Báo cáo khoa học: Aptamers toEscherichia colicore RNA polymerase that sense its interaction with rifampicin, r-subunit and GreB

doi:10.1111/j.1432-1033.2004.04461.x

Eur. J. Biochem. 271, 4921–4931 (2004) (cid:1) FEBS 2004

Aptamers to Escherichiacolicore RNA polymerase that sense its interaction with rifampicin, r-subunit and GreB

Andrey Kulbachinskiy1,2, Andrey Feklistov2,3, Igor Krasheninnikov3, Alex Goldfarb1 and Vadim Nikiforov1,2 1Public Health Research Institute, Newark, New Jersey, USA; 2Institute of Molecular Genetics, Moscow, Russia; 3Department of Molecular Biology, Moscow State University, Moscow, Russia

of the aptamers to RNAP core but did not affect the disso- ciation rate of preformed RNAP–aptamer complexes. We argue that these ligands sterically block access of the aptamers to their binding sites within the main RNAP channel. In contrast, transcript cleavage factor GreB increased the rate of dissociation of preformed RNAP– aptamer complexes. This suggested that GreB that binds RNAP outside the main channel actively disrupts RNAP– aptamer complexes by inducing conformational changes in the channel. We propose that the aptamers obtained in this work will be useful for studying the interactions of RNAP with various ligands and regulatory factors and for investi- gating the conformational flexibility of the enzyme.

Keywords: aptamers; conformational changes; elongation complex; GreB; RNA polymerase.

Bacterial RNA polymerase (RNAP) is the central enzyme of gene expression that is responsible for the synthesis of all types of cellular RNAs. The process of transcription is accompanied by complex structural rearrangements of RNAP. Despite the recent progress in structural studies of RNAP, detailed mechanisms of conformational changes of RNAP that occur at different stages of transcription remain unknown. The goal of this work was to obtain novel ligands to RNAP which would target different epitopes of the enzyme and serve as specific probes to study the mech- anism of transcription and conformational flexibility of RNAP. Using in vitro selection methods, we obtained 13 classes of ssDNA aptamers against Escherichia coli core RNAP. The minimal nucleic acid scaffold (an oligonucleo- tide construct imitating DNA and RNA in elongation complex), rifampicin and the r70-subunit inhibited binding

DNA-directed RNA polymerase (RNAP, EC 2.7.7.6) is a complex molecular machine undergoing multiple intra- molecular rearrangements in the process of RNA synthesis [1–3]. During the transcription cycle, RNAP makes specific and nonspecific contacts with double and single stranded (ss) DNA, the RNA/DNA hybrid and nascent RNA. Recent advances in structural studies of bacterial and yeast RNAPs [4–8] made it possible to create three-dimensional models of the promoter and elongation complexes and to propose the roles for various RNAP domains in interactions with DNA and RNA [6,8–11].

RNA behind the hybrid. The 8-bp-long DNA/RNA hybrid is lodged between the catalytic Mg2+ ion and a structural element of b¢ called the rudder (Fig. 1) [9]. The downstream DNA duplex is placed in a (cid:1)trough(cid:2) formed by several domains of b¢ (clamp and jaw) and b (b2 lobe). The b-subunit flexible flap domain closes the main channel from the upstream side leaving a narrow RNA exit channel. Rifampicin (Rif), one of the most efficient inhibitors of RNAP, binds the enzyme near the active center at a pocket formed by the b-subunit and sterically blocks RNA synthesis [13]. The b¢ F-bridge helix crosses the cleft in the vicinity of the catalytic Mg2+ separating the main and secondary channels (Fig. 1B). The secondary channel gives access to the active site for nucleotide substrates [9,14] and for elongation factors GreA and GreB (Fig. 1B) [15,16].

The most striking structural feature of RNAP is a deep cleft (the main channel) formed by the two largest RNAP subunits (b and b¢ in the bacterial enzyme) that runs along the full length of the molecule [4,12]. In the elongation complex, the main channel accommodates the RNA/DNA hybrid, duplex DNA downstream from the hybrid and

Despite the great progress of the past few years in structural studies of transcription, many molecular details of the RNAP–nucleic acid interactions remain unknown. Little is also known about the mechanisms of conforma- tional changes of RNAP that occur at different stages of transcription. Comparisons of homologous bacterial [17] and yeast RNAP structures [5] suggest significant conform- ational flexibility of RNAP domains that allows for the opening and closing of the main channel. The closure of RNAP around the DNA/RNA framework was proposed to be of crucial importance for the formation of stable elongation complexes [4–6,18]. More local conformational changes are thought to occur in the vicinity of the RNAP active center. In particular, the movement of the F-bridge helix was hypothesized to accompany the translocation step

Correspondence to A. Kulbachinskiy, Laboratory of Molecular Gen- etics of Microorganisms, Institute of Molecular Genetics, Kurchatov Sq. 2, Moscow 123182, Russia. Fax:/Tel.: + 7095 1960015, E-mail: akulb@img.ras.ru Abbreviations: RNAP, DNA-directed RNA polymerase; Eco, Escherichia coli; Taq, Thermus aquaticus; SELEX, systematic evolu- tion of ligands by exponential enrichment; MS, minimal nucleic acid scaffold; Rif, rifampicin; ss, single stranded. Enzymes: DNA-directed RNA polymerase (EC 2.7.7.6). (Received 25 May 2004, revised 19 October 2004, accepted 25 October 2004)

4922 A. Kulbachinskiy et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

Aptamers to several enzymes were shown to affect the conformation of the target protein [31–33]. For example, ssDNA aptamers to Ile-tRNA synthetase stimulated the editing activity of the enzyme, which is normally induced by tRNAIle [31], while aptamers to hepatitis C virus RNA- dependent RNAP allosterically prevented the entry of an RNA substrate into the enzyme’s active site [32].

during each cycle of nucleotide addition [6,8,14]. Several inhibitors of RNAP such as streptolydigin, a-amanitin, microcin J25 and CBR703 which bind at different sites near the F-bridge have recently been proposed to act by restricting the intramolecular mobility of the enzyme [14,19–21]. Thus, the analysis of different ligands that bind RNAP and stabilize alternative structural states of the enzyme could open the way for a better understanding of the conformational flexibility of RNAP.

Aptamers are synthetic RNA and ssDNA ligands that can be obtained to virtually any desired target [22]. The affinities and specificities of aptamers to different protein targets are comparable to those of monoclonal antibodies. Not surprisingly, aptamers have drawn significant attention as very promising ligands that can be used in a variety of biological applications. Aptamers to various nucleic acid binding proteins (including proteins that do not recognize their substrates sequence specifically) usually bind their targets at natural RNA or DNA recognition sites [22–26]. Structural analysis of several aptamer–protein complexes has shown that aptamers mimic the natural nucleic acid ligand of a protein and bind at the same place even if they have an unrelated nucleotide sequence and secondary structure (e.g. aptamers to the MS2 phage coat protein [27], NF-jB [28], reverse transcriptase [24]). As a result, many aptamers are very effective and highly specific inhibitors of their targets [29,30].

Here, we describe the isolation of aptamers to Escheri- chia coli (Eco) core RNAP. All selected aptamers are highly potent inhibitors of RNAP and are likely to bind within the main channel of the enzyme. We also developed a site-directed SELEX (systematic evolution of ligands by exponential enrichment [22]) procedure that allowed iden- tification of several aptamers that interact specifically with the Rif-binding pocket of RNAP. The RNAP–aptamer complexes were compared with the complex of the core enzyme with the minimal RNA/DNA scaffold (Fig. 1) [34], which mimics the natural elongation complex. We found that the aptamers and the minimal scaffold bind to overlapping sites on the core enzyme and that the resulting complexes have many similar features. Finally, we showed that the aptamers sensed interactions of core RNAP with the r70-subunit and transcript cleavage factor GreB. The results indicate that stimulation of the RNAP endonuclease activity by GreB may be accompanied by significant conformational changes of the enzyme. We propose that in studying the the selected aptamers may be useful

Fig. 1. Structural features of RNAP clasping minimal nucleic acid scaffold. (A) Minimal nucleic acid scaffold (MS) used in this study. (B) Model of MS in complex with Taq core RNAP [9]. MS (ball and stick representation: nontemplate DNA strand, black; template DNA strand, light violet; RNA, red) is placed inside the main RNAP channel. Also shown are active site magnesium (light blue), rifampicin (orange) clashing with RNA, b¢ F-bridge helix (green), rudder (red), b¢ coiled-coil r-subunit binding protrusion (dark violet), b flexible flap (blue; region cor- responding to Eco amino acids 885–914 is shown in red), and b¢ elements from the downstream part of the main channel (jaw and a part of the clamp, light blue; W217His6 insertion site is red). The b2 (lobe) domain of the b-subunit (amino acids 174–314, corres- ponding to Eco186–433) located above the MS is outlined by a thick brown line. The secondary channel is located just behind the F-bridge. The location of the GreB binding site [15] is shown schematically as a yellow oval. The ochre contour corresponds to the r-subunit, the position of which was taken from the T. thermophilus holoenzyme struc- ture [8]. r-Induced conformational changes are not shown. The semitransparent area shows the position of r region 3.2.

Aptamer probes of RNA polymerase structure (Eur. J. Biochem. 271) 4923

(cid:1) FEBS 2004

mechanism of transcription and conformational flexibility of RNAP.

Materials and methods

Proteins

Eco core RNAP with a His6 tag in the C terminus of the b¢-subunit and the r70-subunit were purified as described [35,36]. Eco core RNAP bearing the insertion of six histidine residues at position 217 of the b¢-subunit was reconstituted in vitro from individual subunits [37]. Mutant Eco core enzymes with deletions of the b2 (bD186–433) and the flexible flap (bD885–914) domains were kindly provided by K. Severinov and K. Kuznedelov [38,39]. Thermus aquaticus (Taq) core RNAP was purified from Eco cells expressing all four core subunits from plasmid pET28ABCZ as described [40]. The GreB protein was a generous gift of S. Borukhov (The State University of New York).

Selection of aptamers to Eco core RNAP

the initial

treated with phenol and chloroform and ethanol precipi- tated. The resulting enriched library was incubated with the core enzyme (taken in twofold excess relative to the first selection step) in the presence of 20 lgÆmL)1 Rif (rifamycin SV, Sigma, St Louis, MO, USA). DNA–protein complexes were adsorbed on Ni2+–agarose and discarded, while unbound oligonucleotides remaining in the solution were ethanol precipitated, PCR amplified and used in the next SELEX round. After the final round of selection, the enriched libraries were amplified with primers containing EcoRI and HindIII sites and cloned into the pUC19 plasmid. The sequences of individual aptamers were determined using the standard sequencing protocol. Indi- vidual ssDNA aptamers were obtained by PCR with the primers corresponding to aptamer flanks; the DNA strands

A ssDNA library (Fig. 2A) was purchased from Operon Technologies Inc. The amounts of ssDNA and the core enzyme varied from 5 nmol and 100 pmol, respectively, in the first round of selection to 100 pmol and 10 pmol in subsequent rounds. Prior to each round of selection, a 10-pmol aliquot of ssDNA was labeled with -[32P]ATP[cP] (7000 CiÆmmol)1, ICN, Costa Mesa, CA, USA) and T4 polynucleotide kinase (New England BioLabs, Beverly, MA, USA), purified by 10% PAGE and added to the bulk DNA sample to monitor the binding of the library to RNAP. ssDNA was then diluted in 1 mL binding buffer (20 mM Tris/HCl pH 7.9, 10 mM MgCl2, 300 mM NaCl, 30 mM KCl; in subsequent rounds NaCl and KCl concen- trations were increased to 400 and 40 mM, respectively), heated for 5 min at 95 (cid:2)C and cooled rapidly to 0 (cid:2)C. The DNA solution was passed through a 50-lL Ni2+–nitrilo- triacetic acid–agarose (Qiagen) microcolumn pre-equili- brated with the binding buffer. The core enzyme was then added to the solution and the mixture was incubated for 15 min at room temperature. Thirty microliters Ni–nitrilo- triacetic acid–agarose was added and the incubation was continued for a further 20 min with occasional shaking. The solution containing unbound DNA was removed and the sorbent was washed two to four times with 1 mL of binding buffer (for a total time of 30–60 min). ssDNA–RNAP complexes were eluted with 300 lL binding buffer contain- ing 200 mM imidazole. The solution was treated with 300 lL phenol and 300 lL chloroform. DNA was ethanol precipitated, dissolved in water and amplified using Vent DNA polymerase (New England BioLabs) and primers corresponding to fixed regions of library (5¢-GGGAGCTCAGAATAAACGCTCAA-3¢ and BBB- 5¢-GATCCGGGCCTCATGTCGAA-3¢, where B is a bio- tin residue). Two DNA strands were separated by size on 10% denaturing PAGE, the nonbiotinilated strand was eluted and used for the next SELEX round. In the Rif- directed SELEX experiment each round of the selection included two successive partitioning steps. The initial selection of oligonucleotides was carried out as described above. DNA eluted from the complexes with RNAP was

Fig. 2. Selection of aptamers to Eco core RNAP. (A) Random ssDNA library used in selection experiments. (B) The effect of Rif on the binding of round 11 libraries (0.1 nM) to Eco core RNAP (10 nM) in binding buffer containing 440 mM salt. Binding was measured as described in Materials and methods. One hundred per cent corres- ponds to the binding in the absence of Rif. (C) Sequences of repre- sentative aptamers from 13 different classes described. Shown are the central 32-nt-long regions of the aptamers. Aptamer E3 contains a T fi A change at the first position of the right constant region; aptamer E13 contains a single nucleotide deletion at the same site. The sequence motif identical in aptamers E9 and E12 is underlined.

4924 A. Kulbachinskiy et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

10 min by the addition of a formamide-containing stop buffer and applied to 23% urea PAGE. The amount of radioactively labeled 9-nt RNA product was quantified by using a PhosphorImager.

were separated on denaturing PAGE as described above. Control experiments demonstrated that aptamers did not bind to the Ni-affinity sorbent and therefore the SELEX protocol was highly efficient in selecting specific aptamer sequences.

Results

Quantitation of the binding of aptamers to RNAP

Selection of aptamers to Eco core RNAP – conventional vs. site-directed SELEX

Both the core and holo enzymes of bacterial RNAP bind nucleic acids [42–45]. While the holoenzyme is able to recognize specific DNA sequences, the interactions of the core enzyme with DNA and RNA are generally nonspecific. There are numerous reports on interactions of the core with total cellular DNA and RNA [43,46], tRNA [47,48], ssDNA [43] and also some individual RNA sequences [49]. Repor- ted Kd values for some of these interactions are in the range of 10)8 to 10)10 M [43,48,49] and are comparable to the affinities of known aptamers to their protein targets.

We used a library of 75-nt long ssDNA containing a 32-nt central region of random sequence to select aptamers that would specifically interact with the RNAP core (Fig. 2A). We found that at low ionic strength, molecules from the unenriched library bound the core enzyme very tightly (Kd (cid:1) 0.2 nM at 40 mM salt). Such a high level of nonspe- cific affinity of RNAP to nucleic acids could be a serious obstacle for the selection of specific aptamer sequences. However, we observed that the nonspecific binding of ssDNA to core RNAP was considerably reduced at increased ionic strength (Kd > 100 nM at 300 mM salt). Therefore, we performed all selection procedures at elevated monovalent salt concentrations (300–440 mM).

(1 lM) or GreB (3 lM) and aliquots of

Determination of Kd values for the binding of oligo- nucleotides to RNAP was achieved by using the nitro- cellulose filtration method as described [41]. All measurements were performed in binding buffer contain- ing 400 mM NaCl and 40 mM KCl unless otherwise indicated. A 5¢-end labeled oligonucleotide (0.003 nM) was incubated with a series of dilutions of core RNAP (from 0.01 to 100 nM) in binding buffer containing 50 lgÆmL)1 BSA for 45–60 min at 22 (cid:2)C and then filtered through 0.45-lm nitrocellulose filters (HAWP, Millipore) prewetted in the same buffer. The filters were washed with 5 mL buffer and quantified on a Phosphor- Imager (Molecular Dynamics, Sunnyvale, CA, USA). Ki measurements were carried out at fixed core (1–3 nM) and aptamer (0.1 nM) concentrations. Rif, r70 or GreB were included in the binding reactions 5 min prior to the addition of oligonucleotides; the samples were incubated for 1 h at room temperature and passed through nitrocellulose filters. Kd and Ki values were calculated from the binding curves using KALEIDAGRAPH software (Synergy, Reading, PA, USA). To measure dissociation kinetics of RNAP–aptamer complexes, the core polym- erase (3 nM) was preincubated with a labeled aptamer (0.1 nM) for 60 min, the complex was challenged with the (100 nM), minimal corresponding unlabeled aptamer (2 lgÆmL)1), r70- nucleic acid scaffold (500 nM), Rif subunit the increasing time intervals. sample were filtered after Control experiments demonstrated that the level of RNAP–aptamer binding did not change if the measure- ments were done in the absence of the inhibitors.

Minimal nucleic acid scaffold (MS)

We conducted two types of experiments to select aptamers to Eco core RNAP. In the first type of experiment (I), the SELEX procedure was performed in a conventional way. In brief, in each round of the selection the ssDNA library was incubated with core RNAP immobilized on a Ni-affinity sorbent via the hexahistidine tag present at the C-terminal end of the b¢-subunit. Then unbound DNA was extensively washed out to select sequences that formed stable complexes with RNAP. RNAP–DNA complexes were eluted with imidazole, recovered oligonucleotides were amplified by PCR and used in the next round of selection. To avoid selection of nonspecific sequences that bind to the affinity sorbent used in the reaction, the library was passed through Ni–agarose column in the absence of RNAP before each SELEX round.

The second type of experiment (II) aimed to identify ligands that bound specifically to the Rif-binding pocket of RNAP. Rif is one of the most potent inhibitors of the enzyme and is used as a drug in the therapy of several infectious diseases. However, a large number of mutations in core RNAP conferring resistance to this drug have been described. Identification of new ligands that can mimic the effect of Rif is therefore of great importance. Each round of site-directed SELEX consisted of two consecutive binding reactions. First, we selected sequences that bound to free core RNAP. Second, DNA molecules that interacted with RNAP were incubated with core RNAP in the presence of excess Rif. DNA molecules that were unable to bind RNAP in complex with Rif were used in the next round of selection.

The sequences of DNA and RNA oligos used to recons- titute MS are shown in Fig. 1A. MS was prepared as described [9]. The RNA oligo (200 pmol, final concentra- tion 10 lM) was labeled with 10 U T4 polynucleotide kinase [32P]ATP[cP], mixed with template and and 0.5 mCi nontemplate DNA oligonucleotides (final concentrations of the oligonucleotides were 1, 1 and 2 lM, respectively) in the binding buffer, heated to 65 (cid:2)C and slowly cooled to 20 (cid:2)C. Determination of Kd for the binding of MS to RNAP was performed as described above. In some cases, the binding was measured in the buffer containing 200 mM salt (20 mM Tris/HCl pH 7.9, 10 mM MgCl2, 160 mM NaCl, 40 mM KCl). When studying the inhibitory effect of aptamers on RNAP activity, Eco core enzyme (10 nM) was added to the mixture of unlabeled MS (10 nM) and aptamers (30 nM) in binding buffer containing 400 mM NaCl and 40 mM KCl. The samples were incubated for 30 min at room temperature and supplemented with (0.1 lM, 3000 CiÆmmol)1, Perkin Elmer, [32P]UTP[aP] Wellesley, MA, USA). The reaction was stopped after

Aptamer probes of RNA polymerase structure (Eur. J. Biochem. 271) 4925

(cid:1) FEBS 2004

Table 1. Properties of the aptamers to Eco core RNAP. Kd values were measured in the binding buffer containing 400 mM NaCl and 40 mM KCl.

Inhibition by Binding to mutant RNAPsa Clones (n)

II Dflap D186–433 I Rif GreBc rb Aptamer SELEX Kd (nM) WHis6

2.5 3.2

6.3 3.3

1.1

5.5 6.5

a The increase in Kd for aptamer binding to mutant variants of core RNAPs over Kd values for the wild-type enzyme: +, 1–5 times; +/), 5–20 times; –, more than 20 times. b The increase in Kd for aptamer binding to the core polymerase in the presence of 0.5 lM r-subunit: +, approximate change in Kd is 10–30 times. c The increase in Kd for aptamer binding in the presence of 1.5 lM GreB. Blank cells, no data.

We compared the RNAP–aptamer complexes with a complex of the RNAP core bound to the minimal nucleic acid scaffold (MS) (Fig. 1A) – a model of the elongation complex [34]. The contacts of MS with Eco core RNAP were mapped previously by nucleic acid–protein crosslink- ing techniques and the results were used to position MS on structure of Taq core RNAP the three-dimensional (Fig. 1B) [9]. The interaction of MS with RNAP was shown to be independent of the MS sequence [9,13]. The MS used in our study consisted of an 18-nt-long down- stream DNA duplex and an 8-nt-long RNA–DNA hetero- duplex separated by two unpaired DNA bases (Fig. 1A). Unlike the aptamers, MS bound both Eco and Taq core RNAPs with comparable affinities (with a Kd value of (cid:1) 1 nM in binding buffer containing 40 mM salt). The complex of MS with Eco core polymerase was transcrip- tionally active at both low (40 mM) and high (440 mM) salt concentrations (data not shown). Remarkably, the affinity of MS to RNAP at 440 mM salt (Kd ‡ 50 nM) was lower than the affinities of the aptamers at these conditions.

After 11 rounds of selection, the enriched libraries obtained by both protocols bound core polymerase with high affinity (Kd (cid:1) 5 nM in binding buffer containing 440 mM salt) but exhibited substantially different sensitivity to Rif addition (Fig. 2B). While the RNAP binding of the (cid:1)conventionally(cid:2) enriched library was essentially resistant to Rif, binding of the site-specifically selected library was severely inhibited by the antibiotic. Both libraries were cloned and 50 individual clones were sequenced in each case. Analysis of individual clones allowed us to identify 13 different classes of sequences, designated E1–E13 (Fig. 2C; the total number of clones within each class is shown in Table 1). Each class consisted of several clones with identical or closely related sequences. Sequences from classes E1–E4 were found only in the conventionally enriched library, sequences from classes E5–E8 were present in both types of libraries and sequences from classes E9–E13 were unique to the library obtained by Rif-directed selection. All of the aptamers were predicted to fold into distinct secondary structures, such as hairpins and G-quartets (e.g. aptamers E1, E3, E4, E5) (data not shown). One aptamer representative of each class was chosen for further investigation (Fig. 2C).

Aptamers bind Eco core RNAP with high affinity and inhibit the enzyme’s activity

All selected aptamers competed with MS for binding to core RNAP and efficiently inhibited RNAP activity in the transcription assay (most probably by preventing the formation of the RNAP–MS complex, see below) (Fig. 3). The inhibition of the core polymerase activity by aptamers was specific as much weaker inhibition was observed in the case of the initial random library (Fig. 3).

Aptamers interact with distinct sites inside the main channel of core RNAP

Individual aptamers from all 13 classes proved to be high- affinity ligands to Eco core RNAP with apparent Kd values ranging from 0.13 nM for aptamer E1 to 6.3 nM for aptamer E8 at 440 mM salt (Table 1). These affinities are comparable to the affinity of Rif to Eco core polymerase [50,51] and greatly exceed those of other small molecule ligands of RNAP such as streptolydigin [52], microcin J25 [20] and CBR703 [21]. Neither the initial library nor any other nonspecific oligonucleotide tested appreciably bound RNAP at these conditions. All of the aptamers competed with each other for the binding to core RNAP which indicated that they interacted with overlapping sites on the RNAP surface (data not shown).

In order to locate the aptamer binding sites more precisely we checked the ability of the aptamers to interact with Eco core RNAP bearing insertion–deletion mutations in several sites on the periphery of the main channel (Table 1 and Fig. 1B). The mutations were a deletion of the flexible flap domain in the b-subunit (bD885–914), a deletion of the domain b2 in the b-subunit (bD186–433) and an insertion of six histidine residues at position 217 of the b¢-subunit

5.8 E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 E11 E12 E13 10 16 12 6 2 2 2 1 – – – – – – – – – 3 4 1 4 8 4 5 9 7 0.13 1.04 1.23 2.23 0.72 1.62 2.21 6.32 1.43 2.82 3.56 4.00 4.93 I I I I I + II I + II I + II I + II II II II II II +/– + + + + + + + + + +/– + + – – – – +/– – + – – – – – – + +/– – +/– +/– +/– + + – +/– + +/– + – – – + + + + + + + + + + 8.1 19.9 28.8 + 10.4 + + + 14.2 + + + +

4926 A. Kulbachinskiy et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

Fig. 4. Effect of Rif on the binding of aptamers and MS to core RNAP. Binding reactions contained 10 nM of the core enzyme, 0.1 nM oligo- nucleotides and varied amounts of Rif. Monovalent salt concentration in the binding buffer was 440 mM in the case of aptamers and 200 mM in the case of MS. Binding was measured as described in Materials and methods and normalized to the binding in the absence of Rif. The experiment was performed with the wild-type core enzyme (S) and Rif- resistant mutant RNAP (S531F, R).

of the aptamers unique to the conventional selection procedure (E1–E3) were insensitive to Rif (RifR, for Rif- resistant, aptamers, Table 1 and Fig. 4) and only one of them (E4) was found to be RifS. RifS sequences from classes E5–E8 which were identified in both selection experiments comprised only a small fraction of all sequences in the first SELEX population (Table 1). Thus, conventional SELEX produced mainly RifR aptamers whereas Rif-directed SELEX succeeded in identifying only RifS sequences. The high efficiency of the site-directed SELEX protocol used in our work suggests that similar procedures can be used to obtain high affinity aptamers to antibiotic-binding sites of many proteins of interest.

We repeated the binding assay using Rif-resistant core RNAP carrying an S531F substitution in the b-subunit. In this case, the effect of Rif was much weaker with Ki (cid:1) 0.5 lM (Fig. 4). At the same time, the mutation did not affect the binding of aptamers. Thus, the core mutation conferring Rif resistance weakened Rif binding to RNAP by more than three orders of magnitude while having little or no effect on RNAP–aptamer interactions.

(b¢W217His6). The aptamers differed in their affinity to the mutants (Table 1). The flap deletion had the least pro- nounced effect on the interactions of the aptamers with RNAP, significantly affecting the binding of only two of them, E1 and E11 (their Kd values were increased 5.6- and 11.2-fold, respectively). In contrast, the binding of most of the aptamers was disturbed severely by the b2 domain deletion (for example, Kd for E9 increased about 250-fold) and the only aptamer that bound this mutant with considerable affinity was E7. The most interesting results were obtained with the b¢W217His6 insertion mutant. While some of the aptamers (E13, E8) interacted with the mutant with unchanged affinity, binding of the others was weak- ened to different degrees (Table 1). The strongest effect was for aptamer E3 (Kd increased (cid:1) 100-fold). The simplest interpretation of the observed effects is that the regions of RNAP changed by the mutations are parts of the aptamers’ binding sites.

Effect of rifampicin on the binding of aptamers

Rif binds near the RNAP active center at the so-called Rif- binding pocket of the b-subunit and sterically prevents the synthesis of RNAs longer than a dinucleotide (Fig. 1B). Rif also prevents the binding of MS to the core enzyme [13]. We confirmed this result and found that Rif inhibited MS binding with an apparent Ki of < 0.5 nM (Fig. 4). This value is in good agreement with earlier reports on Rif Kd for binding to RNAP (0.5–2 nM) [50,51].

The r70-subunit and GreB suppress the interaction of the core RNAP with aptamers The r70-subunit inhibited the binding of all the aptamers to the core polymerase. Apparent Kds for the binding of different aptamers to the holoenzyme of RNAP were increased in the range 8–30 times in comparison with those for the core enzyme (Table 1). When the binding of the E2 aptamer was measured at fixed core and increasing r70- subunit concentrations, r inhibited the interaction with an observed Ki of (cid:1) 10 nM (Fig. 5A). This value apparently corresponded to Kd for the r70–core interaction at these conditions. The r70-subunit also suppressed the interaction of the core enzyme with MS (Fig. 5A). This result is in agreement with previous studies which demonstrated that the binding of r and RNA in the elongation complex was

Rif exhibited different effects on the interaction of various aptamers with RNAP (Table 1). The binding of all the aptamers obtained through Rif-directed selection (E5–E13) was inhibited by Rif with the same efficiency as the binding of MS (these aptamers were therefore called RifS, for Rif- sensitive, aptamers, Table 1 and Fig. 4). In contrast, most

Fig. 3. Inhibition of the Eco core polymerase activity by aptamers. RNAP activity was measured as described in Materials and methods. The core enzyme was added to the mixture of MS and aptamers in binding buffer containing 440 mM salt and transcription was initiated by adding [32P]UTP[aP]. The amount of radioactively labeled 9-nt-long RNA product was quantified and normalized to the activity in the absence of the inhibitor. I, Aptamers found only in the con- ventional selection experiment; II, aptamers unique to the Rif-directed experiment; I + II, aptamers identified in both selections; N, the initial library.

Aptamer probes of RNA polymerase structure (Eur. J. Biochem. 271) 4927

(cid:1) FEBS 2004

Fig. 5. Inhibition of aptamer binding by the r70-subunit and GreB. (A) Inhibition of the binding of aptamer E2 and MS (0.1 nM) to the core polymerase (1 and 2 nM, respectively) by increasing concentrations of the r-subunit. Binding buffer contained 440 mM salt in the case of the aptamers and 200 mM salt in the case of MS. (B) Inhibition of the binding of aptamer E9 (0.03 nM) to the core enzyme (3 nM) by increasing amounts of GreB. Binding was measured in buffer con- taining 440 mM salt.

mutually exclusive [53,54]. In the three-dimensional struc- ture of the holoenzyme polymerase, region 3.2 of r seems to clash with the 5¢ end of growing RNA during initiation (Fig. 1B) [7,8]. Thus, it is possible that r70 interferes with MS binding by competing with its RNA component for the same site on core RNAP.

GreB concentration resulted in complete inhibition of aptamer binding (Fig. 5B). The apparent Ki value for GreB action calculated from the inhibition curve was (cid:1) 100 nM. This value is in good agreement with Kd reported for the GreB–core interaction [56].

MS, Rif and the r70-subunit do not affect the stability of RNAP–aptamer complexes while GreB promotes their rapid dissociation To investigate the nature of the effects of MS, Rif, r70 and GreB on RNAP–aptamer interactions, we measured the dissociation kinetics of several RNAP–aptamer complexes in the presence of these ligands (Fig. 6). When the complexes containing radioactively labeled aptamers were

GreB exerts its effect on the elongation complex in a backtracked state stimulating the nuclease activity of the RNAP active center [55]. We found that the binding of MS to the core polymerase was not affected by GreB. GreB also failed to stimulate the cleavage of the RNA component of MS (data not shown). This, as well as resistance of MS to pyrophosphorolysis (N. Korzheva, personal communica- tion), suggested that MS was captured by RNAP in a post- translocated state. At the same time, GreB suppressed the interaction of Eco RNAP with all the aptamers tested except E7, increasing their apparent Kd values three- to sixfold, when present at 1.5 lM (Table 1). The weaker effect of GreB in comparison with the r-subunit is probably due to its lower affinity to core RNAP. Indeed, the increase of

Fig. 6. Dissociation kinetics of RNAP–aptamer complexes in the pres- ence of various competitors. The core enzyme was preincubated with a labeled aptamer and the complex was challenged with the corres- ponding unlabeled aptamer, MS, Rif, the r70-subunit or GreB. Aptamer binding was measured in buffer containing 440 mM salt. The dissociation kinetics is shown for aptamers E4 (A), E7 (B) and E10 (C).

4928 A. Kulbachinskiy et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

incubated with an excess of the corresponding unlabeled aptamers, they dissociated with half-life times of more than 1 h. The dissociation kinetics of the RNAP–aptamer complexes measured in the presence of MS, Rif (in case of RifS aptamers) or the r-subunit followed the kinetics observed when the unlabeled aptamer was used as a competitor (Fig. 6). Control experiments demonstrated that when these ligands were added to RNAP before the aptamers, they completely suppressed complex formation (data not shown).

Several facts indicated that the aptamers interact with the main channel of RNAP where nucleic acids in natural transcription complexes are held. All of the aptamers competed with MS for binding to RNAP and inhibited core polymerase activity. Binding of the aptamers was affected by mutations at several sites in the main channel that were previously implicated in the interactions with nucleic acids in transcription complexes. Furthermore, the binding of 10 out of 13 aptamers was sensitive to Rif. As Rif does not cause any significant conformational changes of the core polymerase [13], its effect must result from direct compe- tition with aptamers for the Rif pocket of the b-subunit. Finally, the dissociation kinetics of the RNAP–aptamer complexes measured in the presence of MS and Rif followed the same time course as the kinetics measured in the presence of the unlabeled aptamers. This indicated that these ligands acted by simple trapping of free RNAP and preventing reassociation of the complexes. Thus, both MS and Rif are likely to compete with the aptamers for the binding sites in the main channel.

In contrast, GreB greatly reduced the stability of several RNAP–aptamer complexes (Fig. 6 and data not shown). In agreement with Kd measurements, GreB did not affect the stability of the E7–RNAP complex (Fig. 6B). At the same time, when GreB was added to the preformed complexes of RNAP with E4 and E10 aptamers, it caused their rapid dissociation; half-life times of the complexes were reduced by more than 10 times (5 min in comparison with > 1 h when the kinetics was measured without GreB) (Fig. 6A and C). The residual binding of aptamers measured at large time intervals corresponded to the maximum inhibition observed when GreB was added before the aptamers (Fig. 5B and data not shown).

Specific and nonspecific interactions of aptamers with the RNAP main channel

In contrast

The interaction of the aptamers with RNAP was found to be highly dependent on the ionic strength of the solution. At elevated ionic strength (440 mM), the binding of the aptamers was very sequence specific as even point mutations of aptamers’ sequences disrupted their interaction with RNAP. The aptamers were also specific to Eco core RNAP and neither of them bound Taq RNAP (data not shown). At lower ionic strength (< 200 mM), RNAP still bound the aptamers but sequence specificity was apparently lost. Under these conditions all the sequences tested, including the random DNA library, bound the core enzyme with equal affinities (Kd (cid:1) 1 nM). MS suppressed the binding of all the oligonucleotides which suggested that the nonspecific bind- ing of ssDNA also occurred at RNAP sites involved in the interaction with RNA and DNA in the elongation complex. At elevated ionic strength, Rif and r70 suppressed RNAP–aptamer interactions (above). Under low ionic strength conditions, Rif and r70 had no effect on the binding of RifS aptamers to core RNAP (data not shown). Therefore, the structure of nonspecific complexes of RNAP the with the aptamers differs from the structure of complexes formed at high ionic strength.

Discussion

The r-subunit also binds within the main channel of RNAP. The main docking sites of r on the core polymerase include the clamp domain of b¢ and the flexible flap domain of the b-subunit (Fig. 1B) [7,8]. In addition, the N-terminal region of r, which is not visible in the holoenzyme structure, was shown to occupy the downstream portion of the main channel [57]. The binding of r to the core polymerase causes repositioning of several structural modules of the core, including the clamp, b1, b2 and flap domains, which results in partial closure of the main channel [7]. Thus, the inhibition of aptamer binding by r could occur by both steric and allosteric mechanisms. We found that, similarly to MS and Rif, r did not affect the dissociation rate of RNAP–aptamer complexes. Thus, the most likely inter- pretation of the inhibitory effect of r is that it also directly blocks RNAP sites involved in aptamer binding. The steric competition between aptamers and r is not surprising, when taking into account the extensive interaction interface between r and the core polymerase. Hopefully, further studies of mutant variants of r as well as testing various alternative r-subunits will help to establish the regions of r which are responsible for the inhibition of aptamer binding. to MS, Rif and the r-subunit, GreB dramatically increased the dissociation rate of RNAP– aptamer complexes and therefore actively disrupted RNAP–aptamer interactions. As opposed to r70, GreB binds RNAP from the secondary channel side of the i.e. at the side opposite to the aptamers (see enzyme, Fig. 1B) [15]. The binding site of the C-terminal domain of GreB near the entrance of the secondary channel is located outside of the enzyme’s catalytic cleft and seems unlikely to be involved in aptamer binding. The GreB N-terminal coiled-coil domain protrudes deep into the secondary channel, providing two conserved acidic residues which play a key role in the RNA cleavage reaction [15,16,58]. Based on these observations, one could suggest two mechanisms of GreB action on the binding of the aptamers. One possibility is that the aptamers bound in the main channel might occupy the mouth of the secondary channel and directly interfere with GreB binding. Alternatively, the aptamers could sense GreB-induced conformational chan- ges inside the RNAP main channel.

The principal result of this work is that the aptamers sense the interaction of RNAP with various ligands, including nucleic acids, antibiotics and protein factors. Based on the mechanism of the inhibition of aptamer binding, these ligands can be divided into two groups. The minimal nucleic acid scaffold, Rif and the r70-subunit seem to inhibit RNAP–aptamer interactions by steric blocking of the aptamer binding sites on the RNAP molecule, while GreB is likely to affect aptamer binding in an allosteric manner.

Aptamer probes of RNA polymerase structure (Eur. J. Biochem. 271) 4929

(cid:1) FEBS 2004

5. Cramer, P., Bushnell, D.A. & Kornberg, R.D. (2001) Structural basis of transcription: RNA polymerase II at 2.8 angstrom resolution. Science 292, 1863–1876.

6. Gnatt, A.L., Cramer, P., Fu, J., Bushnell, D.A. & Kornberg, R.D. (2001) Structural basis of transcription: an RNA polymerase II elongation complex at 3.3 A˚ resolution. Science 292, 1876– 1882.

7. Murakami, K.S., Masuda, S. & Darst, S.A. (2002) Structural basis of transcription initiation: RNA polymerase holoenzyme at 4 A˚ resolution. Science 296, 1280–1284.

and seems

to utilize

channel

very

8. Vassylyev, D.G., Sekine, S., Laptenko, O., Lee, J., Vassylyeva, M.N., Borukhov, S. & Yokoyama, S. (2002) Crystal structure of a bacterial RNA polymerase holoenzyme at 2.6 A˚ resolution. Nature 417, 712–719.

9. Korzheva, N., Mustaev, A., Kozlov, M., Malhotra, A., Nikiforov, V., Goldfarb, A. & Darst, S.A. (2000) A structural model of transcription elongation. Science 289, 619–625.

The strong stimulatory effect of GreB on the disso- ciation of RNAP–aptamer complexes provides serious evidence in support of the allosteric mechanism of GreB action. Sensing of GreB binding by several aptamers, each interacting with RNAP in a different way, as well as different strengths of GreB effect on various aptamers (Table 1) is also consistent with the allosteric mechanism. Our data thus give evidence that the interaction of GreB with RNAP may result in structural changes of the core polymerase. The resolution of current structural data does not allow us to verify such changes [15]. However, conformational rearrangements in the main channel were observed in the complex of yeast RNAPII with elonga- tion factor TFIIS, which also protrudes into the secon- similar dary mechanisms to stimulate RNA cleavage [59]. GreB- induced conformational changes of RNAP detected with the aptamers may be essential for the stimulation of the endonuclease activity of the enzyme.

10. Naryshkin, N., Revyakin, A., Kim, Y., Mekler, V. & Ebright, R.H. (2000) Structural organization of the RNA polymerase- promoter open complex. Cell 101, 601–611.

11. Murakami, K.S., Masuda, S., Campbell, E.A., Muzzin, O. & Darst, S.A. (2002) Structural basis of transcription initiation: an RNA polymerase holoenzyme-DNA complex. Science 296, 1285– 1290. 12. Darst, S.A. (2001) Bacterial RNA polymerase. Curr. Opin. Struct. Biol. 11, 155–162.

13. Campbell, E.A., Korzheva, N., Mustaev, A., Murakami, K., Nair, S., Goldfarb, A. & Darst, S.A. (2001) Structural mechanism for rifampicin inhibition of bacterial RNA polymerase. Cell 104, 901–912.

14. Epshtein, V., Mustaev, A., Markovtsov, V., Bereshchenko, O., Nikiforov, V. & Goldfarb, A. (2002) Swing-gate model of nucleotide entry into the RNA polymerase active center. Mol. Cell 10, 623–634.

conformational

to verify. The aptamers

Recent studies demonstrated that other protein factors (e.g. DksA) and antibiotics (microcin) also bind RNAP within the secondary channel and seem to affect RNAP conformation [60–62]. We propose that the aptamers could be used to study the conformational changes of RNAP induced by the binding of these regulatory factors. The aptamers could also be useful in studies of various RNAP mutations that are thought to change the conformation of the enzyme. The examples of such mutations include the substitution at position 934 near the F-bridge helix in the b¢-subunit that was proposed to shift the conformation of the F-bridge toward the bent form [14] and mutations on the surface of the b-subunit that impair Q-protein mediated anti-termination (pre- sumably by changing the conformation of the interior of the main channel) [63]. It should be noted that such changes of RNAP are hypothetical usually very difficult thus represent a very useful tool to probe RNAP structure in many experimental systems.

15. Opalka, N., Chlenov, M., Chacon, P., Rice, W.J., Wriggers, W. & Darst, S.A. (2003) Structure and function of the transcription elongation factor GreB bound to bacterial RNA polymerase. Cell 114, 335–345.

16. Sosunova, E., Sosunov, V., Kozlov, M., Nikiforov, V., Goldfarb, A. & Mustaev, A. (2003) Donation of catalytic residues to RNA polymerase active center by transcription factor Gre. Proc. Natl Acad. Sci. USA 100, 15469–15474.

Acknowledgements

17. Darst, S.A., Opalka, N., Chacon, P., Polyakov, A., Richter, C., Zhang, G. & Wriggers, W. (2002) Conformational flexibility of bacterial RNA polymerase. Proc. Natl Acad. Sci. USA 99, 4296–4301. 18. Landick, R. (2001) RNA polymerase clamps down. Cell 105, 567–570.

19. Bushnell, D.A., Cramer, P. & Kornberg, R.D. (2002) Structural transcription: alpha-amanitin-RNA polymerase II resolution. Proc. Natl Acad. Sci. USA 99, basis of cocrystal at 2.8 A˚ 1218–1222. We thank K. Severinov for protein and plasmid samples and for reading the manuscript, K. Kuznedelov and S. Borukhov for materials, A. Stolyarenko for reading the manuscript. A.K. is especially grateful to N. Korzheva, V. Epshtein and A. Mustaev for help in doing some experiments. This work was supported by the NIH grant GM30717 to A.G. and by the Russian Foundation for Basic Research grant 02-04- 48525.

References

1. Gelles, J. & Landick, R. (1998) RNA polymerase as a molecular 20. Yuzenkova, J., Delgado, M., Nechaev, S., Savalia, D., Epshtein, V., Artsimovitch, I., Mooney, R.A., Landick, R., Farias, R.N., Salomon, R. & Severinov, K. (2002) Mutations of bacterial RNA polymerase leading to resistance to microcin j25. J. Biol. Chem. 277, 50867–50875. motor. Cell 93, 13–16.

21. Artsimovitch, I., Chu, C., Lynch, A.S. & Landick, R. (2003) A new class of bacterial RNA polymerase inhibitor affects nucleotide addition. Science 302, 650–654. 2. Erie, D.A. (2002) The many conformational states of RNA polymerase elongation complexes and their roles in the regulation of transcription. Biochim. Biophys. Acta 1577, 224–239.

22. Gold, L., Polisky, B., Uhlenbeck, O. & Yarus, M. (1995) Diversity of oligonucleotide functions. Annu. Rev. Biochem. 64, 763– 797. 3. Korzheva, N. & Mustaev, A. (2003) RNA and DNA polymerases. In Molecular Motors (Schliwa, M., ed.), pp. 153–177. Wiley, John & Sons Inc., Hoboken, NJ, USA.

23. Tuerk, C., MacDougal, S. & Gold, L. (1992) RNA pseudoknots that inhibit human immunodeficiency virus type 1 reverse tran- scriptase. Proc. Natl Acad. Sci. USA 89, 6988–6992. 4. Zhang, G., Campbell, E.A., Minakhin, L., Richter, C., Severinov, K. & Darst, S.A. (1999) Crystal structure of Thermus aquaticus core RNA polymerase at 3.3 A˚ resolution. Cell 98, 811–824.

4930 A. Kulbachinskiy et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

polymerase and their effect on DNA-dependent RNA synthesis. FEBS Lett. 58, 359–362. 24. Jaeger, J., Restle, T. & Steitz, T.A. (1998) The structure of HIV-1 reverse transcriptase complexed with an RNA pseudoknot inhibitor. EMBO J. 17, 4535–4542.

25. Dang, C. & Jayasena, S.D. (1996) Oligonucleotide inhibitors of Taq DNA polymerase facilitate detection of low copy number targets by PCR. J. Mol. Biol. 264, 268–278. 43. deHaseth, P.L., Lohman, T.M., Burgess, R.R. & Record, M.T. Jr (1978) Nonspecific interactions of Escherichia coli RNA polymerase with native and denatured DNA: differences in the binding behavior of core and holoenzyme. Biochemistry 17, 1612– 1622.

26. Allen, P., Worland, S. & Gold, L. (1995) Isolation of high-affinity RNA ligands to HIV-1 integrase from a random pool. Virology 209, 327–336.

44. Strauss, H.S., Burgess, R.R. & Record, M.T. Jr (1980) Binding of Escherichia coli ribonucleic acid polymerase holoenzyme to a bacteriophage T7 promoter-containing fragment: selectivity exists over a wide range of solution conditions. Biochemistry 19, 3496– 3504.

27. Rowsell, S., Stonehouse, N.J., Convery, M.A., Adams, C.J., Ellington, A.D., Hirao, I., Peabody, D.S., Stockley, P.G. & Phillips, S.E. (1998) Crystal structures of a series of RNA aptamers complexed to the same protein target. Nat. Struct. Biol. 5, 970– 975.

45. Wheeler, A.R., Woody, A.Y. & Woody, R.W. (1987) Salt- dependent binding of Escherichia coli RNA polymerase to DNA and specific transcription by the core enzyme and holoenzyme. Biochemistry 26, 3322–3330. 46. Tissieres, A., Bourgeois, S. & Gros, F. (1963) Inhibition of RNA polymerase by RNA. J. Mol. Biol. 7, 100–103. 28. Huang, D.B., Vu, D., Cassiday, L.A., Zimmerman, J.M., Maher, L.J. & 3rd & Ghosh, G. (2003) Crystal structure of NF-kappaB (p50) 2 complexed to a high-affinity RNA aptamer. Proc. Natl Acad. Sci. USA 100, 9268–9273. 29. Gold, L. (1995) Oligonucleotides as research, diagnostic, and therapeutic agents. J. Biol. Chem. 270, 13581–13584. 30. Brody, E.N. & Gold, L. (2000) Aptamers as therapeutic and 47. Bremer, H., Yegian, C. & Konrad, M. (1966) Inactivation of purified Escherichia coli RNA polymerase by transfer RNA. J. Mol. Biol. 16, 94–103. diagnostic agents. J. Biotechnol. 74, 5–13. 31. Hale, S.P. & Schimmel, P. (1996) Protein synthesis editing by a 48. Spassky, A., Busby, S.J., Danchin, A. & Buc, H. (1979) On the binding of tRNA to Escherichia coli RNA polymerase. Eur. J. Biochem. 99, 187–201. DNA aptamer. Proc. Natl Acad. Sci. USA 93, 2755–2758.

49. Altmann, C.R., Solow-Cordero, D.E. & Chamberlin, M.J. (1994) RNA cleavage and chain elongation by Escherichia coli DNA- dependent RNA polymerase in a binary enzyme. RNA complex. Proc. Natl Acad. Sci. USA 91, 3784–3788. 32. Biroccio, A., Hamm, J., Incitti, I., De Francesco, R. & Tomei, L. (2002) Selection of RNA aptamers that are specific and high- affinity ligands of the hepatitis C virus RNA-dependent RNA polymerase. J. Virol. 76, 3688–3696.

33. Hamm, J., Alessi, D.R. & Biondi, R.M. (2002) Bi-functional, substrate mimicking RNA inhibits MSK1-mediated cAMP- response element-binding protein phosphorylation and reveals magnesium ion-dependent conformational changes of the kinase. J. Biol. Chem. 277, 45793–45802. 50. Handschin, J.C. & Wehrli, W. (1976) On the kinetics of the rifampicin-RNA-polymerase complex. Differences between crude and purified enzyme fractions. Eur. J. Biochem. 66, 309–317. 51. Yarbrough, L.R., Wu, F.Y. & Wu, C.W. (1976) Molecular mechanism of the rifampicin–RNA polymerase interaction. Bio- chemistry 15, 2669–2676.

34. Korzheva, N., Mustaev, A., Nudler, E., Nikiforov, V. & Gold- farb, A. (1998) Mechanistic model of the elongation complex of Escherichia coli RNA polymerase. Cold Spring Harb. Symp Quant. Biol. 63, 337–345. 52. Heisler, L.M., Suzuki, H., Landick, R. & Gross, C.A. (1993) Four contiguous amino acids define the target for streptolydigin resistance in the beta subunit of Escherichia coli RNA polymerase. J. Biol. Chem. 268, 25369–25375.

53. Sidorenkov, I., Komissarova, N. & Kashlev, M. (1998) Crucial role of the RNA: DNA hybrid in the processivity of transcription. Mol. Cell 2, 55–64. 35. Kashlev, M., Martin, E., Polyakov, A., Severinov, K., Nikiforov, V. & Goldfarb, A. (1993) Histidine-tagged RNA polymerase: dissection of the transcription cycle using immobilized enzyme. Gene 130, 9–14.

36. Borukhov, S. & Goldfarb, A. (1993) Recombinant Escherichia coli individually overexpressed 54. Daube, S.S. & von Hippel, P.H. (1999) Interactions of Escherichia coli sigma (70) within the transcription elongation complex. Proc. Natl Acad. Sci. USA 96, 8390–8395. RNA polymerase: purification of subunits and in vitro assembly. Protein Expr. Purif. 4, 503–511.

55. Borukhov, S., Laptenko, O. & Lee, J. (2001) Escherichia coli functions and transcript cleavage factors GreA and GreB: mechanisms of action. Methods Enzymol. 342, 64–76.

37. Kulbachinskiy, A.V., Ershova, G.V., Korzheva, N.V., Brodolin, K.L. & Nikiforov, V.G. (2002) Mutations in b¢ subunit of the Escherichia coli RNA polymerase influence interaction with the downstream DNA duplex in the elongation complex. Genetika [Russian] 38, 1207–1211.

56. Koulich, D., Orlova, M., Malhotra, A., Sali, A., Darst, S.A. & Borukhov, S. (1997) Domain organization of Escherichia coli transcript cleavage factors GreA and GreB. J. Biol. Chem. 272, 7201–7210. 38. Severinov, K. & Darst, S.A. (1997) A mutant RNA polymerase that forms unusual open promoter complexes. Proc. Natl Acad. Sci. USA 94, 13481–13486.

57. Mekler, V., Kortkhonjia, E., Mukhopadhyay, J., Knight, J., Revyakin, A., Kapanidis, A.N., Niu, W., Ebright, Y.W., Levy, R. & Ebright, R.H. (2002) Structural organization of bacterial RNA polymerase holoenzyme and the RNA polymerase-promoter open complex. Cell 108, 599–614. 39. Kuznedelov, K., Minakhin, L., Niedziela-Majka, A., Dove, S.L., Rogulja, D., Nickels, B.E., Hochschild, A., Heyduk, T. & Sever- inov, K. (2002) A role for interaction of the RNA polymerase flap domain with the sigma subunit in promoter recognition. Science 295, 855–857.

58. Laptenko, O., Lee, J., Lomakin, I. & Borukhov, S. (2003) Transcript cleavage factors GreA and GreB act as transient catalytic components of RNA polymerase. EMBO J. 22, 6322– 6334. 40. Minakhin, L., Nechaev, S., Campbell, E.A. & Severinov, K. (2001) Recombinant Thermus aquaticus RNA polymerase, a new tool for structure-based analysis of transcription. J. Bacteriol. 183, 71–76.

59. Kettenberger, H., Armache, K.J. & Cramer, P. (2003) Archi- tecture of the RNA polymerase II-TFIIS complex and implica- tions for mRNA cleavage. Cell 114, 347–357. 41. Carey, J., Cameron, V., de Haseth, P.L. & Uhlenbeck, O.C. (1983) Sequence–specific interaction of R17 coat protein with its ribo- nucleic acid binding site. Biochemistry 22, 2601–2610.

60. Perederina, A., Svetlov, V., Vassylyeva, M.N., Tahirov, T.H., Yokoyama, S., Artsimovitch, I. & Vassylyev, D.G. (2004) Regulation through the secondary channel – structural framework 42. Yefimova, L.Y., Knorre, V.L., Savinkova, L.K. & Salganik, R.I. (1975) Selective binding of oligoribonucleotides by E. coli RNA

Aptamer probes of RNA polymerase structure (Eur. J. Biochem. 271) 4931

(cid:1) FEBS 2004

62. Nickels, B.E. & Hochschild, A. (2004) Regulation of RNA poly- for ppGpp-DksA synergism during transcription. Cell 118, 297– 309. merase through the secondary channel. Cell 118, 281–284.

63. Santangelo, T.J., Mooney, R.A., Landick, R. & Roberts, J.W. (2003) RNA polymerase mutations that impair conversion to a termination-resistant complex by Q antiterminator proteins. Genes Dev. 17, 1281–1292. 61. Adelman, K., Yuzenkova, J., La Porta, A., Zenkin, N., Lee, J., Lis, J.T., Borukhov, S., Wang, M.D. & Severinov, K. (2004) Molecular mechanism of transcription inhibition by peptide antibiotic microcin j25. Mol. Cell 14, 753–762.