Stepwise mechanism for transcription fidelity
© Yuzenkova et al; licensee BioMed Central Ltd. 2010
Received: 1 April 2010
Accepted: 7 May 2010
Published: 7 May 2010
Transcription is the first step of gene expression and is characterized by a high fidelity of RNA synthesis. During transcription, the RNA polymerase active centre discriminates against not just non-complementary ribo NTP substrates but also against complementary 2'- and 3'-deoxy NTPs. A flexible domain of the RNA polymerase active centre, the Trigger Loop, was shown to play an important role in this process, but the mechanisms of this participation remained elusive.
Here we show that transcription fidelity is achieved through a multi-step process. The initial binding in the active centre is the major discrimination step for some non-complementary substrates, although for the rest of misincorporation events discrimination at this step is very poor. During the second step, non-complementary and 2'-deoxy NTPs are discriminated against based on differences in reaction transition state stabilization and partly in general base catalysis, for correct versus non-correct substrates. This step is determined by two residues of the Trigger Loop that participate in catalysis. In the following step, non-complementary and 2'-deoxy NTPs are actively removed from the active centre through a rearrangement of the Trigger Loop. The only step of discrimination against 3'-deoxy substrates, distinct from the ones above, is based on failure to orient the Trigger Loop catalytic residues in the absence of 3'OH.
We demonstrate that fidelity of transcription by multi-subunit RNA polymerases is achieved through a stepwise process. We show that individual steps contribute differently to discrimination against various erroneous substrates. We define the mechanisms and contributions of each of these steps to the overall fidelity of transcription.
Mutations in TL and surrounding amino acids were found to alter discrimination against non-complementary NTP (ncNTP) and complementary 2'- and 3'-deoxy NTPs (c2'dNTPs and c3'dNTPs) [6, 10, 11, 13, 14]. These observations led to suggestions that TL could participate in substrate discrimination by either: (i) selectively facilitating incorporation of correct substrates; or (ii) by providing sufficient time for accurate NTP selection prior to the act of catalysis. By analogy with DNAPs and ssRNAPs, the existence of a conformational change in the active centre of multi-subunit RNAPs (TL folding) that brings the active centre to a catalytically active conformation suggests that substrate discrimination may take place via differential efficiency of conformational change occurrence upon the binding of cNTPs versus ncNTPs. In support of this idea, inhibition of TL folding of yeast RNAP II by α-amanitin resulted in a substantial decrease in the accuracy of NTP addition . In contrast, the recent study on E. coli RNAP suggested that the open state of the active centre (with unfolded TL) may be a major checkpoint for discrimination against ncNTPs and c2'dNTPs .
Here, by using structure-based mutagenesis and fast kinetic analysis of phosphodiester bond formation by wild-type (WT) and mutant T. aquaticus RNAPs we uncover a stepwise mechanism of transcription fidelity, and dissect the role of TL in this process. Our results also reveal the role played by TL in the catalysis of phosphodiester bond formation.
Experimental set up and kinetic analysis
In order to establish the roles of the open and closed states of the active centre in catalysis and fidelity, we have constructed RNAP lacking TL (ΔTL RNAP)  and several RNAPs bearing substitutions of TL amino acids that directly face/contact the cNTP bound in the i+1 site when TL is folded . These amino acids included: Q1235, M1238, R1239, F1241, H1242 and T1243 (Figure 1). Compared to WT, F1241A and T1243A substitutions caused only minor, three- to fourfold defects in all reactions catalysed by the RNAP active centre (cNTP addition, pyrophosphorolysis, hydrolysis; see Additional File 1: Figure S1A, B, C, respectively). We considered these effects non-specific and did not investigate them further.
Given that the rates of the reactions catalysed by RNAP can be influenced by the translocation state of the elongation complex (pre- versus post-translocated), we used elongation complexes in which the translocation state equilibriums were not significantly affected by RNAP mutations (Additional File 1: Figure S3).
We have previously shown that misincorporated ncNMPs are efficiently removed from nascent RNA through transcript-assisted hydrolysis of the second (with respect to the 3' end of the RNA) phosphodiester bond of the transcript . Therefore, in order to measure the rate of misincorporation we needed to monitor both the extension of the nascent RNA with ncNMP and the subsequent transcript cleavage. In order to achieve this goal, initial transcripts were labelled at the 3' end allowing us to simultaneously observe single-nucleotide addition upon misincorporation and the generation of dinucleotide through transcript-assisted proofreading (Figure 2b). Note, for example, that, as seen from Figure 2b, it would have been virtually impossible to compare misincorporation by R1239A and H1242A RNAPs if only the RNA extension is monitored. Although they have similar rates of misincorporation, these RNAPs have very different rates of proofreading: the R1239A substitution does not influence intrinsic cleavage, while H1242A strongly diminishes it. This result also suggests that TL plays an important role in intrinsic RNA hydrolysis, and this is being investigated in a separate study.
Active centre of ΔTL RNAP is stabilized in the open conformation
Kd and kpol for incorporation and misincorporation by wild-type (WT) and ΔTL RNA polymerase (RNAP)
1.6 ± 0.1 × 10-3
36 ± 6
2.7 ± 0.3 × 10-2
2800 ± 600
4.2 ± 0.5 × 10-4
840 ± 190
5.5 ± 0.2 × 10-2
400 ± 30
1.3 ± 0.1 × 10-3
50 ± 9
1.4 ± 0.1× 10-1
55 ± 8
2.7 ± 0.2 × 10-3
57 ± 11
Kd and kpol for incorporation and misincorporation by mutant RNA polymerase (RNAP)
3.6 ± 0.2 × 10-3
24 ± 6
1.1 ± 0.4 × 10-4
710 ± 70
9.8 × 10-1
1.7 ± 0.2 × 10-3
2300 ± 500
2.7 ± 0.3 × 10-3
2700 ± 500
6.9 ± 0.5 × 10-2
37 ± 11
4.0 ± 0.4 × 10-4
2900 ± 500
5.6 ± 0.5 × 10-2
47 ± 17
3.6 ± 0.3 × 10-4
990 ± 230
Fidelity of RNAP active centre in the open state
Discrimination against various misincorporation events by ΔTL and wild-type RNA polymerase (WT RNAP).
WT RNAP incorporated ncNTP three orders of magnitude slower than cNTP. The rate of misincorporation of ncNTP by ΔTL RNAP was only slightly slower than incorporation of the cNTP (Table 1). WT/Stl RNAP misincorporated ncNTP at a rate similar to that of ΔTL RNAP (Table 2). Given that the active centres of ΔTL and WT/Stl RNAPs are stabilized in the open state, we conclude that the open state does not allow efficient discrimination based on the kinetics of incorporation of cNTP versus ncNTP misincorporation. This important result was not influenced by the differences in the elongation complexes we used for incorporation and misincorporation analysis (Additional File 1: Supplementary Text).
Induced fit discrimination and active expulsion of ncNTP from the active centre
As mentioned above, TL does not influence the affinity of the RNAP active centre for cNTP. We measured Kd [ncNTP] for WT and ΔTL RNAPs. The affinity of ΔTL (and WT/Stl) RNAP for ncNTP was ~20 times lower than that for cNTP (Table 1). The difference in Kd [cNTP] and Kd [ncNTP] is probably due to improper hydrogen bonding with the template base in the case of ncNTP and is in agreement with theoretical calculations of base pair formation between non-cognate nucleotide bases . Unexpectedly, Kd [ncNTP] for WT RNAP was significantly higher than that for ΔTL (Table 1) or WT/Stl (Table 2) RNAPs, indicating that ncNTP bound in the i+1 site and the folded TL compete with each other. According to the structural data on T. thermophilus RNAP and yeast RNAP II, cNTP and folded TL are structurally complementary (Figure 1) [9, 11]. The competition between folded TL and ncNTP can, therefore, be explained by the absence of such complementarity, probably caused by an altered geometry of a non Watson-Crick (G-T) base pair and, as a result, by a steric collision between them. Given that the folded TL participates in phosphodiester bond formation (see below), steric competition of ncNTP bound in the i+1 site with the folded TL leads to a diminished rate of phosphodiester bond synthesis - to a slow misincorporation. No such competition takes place in the case of cNTP bound in the active centre (Table 1), which allows efficient TL folding and catalysis. Therefore, TL-dependent cNTP incorporation proceeds much faster than TL-dependent ncNTP misincorporation, leading to kinetic discrimination against ncNTP.
The mainly kinetic (rather than affinity) based selection of ncNTPs by RNAP suggests that discrimination takes place via an induced fit mechanism . From our results it follows that: (i) TL is required for catalysis (and participates in it, see below); and (ii) TL is sterically conflicting with the ncNTP bound in the active centre (as follows from their competition in the active centre). This suggests that the conformational change (TL closing) in the active centre required for incorporation of cNTPs is obstructed in the presence of ncNTPs, thus supporting the induced fit discrimination hypothesis. TL-dependent induced-fit kinetic discrimination, therefore, is another step assuring fidelity of transcription.
M1238 governs TL folding and expulsion of ncNTP from the active centre
Discrimination against 2' and 3'-deoxy substrates by ΔTL RNA polymerase.
Role of TL in catalysis
We further investigated the basis for the TL-dependent kinetic step of transcription fidelity. In the crystal structure, R1239 and H1242 of the folded TL contact phosphates of the triphosphate moiety of cNTP bound in the i+1 site  (Figure 1). Alanine substitutions of either R1239 or H1242 reduce the rate of cNTP incorporation 50-100-fold (Tables 1 and 2). The effects of these substitutions were cumulative: the double alanine substitution (R1239A/H1242A) led to a further decrease in the rate of cNTP incorporation (Table 2), which was approximately the multiplicative product of individual defects observed for R1239A and H1242A mutants. This indicates that R1239 and H1242 act at the same kinetic step of catalysis. Furthermore, the rate of pyrophosphorolysis was also affected similarly by R1239A and H1242A substitutions (Figure 3b, Additional File 1: Table S3), suggesting that R1239 and H1242 play functionally analogous roles during catalysis of phosphotransfer reactions. In order to test if the decreased rate of cNTP incorporation by mutant RNAPs is caused by distortion of TL folding, we analysed the ability of R1239A/H1242A TL to displace ncNTP from the active centre. As can be seen from Table 1, R1239A/H1242A RNAP has Kd [ncNTP] similar to that of WT (and significantly higher than those of ΔTL or WT/Stl). Furthermore, R1239A substitution had no effect on intrinsic RNA hydrolytic activity (Figure 3c), which also requires TL folding . These data indicate that TL with R1239A/H1242A double substitution has retained its ability to fold and that the defect caused by this mutation is caused by deficiency in catalysis per se. The magnitude of R1239A/H1242A substitution effect on phosphodiester bond formation indicates that these two amino acids are the main determinants of TL function in catalysis.
Therefore, in agreement with crystallographic studies of T. thermophilus RNAP elongation complex , our results suggest that R1239 and H1242 coordinate the triphosphate moiety of cNTP in the i+1 site to allow efficient transition state formation (transition state stabilization and/or orientation catalysis), although they are also indirectly participating in general base catalysis. We therefore propose that the kinetic induced-fit discrimination against ncNTP is based on different efficiencies of transition state stabilization and/or reactants orientation and, in part, of base catalysis during cNTP incorporation versus ncNTP misincorporation.
Discrimination against 2'-deoxyNTP
Incorporation of 2'-deoxyNTPs in RNA may be detrimental to cells as 2' hydroxyls are frequently involved in RNA ternary structure and function and may influence mRNA translation. We therefore investigated the mechanisms assuring discrimination against complementary 2'-deoxyNTPs (c2'dNTPs) during transcription. As can be seen from Table 1, ΔTL RNAP only poorly distinguished between c2'dNTP and cNTP, while discrimination by WT RNAP was strong (~2000 fold, see Tables 1, 4 and 5). The observed poor kinetic discrimination by ΔTL RNAP was not due to differences in the elongation complexes used for cNTP and c2'dNTP incorporation (Additional File 1: Supplementary Text). Low kinetic discrimination by ΔTL RNAP was observed for all c2'dNTPs (Table 4). The highest discrimination was achieved against cTTP, which was incorporated only 17 times slower than cUTP (Table 4). The result indicates that in the open state of the active centre, discrimination against c2'dNTP is ineffective; in other words, discrimination against c2'dNTP mainly depends on the presence of TL.
The kinetic discrimination against c2'dNTPs implies that the induced fit mechanism is involved . We tested if TL folding, that apparently determines induced fit discrimination, would result in expulsion of c2'dNTP from the active centre - that is, if Kd [c2'dNTP] of WT RNAP would be higher than that of ΔTL RNAP. As can be seen from Table 1, WT RNAP indeed had ~10-fold higher Kd [c2'dNTP] than ΔTL RNAP, indicating that TL competes with c2'dNTP in the i+1 site. The competition between c2'dNTP and TL leads to the inhibition of TL folding and, thus, of productive catalysis compared to cNTP which does not compete with TL. The result supports the idea that discrimination against c2'dNTP proceeds via induced fit mechanism. The result also suggests that, as in the case of ncNTPs, TL participates in the active removal of c2'dNTP from the active centre.
Contribution of individual fidelity steps.
Open active centre
Trigger loop folding
(kpol [cGTP]/Kd [cGTP])/(kpol [ncGTP]/Kd [ncGTP])
Kinetic kpol [cGTP]/kpol [ncGTP]
Affinity Kd [ncGTP]/Kd [cGTP]
(kpol [cGTP]/Kd [cGTP])/(kpol [c2'dATP]/Kd [c2'dATP])
Kinetic kpol [cGTP]/kpol [c2'dATP]
Affinity Kd [c2'dATP]/Kd [cGTP]
(kpol [cGTP]/Kd [cGTP])/(kpol [c3'dATP]/Kd [c3'dATP])
Kinetic kpol [cGTP]/kpol [c3'dATP]
Affinity Kd [c3'dATP]/Kd [cGTP]
Discrimination against 3'-deoxyNTP
Incorporation of NTPs bearing modifications of the 3' hydroxyl or lacking this group abolishes further transcription. We therefore analysed the mechanism(s) used by RNAP to discriminate against complementary 3'-deoxyNTPs (c3'dNTPs). WT RNAP incorporated c3'dNTP ~700 times slower than cNTP (Table 1). In contrast, ΔTL RNAP practically failed to discriminate against c3'dNTP (Tables 1, 4 and 5), suggesting that discrimination strictly depends on the presence of TL. The phenomenon was not due to the differences in elongation complexes or the identity of c3'dNTP (Additional File 1: Supplementary Text). Next, we measured Kd [c3'dNTP] for WT and ΔTL RNAPs. Surprisingly, WT RNAP had the same Kd [c3'dNTP] as ΔTL RNAP (Table 1), indicating that c3'dNTP does not compete with, and is therefore not expelled by, the folded TL. Therefore, only kinetic TL-dependent discrimination is utilized against c3'dNTPs.
We investigated the mechanism of TL-dependent kinetic discrimination against c3'dNTP. Based on crystallographic studies of yeast RNAP II elongation complexes, it was suggested that Q1235 of TL and β'N737 may act in concert to check for the presence of the 3' OH . We tested RNAP bearing Q1235A substitution for its ability to discriminate against c3'dNTP. As can be seen from Figure 4a, Q1235A RNAP showed only ~fivefold discrimination against c3'dNTP, compared to ~700-fold discrimination exhibited by WT RNAP. Therefore, Q1235 of TL is the primary determinant of kinetic discrimination against c3'dNTP. In line with this idea, N737A RNAP had a slight (fivefold) decrease in c3'dNTP discrimination which was, however, TL-dependent, because in the absence of TL, N737A substitution had no effect (ΔTL/N737A RNAP) (Additional File 1: Table S1).
We investigated the mechanism of Q1235-dependent kinetic discrimination. Q1235A substitution slowed down cNTP incorporation ~100 fold as compared to WT (Figure 4a). However, this effect was not due to altered TL folding as inferred from Kd [ncNTP] (4 ± 1 mM). Furthermore, the Q1235A substitution did not significantly influence intrinsic transcript hydrolysis (Figure 3c), which also requires TL folding . In order to test if Q1235 participates, along with R1239 and H1242, in catalysis, we analysed Q1235A/R1239A/H1242A triple mutant RNAP. Curiously, while the single Q1235A substitution had a significant effect on the rate of reaction, the triple mutant did not have any further defect in catalysis than the double R1239/H1242 mutant (Figure 4b). The result suggests that Q1235 does not directly participate in catalysis along with R1239 and H1242 but is somehow linked to the function of these amino acids, possibly acting at the step preceding transition state stabilization by R1239/H1242. Note also that Q1235A and R1239A substitutions have similar effects on catalysis and, in the crystal structure of yeast RNAP II elongation complex, Q1078 (a homologue of Q1235 of T. aquaticus RNAP) interacts with both N1082 (a homologue of R1239 of T. aquaticus RNAP) and the 3'OH of the substrate . These data suggest that Q1235 may participate in orientation of R1239 into a catalytically active conformation. When the incoming NTP lacks the 3' OH, Q1235 fails to perform this role, leading to inefficient catalysis and inefficient kinetic discrimination against c3'dNTPs. The contributions of different steps in discrimination against c3'dNTP are summarized in Table 5 and are also shown schematically in Figure 5.
Taken together, our results suggest that the fidelity of transcription is achieved via a stepwise process and that each step contributes differently to discrimination against non-complementary, 2'-deoxy, and 3'-deoxy erroneous substrates. The scheme summarizing our findings is presented in Figure 5. These steps are based on two conformational states of the RNAP active centre (open and close) and on a structural rearrangement of the active centre during transition between these two states. The open active centre can efficiently discriminate only against unNTPs (which, apparently, cannot base pair with template DNA) but is error-prone in the case of ncNTPs, c2'dNTP and c3'dNTPs. Similar conclusions were made during analysis of yeast RNAP II . That a 'motionless' active centre (that is, which lacks additional conformational changes) is error-prone appears to be a general concept, as illustrated, for example, by primases, which have error rates similar to that of ΔTL RNAP [20, 21].
The poor fidelity achieved in the open state of the RNAP active centre does not have a strong influence on the accuracy of transcription because this state is catalytically very inefficient. Folding of TL that makes the active centre catalytically efficient therefore determines the following steps that contribute to transcription fidelity. Our results indicate that the catalytic role of TL is primarily determined by R1239 and H1242, which stabilize reaction transition state (or orient the reactants) and also indirectly contribute to general base catalysis. Thus, folding of TL, which is governed by M1238, is mainly required in order to bring R1239 and H1242 to the position from which they can accomplish this function. The proposed steric competition between folded TL and ncNTP or c2'dNTP bound in i+1 site makes catalysis with these incorrect nucleotides inefficient. There is no such competition between folded TL and cNTP. Therefore, though the rate of misincorporation is increased during TL folding (compared to the open state of the active centre), the increase of cNTP incorporation rate is much higher, leading to kinetic discrimination against ncNTPs and c2'dNTPs. Interestingly, in the case of c3'dNTP, TL is proposed to fold but fails to properly orient its catalytic residues in the absence of the functional relationship between Q1235 and the 3'OH.
The TL-controlled fidelity checkpoint is an example of an induced fit mechanism of enzymatic accuracy. DNAPs and single-subunit RNAPs also use an induced fit mechanism to ensure fidelity. In these enzymes, the binding of correct substrate induces a large conformational change converting the complex to a catalytically active, closed, conformation [22, 23] (see also Introduction).
The principal finding of our study is that the apparent competition of the folded TL with ncNTP and c2'dNTP bound in the i+1 site results in the erroneous substrates' displacement from the active centre, further improving the fidelity by another order of magnitude (Table 5, Figure 5). The contribution of this step to transcription fidelity may be more important for c2'dNTPs, because these nucleotides bind in the active site with the same affinity as cNTPs and, therefore, should compete with correct substrates much more efficiently than ncNTPs. However, it should be noted that this step has no effect on the discrimination against c3'dNTPs.
Misincorporation rates by T. aquaticus WT RNAP measured in our study (Table 3) were qualitatively similar to those observed for yeast WT RNAP II , with two exceptions. T. aquaticus WT RNAP misincorporated ncGTP in place of cATP and ncUTP in place of cCTP much faster (relative to other misincorporation events) than did yeast WT RNAP II. This suggests that discrimination against some non-cognate NTPs may depend on the sequence of elongation complex or may vary between bacterial and eukaryotic RNAPs.
Curiously, results from a recent study on E. coli RNAP contradict some of our conclusions . By studying E. coli ΔTL RNAP, the authors concluded that the open state of the active centre is able to efficiently discriminate against ncNTP. Although our conclusions agree on some misincorporation events (unNTPs: ncCTP/cATP and ncUTP/cATP), there is a discrepancy in the case of ncGTP/cATP misincorporation. Zhang et al.  argue that they failed to observe meaningful misincorporation at 10 mM ncGTP, while T. aquaticus RNAP misincorporated only six times slower than cNTP (Table 3, Additional File 1: Figure S5B). The elongation complexes used in both studies were of the same sequence. This factor can, therefore, be excluded when analysing the possible source of this discrepancy. We have constructed an E. coli ΔTL RNAP with a TL deletion identical to the one used by Zhang et al.  and tested it for ncGTP misincorporation (Additional File 1: Figure S5). We observed efficient misincorporation even in 1 mM ncGTP by the mutant enzyme, which was ~50 times slower than incorporation rate in the presence of 1 mM cGTP. Taking into account the higher Kd [ncNTP] value, ~25-fold kinetic discrimination is expected in the open active centre, which is slightly higher than that observed for T. aquaticus ΔTL RNAP for the same misincorporation event. We also observed higher kinetic discrimination against c2'dNTP and c3'dNTP by E. coli ΔTL RNAP compared to T. aquaticus ΔTL RNAP in the same elongation complexes (Additional File 1: Figure S5).
Better discrimination by open-state active centre of E. coli RNAP is achieved due to the relatively faster TL-independent incorporation of cNTP by E. coli ΔTL RNAP compared to T. aquaticus ΔTL RNAP (compare Additional File 1: Figure S5, Table 1 and Figures 4 and 5 in ). This suggests that some amino acids of the E. coli RNAP active centre that lie outside TL may support catalysis and, thus, contribute to kinetic discrimination against erroneous NTPs in the open state of the active centre. This idea is supported by the fact that alanine substitutions of E. coli homologues of R1239 and H1242 have a far less drastic effect on catalysis (a four- to sixfold decrease) than the corresponding T. aquaticus RNAP substitutions (50-to 100-fold decrease; Tables 2). Therefore, E. coli possesses some additional catalytic mechanism that is lacking or diminished in T. aquaticus RNAP. Our results indicate that, in addition to TL-dependent transition state stabilization, the active centre performs general base catalysis which only partly depends on TL. It is therefore possible that E. coli RNAP relies on TL-independent base catalysis more than T. aquaticus RNAP and that this determines better kinetic discrimination in the open active centre. The drastic catalytic effect of the TL deletion in E. coli RNAP, however, indicates that this additional catalytic mechanism is still dependent on the TL, possibly through allosteric action of TL on amino acids of the active centre. Overall, our results suggest that the low accuracy of the open state of the active centre (except for unNTPs) is a general feature of multi-subunit RNAPs, though some species-specific variations are possible.
According to our results, the total fidelity of transcription that can be achieved through steps dissected in our study is ~10-6 for ncNTPs, ~10-5 for c2'dNTPs and ~10-3 for c3'dNTPs (Tables 3, 4, 5). At high NTP concentrations present in cells, accuracy is expected to drop by two orders of magnitude in the case of ncNTPs and one order of magnitude for c2'dNTPs, but to remain the same for c3'dNTPs. However, the fidelity of transcription can be further improved by a factor of 10-1-10-2 during the proofreading of transcription via transcript-assisted and cleavage factor-assisted (Gre in bacteria, TFIIS in eukaryotes and TFS in archaea) removal of erroneous nucleotides [15, 25–27].
Transcription is characterized by high fidelity of copying of genetic information into RNA. Discrimination against non-cognate and 2'- and 3'-deoxy NTPs entering the RNAP active centre proceeds via a stepwise mechanism involving not only discrimination based on Watson-Crick pairing but also active recognition of substrates by the catalytic domain of the RNAP active centre, the TL. Each step contributes differently to the overall fidelity against various erroneous substrates. A unique property of the multi-subunit RNAP active centre, compared to single-subunit RNAPs and DNAPs active centres, is the ability of the former to actively expel wrong substrates upon TL folding. This property may be particularly important in the case of 2'-deoxy NTPs, which are not discriminated during initial binding and, therefore, compete with the cognate rNTPs.
Induced fit discrimination based on kinetics of incorporation versus misincorporation is determined by the direct involvement of TL in catalysis along with the two Mg2+ ions of the RNAP active centre. Interestingly, such combinatorial way of catalysis (two Mg2+ ion catalysis + TL dependent transition state stabilization/base catalysis) indicates that the regulation of the activity of the enzyme can be accomplished through controlling TL folding, rather than the orientation of the catalytic Mg2+ ions. These effects were observed during analysis of RNAPs with mutations in non-catalytic amino acids surrounding the active site [5, 6, 10, 14, 28]. Regulation of catalysis via TL may also be accomplished by external factors . 'Combinatorial' catalysis also suggests that RNAPs of different organisms can adapt to various conditions by changing the amino acid composition of TL and surrounding amino acids, while leaving the two Mg2+ binding sites virtually intact . For example, changes in the F-loop, a domain close to TL, determine the temperature optima of catalysis by closely related RNAPs from thermophilic Thermus aquaticus and mesophilic Deinococcus radiodurans .
Mutant RNAP construction and purification
T. aquaticus RNAP lacking TL (β' residues 1238-1254 replaced with a glycine residue) was constructed as described . Single alanine substitutions in the TL of T. aquaticus RNAP (β' M1238A, R1239A, F1241A, H1242A and T1243A) were obtained by the site directed mutagenesis in a co-expression system for recombinant RNAP . WT and mutant core RNA polymerases were purified as described . In order to obtain E. coli ΔTL RNAP (having β' residues 931-1137 replaced with three alanine residues ) the deletion was introduced in the β' subunit coded under inducible promoter in plasmid pRL663. Mutant RNAP production was performed as described  and its purification as described .
Elongation complexes (Additional File 1: Figure S2) [3, 16] were assembled and immobilized on Ni-NTA beads (Qiagen, Venlo, The Netherlands) as described [15, 16]. The beads were washed with 1 M KCl to remove incorrectly assembled complexes. For fast kinetics experiments the immobilization step was omitted (which results in incomplete incorporation, see Figure 2a). The presence of non-template strand in all complexes was confirmed by treatment with Exo III (digests only double stranded DNA, not shown). When specified, RNAP was walked to a desired position by the addition of required substrates with their subsequent removal through beads washing. RNA was either 32P labelled at the 5'end  or labelled at 3'-end after elongation complex assembly by incorporation of [α-32P]GTP (PerkinElmer, MA, USA) with subsequent removal of unincorporated [α-32P]GTP through beads washing. Transcription buffer contained 40 mM KCl and 20 mM Tris pH 8.0, unless otherwise specified. Intrinsic RNA hydrolysis was initiated by the addition of 10 mM MgCl2. Incorporation and pyrophosphorolysis reactions were initiated by the addition of a mixture 10 mM MgCl2 with NTPs or PPi, respectively. For misincorporation, 20 mM MgCl2 was used in order to avoid possible titrating out of the Mg2+ by high NTP concentrations. Reactions were incubated at 40°C (20°C for pH dependences), stopped by the addition of 1 M HCl (final concentration) or formamide-containing buffer, neutralized if required and analysed by Urea-PAGE and Phosphorimaging (GE Healthcare, Buckinghamshire, UK). Exo III footprinting and hydroxyl radical probing of Additional File 1: Figure S3 was performed as described in the figure legend.
Fast flow kinetics
Fast kinetic experiments were performed using a BioLogic Quenched-Flow QFM-400 instrument (Bio-logic SAS, Claix, France) equipped with 3.5 μL ageing line. The settings and calibration of the QFM-400 were accomplished according to the standard procedure suggested by the manufacturer. Experiments were conducted at 40°C (20°C for pH dependences) by using a circulating water bath. Assembled elongation complexes were loaded in the syringe 1 and a solution containing various concentrations of correct nucleotide and 10 mM MgCl2 were loaded in syringe 2. 30 μL samples from syringes 1 and 2 were mixed and the incorporation was allowed to proceed for the indicated times. Reactions were quenched by addition of HCl to a final concentration of 1 M from syringe 4. Immediately after the addition of HCl, the solution was neutralized by the addition of 1 M KOH and 300 mM Tris base (final concentration). Products were analysed as described above.
For kinetics that are described by a single exponent (for other case see Additional File 1: Supplementary Information), we used P = A × (1-ekt), where P is the fraction of incorporated RNA, A is the fraction of active RNAP complexes, k is the reaction rate and t is the time. The rates obtained for various concentrations were fitted to the Michaelis-Menten equation; k = kpol × [NTP]/(KMobs+ [NTP]), where [NTP] is the substrate concentration, kpol is the maximum rate of the enzyme and KMobs is the Michaelis constant. Prism 5 (GraphPad Software, CA, USA) was used to perform the single exponential fits to misincorporation and incorporation data. The kinetic simulations were performed using the ISRES algorithm [34, 35] implemented in MATLAB (The MathWorks, MA, USA).). Nonlinear regression with least squares was used to obtain the best fit of the model to the data. We minimized the sum of squares between the experimental data and the data predicted by the model for a particular set of parameters to obtain the best fit parameters.
We thank Sergei Nechaev and David Adams for their critical reading of the manuscript. The work was supported by an NIH grant (GM64530; KS), BBSRC grant (BB/F006462/1; NZ), BBSRC grant (BB/I004564/1; NZ) in the framework of the SYSMO programme and an ERC grant (ERC-2007-StG 202994-MTP; NZ).
- Steitz TA: A mechanism for all polymerases. Nature. 1998, 391: 231-232. 10.1038/34542.View ArticlePubMedGoogle Scholar
- Steitz TA, Steitz JA: A general two-metal-ion mechanism for catalytic RNA. Proc Natl Acad Sci USA. 1993, 90: 6498-6502. 10.1073/pnas.90.14.6498.PubMed CentralView ArticlePubMedGoogle Scholar
- Sosunov V, Sosunova E, Mustaev A, Bass I, Nikiforov V, Goldfarb A: Unified two-metal mechanism of RNA synthesis and degradation by RNA polymerase. Embo J. 2003, 22: 2234-2244. 10.1093/emboj/cdg193.PubMed CentralView ArticlePubMedGoogle Scholar
- Sosunov V, Zorov S, Sosunova E, Nikolaev A, Zakeyeva I, Bass I, Goldfarb A, Nikiforov V, Severinov K, Mustaev A: The involvement of the aspartate triad of the active center in all catalytic activities of multisubunit RNA polymerase. Nucleic Acids Res. 2005, 33: 4202-4211. 10.1093/nar/gki688.PubMed CentralView ArticlePubMedGoogle Scholar
- Tan L, Wiesler S, Trzaska D, Carney HC, Weinzierl RO: Bridge helix and trigger loop perturbations generate superactive RNA polymerases. J Biol. 2008, 7: 40-10.1186/jbiol98.PubMed CentralView ArticlePubMedGoogle Scholar
- Kireeva ML, Nedialkov YA, Cremona GH, Purtov YA, Lubkowska L, Malagon F, Burton ZF, Strathern JN, Kashlev M: Transient reversal of RNA polymerase II active site closing controls fidelity of transcription elongation. Mol Cell. 2008, 30: 557-566. 10.1016/j.molcel.2008.04.017.View ArticlePubMedGoogle Scholar
- Temiakov D, Zenkin N, Vassylyeva MN, Perederina A, Tahirov TH, Kashkina E, Savkina M, Zorov S, Nikiforov V, Igarashi N, Matsugaki N, Wakatsuki S, Severinov K, Vassylyev DG: Structural basis of transcription inhibition by antibiotic streptolydigin. Mol Cell. 2005, 19: 655-666. 10.1016/j.molcel.2005.07.020.View ArticlePubMedGoogle Scholar
- Kettenberger H, Armache KJ, Cramer P: Complete RNA polymerase II elongation complex structure and its interactions with NTP and TFIIS. Mol Cell. 2004, 16: 955-965. 10.1016/j.molcel.2004.11.040.View ArticlePubMedGoogle Scholar
- Vassylyev DG, Vassylyeva MN, Zhang J, Palangat M, Artsimovitch I, Landick R: Structural basis for substrate loading in bacterial RNA polymerase. Nature. 2007, 448: 163-168. 10.1038/nature05931.View ArticlePubMedGoogle Scholar
- Kaplan CD, Larsson KM, Kornberg RD: The RNA polymerase II trigger loop functions in substrate selection and is directly targeted by alpha-amanitin. Mol Cell. 2008, 30: 47-556. 10.1016/j.molcel.2008.04.023.View ArticleGoogle Scholar
- Wang D, Bushnell DA, Westover KD, Kaplan CD, Kornberg RD: Structural basis of transcription: role of the trigger loop in substrate specificity and catalysis. Cell. 2006, 127: 941-954. 10.1016/j.cell.2006.11.023.PubMed CentralView ArticlePubMedGoogle Scholar
- Castro C, Smidansky ED, Arnold JJ, Maksimchuk KR, Moustafa I, Uchida A, Gotte M, Konigsberg W, Cameron CE: Nucleic acid polymerases use a general acid for nucleotidyl transfer. Nat Struct Mol Biol. 2009, 16: 212-218. 10.1038/nsmb.1540.PubMed CentralView ArticlePubMedGoogle Scholar
- Zhang J, Palangat M, Landick R: Role of the RNA polymerase trigger loop in catalysis and pausing. Nat Struct Mol Biol. 2009, 17: 99-104. 10.1038/nsmb.1732.PubMed CentralView ArticlePubMedGoogle Scholar
- Bar-Nahum G, Epshtein V, Ruckenstein AE, Rafikov R, Mustaev A, Nudler E: A ratchet mechanism of transcription elongation and its control. Cell. 2005, 120: 183-193. 10.1016/j.cell.2004.11.045.View ArticlePubMedGoogle Scholar
- Zenkin N, Yuzenkova Y, Severinov K: Transcript-assisted transcriptional proofreading. Science. 2006, 313: 518-520. 10.1126/science.1127422.View ArticlePubMedGoogle Scholar
- Sidorenkov I, Komissarova N, Kashlev M: Crucial role of the RNA:DNA hybrid in the processivity of transcription. Mol Cell. 1998, 2: 55-64. 10.1016/S1097-2765(00)80113-6.View ArticlePubMedGoogle Scholar
- Toulokhonov I, Zhang J, Palangat M, Landick R: A central role of the RNA polymerase trigger loop in active-site rearrangement during transcriptional pausing. Mol Cell. 2007, 27: 406-419. 10.1016/j.molcel.2007.06.008.View ArticlePubMedGoogle Scholar
- Mathews DH, Sabina J, Zuker M, Turner DH: Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. J Mol Biol. 1999, 288: 911-940. 10.1006/jmbi.1999.2700.View ArticlePubMedGoogle Scholar
- Svetlov V, Vassylyev DG, Artsimovitch I: Discrimination against deoxyribonucleotide substrates by bacterial RNA polymerase. J Biol Chem. 2004, 279: 38087-38090. 10.1074/jbc.C400316200.PubMed CentralView ArticlePubMedGoogle Scholar
- Swart JR, Griep MA: Primer synthesis kinetics by Escherichia coli primase on single-stranded DNA templates. Biochemistry. 1995, 34: 16097-16106. 10.1021/bi00049a025.View ArticlePubMedGoogle Scholar
- Frick DN, Richardson CC: DNA primases. Annu Rev Biochem. 2001, 70: 39-80. 10.1146/annurev.biochem.70.1.39.View ArticlePubMedGoogle Scholar
- Kunkel TA, Bebenek K: DNA replication fidelity. Annu Rev Biochem. 2000, 69: 497-529. 10.1146/annurev.biochem.69.1.497.View ArticlePubMedGoogle Scholar
- Temiakov D, Patlan V, Anikin M, McAllister WT, Yokoyama S, Vassylyev DG: Structural basis for substrate selection by t7 RNA polymerase. Cell. 2004, 116: 381-391. 10.1016/S0092-8674(04)00059-5.View ArticlePubMedGoogle Scholar
- Sydow JF, Brueckner F, Cheung AC, Damsma GE, Dengl S, Lehmann E, Vassylyev D, Cramer P: Structural basis of transcription: mismatch-specific fidelity mechanisms and paused RNA polymerase II with frayed RNA. Mol Cell. 2009, 34: 710-721. 10.1016/j.molcel.2009.06.002.View ArticlePubMedGoogle Scholar
- Lange U, Hausner W: Transcriptional fidelity and proofreading in Archaea and implications for the mechanism of TFS-induced RNA cleavage. Mol Microbiol. 2004, 52: 1133-1143. 10.1111/j.1365-2958.2004.04039.x.View ArticlePubMedGoogle Scholar
- Thomas MJ, Platas AA, Hawley DK: Transcriptional fidelity and proofreading by RNA polymerase II. Cell. 1998, 93: 627-637. 10.1016/S0092-8674(00)81191-5.View ArticlePubMedGoogle Scholar
- Erie DA, Hajiseyedjavadi O, Young MC, von Hippel PH: Multiple RNA polymerase conformations and GreA: control of the fidelity of transcription. Science. 1993, 262: 867-873. 10.1126/science.8235608.View ArticlePubMedGoogle Scholar
- Malagon F, Kireeva ML, Shafer BK, Lubkowska L, Kashlev M, Strathern JN: Mutations in the Saccharomyces cerevisiae RPB1 gene conferring hypersensitivity to 6-azauracil. Genetics. 2006, 172: 2201-2209. 10.1534/genetics.105.052415.PubMed CentralView ArticlePubMedGoogle Scholar
- Symersky J, Perederina A, Vassylyeva MN, Svetlov V, Artsimovitch I, Vassylyev DG: Regulation through the RNA polymerase secondary channel. Structural and functional variability of the coiled-coil transcription factors. J Biol Chem. 2006, 281: 1309-1312. 10.1074/jbc.C500405200.PubMed CentralView ArticlePubMedGoogle Scholar
- Miropolskaya N, Artsimovitch I, Klimasauskas S, Nikiforov V, Kulbachinskiy A: Allosteric control of catalysis by the F loop of RNA polymerase. Proc Natl Acad Sci USA. 2009, 106: 18942-19847. 10.1073/pnas.0905402106.PubMed CentralView ArticlePubMedGoogle Scholar
- Kuznedelov K, Minakhin L, Severinov K: Preparation and characterization of recombinant Thermus aquaticus RNA polymerase. Methods Enzymol. 2003, 370: 94-108. full_text.View ArticlePubMedGoogle Scholar
- Yuzenkova J, Delgado M, Nechaev S, Savalia D, Epshtein V, Artsimovitch I, Mooney RA, Landick R, Farias RN, Salomon R, Severinov K: Mutations of bacterial RNA polymerase leading to resistance to microcin j25. J Biol Chem. 2002, 277: 50867-50875. 10.1074/jbc.M209425200.View ArticlePubMedGoogle Scholar
- Kashlev M, Nudler E, Severinov K, Borukhov S, Komissarova N, Goldfarb A: Histidine-tagged RNA polymerase of Escherichia coli and transcription in solid phase. Methods Enzymol. 1996, 274: 326-334. full_text.View ArticlePubMedGoogle Scholar
- Runnarson TP, Yao X: Search biases on constrained evolutionary optimization. IEEE Ttrans Systems, Man and Cybernetics - Part C: Applications Rev. 2005, 35: 233-243. 10.1109/TSMCC.2004.841906.View ArticleGoogle Scholar
- Runnarson TP, Yao X: Stochastic ranking for constrained evolutionary optimization. IEEE Trans Evolutionary Computation. 2000, 4: 284-294. 10.1109/4235.873238.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.