54 is an alternative bacterial transcription factor that directs transcription of specific subsets of genes (reviewed by Magasanik(1989), Kustu et al.(1989), and Merrick(1993)). 54 is the only sigma factor that is not a member of the 70 family of proteins, as defined by sequence similarity (Merrick and Gibbins, 1985; Lonetto et al., 1992). Regulation of 54-dependent promoters differs from regulation of 70-dependent promoters (Gralla, 1991; Collado-Vides et al., 1991) despite the fact that both sigma factors bind the same core RNA polymerase. All known 54-dependent promoters are controlled by activator proteins rather than by repressors (Collado-Vides et al., 1991). The activators are generally enhancer-binding proteins (Reitzer and Magasanik, 1986), which work from locations that are too remote to activate transcription in 70-dependent systems.

Certain steps in the transcription cycle at 54 promoters differ from analogous steps at 70 promoters. 54 can bind certain promoters without being part of holoenzyme (Buck and Cannon, 1992), whereas 70 cannot (Dombroski et al., 1992). The two holoenzymes recognize different sequence elements (see review by Merrick, 1993). In addition the 54 promoter elements are located in different positions near -12 and -24 (Morett and Buck, 1989). ATP hydrolysis is required to form open complexes at the 54 promoters (Popham et al., 1989; Weiss et al., 1991) but not at 70 promoters.

Recently we identified another difference; 54 can remain bound to a promoter after RNA polymerase begins transcription elongation (Tintut etal., 1995). This is in contrast to the well established 70 transcription cycle, in which 70 is released when an elongation complex is formed. The observation has led us to attempt to investigate the transition from open promoter complex to elongation complex for 54 holoenzyme. The process is well known for 70 promoters (Hansen and McClure 1980; Carpousis and Gralla, 1985; Krummel and Chamberlin, 1989). At those promoters RNA synthesis begins with an abortive cycling phase, in which promoter-bound holoenzyme synthesizes short RNAs (Carpousis and Gralla, 1980). Subsequently, a longer RNA of length 10 or 11 is made, and at this point polymerase is released from the promoter (Carpousis and Gralla, 1985; Krummel and Chamberlin, 1989). 70 is released both from the polymerase and from the DNA, and the complex becomes fully committed to elongation.

In this paper we investigate the transition of 54 holoenzyme from open complex to elongation complex at the Escherichia coli glnAp2 promoter. Popham et al(1989) showed that open complexes formed at an analogous promoter are capable of synthesizing short, probably abortive, RNAs. We confirm this finding in a different context and go on to characterize the transition from abortive to productive transcription. The results indicate that several important aspects of this transition differ from those observed in prior studies of 70 systems.

EXPERIMENTAL PROCEDURES

Materials

Methods

Footprinting

For DNase I digestion, 2 µl of DNase I (0.45 µg/ml, including 45 mM MgCl and 22.5 mM CaCl) was added for 30 s at 37 °C, followed by an addition of 2 µl of 0.5 M EDTA to stop the reaction. For KMnO footprinting, 4 µl of 92.5 mM KMnO was added for 1 min at 37 °C, and the reactions were quenched by adding 6 µl of -mercaptoethanol.

The modified DNA was isolated by extracting with phenol and passed through a 1-ml syringe packed with G50-80 (Sigma) equilibrated with water. The samples were subjected to primer extensions using P-labeled primers (Sasse-Dwight and Gralla, 1991). The DNA products of the Taq reaction was isolated on 6% denatured polyacrylamide gel.

Abortive Initiation

RESULTS

The initial goal was to monitor the progress of 54 holoenzyme as it transcribed from the glnAp2 promoter. In order to do that, the polymerase was stalled at different positions along the template as it attempted to move downstream during transcription initiation. The stalling was accomplished by omitting a subset of the required nucleotides. The expected positions of stalling are shown in Table 1along with the nucleotide combinations used. Three assays were used to characterize these stalled complexes: DNase I footprinting to locate the polymerase, abortive initiation to assay abortive transcripts, and permanganate probing to locate melted regions.

Plasmid pLR1, which contains the 54-dependent promoter glnAp2 with upstream NtrC enhancer sites, was used (Reitzer and Magasanik, 1986). Primer extension footprinting methods, which allow probing of both DNA strands of the same sample, were used (see Gralla(1985)). This method also allows the use of supercoiled DNA, which may be an important parameter in this system.

DNase I Footprinting

Top Strand

The footprint pattern of the top strand of the glnAp2 promoter is shown in Fig. 1A. Lane 1 shows the protection pattern of a closed complex formed from 54 and core polymerase. The closed complex footprint covers from -34 to -2 (compare lane1versus the lane2 control pattern). This encompasses both the -12 and -24 promoter recognition elements but not the transcribed region, in agreement with prior experiments (Popham et al., 1989). When phosphorylated NtrC and ATP were also present, the open complex formed extends into the transcribed region to position +23 (lane3), covering at total of approximately 57 bases.

Figure 1: DNase I footprints of the glnAp2 promoter. A, top strand; B, bottom strand. Lane1, closed complex (cc); lane2, control; lane3, open complex (oc); lane4, polymerase stalled at +2 position; lane5, +7 position; lane6, +18 position; lane7, >+18 position. The arrow indicates the +1 start site and the direction of transcription. Dideoxy sequencing reactions are shown on the left.

Next, the holoenzyme was allowed to transcribe to different positions by adding different combinations of nucleotides (see Table 1). In order to prevent binding by new polymerases from solution, free proteins were titrated with a competitor promoter DNA. The competitor contains the 54-dependent tightly binding R. meliloti nifH promoter (Buck and Cannon, 1992) carried on a 160-base pair fragment. A 100-fold excess of this competitor was added after open complexes were formed with glnAp2 DNA, but before the addition of nucleotides, which caused the polymerase to begin RNA synthesis. The lack of protection in control lane2, where competitor was mixed with glnAp2 before addition of proteins, confirms that a sufficient amount of the competitor is present to titrate excess proteins.

In the first experiment dinucleotide primer UpA and CTP were added to the open complex to allow the polymerase to form a single bond, creating the product UpApC, where ``A'' is the initiating nucleotide (see ``Transcription'' below) (Popham et al., 1989). This corresponds to incorporating the nucleotide at position +2 in the natural mRNA. Under these conditions the result shows no change in the DNase protection pattern compared to the open complex (lane4versuslane3). Not surprisingly, the polymerase can make the first bond of the mRNA without moving along the DNA.

By contrast, the polymerase did change its interaction with DNA when GTP was added to allow transcription as far as the +7 position (see Table 1). This is demonstrated by the shortened protection seen in lane5. The polymerase now fully protects the DNA only from -2 to +23 with additional partial protection as far as -11 upstream and perhaps to +27 downstream. The overall protection pattern corresponds to less than a 40-base pair region, compared to open and first bond-making complexes, both of which protect approximately 57 base pairs. The comparison indicates that when the polymerase transcribes to the +7 position, it is released from its upstream promoter contacts. The important promoter element near -24 is no longer protected, and the protection near the -12 element is quite weak.

Next, the polymerase is allowed to transcribe to position +18 (see Table 1). The primary protection (lane6) now extends from +3 to +23 and is not as complete as in other cases. In addition there is partial protection from +23 to +39. Thus, the stalled polymerase covers a total of 36 bases, approximately the same extent as seen at the +7 position, but the protection is weaker. Finally, when all four nucleotides are present no protection is seen (lane7), as expected for a polymerase that has moved to downstream positions.

Bottom Strand

As also observed on the top strand, a drastic change in the footprint occurs when polymerase is allowed to move to position +7 (lane5). The protection is strong from -13 to +22. The result is similar to that from the top strand in that polymerase no longer protects the critical -24 promoter element. A transition indicative of loss of promoter interaction has occurred on both strands. The footprints shorten a little more when transcription is halted at +18 (lane6), at which positions from -3 to +31 are partially protected.

In addition, lanes 5, 6, and 7, corresponding to stalling at +7, at +18, and steady-state transcription, show partial protection of the region upstream from the start site (lesser intensity in these lanes compared to the control lane2). This is a reflection of the weak DNase protection due to not being released from this region, as shown by Tintut et al.(1995).

A pictorial summary of the DNase I footprinting results is presented in Fig. 7. The main point at this stage is that a transition releasing polymerase from the promoter elements has occurred before polymerase reaches position +7.

Figure 7: Summary of data. Solidbar, strong protection; stripedbar, weak protection; openoval, melted region.

Transcription

The production of UpApC was used as a measure of abortive initiation. This product is made using dinucleotide primer UpA and P-labeled CTP as the only source of nucleotides (Fig. 2). When all required abortive transcription components were present, an abortive initiation product was seen (lane7 with ATP). The specificity of this reaction was confirmed by showing its dependence on various required components. No signal is seen with the omission of: NtrC (lane1), carbamyl phosphate to phosphorylate NtrC (lane2), 54 (lane3), core polymerase (lane4), ATP (lane5), and UpA (lane6). Parallel experiments (data not shown) to produce long transcripts under these same conditions confirmed the specificity of glnA transcription.

Figure 2: Abortive initiation. The component that is left out from the transcriptions is as follows: lane1, NtrC; lane2, CBP; lane3, 54; lane4, core polymerase; lane5, ATP; lane6, UpA. In lane7 all the required transcription components are present.

Next we characterized the amount of abortive initiation that occurs when polymerase is permitted to move to various positions (see Table 1) as described in the DNase footprinting experiments. The goal is to learn at what position the polymerase ceases to make large amounts of abortive initiation products. Open complexes were formed, and a subset of nucleotides containing radioactive CTP (see Table 1) were added. The reactions were allowed to proceed for 10 min to achieve the steady state conditions. Reaction products were loaded directly onto an acrylamide gel, separated by electrophoresis, and detected by autoradiography.

In the presence of dinucleotide UpA and labeled CTP, the polymerase produced a large amount of abortive product UpApC (see band indicated by the arrow in Fig. 3A, lane 1). When the polymerase was allowed to move as far as position +7, a much smaller amount of abortive product was made (much lighter band in the same position of lane2, quantified below). In addition small amounts of slightly longer products were made, as expected; we have not definitively characterized the size of these longer products. The data indicate that few abortive products are made when polymerase is allowed to move as far as position +7.

Figure 3: Abortive initiation at certain stalled positions. A, lane1, +2 position; lane2, +7 position; lane3, +18 position; lane 4, >+18 position. B, lane5, +2 position; lane6, +3 position; lane7, +18 position. The arrow indicates the position of the abortive product UpApC.

When additional nucleotides were present (see Table 1) to promote further downstream movement of the polymerase, the amount of abortive product was only slightly lessened (lane3; band in same position denoted by arrow). Two major longer products were also produced under these conditions. Calibration using short RNAs (data not shown) suggests that these two products correspond to RNAs of the expected length, 19, and also of length 22; the 22-mer probably occurs from the mis-incorporating the omitted UTP, followed by stalling at the next position of UTP where two UTPs must be incorporated. This has been seen previously (Carpousis and Gralla, 1985). Finally, when all four nucleotides were present, the amount of abortive product was essentially unchanged (lane4).

This qualitative analysis indicates that a transition from a frequent abortive initiation state to a lower abortive state has occurred between position +2 and +7 (lane1versuslane2). Note that the +7 position is the same point at which DNase footprinting revealed a physical change in the state of the transcription complex.

One further combination of nucleotides was used to narrow the position at which the transition occurs. By the use of UpA, CTP, and 3-methyl-GTP the polymerase could be stalled at position +3. The result (Fig. 3B, lane 6) indicates that the polymerase stalled at +3 still produces large amounts of abortive products (quantified below). That is, the amount is comparable to that seen when transcription is stalled at +2 (lane5) but much more than that seen when transcription is stalled later (lane7). Thus the transition out of frequent abortive mode seems to occur after the polymerase reaches +3 but before it reaches +7.

For each of several samples where the polymerase stalled at positions +2, +3, +7, +18, and >+18, the amount of abortive products and productive transcripts were quantified using a PhosphorImager. From the quantitative data, the relative molar amount was determined by normalizing to the number of radiolabeled CTPs incorporated (Table 2). Note that the longer products incorporate more than one radioactive cytidine, and thus their molar amount is over-represented on the autoradiographs. The molar excess of abortive RNA UpApC (compared to productive RNA, 18 + 19 nucleotides long, produced in a parallel experiment) was determined. The data (Table 2) showed that during the 10-min abortive initiation period, a 50-fold excess of abortive product UpApC was made from polymerase stalled at the +2 and +3 positions.

The relative amount of abortive products associated with stalling at each position is plotted in Fig. 4to display the transition from frequent abortive phase to more stable elongation complex. When transcription is artificially stalled at position +2, the amount of abortive product UpApC made is approximately 15-fold higher than the amount of abortive product made at position +7. This huge overproduction of UpApC is maintained when polymerase is artificially stalled at position +3. Apparently, a transition out of abortive mode has occurred at position +7, as the polymerase aborts transcription rarely after reaching this position. This much lower production of UpApC is maintained under conditions in which polymerase can reach position +18 and, indeed, under conditions of free transcription. The quantitative analysis confirms that a transition away from abortive mode has occurred between positions +3 and +7.

Figure 4: The total of all abortive RNAs produced (see Table 2) is plotted against the position at which transcription is stalled.

A key parameter in the synthesis of productive transcripts is the probability of aborting synthesis at a particular position (Carpousis, 1983; Carpousis and Gralla, 1985). The probability can be defined as the amount of RNA shorter than a certain length divided by the sum of all RNA products made (abortive plus longer). For example, the data showed that UpApC constitutes approximately 75% of all RNAs made under various conditions (Table 2). That is, in several experiments where the polymerase was stalled at position +7 or +18, the probability of aborting synthesis at +2 was calculated to be 0.75. Thus, approximately three out of four times, polymerase aborts synthesis after forming UpApC and must restart; the remaining time, it goes forward to make productive transcript. This predicts that only a small (roughly 2-fold) excess of abortive product will be seen during productive transcription, as is observed. Note that huge amounts of abortive products are only produced when polymerase is artificially stalled at a position that is not far enough to be associated with the transition to elongation mode. Significant amounts of abortive transcripts have been seen previously under productive transcription conditions, but this is probably a consequence of the use of low concentrations of CTP, GTP, and UTP (Popham et al., 1989).

In order to confirm that this abortive phase is not restricted to dinucleotide-primed synthesis, we repeated selected experiments using the normal initiating nucleotide ATP. Abortive transcripts could not be detected directly because the presence of the triphosphate end caused them to migrate within a region dominated by radioactive products from the unincorporated labeled CTP. Thus after transcription the samples were subjected to calf intestine alkaline phosphatase to cleave off the 5`-phosphate groups (Jacob et al., 1994) before loading onto gels. Under these conditions abortive product can be seen. Lane1 of Fig. 5shows that a large amount of abortive product is made when ATP and labeled CTP are used to form pppApC. When the same experiment is done in the presence of additional GTP and UTP to allow elongation, the result shown in lane2 is obtained. The much reduced amount of abortive product under these conditions is similar to the result obtained with dinucleotide-primed synthesis. This confirms that abortive initiation can occur using the natural initiating nucleotide ATP.

Figure 5: Abortive initiation using the initiating nucleotide ATP. Lane1, +2 position; lane2, >+18 position.

KMnOFootprinting

In order to follow the single-stranded regions in this transition, the exposed single-stranded DNA was probed with KMnO. KMnO reacts selectively with single-stranded thymines and has been used previously to detect the melted transcription bubble at this promoter (Sasse-Dwight and Gralla, 1988; 1990). In these experiments the bottom strand of the glnAp2 promoter was probed. As a control, open complexes, formed as described above, yield a strong permanganate signal (Fig. 6, lane 2) corresponding to bottom strand thymines at positions -9 and +1. As expected, this reactivity was absent in closed complexes (Fig. 6, lane 1).

Figure 6: KMnO footprinting of the glnAp2 promoter. Lane1, closed complex (CC); lane2, open complex (OC); lane3, polymerasestalled at +2 position; lane4, +7 position; lane5, +18 position; lane6, >+18 position. The bottom strand was probed. Arrows indicate the positions of thymines.

Next, samples were probed with permanganate using the same conditions described above to move the polymerase to various positions along the template. No change in pattern was observed when complexes forming product UpApC were probed (Fig. 6, compare lane3 with the open complex signal of lane2). Recall that at this stage the polymerase has not moved (as assayed using DNase footprinting; Fig. 1) and is in abortive mode (as assayed by abortive initiation; Fig. 3A).

By contrast, when the polymerase is stalled at position +7, new permanganate hypersites extended to positions +7 and +9 (Fig. 6, lane4). Recall that under these conditions the polymerase has been largely released from abortive mode (see above). When polymerase is stalled further downstream in the +18 to +21 region, additional permanganate-sensitive sites are seen corresponding to thymines within this region (lane5). As expected, no hypersites are seen in the transcribed region when polymerase transcribes in the presence of all nucleotides (lane6). The results provide evidence that a melted bubble moves downstream with the elongating polymerase. The differences in intensities of permanganate-sensitive bands is probably due to different environments surrounding thymines at different positions.

Note that under all conditions the transcription start site remains open (Fig. 6, bracket indicating bands in lanes 2-6). This is in agreement with prior experiments, not involving abortive initiation, analogous to those in lanes 2,5, and 6 but probing the top strand with permanganate (Tintut et al., 1995). The result supports the view that, as polymerase is released from the promoter, the original open complex bubble splits into two bubbles; one moves with the elongating polymerase and the other remains behind with 54.

DISCUSSION

These experiments describe the transition from abortive initiation mode to elongation mode for 54 holoenzyme at the glnAp2 promoter (see summary in Fig. 7). The holoenzyme makes the first mRNA bond without moving from the promoter. After making this short RNA product, the data show that there is a 75% chance that the synthesis will be aborted and started anew. Thus, even when all elongation substrates are present, this abortive product accumulates in modest excess over long RNA. However, longer abortive products do not accumulate in significant amount under these conditions, leading to relatively efficient productive transcription.

The polymerase is capable of abortive initiation when RNA synthesis reaches position +3 but loses this property prior to position +7 (Fig. 4). At this stage, footprinting shows that the polymerase is released from the contacts that hold it to the promoter region (see summary in Fig. 7). This leads to a reduction in the size of the protected region by approximately 20 base pairs. Permanganate probing shows that the transcription bubble has moved forward to cover the position that is transcribed. Thus the polymerase appears to have reached elongation mode prior to +7 position of the glnA gene. Probing of complexes stalled further downstream also show elongation complexes with characteristic of shortened footprints and transcription bubbles covering the point of synthesis.

This pathway may be compared to the analogous pathway used by the common 70 form of RNA polymerase holoenzyme. The comparison reveals both similarities and differences. The main similarity is that both pathways require the polymerase to pass through an abortive initiation mode prior to being committed to transcription elongation (Carpousis and Gralla, 1985). The molecular basis for this requirement is not known, but it was speculated previously to be related to the primer-independent nature of mRNA initiation (Carpousis and Gralla, 1985). The lack of primers, common to RNA polymerases but not DNA polymerases, may cause difficulty in initiating RNA synthesis. The abortive mode can be thought of as a phase in which a primer is created with some difficulty. For example, cycling to produce abortive RNAs may occur because these short RNAs are associated with the ``loose product site'' and are not translocated to the ``tight product site'' (see Chamberlin(1992-1993)) until a stable ternary complex is formed (Mustaev et al., 1994). Alternatively, the need to break strong contacts between polymerase and promoter may also retard the transition to elongation phase.

There are several differences in this pathway compared to prior studies, which used the 70 form of polymerase. One important difference is that the 54-dependent glnAp2 promoter involves a relatively efficient transition out of abortive cycling mode. The data show that approximately 25% of the glnAp2 RNA 5` ends that are produced will end up in productive transcripts. In the two 70 cases studied, the probabilities are much lower: 2% at lacUV5 (Carpousis and Gralla, 1980; Carpousis, 1983) and less than 8% at T7A1 promoters (Krummel and Chamberlin, 1989). ()This difference is largely a consequence of what happens after the first bond of the mRNA is formed. Although the 54 polymerase is capable of abortive initiation up to position +7, it aborts primarily after forming the first bond. That is, the release of abortive products at glnAp2 during transcription is observed primarily from polymerases that have formed a single bond. By contrast, in prior studies at 70 promoters, longer RNA products are also aborted.

A second difference is that polymerase is released from the contacts that hold it to the promoter sooner in the 54 case studied here than in the 70 cases studied previously (Carpousis and Gralla, 1985; Krummel and Chamberlin, 1989). As discussed above, at glnAp2 this occurs after formation of the second bond but prior to formation of the sixth bond. In the prior cases elongation mode is not attained until the RNA is approximately twice as long, and abortive cycling occurs at several positions. One possible cause of these differences might be a lower affinity of core polymerase for 54 (see Lesley et al., 1991), allowing polymerase to dissociate more readily; the two sigma factors use different determinants to bind polymerase (Tintut et al., 1994; Tintut and Gralla, 1995).

Another difference is that the initial transcription bubble is split into two parts only during initiation by 54 holoenzyme. That is, as RNA synthesis begins the initial bubble separates into two bubbles; one moves downstream with the polymerase and the other remains transiently over the start site. We show elsewhere that this is because 54 remains promoter-bound during initial transcription due to its stronger affinity for DNA (Tintut et al., 1995).

The mechanism of 54-dependent transcription has been described as a hybrid between prokaryotic and eukaryotic mechanisms (Gralla, 1991; North et al., 1993). If this analogy extends to the pathway studied here, then one might expect eukaryotic mRNA synthesis to go through an analogous, quite brief, abortive to elongation transition (see Jacob et al.(1994), and references therein). Indirect experiments ()support this possibility, but more direct experiments will be required to test this further. 54 is also used more commonly in other bacteria, and it will be useful to learn if there are promoter- or species-specific differences in the mechanism of transcription initiation by 54 holoenzyme.

FOOTNOTES

*

This research was supported by United States Department of Health and Human Services Grant GM35754 (to J. D. G.), and by National Institutes of Health Traineeship Grant GM07185 (to Y. T.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked ``advertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§

These authors contributed equally to this work.

()

L. Hsu and M. J. Chamberlin, personal communication.

()

Y. Jiang, M. Yan, and J. D. Gralla, unpublished data.

ACKNOWLEDGEMENTS

REFERENCES

1. Buck, M., and Cannon, W. (1992) Nature 358,422-424 [CrossRef][Medline] [Order article via Infotrieve]
2. Carpousis, A. J. (1983) Abortive Initiation and Cycling in Vitro at Mutant lac Promoters , Ph.D. thesis, pp. 70-107, University of California, Los Angeles
3. Carpousis, A. J., and Gralla, J. D. (1980) Biochemistry 19,3245-3253 [CrossRef][Medline] [Order article via Infotrieve]
4. Carpousis, A. J., and Gralla, J. D. (1985) J. Mol. Biol. 183,165-177 [CrossRef][Medline] [Order article via Infotrieve]
5. Chamberlin, M. J. (1992-1993) Harvey Lect. 88,1-21
6. Collado-Vides, J., Magasanik, B., and Gralla, J. D. (1991) Microbiol. Rev. 55,371-394 [Abstract/Free Full Text]
7. Dombroski, A. J., Walter, W. A., Record, M. T., Jr., Siegele, D. A., and Gross, C. A. (1992) Cell 70,501-512 [CrossRef][Medline] [Order article via Infotrieve]
8. Feng, J., Atkinson, M. R., McCleary, W., Stock, J. B., Wanner, B. L., and Ninfa, A. J. (1992) J. Bacteriol. 174,6061-6070 [Abstract/Free Full Text]
9. Gralla, J. D. (1985) Proc. Natl. Acad. Sci. U. S. A. 82,3078-3081 [Abstract/Free Full Text]
10. Gralla, J. D. (1991) Cell 66,415-418 [CrossRef][Medline] [Order article via Infotrieve]
11. Hansen, U. M., and McClure, W. R. (1980) J. Biol. Chem. 255,9564-9570 [Abstract/Free Full Text]
12. Hunt, T. P., and Magasanik, B. (1985) Proc. Natl. acad. Sci. U. S. A. 82,8453-8457 [Abstract/Free Full Text]
13. Jacob, G. A., Kitzmiller, J. A., and Luse, D. S. (1994) J. Biol. Chem. 269,3655-3663 [Abstract/Free Full Text]
14. Krummel, B., and Chamberlin, M. J. (1989) Biochemistry 28,7829-7842 [CrossRef][Medline] [Order article via Infotrieve]
15. Kustu, S., Santero, E., Keener, J., Popham, D., and Weiss, D. (1989) Microbiol. Rev. 53,367-376 [Free Full Text]
16. Lesley, S. A., Brow, M. A., and Burgess R. R. (1991) J. Biol. Chem. 266,2632-2638 [Abstract/Free Full Text]
17. Lonetto, M., Gribskov, M., and Gross, C. A. (1992) J. Bacteriol. 174,3843-3849 [Free Full Text]
18. Magasanik, B. (1989) New Biol. 1,247-251 [Medline] [Order article via Infotrieve]
19. Merrick, M. J. (1993) Mol. Microbiol. 10,903-909 [Medline] [Order article via Infotrieve]
20. Merrick, M. J., and Gibbins, J. R. (1985) Nucleic Acids Res. 13,7607-7620 [Abstract/Free Full Text]
21. Morett, E., and Buck, M. (1989) J. Mol. Biol. 210,65-77 [CrossRef][Medline] [Order article via Infotrieve]
22. Mustaev, A., Zaychikov, E., Severinov, K., Kashlev, M., Polyakov, A., Nikiforov, V., and Goldfarb, A. (1994) Proc. Natl. Acad. Sci. U. S. A. 91,12036-12040 [Abstract/Free Full Text]
23. Ninfa, A. J., Reitzer, L., and Magasanik, B. (1987) Cell 50,1039-1046 [CrossRef][Medline] [Order article via Infotrieve]
24. North, A. K., Klose, K. E., Stedman, K. M., and Kustu, S. (1993) J. Bacteriol. 175,4267-4273 [Free Full Text]
25. Popham, D. L., Szeto, D., Keener, J., and Kustu, S. (1989) Science 243,629-635 [Abstract/Free Full Text]
26. Reitzer, L., and Magasanik, B. (1986) Cell 45,785-792 [CrossRef][Medline] [Order article via Infotrieve]
27. Sasse-Dwight, S., and Gralla, J. D. (1988) Proc. Natl. Acad. Sci. U. S. A. 85,8934-8938 [Abstract/Free Full Text]
28. Sasse-Dwight, S., and Gralla, J. D. (1990) Cell 62,945-954 [CrossRef][Medline] [Order article via Infotrieve]
29. Sasse-Dwight, S., and Gralla, J. D. (1991) Methods Enzymol. 208,146-168 [CrossRef][Medline] [Order article via Infotrieve]
30. Tintut, Y., Wong, C., Jiang, Y., Hsieh, M., and Gralla, J. D. (1994) Proc. Natl. Acad. Sci. U. S. A. 91,2120-2124 [Abstract/Free Full Text]
31. Tintut, Y., Wang, J. T., and Gralla, J. D. (1995) Genes & Dev. , in press
32. Tintut, Y., and Gralla, J. D. (1995) J. Bacteriol., in press
33. Weiss, D. S., Batut, J., Klose, K. E., Keener, J., and Kustu, S. (1991) Cell 67,155-167 [CrossRef][Medline] [Order article via Infotrieve]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				P. Gong and C. T. Martin Mechanism of Instability in Abortive Cycling by T7 RNA Polymerase J. Biol. Chem., August 18, 2006; 281(33): 23533 - 23544. [Abstract] [Full Text] [PDF]

				Q. Guo and R. Sousa Weakening of the T7 Promoter-Polymerase Interaction Facilitates Promoter Release J. Biol. Chem., April 15, 2005; 280(15): 14956 - 14961. [Abstract] [Full Text] [PDF]

				Q. Guo and R. Sousa Multiple Roles for the T7 Promoter Nontemplate Strand during Transcription Initiation and Polymerase Release J. Biol. Chem., February 4, 2005; 280(5): 3474 - 3482. [Abstract] [Full Text] [PDF]

				P. C. Burrows, K. Severinov, A. Ishihama, M. Buck, and S. R. Wigneshweraraj Mapping {sigma}54-RNA Polymerase Interactions at the -24 Consensus Promoter Element J. Biol. Chem., August 8, 2003; 278(32): 29728 - 29743. [Abstract] [Full Text] [PDF]

				P. Bhargava and G. A. Kassavetis Abortive Initiation by Saccharomyces cerevisiae RNA Polymerase III J. Biol. Chem., September 10, 1999; 274(37): 26550 - 26556. [Abstract] [Full Text] [PDF]

				T.-J. Fu, E. P. Geiduschek, and G. A. Kassavetis Abortive Initiation of Transcription at a Hybrid Promoter. AN ANALYSIS OF THE SLIDING CLAMP ACTIVATOR OF BACTERIOPHAGE T4 LATE TRANSCRIPTION, AND A COMPARISON OF THE sigma 70 AND T4 GP55 PROMOTER RECOGNITION PROTEINS J. Biol. Chem., December 18, 1998; 273(51): 34042 - 34048. [Abstract] [Full Text] [PDF]

				J. T. Wang and J. D. Gralla The Transcription Initiation Pathway of Sigma 54Mutants That Bypass the Enhancer Protein Requirement. IMPLICATIONS FOR THE MECHANISM OF ACTIVATION J. Biol. Chem., December 20, 1996; 271(51): 32707 - 32713. [Abstract] [Full Text] [PDF]

				J. T. Wang, A. Syed, M. Hsieh, and J. D. Gralla Converting Escherichia coli RNA Polymerase into an Enhancer-Responsive Enzyme: Role of an NH2-Terminal Leucine Patch in sigma54 Science, November 10, 1995; 270(5238): 992 - 994. [Abstract] [PDF]

				A. L. Gnatt, P. Cramer, J. Fu, D. A. Bushnell, and R. D. Kornberg Structural Basis of Transcription: An RNA Polymerase II Elongation Complex at 3.3 A Resolution Science, June 8, 2001; 292(5523): 1876 - 1882. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tintut, Y. Articles by Gralla, J. D. Search for Related Content PubMed PubMed Citation Articles by Tintut, Y. Articles by Gralla, J. D. Social Bookmarking                      What's this?