The role of RNA polymerase σ subunit in promoter-independent initiation of transcription

March 15, 2004
101 (13) 4396-4400


In bacteria, initiation of transcription depends on the RNA polymerase σ subunit, which brings catalytically proficient RNA polymerase core to promoters by binding to specific DNA elements located upstream of the transcription start point. Here, we study σ-dependent synthesis of a transcript that is used to prime replication of the single-stranded genome of bacteriophage M13. We show that, in this system, σ plays no role in DNA recognition, which is accomplished solely through RNA polymerase core interaction with DNA downstream of the transcription start point. However, σ is required for full-sized transcript synthesis by allowing RNA polymerase core to escape into productive elongation. RNA polymerase σ may play a similar role during replication primer synthesis in other bacterial mobile elements whose life cycle involves a single-stranded DNA stage.
DNA-dependent RNA polymerase (RNAP) is the central enzyme of gene expression and a major target for regulation. The functional cycle of RNAP consists of transcription initiation, transcription elongation, and transcription termination. Transcription initiation can be further subdivided into promoter complex formation, abortive initiation, and promoter escape. Bacterial RNAP core (subunit composition α2ββ′ω) is catalytically proficient but is unable to initiate transcription from promoters. Binding of one of the several σ factors to RNAP core results in the formation of the holoenzyme, which can recognize and initiate transcription from promoters. The main σ factor of Escherichia coli, σ70, is the best understood protein of its class (1). The role of σ in promoter recognition is well-established: in the context of the holoenzyme, σ regions 4.2 and 2.4 directly bind the –35 and the –10 promoter consensus elements, respectively. Accumulating evidence suggests that σ factors may be involved in later stages of transcription initiation such as abortive RNA synthesis (2) and promoter escape (3), and might even participate in transcript elongation (4, 5). However, these additional roles are poorly understood, partially because of the difficulties in separating the “conventional” promoter-recognition/melting functions of σ from those that come into play later in transcription process.
The initiation of M13 minus-strand DNA synthesis is catalyzed by E. coli RNAP σ70 holoenzyme (Eσ70) (6), which acts as a primase and synthesizes an 18- to 20-nt-long primer RNA (pRNA) (79). RNAP core enzyme is inactive in this reaction (68). Eσ70 specifically recognizes a site on M13 DNA plus strand, called the minus-strand origin. The minus-strand origin contains two inverted repeats that can form two hairpins, resulting in the formation of partially double-stranded structure (7, 8) schematically shown in Fig. 1a. The σ70-dependence of pRNA synthesis from the partially double-stranded minusstrand origin DNA was explained by Higashitani and colleagues. (7, 8), who identified sequences with similarities to the –10 and –35 promoter consensus elements located at appropriate distances from the pRNA start point. Hence, the recognition of the minus-strand origin was regarded as conventional bacterial promoter recognition. Here, we show that this view is incorrect and that both RNAP core and holo enzymes bind minus-strand origin DNA and that this interaction does not involve putative promoter consensus elements. The presence of the σ70 subunit allows RNAP core to escape into elongation and complete pRNA synthesis.
Fig. 1.
σ70-dependent pRNA synthesis from M13 minus-strand origin DNA fragments. (a) The 128 fragment of M13 minus-strand origin (template 1) is presented in the pseudo double-stranded conformation. The putative –10 and –35 promoter consensus elements are indicated. The positions of 5′ ends of minus-strand origin fragments used as templates 2 and 3 in experiment of Fig. 1c are shown by vertical arrows. The minimal fragment capable of supporting pRNA synthesis is shown in red. The pRNA transcript starts at a position indicated as +1; pRNA endpoints are indicated by horizontal arrows. (b) Results of in vitro transcription reactions by σ70 RNAP holoenzyme from the entire M13 genome or from 128 base-long minus origin fragment shown in a. Reactions were performed in the presence or the absence of E. coli SSB, as indicated. (c) The pRNA synthesis reactions were performed by RNAP core in the presence or in the absence of σ70 from minus origin fragments of decreasing length.

Materials and Methods

Proteins. Wild-type or mutant RNAP core enzymes were purified as described (10). The recombinant wild-type σ70 subunit was purified as described in ref. 11. Plasmid overexpressing σ70 subunit lacking region 3.2 residues 507–519 was constructed by site-directed mutagenesis, and recombinant mutant σ70 was purified as described (11).
In Vitro Transcription. Three pmols of RNAP core (wild-type or mutant) with or without 10 pmols of recombinant σ70 (wild-type or mutant) and 300 ng of M13 plus-strand genomic DNA (New England Biolabs) or 3 pmols of synthetic DNA templates corresponding to different portions of M13 minus-strand origin were incubated at 37°C for 10 min in 10 μl of transcription buffer (20 mM Tris·HCl, pH 7.9/40 mM KCl/10 mM MgCl2) in the presence or absence of 2.5 μg of single-stranded DNA-binding protein (SSB) (Sigma). Abortive synthesis was initiated by the addition of 1 mM ATP and 0.3 μM[α-32P]GTP [3,000 Ci/mmol) (1 Ci = 37 GBq)]; full-sized pRNA synthesis was initiated by the addition of 100 μM ATP, CTP, UTP, and [α-32P]GTP (10 Ci/mmol). For the experiment shown in Fig. 3a, reactions also contained 0.5 mM ApG or ApGpG primers. Reactions were terminated after a 10-min incubation at 37°C by the addition of formamide-containing loading buffer, and products were separated on denaturing (7 M urea) polyacrylamide gels and revealed with PhosphorImager (Molecular Dynamics).
Transcription from the T7 A1 promoter template was performed at similar conditions, except that 0.3 pmols of PCR fragment, containing the T7 A1 promoter (12), were used as a template. Abortive synthesis was initiated by the addition of 100 μM CpA and 0.3 μM [α-32P]UTP (3,000 Ci/mmol); in run-off assays, the reactions were also supplemented with 100 μM NTPs.
DNase I Footprinting. A synthetic, gel-purified 128-nt-long DNA fragment (template 1) was 32P-labeled at its 5′ end with T4 polynucleotide kinase. Complexes were formed as described in the previous section and in the Fig. 2a legend and footprinted with DNase I exactly as described (13).
Fig. 2.
Recognition of M13 minus-strand origin DNA by RNAP core. (a) Three pmols of RNAP core was combined with 3 pmols of template 1 formed in the presence or in the absence of 10 pmols of σ70, and complexes were footprinted with DNase I. Lane 3 is a control (no RNAP added to reaction). Horizontal arrows indicate DNA positions that become protected from DNase I attack in the presence of RNAP (black arrows). Vertical arrows on the right schematically illustrate palindromic sequences that form the right and the left shoulders of the minus-strand origin (the top and the bottom pair of arrows, respectively). (b) Complexes were formed as in a and crosslinked with initiating AMP analogue. The crosslinks were revealed with the indicated radioactive nucleotides, and the products of affinity labeling were revealed, after SDS/PAGE separation, by autoradiography. (c) Results of in vitro transcription reactions performed by using template 1 complexes formed as in a. Lane 1 is control (no RNAP added to transcription reactions). (d) Results of in vitro abortive transcription reactions performed by using RNAP core and either wild-type template 2 or a modified template plus 2G → C that specifies a CMP instead of GMP in the second position of pRNA. (e) A structural model of Thermus aquaticus RNAP core (backbone representation) interacting with downstream DNA corresponding to the minimal functional M13 minus origin fragment [shown in spacefill, the template strand is in pink, the non-template strand is in gold, and the trajectory of DNA is based on a structural model of an elongation complex (18)]. β′ is in cyan, β is in green, and α and ω are in gray. The view is roughly perpendicular to the axis of the DNA-binding channel of the enzyme. The active-center Mg2+ is in red. The β′ amino acids corresponding to E. coli residues removed by the JE1134 mutation (15) are shown in spacefill. (f) In vitro abortive transcription initiation by the wild-type Eσ70 and mutant Eσ70 that lacks the β′ downstream jaw domain [Δ(1149–1190)]. In lanes 1–3, M13 minus-strand origin fragment was used as template, and no RNAP was added to reaction in lane 1. Control lanes 4 and 5 show the products of abortive transcription initiation by the wild-type and the mutant enzymes from –10/–35 class promoter T7 A1.
Affinity Labeling. Affinity labeling of minus-strand origin complexes formed as described above was performed by using AMP(1065) (o-formylphenyl ester of AMP) exactly as described (14).
Microbiological Techniques. F′ JE1134 or control MG1655 rpoC+ E. coli (15) were conjugated with BG389 E. coli containing an F′ factor marked with kanamycin resistance (16). M13 was plated on kanamycin-resistant, F′+ JE1134 and MG1655 E. coli exactly as described (17).


Defining a Minimal Minus-Strand Origin Fragment Sufficient for pRNA Synthesis. To better understand sequence requirements for minus-strand origin recognition by Eσ70, we delineated a minimal M13 DNA fragment sufficient for pRNA synthesis. Our starting template, template 1, was a 128-nt DNA fragment corresponding to M13 positions 5624–5751 and containing both pairs of inverted repeats (Fig. 1a). A slightly longer 137-nt fragment of highly similar f1 phage was previously studied by Higashitani et al., (7). The pRNA synthesis on the entire plus-strand M13 DNA required E. coli SSB (Fig. 1b, compare lanes 1 and 2). In the absence of SSB, nonspecific RNA synthesis was observed, which is likely due to transcription initiation on multiple sites on single-stranded DNA. Eσ70 also synthesized pRNA from template 1, but the synthesis was SSB-independent (Fig. 1b, compare lanes 3 and 4). The observed lack of SSB dependence for the shorter DNA fragment suggests that SSB functions in pRNA synthesis to simply prevent nonspecific RNAP binding to single-stranded DNA, as suggested earlier (6, 7). The length distribution of pRNA products was also changed, with the 18-nt transcript becoming the major species on template 1 (the 20-nt transcript predominated when the entire M13 plus-strand DNA was used a template, Fig. 1b, compare lane 2 with lanes 3 and 4). The reasons for changes in pRNA length distribution are not known although similar behavior was observed previously (7). We note that all transcripts initiated from the same start point and the observed heterogeneity in pRNA lengths were due to differences in pRNAs' 3′ ends as indicated by experiments that involved various combinations of missing nucleotides or 5′ end-labeling of pRNA transcript with [γ-32P]ATP (data not shown).
The putative Eσ70 –35 promoter element is located in the left shoulder of the pseudo double-stranded M13 minus-strand origin (Fig. 1a). To check the role of this element in pRNA synthesis, the ability of a 50-nt minus-strand origin fragment that lacked the entire left shoulder (template 2, Fig. 1a) to support pRNA synthesis was investigated. Eσ70 efficiently synthesized pRNA from this template (Fig. 1c, lane 4), indicating that the putative –35 promoter element is dispensable for origin recognition and pRNA synthesis. Even more strikingly, deletion of the putative –10 promoter element (template 3, Fig. 1a) did not lead to changes in pRNA synthesis efficiency (Fig. 1c, lane 6). Further analysis defined the minimal 33-nt DNA fragment that functioned as a template for pRNA synthesis. The minimal fragment could be folded into an imperfect 12-bp hairpin with a 3-nt terminal loop and a 6-nt single-strand extension on the 3′ end (indicated in red in Fig. 1a). Shortening the minimal 33-nt fragment from either the 3′ or the 5′ ends abolished primer synthesis (data not shown).
The synthesis of pRNA initiated 4 nt downstream of the 3′ end of the minimal template. This initiation start point corresponded to the start point observed on the intact M13 DNA (9) and was highly specific: pRNA synthesis initiated with ATP even though transcription reactions contained high concentrations of all four NTPs that could have been used to initiate transcription from other positions of the template.
Efficient pRNA synthesis from templates that almost totally lacked upstream promoter elements raised the question whether pRNA synthesis from these templates was σ70-dependent. Comparison of odd and even lanes of Fig. 1c shows that pRNA synthesis was strictly dependent on σ70 addition on templates 1–3 and on the minimal template (data not shown).
RNAP Core Forms a Complex with Minus-Strand Origin DNA. The strict σ70-dependence of pRNA synthesis on templates lacking sequences upstream of the transcription start point could be due to the absence of template DNA recognition by RNAP core. Alternatively, σ70 may not participate in the minus-strand origin recognition but may be required at later stages of pRNA synthesis. The ability of RNAP core and Eσ70 to bind the minus-strand origin DNA was studied by DNase I footprinting of complexes formed at a stoichiometric ratio of RNAP and template 1 (Fig. 2a). Both RNAP core and Eσ70 afforded a similar degree of protection of template DNA from DNase I digestion (Fig. 2a, compare lanes 4 and 5 with control lane 3, where no RNAP has been added). The result is in contrast with results of Higashitani et al. (7), who reported that RNAP core did not bind the minus-strand origin. We do not know the reason for this discrepancy. We note that in our experiments (i) RNAP core was fully inactive in pRNA synthesis reaction (Fig. 1c), indicating that the core enzyme preparation was free of contaminating holoenzyme and (ii) Eσ70 was reconstituted directly in the reaction by combining σ70 with RNAP core, thus allowing direct comparisons of reactions containing core and holoenzymes. We therefore conclude that RNAP core interacts with the minus-strand origin DNA, but this complex is inactive in pRNA synthesis.
RNAP Core Bound to Minus-Strand Origin Is Unable to Escape in Elongation. The absence of pRNA synthesis by RNAP core could be due to its inability to catalyze phosphodiester bond formation or can result from problems at the initiation-to-elongation transition. To address this alternative, we used highly selective affinity labeling with initiation substrate analogue (14). In this protocol, RNAP in the promoter complex is first crosslinked to an initiating (+1) nucleotide derivative, and the crosslinked nucleotide is next extended in a template-dependent manner, with radiolabeled nucleoside triphosphate specified by position +2 of the template. As a result, a radioactive dinucleotide is covalently attached to RNAP subunit(s); the labeled subunit(s) is visualized, after SDS/PAGE, by autoradiography. Affinity labeling of core and Eσ70 minus-strand origin complexes was performed by using a crosslinkable initiating AMP derivative that is known to target a conserved residue in RNAP β (14), and the crosslink was extended in the presence of [α-32P]GTP (note that the 5′ end of pRNA is 5′-ApG...). RNAP β was efficiently labeled in reactions containing either core or holo enzymes (Fig. 2b, compare lanes 1 and 2). In the Eσ70 complexes, σ70 was also labeled with low efficiency (Fig. 2b, lane 2). The labeling of β in reactions containing either core or holoenzyme must reflect transcription initiation at the true pRNA start point because the addition of “non-cognate” nucleotides [α-32P]ATP, [α-32P]UTP, or [α-32P]CTP did not result in any labeling (Fig. 2b, compare lanes 1 and 2 with lanes 3–8).
The affinity labeling result shows that RNAP core complexed with minus-strand origin can synthesize the first phosphodiester bond. Indeed, analysis of pRNA synthesis reaction products on high-resolution gels revealed that both RNAP core and Eσ70 produced copious amounts of abortive dinucleotide pppApG but only Eσ70 synthesized the full-sized pRNA, as expected (Fig. 2c, lanes 2 and 3). Eσ70 also synthesized the expected trinucleotide abortive product, which was absent from the RNAP core reaction.
To definitively establish that RNAP core synthesizes abortive transcripts from the pRNA start point, the following experiment was performed. The DNA template was modified such that the GC base pair immediately downstream of the pRNA start point was substituted for a CG base pair. The resulting template encodes a mutant pRNA containing a C at position +2 and is used by Eσ70 to synthesize a full-length pRNA (data not shown). The ability of RNAP core to synthesize abortive products pppApG and pppApC from the wild-type and the mutant template was determined. The result is presented in Fig. 2d. As can be seen, RNAP core synthesized abortive products from both templates in the presence of the “correct” combination of nucleotide substrates (Fig. 2d, lanes 2 and 5) and was totally inactive when a non-cognate nucleotide combination was used (Fig. 2d, lanes 3 and 4). We therefore conclude that RNAP core is able to correctly locate the pRNA transcription start point in the absence of σ70 but is unable to synthesize pRNA because it cannot extend initiated transcripts beyond the second position.
The β′ Jaw Domain Is Required for Minus-Strand Origin Recognition. The results presented above indicate that the minus-strand origin DNA is specifically recognized by RNAP core even in the absence of σ70. The minimal functional minus-strand origin template is likely bound in RNAP trough that is normally occupied by double-stranded DNA downstream of the +1 position (18) (Fig. 2e). Eσ70 that lacked the β′ downstream jaw domain that protrudes into the trough and cradles the downstream DNA (15) (Fig. 2e) transcribed well from conventional promoters (Fig. 2f, compare lanes 4 and 5) but did not transcribe from minus-strand origin DNA, indicating that the β′ downstream jaw is involved in recognition of the minus-strand origin and/or stability of the resulting complex. Both core and holoenzymes lacking the β′ downstream jaw were transcriptionally inactive on minus-strand origin DNA (Fig. 2f, lanes 2 and 3), underscoring the requirement for intact β′ downstream jaw for pRNA synthesis. In accordance with this in vitro result, E. coli cells that harbor the rpoC deletion that removed the β′ jaw were found to be nonpermissive for M13 (data not shown).
σ70 Region 3.2 Is Required for pRNA Synthesis. How does σ70 allow RNAP core to escape into productive pRNA synthesis? RNAP core did not synthesize full-sized pRNA even at very high (1 mM) concentrations of NTPs (data not shown), making it unlikely that σ70 acts by decreasing Km for incoming NTPs. Alternatively, σ70 may stabilize the retention of short abortive transcripts during early stages of pRNA synthesis and allow them to be extended beyond the second position. To check this idea, pRNA reaction was initiated with ApG dinucleotide, which corresponds to the first abortive transcript that both core and holoenzymes can synthesize, or with ApGpG trinucleotide that can be synthesized only by Eσ70 (Fig. 2c). As can be seen from Fig. 3a, RNAP core acquired the ability to synthesize full-sized pRNA when reaction was initiated with ApGpG but not with ApG. However, even in the presence of ApGpG, RNAP core was only 30% as active as Eσ70 in pRNA synthesis (Fig. 3a, compare lanes 3 and 4). Thus, σ70 becomes partially dispensable for pRNA synthesis once the second phosphodiester bond has formed.
Fig. 3.
Requirements for pRNA synthesis. (a) The pRNA synthesis reaction was performed from template 2 in the presence of ApG (lanes 1 and 2) or ApGpG (lanes 3 and 4) as primers by using RNAP core or Eσ70 as indicated. (b) In vitro transcription of the wild-type Eσ70 and Eσ70 mutant reconstituted from wild-type RNAP core and σ70 lacking region 3 [Δ(507–519)]. Lanes 1–4 show products of run-off transcription from a DNA fragment containing –10/–35 class promoter T7 A1 (lanes 1 and 2) and M13 minus-strand origin fragment. Lanes 5 and 6 show the products of abortive initiation from the M13 minus-strand origin fragment.
Functional and structural analyses indicate that a part of σ, region 3.2, protrudes toward the catalytic center and can contact the 5′ end of short transcripts (1921). This interaction could stabilize short transcripts and allow them to be extended. Eσ70 reconstituted from σ70 mutant lacking region 3.2 did not produce pRNA (Fig. 3b, lane 4). However, the mutant enzyme was active in abortive synthesis of pppApG dinucleotide from the minusstrand origin template (Fig. 3b, compare lanes 5 and 6) and also transcribed well from a –10/–35 promoter T7 A1 (Fig. 3b, compare lanes 1 and 2). Control experiments showed that RNAP holoenzymes reconstituted from sigmas that lacked regions 1 or 4 synthesized pRNA normally (data not shown). We therefore conclude that σ70 region 3.2 is specifically required for pRNA synthesis.


Our principal finding is the demonstration that production of pRNA from M13 minus-strand origin, although dependent on σ70, is independent of sigma's prototypical ability to specifically recognize DNA. Instead, σ70 allows RNAP core to escape into productive elongation. Our results suggest that σ70 region 3.2 plays an essential role in pRNA synthesis, possibly by contacting and stabilizing short abortive transcripts and allowing their extension.
The recognition of minus-strand origin is evidently accomplished by RNAP core, which heretofore was not known to initiate transcription specifically. Whereas our results show that the β′ downstream jaw is required for minus-strand origin complex formation, it is possible that the jaw stabilizes the complex once it is formed and is therefore not involved in recognition per se. It remains to be determined which features of the minus-strand origin allow specific recognition and correct positioning of RNAP catalytic center to allow transcript initiation at +1.
Sigma subunits are thought to be responsible for DNA recognition during RNAP holoenzyme-dependent priming of replication from single-stranded origins of plasmids (22, 23), episomes (24), and phages (25) from evolutionarily distant groups of bacteria. Our results suggest that RNAP core may be the primary determinant of origin recognition in these and possibly other systems whereas sigma subunits may perform essential functions that are unrelated to their prototypical promoter recognition function.
The priming of replication by RNAP holoenzyme has been known for some time (6). In addition to issues of specific origin recognition by RNAP that have been studied in this work, there is a fascinating question of how the transfer of pRNA 3′ end from RNAP to DNA polymerase is accomplished. What determines the length of the pRNA transcript? Is RNAP part of the priming complex and if yes, how does DNA polymerase gain access to the primer's 3′ end? The minimal system described here should facilitate investigation of these important questions in the future.


Abbreviations: RNAP, RNA polymerase; pRNA, primer RNA; SSB, single-stranded DNA-binding protein.


This work is dedicated to the memory of Dmitriy Salonin. We thank Ann Hochschild for materials, Yulia Yuzenkova for advice, and Leonid Minakhin for providing the σ70 mutant used in this work. This work was supported by a Borroughs Welcome Fund Career Award and National Institutes of Health Grant GM64530 (to K.S.).


Gross, C., Chan, C., Dombroski, A., Gruber, T., Sharp, M., Tupy, J. & Young, B. (1998) Cold Spring Harbor Symp. Quant. Biol. 63, 141–155.
Hernandez, V. J., Hsu, L. M. & Cashel, M. (1996) J. Biol. Chem. 271, 18775–18779.
Roberts, J. W., Yarnell, W., Bartlett, E., Guo, J., Marr, M., Ko, D. C., Sun, H. & Roberts, C. W. (1998) Cold Spring Harbor Symp. Quant. Biol. 63, 319–325.
Mukhopadhyay, J., Kapanidis, A. N., Mekler, V., Kortkhonjia, E., Ebright, Y. W. & Ebright, R. H. (2001) Cell 106, 453–463.
Bar-Nahum, G. & Nudler, E. (2001) Cell 106, 443–451.
Kaguni, J. M. & Kornberg, A. (1982) J. Biol. Chem. 257, 5437–5443.
Higashitani, N., Higashitani, A., Guan, Z. W. & Horiuchi, K. (1996) Genes Cells 1, 829–841.
Higashitani, A., Higashitani, N. & Horiuchi K. (1997) Proc. Natl. Acad. Sci. USA 94, 2909–2914.
Higashitani, N., Higashitani, A. & Horiuchi, K. (1993) J. Virol. 67, 2175–2181.
Kashlev, M., Nudler, E., Severinov, K., Borukhov, S., Komissarova, N. & Goldfarb, A. (1996) Methods Enzymol. 274, 326–334.
Borukhov, S. & Goldfarb, A. (1993) Protein Expression Purif. 4, 503–511.
Zakharova, N., Bass, I., Arsenieva, E., Nikiforov, V. & Severinov, K. (1998) J. Biol. Chem. 273, 24912–24920.
Severinova, E., Severinov, K. & Darst, S. A. (1998) J. Mol. Biol. 279, 9–18.
Severinov, K., Mustaev, A., Severinova, E., Kozlov, M., Darst, S. A. & Goldfarb, A. (1995) J. Biol. Chem. 270, 29428–29432.
Ederth, J., Artsimovitch, I., Isaksson, L. A. & Landick, R. (2002) J. Biol. Chem. 277, 37456–37463.
Whipple, F. W. (1998) Nucleic Acids Res. 26, 3700–3705.
Sambrook, J., Fritch, E. F. & Maniatis, T. (1989) Molecular Cloning (Cold Spring Harbor Lab. Press, Plainview, NY).
Korzheva, N., Mustaev, A., Kozlov, M., Malhotra, A., Nikiforov, V., Goldfarb, A. & Darst, S. A. (2000) Science 289, 619–625.
Severinov, K., Fenyö, D., Severinova, E., Mustaev, A., Chait, B. T., Goldfarb, A. & Darst, S. A. (1994) J. Biol. Chem., 269, 20826–20828.
Murakami, K. S., Masuda, S. & Darst, S. A. (2002) Science 296, 1280–1284.
Vassylyev, D. G., Sekine, S., Laptenko, O., Lee, J., Vassylyeva, M. N., Borukhov, S. & Yokoyama, S. (2002) Nature 417, 712–719.
Kramer, M. G., Khan, S. A. & Espinosa, M. (1997) EMBO J. 18, 5784–5795.
Gruss, A. D., Ross, H. F. & Novick, R. P. (1987) Proc. Natl. Acad. Sci. USA 84, 2165–2169.
Masai, H. & Arai, K. (1997) Cell 89, 897–907.
Peeters, B. P., Peters, R. M., Schoenmakers, J. G. & Konings, R. N. (1985) J. Mol. Biol. 181, 27–39.

Information & Authors


Published in

Go to Proceedings of the National Academy of Sciences
Go to Proceedings of the National Academy of Sciences
Proceedings of the National Academy of Sciences
Vol. 101 | No. 13
March 30, 2004
PubMed: 15070729


Submission history

Received: January 4, 2004
Published online: March 15, 2004
Published in issue: March 30, 2004


This work is dedicated to the memory of Dmitriy Salonin. We thank Ann Hochschild for materials, Yulia Yuzenkova for advice, and Leonid Minakhin for providing the σ70 mutant used in this work. This work was supported by a Borroughs Welcome Fund Career Award and National Institutes of Health Grant GM64530 (to K.S.).



Nikolay Zenkin
The Waksman Institute and Department of Molecular Biology and Biochemistry, Rutgers, The State University of New Jersey, Piscataway, NJ 08854; and Institute of Molecular Genetics, Russian Academy of Sciences, Moscow 142292, Russia
Konstantin Severinov§
The Waksman Institute and Department of Molecular Biology and Biochemistry, Rutgers, The State University of New Jersey, Piscataway, NJ 08854; and Institute of Molecular Genetics, Russian Academy of Sciences, Moscow 142292, Russia


To whom correspondence should be addressed at: The Waksman Institute, 190 Frelinghuysen Road, Piscataway, NJ 08854. E-mail: [email protected].
Communicated by E. Peter Geiduschek, University of California at San Diego, La Jolla, CA, February 6, 2004

Metrics & Citations


Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.

Citation statements



If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by


    View Options

    View options

    PDF format

    Download this article as a PDF file


    Get Access

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Personal login Institutional Login

    Recommend to a librarian

    Recommend PNAS to a Librarian

    Purchase options

    Purchase this article to get full access to it.

    Single Article Purchase

    The role of RNA polymerase σ subunit in promoter-independent initiation of transcription
    Proceedings of the National Academy of Sciences
    • Vol. 101
    • No. 13
    • pp. 4333-4719







    Share article link

    Share on social media