kornberg, R. D. (2004) Science 303, 1014–1016[Abstract/Free Full Text]
  20. Keene, R. G., and Luse, D. S. (1999) J. Biol. Chem. 274, 11526–11534[Abstract/Free Full Text]
  21. Pal, M., and Luse, D. S. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 5700–5705[Abstract/Free Full Text]
  22. Pal, M., and Luse, D. S. (2002) Mol. Cell. Biol. 22, 30–40[Abstract/Free Full Text]
  23. Ferguson, H. A., Kugel, J. F., and Goodrich, J. A. (2001) J. Mol. Biol. 314, 993–1006[CrossRef][Medline] [Order article via Infotrieve]
  24. Sugimoto, N., Nakano, S., Katoh, M., Matsumura, A., Nakamuta, H., Ohmichi, T., Yoneyama, M., and Sasaki, M. (1995) Biochemistry 34, 11211–11216[CrossRef][Medline] [Order article via Infotrieve]
  25. Santa Lucia, J., Jr., Allawi, H. T., and Seneviratne, P. A. (1996) Biochemistry 35, 3555–3562[CrossRef][Medline] [Order article via Infotrieve]
  26. Martin, F. H., and Tinoco, I., Jr. (1980) Nucleic Acids Res. 8, 2295–2299[Abstract/Free Full Text]
  27. Felsenfeld, G., and Miles, H. T. (1967) Annu. Rev. Biochem. 36, 407–448[CrossRef]
  28. Nudler, E., Mustaev, A., Lukhtanov, E., and Goldfarb, A. (1997) Cell 89, 33–41[CrossRef][Medline] [Order article via Infotrieve]
  29. Javahery, R., Khachi, A., Lo, K., Zenzie-Gregory, B., and Smale, S. T. (1994) Mol. Cell. Biol. 14, 116–127[Abstract/Free Full Text]
  30. Lo, K., and Smale, S. T. (1996) Gene (Amst.) 182, 13–22[CrossRef][Medline] [Order article via Infotrieve]
  31. Smale, S. T., Jain, A., Kaufmann, J., Emami, K. H., Lo, K., and Garraway, I. P. (1998) Cold Spring Harbor Symp. Quant. Biol. 63, 21–31[CrossRef][Medline] [Order article via Infotrieve]
  32. Kaufmann, J., and Smale, S. T. (1994) Genes Dev. 8, 821–829[Abstract/Free Full Text]
  33. Verrijzer, C. P., Chen, J.-L., Yokomori, K., and Tjian, R. (1995) Cell 81, 1115–1125[CrossRef][Medline] [Order article via Infotrieve]
  34. Kaufmann, J., Verrijzer, C. P., Shao, J., and Smale, S. T. (1996) Genes Dev. 10, 873–886[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 RNAHome page
C. A. Espinoza, J. A. Goodrich, and J. F. Kugel
Characterization of the structure, function, and mechanism of B2 RNA, an ncRNA repressor of RNA polymerase II transcription
RNA, April 1, 2007; 13(4): 583 - 596.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
E. V. Mirkin, D. Castro Roa, E. Nudler, and S. M. Mirkin
Transcription regulatory elements are punctuation marks for DNA replication
PNAS, May 9, 2006; 103(19): 7276 - 7281.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/48/39860    most recent
M509376200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Weaver, J. R.
Right arrow Articles by Goodrich, J. A.