Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jul 22.
Published in final edited form as: Mol Cell. 2011 Jul 22;43(2):253–262. doi: 10.1016/j.molcel.2011.05.026

The β subunit gate loop is required for RNA polymerase modification by RfaH and NusG

Anastasia Sevostyanova 1, Georgiy A Belogurov 1,4, Rachel A Mooney 2, Robert Landick 2,3, Irina Artsimovitch 1,*
PMCID: PMC3142557  NIHMSID: NIHMS306005  PMID: 21777814

SUMMARY

In all organisms, RNA polymerase (RNAP) relies on accessory factors to complete synthesis of long RNAs. These factors increase RNAP processivity by reducing pausing and termination, but their molecular mechanisms remain incompletely understood. We identify the β gate loop as an RNAP element required for antipausing activity of a bacterial virulence factor RfaH, a member of the universally conserved NusG family. Interactions with the gate loop are necessary for suppression of pausing and termination by RfaH, but are dispensable for RfaH binding to RNAP mediated by the β′ clamp helices. We hypothesize that, upon binding to the clamp helices and the gate loop, RfaH bridges the gap across the DNA channel, stabilizing RNAP/nucleic acid contacts and disfavoring isomerization into a paused state. We show that contacts with the gate loop are also required for antipausing by NusG and propose that most NusG homologs use similar mechanisms to increase RNAP processivity.

INTRODUCTION

The structure, effects on RNA synthesis, and the major binding site on the RNAP of proteins from the NusG family are universally conserved (Belogurov et al., 2009; Hirtreiter et al., 2010; Mooney et al., 2009b). These proteins are also involved in crosstalk between transcription and other key cellular processes, such as RNA processing and translation (Bies-Etheve et al., 2009; Burmann et al., 2010; Peterlin and Price, 2006; Schneider et al., 2006; Wang and Dennis, 2009). The E. coli NusG and RfaH are the best studied members of this family. NusG is a general transcription factor that is essential in the wild-type E. coli and is bound to nearly all transcribed genes (Mooney et al., 2009a). RfaH, a sequence-specific paralog of NusG, preferentially increases distal expression in operons containing promoter-proximal ops DNA elements. The ops sequence mediates RfaH binding to elongating RNAP and thus restricts RfaH action to just a few E. coli operons (Belogurov et al., 2009); consistently, rfaH is dispensable for cell viability. The N-terminal domains of RfaH and NusG are structurally homologous, are predicted to make similar contacts with RNAP, and are sufficient in vitro for the reduction of transcriptional pausing (Belogurov et al., 2009; Mooney et al., 2009b). However, full-length RfaH and NusG have opposite effects on the expression of poorly translated genes (e.g., foreign DNA) in vivo. NusG acts together with ρ to terminate transcription (Cardinale et al., 2008; Mooney et al., 2009b), whereas RfaH reduces termination via its antipausing (AP) activity and exclusion of NusG (Belogurov et al., 2009; Carter et al., 2004).

RfaH and other antiterminators, such as the phage λN and Q proteins (Roberts et al., 2008), appear to maintain the elongating RNAP complex (EC) in a rapidly-moving mode by preventing its isomerization into an off-pathway state called an elemental pause (Ederth et al., 2006). From this state, the paused EC is thought to isomerize into long-lived paused states (e.g., upon backtracking or formation of a nascent RNA hairpin) or to enter a termination pathway (Artsimovitch and Landick, 2000). Formation of the elemental pause likely involves fraying of the 3′ RNA nucleotide and rearrangement of an active-site element called the trigger loop, and occurs after nucleotide addition but before RNAP translocation (Sydow et al., 2009; Toulokhonov et al., 2007). Thus, a regulator could suppress pausing either by promoting translocation or by preventing active-site conformational changes, but the molecular mechanism of an AP modification remains unknown for any protein.

RfaH is an excellent model to address the question of how a regulator can modify RNAP to resist pause signals. RfaH action can be readily monitored in the cell because it is dispensable and controls just a few operons. In addition, RfaH structure, mechanism of recruitment, DNA- and RNAP-binding regions, and binding site on the RNAP have been determined. RfaH binds to the β′ clamp helices domain (β′CH) and the non-template (NT) DNA strand at the upstream end of the transcription bubble (Belogurov et al., 2007). Unlike the phage antiterminators that also stabilize the EC against dissociation (Rees et al., 1997; Yarnell and Roberts, 1999), RfaH does not affect the EC stability or intrinsic termination, except at one unusual signal (Artsimovitch and Landick, 2002; Carter et al., 2004), allowing direct testing of its AP mechanism. In the simplest scenario, RfaH may favor reannealing of the DNA strands and thus push the RNAP forward; indeed, we showed that RfaH favors forward translocation of the enzyme (Svetlov et al., 2007). However, the location of the RfaH binding site on the EC suggests that the bound RfaH may additionally restrict movements of the β′ clamp that have been proposed to accompany pausing (Landick, 2001). Here we present the evidence in support of the second mechanism. We show that RfaH action depends on two RNAP elements, the β′CH and the β subunit gate loop (βGL), which are located directly across each other on the two sides on the DNA-binding channel. We propose that β′CH, βGL, and RfaH jointly form a clamp on the DNA that allows the RNAP to resist pausing and termination signals.

RESULTS

RfaH’s HTT motif is required for antipausing but not binding to the EC

To identify the determinants required for AP by RfaH, we have previously carried out mutational analysis of the N-terminal domain, RfaHN, which is sufficient for all in vitro activities of RfaH. This analysis (Belogurov et al., 2010) revealed three functional regions (Figure 1A): (i) a cluster of residues that likely interact with the NT DNA, (ii) a hydrophobic surface that is predicted to make van der Waals contacts with the β′CH, and (iii) a cluster of three residues (HTT) distant from the NT DNA and the β′CH. Substitutions in the first two regions are expected to compromise RfaH interactions with the EC; indeed, the defects of these substitutions are suppressed at high RfaH concentrations. In contrast, RfaH variants with substitutions of the HTT residues display dramatically reduced AP activity even when present at high concentrations (Belogurov et al., 2010). These observations are inconsistent with a model in which RfaH binding to the DNA and the β′CH is sufficient for its AP activity and suggest that an additional interaction between the HTT motif and the EC is required. The goal of this study was to characterize this interaction.

Figure 1. RfaH interactions with the EC.

Figure 1

Open in a new tab

(A) AP-defective substitutions in RfaHN (Belogurov et al., 2010). The affected residues are shown as sticks and are colored according to the EC component with which these residues are proposed to interact in panel B.

(B) A model of RfaHN bound to the EC (Belogurov et al., 2007). The T. thermophilus RNAP is shown as tubes with the bridge helix (β′BH) highlighted in cyan, the template DNA – in black, the NT DNA – in yellow, the nascent RNA – in red. Position of the RNAP active site is marked by the catalytic Mg2+ ion (a yellow sphere). RfaHN (blue) is bound to the NT strand and to β′CH (green); Tyr54 (forest green) interacts with the β′CH. The HTT cluster is positioned next to a mobile βGL (magenta). Figures in A and B were prepared with PyMol (DeLano Scientific).

(C) The RfaH retention assay on the linear pIA807 template shown on top with the transcription start site (+1) and end (224), the ops and the -10 elements, and the hisP signal indicated. Halted ECs were incubated with σ70 and the WT or an altered RfaH where indicated. Single-round transcription was restarted upon addition of NTPs and stopped at 10, 20, 40, 90, 180 and 360 sec. In this and other experiments, some halted ECs become arrested and do not resume elongation. Positions of the G37, ops, σP, hisP and run-off (RO) transcripts are indicated with arrows. The fraction of RNA at the σP site after a 360-sec incubation (as % of total RNA) is presented below each panel; the errors (±SD) were calculated from four independent experiments.

First, we wanted to exclude a possibility that, even though the HTT residues are not modeled to interact with either the DNA or the β′CH (Figure 1B), their substitutions compromise RfaH binding to the EC indirectly, and an altered protein dissociates after recruitment. To ascertain that RfaH variants with HTT residues substituted for Ala remain bound to the EC, we used a σ competition assay that relies on the RfaH ability to sterically block σ recruitment to the EC because, as we showed earlier, both proteins bind to the β′CH domain (Sevostyanova et al., 2008).

To test the ability of altered RfaH variants to compete with σ during elongation, we carried out a standard single-round transcription assay (Figure 1C) using a linear DNA template with a strong T7A1 promoter followed by an initial transcribed region that allows for formation of radiolabeled ECs stalled after incorporation of a G residue at position 37 (G37) when transcription is initiated in the absence of UTP. The template also encodes the ops pause signal (opsP), which induces RNAP backtracking and mediates RfaH recruitment to the EC, the consensus -10 element, and a hairpin-stabilized his pause signal (hisP). Upon addition of all NTPs, rifapentin (to block re-initiation), and the excess of free σ (1 μM vs 30 nM EC), RNAP elongated the nascent RNA, transiently pausing at the opsP (U43) and the hisP sites; σ contacts to bases in a -10-like sequence induced a very long pause at the σP site from which RNAP did not escape under the conditions of our experiments (Figure 1C). The wild-type (WT) RfaH, when present, delayed RNAP escape from the C45 position (through interactions with the ops bases) and accelerated transcription at other sites as evidenced by a faster RNAP arrival at the end of the template (RO). However, on this template, AP effects of RfaH are difficult to quantify because the hisP signal, which has been shown to respond to RfaH (Artsimovitch and Landick, 2002), is separated from the halted complex position by ~100 nt encoding two additional strong and many weak pause sites, which desynchronize the RNAP population. We have reported the effects of all RfaH variants on pausing at hisP on a less complex template (Belogurov et al., 2010).

Most importantly, the WT RfaH strongly inhibited pausing at the σP site even at a 15-fold molar excess of σ. We found that H65A and T67A (and T66A, not shown) variants also competed with σ efficiently, indicating that the inability of these variants to accelerate elongation is not due to their failure to remain bound to the EC after recruitment. As a control, we used RfaH Y54F variant with a substitution at the RfaH/β′CH interface (Figure 1B); this substitution confers a dramatic AP defect that can be rescued at 3 μM RfaH (Belogurov et al., 2010). Y54F was initially recruited at the ops site (as reflected by reduced pausing at U43) but apparently failed to maintain stable post-recruitment interactions: it neither accelerated elongation nor competed with σ (Figure 1C). These results suggest that the HTT residues are largely dispensable for RfaH binding to the EC but may be involved in a functional interaction required for the AP modification.

The HTT motif may interact with the β subunit GL

The RfaH/EC model (Belogurov et al., 2007) positions the HTT motif near the βGL (Figure 1B). To evaluate a possible role of the E. coli βGL in RfaH function, we substituted the β residues 368–376 with two glycines. Surprisingly, although the βGL has been proposed to play a key role in DNA loading during initiation (Vassylyev et al., 2002), we found that the ΔβGL variant was active in vitro (Figure 2) and under certain conditions in vivo (Figure S1).

Figure 2. The βGL deletion eliminates AP activity of RfaH.

Figure 2

Open in a new tab

(A) Single-round pause assays on a pIA171 template that encodes the hisP signal, with (50 nM) or without RfaHN. The half-lives of the WT, ΔβGL and Δβ′SI3 RNAPs paused at the hisP site were determined as described in (Landick et al., 1996). Samples were taken at 7, 15, 30, 45, 60, 80, 100, 120, 180, 300 and 540 sec.

(B) Net elongation rates were determined on a “pause-free” pIA146 template as described previously (Svetlov et al., 2007) at 20 μM NTPs. RfaHN (80 nM) was added where indicated. Samples were taken at 2, 3, 5, 7, 10, 20 and 40 min. In panels A and B, the errors (±SD) were calculated from four independent experiments.

We first tested whether the deletion of the βGL would abolish the RNAP response to RfaH in vitro at the hisP signal. In these experiments, we used the isolated RfaHN which does not require an ops element for recruitment (Belogurov et al., 2007), thereby simplifying the kinetic analysis. In a single-round pause assay, RfaHN accelerated RNAP escape from hisP almost four-fold, reducing the pause half-life from 42 to 11 seconds, but had almost no effect on the ΔβGL enzyme (Figure 2A). This result shows that the βGL is necessary for the AP activity of RfaH, at least at one model pause site.

To test whether the βGL deletion confers a general AP defect (as opposed to an effect specific to the his pause), we compared elongation of an rpoB mRNA, which is devoid of strong regulatory pauses but punctuated by many elemental pauses that contribute at least 65% to the overall rate of elongation (Neuman et al., 2003), by the deletion and the WT enzymes. RfaHN increased the apparent elongation rate by the WT RNAP by 1.8-fold, but slowed down the deletion enzyme ~1.4-fold (Figure 2B); RfaHN has a similar inhibitory effect on the WT enzyme at higher NTPs (Svetlov et al., 2007).

One explanation for this phenotype could be that the βGL deletion renders RNAP pause-resistant. We reported previously that RfaH does not accelerate transcription at saturating substrate concentrations or by “fast” RNAP variants with substitutions in β and β′ subunits that render the enzyme insensitive to pause and termination signals (Svetlov et al., 2007). In particular, an enzyme that is missing a large sequence insertion in the β′ trigger loop, Δβ′ SI3 RNAP, fails to respond to RfaH. However, the ΔβGL RNAP was only moderately faster than the WT RNAP and displayed a dramatically different response to the hisP signal in the presence and in the absence of RfaH in a side-by-side comparison with the Δβ′ SI3 enzyme (Figure 2A).

If the βGL deletion removes a contact with the HTT motif, substitutions of the HTT residues, which abolish RfaH effects on transcription by the WT RNAP, should not have any additional effects on the ΔβGL enzyme. Indeed, we found that the in vitro transcription pattern of the ΔβGL enzyme was the same with the WT, H65A, T66A, and T67A RfaH variants (Figure S2). Together, these results demonstrate that the βGL is required in vitro for the AP effects of RfaH and are consistent with a functional βGL/HTT interaction.

The βGL is not required for RfaH binding to the EC

To ascertain that the βGL deletion does not eliminate RfaH binding to the EC, we used σ competition and electrophoretic mobility shift assays (Figure 3). We found that RfaHN, which does not dissociate from the EC in vitro and thus mimics an in vivo pattern (Belogurov et al., 2007), eliminated σ recruitment to both the WT and the ΔβGL ECs (Figure 3A). As expected, RfaHN drastically altered the pattern of transcription by the WT RNAP, delaying escape from the ops site and reducing pausing at the downstream signals, such as hisP. In contrast, RfaHN had no effect on the ΔβGL RNAP, except at the σP site where it abolished pausing as efficiently as with the WT RNAP. The full-length RfaH had similar effects (AS, data not shown).

Figure 3. RfaH binds to the ΔβGL RNAP.

Figure 3

Open in a new tab

(A) The σ competition assay was performed as in Figure 1C.

(B) The gel mobility shift assay was performed as in (Artsimovitch and Landick, 2002). ECs assembled from DNA and RNA oligonucleotides and RNAP variants were incubated with [γ32P]ATP-labeled RfaH or [α32P]GTP (the incoming nucleotide specified by the template), and analyzed on an agarose gel.

We next visualized RfaH binding to the ops-paused ECs by a mobility shift assay (Figure 3B). We assembled ECs on nucleic acid scaffolds composed of the template and NT DNA strands and the RNA (see Experimental Procedures); we used this assay previously to identify the RfaH target on the EC (Artsimovitch and Landick, 2002). To verify that the assembled complexes were active, we incubated unlabeled ECs with radiolabeled α[32P]GTP, the incoming nucleotide specified by the template DNA. Upon α[32P]GMP incorporation, the position of the ECs could be readily visualized in an agarose gel (Figure 3B, left). To monitor RfaH binding to the EC, we radiolabeled RfaH via a kinase recognition sequence at the N-terminus of the protein. RfaH did not enter the gel when present alone but co-migrated with ECs assembled with the WT and ΔβGL RNAPs (Figure 3B, right). By contrast, a β′ I290R substitution at the tip of the β′CH completely abolished RfaH binding (but not GMP incorporation), consistent with a model in which the β′CH is the major affinity determinant for RfaH.

Together, these results argue that the βGL is dispensable for RfaH binding. Consistently, we detected only a very weak (but apparently specific) interaction between RfaHN and the β N-terminal domain in a bacterial two-hybrid assay (Figure S3).

RfaH dramatically reduces ρ-mediated polarity in vivo

Many RfaH-controlled genes are horizontally transferred (Belogurov et al., 2009) and are thus expected to be subject to polarity control by ρ (Cardinale et al., 2008). RfaH has been reported to inhibit the action of ρ in vitro (Belogurov et al., 2009) and in vivo (Stevens et al., 1997). We examined the effect of RfaH on the extent of termination (i.e., polarity) across the 11-gene rfb operon (rfbBDACX-glf-rfc-wbbIJKL; Figure 4A). Using ChIP-chip analysis, we have shown that after recruitment at a promoter-proximal ops site (located 76 bp upstream from the first ORF, rfbB) RfaH remains bound throughout the entire rfb operon and excludes NusG from this and other ops-containing operons (Belogurov et al., 2009). At the same time, we detected significant association of ρ with the distal part of rfb operon (Mooney et al., 2009a), consistent with a notion that NusG is required for efficient RNA release by ρ but not for ρ recruitment.

Figure 4. Polarity in the rfb operon is controlled by RfaH and ρ.

Figure 4

Open in a new tab

(A) The distribution of RNAP, RfaH, NusG and ρ along the rfb operon revealed by the previously published ChIP-chip analyses (Belogurov et al., 2009; Mooney et al., 2009a).

(B) Polarity within the rfb operon evaluated by qRT-PCR. Total RNA was isolated from the WT or ΔrfaH cells grown in the presence or absence of 25 μg/ml BCM for 30 min and the amount of the RNA message in the rfbB region (red) was compared to that in the wbbI region (blue). The errors (±SD) were calculated from three independent experiments.

To test the effects of RfaH and ρ on the rfb expression, we compared expression levels of the first (rfbB; shown in red in Figure 4) and the eighth (wbbI; blue) genes by qRT-PCR as described in Experimental Procedures. We observed strong polarity (defined here as the ratio of rfbB and wbbI messages) in the ΔrfaH strain where wbbI mRNA was barely detectable (rfbB/wbbI ~800; Figure 4B right). Consistent with our expectations, essentially all polarity was caused by ρ, as polarity was eliminated upon the addition of the ρ inhibitor bicyclomycin (BCM; shaded bars in Figure 4B). In the absence of RfaH, addition of BCM increased the absolute level of wbbI RNA more than 700-fold. In the presence of RfaH, ~45% of RNAP molecules reached wbbI gene (Figure 4B; left); this fraction increased to 78% when ρ was inhibited by BCM. These results show that even though RfaH does not completely abolish ρ-dependent termination (e.g., if not all ECs became modified by RfaH), it eliminates most of the ρ-mediated polarity within rfb operon.

The dramatic effect of RfaH on ρ termination is mediated by two mechanisms. First, RfaH modifies RNAP into a pause-resistant state (Svetlov et al., 2007), thus making it less susceptible to ρ. Second, RfaH excludes the ρ-stimulatory factor NusG through competition for binding to the β′ CH. Thus, the levels of NusG associated with the rfb operon are very low (Figure 4A), despite the fact that NusG concentration in the cell far exceeds that of RfaH (Belogurov et al., 2009). In the absence of RfaH, RNAP will be terminated more frequently as a consequence of both (i) increased pausing and (ii) increased binding of NusG with consequent stimulation of ρ. However, the relative contributions of these two mechanisms to the RfaH action in vivo are not known.

Deletion of the βGL compromises RfaH function in vivo

To assess the contribution of AP to the anti-polar effect of RfaH in vivo, we used the RNAP lacking the βGL. Removal of the GL eliminated AP activity (Figure 2) but not RfaH binding to the β′CH (Figure 3). Thus, RfaH would be expected to efficiently exclude NusG from the deletion enzyme and any defect in polarity suppression could be attributed to the loss of AP. We could not construct a chromosomal deletion of the βGL; repeated attempts yielded only gene duplications instead of a clean deletion allele, suggesting that the βGL is essential for cell growth or viability. Instead, to test the effects of the βGL RNAP, we constructed plasmids that encoded the WT or ΔβGL variant under control of an IPTG-inducible Ptrc promoter. The plasmid copy of rpoB also contained a D516V substitution that confers resistance to rifamycins but does not have any other discernible effects on transcription. The plasmid-encoded β was incorporated into the RNAP and supported growth in the presence of rifapentin on plates but not in liquid media (Figure S1). We transformed Ptrc-rpoB plasmids into the rfaH+ and ΔrfaH strains, induced the expression of a plasmid-encoded β by IPTG, and added rifapentin to inhibit the chromosomally-encoded WT RNAP. Polarity in rfb operon was analyzed by qRT-PCR as above.

With the plasmid-encoded WT rpoB, we observed essentially the same trend as with the chromosomally-encoded RNAP (Figure 4B); however, the absolute numbers of mRNAs differed somewhat (Figure S4). The distal wbbI transcript was present at a ~20% the level of rfbB in the presence of RfaH and 0.54% in the absence of RfaH, meaning that RfaH decreased polarity (i.e., increased wbbI mRNA relative to rfbB mRNA) by a factor of ~37 (Figure 5A). Deletion of the βGL reduced the effect of RfaH to ~5.5-fold.

Figure 5. Effects of the βGL deletion on RfaH function in vivo.

Figure 5

Open in a new tab

(A) Total RNA was isolated from cells expressing the WT or the ΔβGL RNAP in rfaH+ or ΔrfaH strain and the amount of wbbI RNA was compared to that in the rfbB region. See Figure S4 for the absolute levels of each message.

(B) Contributions of AP and NusG exclusion activities of RfaH to polarity. In lane 1, the RNAP is colored green to indicate the AP modification by RfaH. In lane 4, the fading NusG/ρ complex indicates its possible loss from the ΔβGL RNAP. Even though ρ may be bound to the RNAP throughout transcription and in the absence of NusG (Epshtein et al., 2010; Peters et al., 2009), here we omitted ρ from the RfaH+ lanes since our data show that ρ is largely inactive under these conditions.

Our results show that both RfaH and the βGL affect polarity in the rfb operon (Figure 5B). When both RfaH and the βGL are present, RfaH modifies RNAP and excludes NusG; this greatly reduces polarity (5 fold, lane 1). In the absence of RfaH, RNAP both is prone to pausing and readily recruits NusG; this causes very strong polarity (185 fold, lane 2). Deletion of the βGL in the rfaH+ background leads to a two-fold increase in polarity (compare lanes 1 and 3), presumably due to the loss of RfaH AP activity in the absence of the βGL. However, the βGL role is not limited to its interactions with RfaH: removal of the GL decreased polarity in the rfaH− strain three fold (compare lanes 2 and 4). This defect could be due to altered kinetic properties of the deletion enzyme or less efficient ρ-mediated, NusG-assisted RNA release. The additional “masking” effect of the βGL prevents precise evaluation of the contribution of AP to the overall (37-fold) effect of RfaH on polarity: it may range from a factor of two (if the anti-polar effect of ΔGL is attributed exclusively to altered interactions with ρ and NusG) to a factor of six (if a somewhat faster elongation rate of the ΔβGL RNAP observed in vitro (Figure 2) accounts for all of its anti-polar effects in vivo).

Contacts with the βGL are required for AP activity of E. coli NusG

One would expect that control of processivity is similar among multi-subunit RNAPs which share a common structural organization. Indeed, RfaH belongs to the NusG family, the only group of universally conserved transcription elongation factors (Spt5 in Archaea and yeast; DSIF in humans). Recent cryoelectron microscopy (Klein et al., 2011) and X-ray (Martinez-Rucobo et al., 2011) structures of archaeal RNAP bound to Spt4/5 complex are fully consistent with our model of RfaH bound to the bacterial EC (Belogurov et al., 2007) and biochemical analysis of T. thermophilus NusG (Sevostyanova and Artsimovitch, 2010): Spt4/5 binds to the same site on the EC and encircles the DNA, suggesting a similar mechanism of action (Svetlov and Nudler, 2011). We wanted to test whether the inferred contact with the βGL plays a functional role in AP by NusG. The low resolution of the NusG/EC model and the lack of data on NusG/NT DNA interactions make it difficult to pinpoint the residues that could mediate this contact. Therefore, we constructed an E. coli NusG variant in which four residues (79–82) that are structurally homologous to the RfaH HTTT motif (Belogurov et al., 2010) were replaced by Ala.

In contrast to the WT NusG, which reduced RNAP pausing at the scrambled ops signal and increased an apparent rate of elongation (Figure 6A), the NusG-4A protein did not affect transcription in vitro. Consistently, deletion of the βGL reduced NusG effect on pausing (Figure S5). However, neither the βGL deletion nor the 4A substitution altered the ability of NusG to enhance ρ-dependent termination (Figure 6B).

Figure 6. NusG/GL interactions are required for AP but not for ρ–dependent termination.

Figure 6

Open in a new tab

(A) Pause assays on the linear pIA392 template shown on top with the scrambled ops (Artsimovitch and Landick, 2002) and the run-off (RO) indicated. Halted WT A38 ECs were incubated with the WT or an altered NusG (at 100 nM) where indicated. Transcription was restarted upon addition of NTPs and rifapentin. Samples were taken at 5, 10, 20, 40, 90, 180, 300 and 600 sec. Positions of the A38 and RO transcripts are indicated with arrows, the pause positions (at 39, 41, 43, and 45) are shown with a bracket. See Figure S5 for additional details. The fractions of the RO and scrambled ops RNAs are presented as % of total RNA on the right; the errors (±SD) were calculated from three independent experiments.

(B) Termination assays were performed on pIA267 template as described (Belogurov et al., 2009). Halted, [α32P]GMP labeled ECs were formed at 40 nM with the WT or the ΔβGL RNAP. ρ (10 nM) and NusG (100 nM) were added where indicated, followed by incubation with NTP substrates for 15 min at 37 °C. NusG facilitates ρ-mediated RNA release, shifting the termination window upstream (marked by a bracket); the fractions of RNA in this region were determined from three independent experiments (±SD).

These results reveal that interactions with the βGL are essential for AP by NusG but are dispensable for ρ-dependent termination in vitro. However, our observations that the βGL deletion reduces polarity (Figure 5) suggest that these interactions may play a role in vivo. The loss of putative NusG/GL interactions may facilitate NusG dissociation from the EC or enable NusG to interact more readily with the ribosomal protein S10 (Burmann et al., 2010), thereby recruiting the ribosome to the rfb transcript. Either effect will decrease ρ-dependent termination. At present, we cannot distinguish between these (and other) scenarios, but a potential role in function of the essential NusG could explain the failure of βGL RNAP to support cell viability (Figure S1).

DISCUSSION

Taken together, our results suggest that RfaH increases RNAP processivity by at least two mechanisms. First, RfaH binds to the β′CH and excludes NusG from the transcribing RNAP, thereby inhibiting ρdependent termination. Second, RfaH makes bridging contacts to the clamp domain and the β lobe domain that may restrict movements of the clamp and other RNAP modules, in turn inhibiting pausing. We propose that the second mechanism is shared by other proteins from the NusG family, which have been shown to regulate transcription in all living organisms.

RfaH makes a functional contact to the βGL

Here, we present evidence that the βGL element, a part of the β lobe, is required for RfaH function in vivo and in vitro. The βGL element has been revealed by structural analysis (Vassylyev et al., 2002), and its position in the RNAP holoenzyme suggested a role in “gating” the entry of the promoter DNA during initiation. The role of the βGL in transcription has not been, to our knowledge, studied. We became interested in this element when our molecular modeling suggested that the βGL is positioned next to the newly identified HTT cluster in RfaH (Figure 1); substitutions of the HTT residues reduced AP activity but were not predicted to interact with either the NT DNA or the β′CH, the two known RfaH targets in the EC (Belogurov et al., 2010). We found that the deletion of the βGL did not confer strong defects in initiation, elongation, termination and RNAP response to accessory transcription factors in vitro (AS and IA, manuscript in preparation). However, the ΔβGL RNAP failed to respond to AP by RfaH and NusG (Figures 2 and 6A).

Using a two hybrid analysis, we detected an interaction between RfaHN and the N-terminal domain of β (Figure S3). This interaction was disrupted by substitutions in the GL and the HTT motif, and thus is likely specific. However, it was at least ten-fold weaker than the interaction between NusGN with a fragment of β′ that includes the β′CH (Nickels, 2009), and we have shown that NusG can t compete with RfaH during elongation (Belogurov et al., 2009). This suggests that the HTT contacts to the βGL do not make a significant contribution to RfaH s affinity for RNAP. Indeed, we show that the βGL is not required for σ exclusion (Figure 3A) and RfaH binding to the static ECs (Figure 3B). Our data argue that the βGL is required for the AP modification of RNAP.

Operon polarity control by RfaH and NusG

In bacteria, if translation is inefficient or interrupted at a stop codon, ρ gains access to the nascent RNA to rapidly terminate transcription (Richardson et al., 1975). ρ establishes polarity along operons: since not all RNAPs reach the end of an operon, promoter-proximal genes are transcribed at higher levels than the promoter-distal genes. Polarity is characteristic for all bacterial operons but is particularly pronounced in poorly translated, e.g., horizontally-transferred operons.

E. coli NusG plays opposite roles in polarity control. The N-terminal domain of NusG anchors the factor to the RNAP whereas the C-terminal domain makes contacts to ρ, thereby enhancing ρdependent termination (Mooney et al., 2009b), or to the ribosomal protein S10 (Burmann et al., 2010), thereby ensuring that translation and transcription are coupled (Proshkin et al., 2010) and ρ is excluded. NusG also participates in the formation of a large nucleoprotein rrn antitermination complex that substitutes for the ribosome in shielding the non-coding rRNA transcripts from ρ (Squires et al., 1993).

Enhancement of ρ-mediated RNA release is hypothesized to constitute the essential function of NusG in E. coli, to limit the expression of foreign DNA (Cardinale et al., 2008). However, some horizontally-transferred genes that play important cellular roles have to be protected from the defensive action of ρ and NusG. The rfb operon, which encodes lipopolysaccharide biosynthesis and transport genes, lacks a recognizable Shine-Dalgarno element in front of the rfbB ORF and has a suboptimal codon distribution, and is thus expected to be a target for ρ.

Here we show that RfaH, a NusG paralog which has lost the ability to bind ρ, antagonizes the action of ρ within the rfb operon (Figure 5). We propose that RfaH reduces polarity by several independent mechanisms. First, RfaH modifies RNAP into a pause-resistant state (Svetlov et al., 2007), thereby inhibiting ρ that preferentially targets paused RNAPs (Morgan et al., 1984). Second, RfaH sterically excludes NusG from the EC (Belogurov et al., 2009). Our results suggest that AP activity suppresses polarity in vivo at least two-fold, in addition to a larger RfaH effect attributed to direct competition with NusG (Figure 5B). Finally, RfaH may promote translation of rfb mRNA. We argued that the stable in vivo association of RfaH with the EC observed by ChIP-chip (Belogurov et al., 2009) requires sequestration of the C-terminal domain, likely by the ribosome. Such an interaction has been detected directly (Bailey et al., 2000) but its regulatory significance remains to be determined.

A structural model for AP by RfaH

In this work, we show that RfaH requires contacts to the β′ clamp and the β lobe domains to mediate AP modification of RNAP. We propose that these bridging contacts restrict mobility of RNAP modules, especially of the clamp, which disfavors pausing and, in turn, termination.

The clamp domain has been observed in different positions in various crystal structures (Figure 7) and proposed to undergo conformational changes throughout the transcription cycle (Davis et al., 2007; Epshtein et al., 2010; Landick, 2001). The clamp may open to permit entry of promoter DNA during initiation, close to establish the tight grip on DNA during elongation, and then open again to allow release of DNA during termination (Tagami et al., 2010). Formation of the elemental pause, a likely target for RfaH, is accompanied by structural rearrangements near the active site that may include opening of the clamp (Sydow et al., 2009; Toulokhonov et al., 2007). RfaH interactions with the β′CH suggested that RfaH may modulate RNAP by controlling clamp movements (Svetlov et al., 2007). However, it remained unclear how binding of RfaH (18 kDa) would hinder movements of the large clamp domain, e.g., as compared to a molecule that binds to the hinge around which the clamp rotates (Mukhopadhyay et al., 2008). Our current genetic data, in combination with in vitro and in vivo analyses suggest a direct and intuitively simple mechanism, in which RfaH binds simultaneously to the β′CH and βGL located on the opposite sides of the main channel to lock the clamp in a closed state (Figure 7).

Figure 7. Model of antipausing by RfaH.

Figure 7

Open in a new tab

Two positions of the clamp domain are shown on a structural outline derived from the NTP-bound T. thermophilus EC (PDB id 2o5j (Vassylyev et al., 2007)) containing the helical hairpin form of the trigger loop (orange) packed against the bridge helix (cyan). The nucleic acid chains are omitted for clarity. The alternative trigger loop position observed in T. thermophilus holoenzyme (PDB id 1i6h (Vassylyev et al., 2002)) is also shown. An antitermination protein (AT, blue) can make bridging contacts between the GL (magenta) and clamp helices (green) across the DNA-binding channel. These contacts will prevent clamp opening observed in the T. aquaticus core RNAP (darker grey; PDB id 1i6v (Zhang et al., 1999)) to stabilize RNAP interactions with the DNA and increase its processivity.

These bridging contacts would allow RfaH to prevent structural rearrangements that occur during transcriptional pausing. These rearrangements appear to involve movements of the trigger loop/trigger helices (Toulokhonov et al., 2007; Zhang et al., 2010), which contact the clamp as part of a three-helix bundle with the bridge helix, mediated in an anchor region by switch 1. They also involve the β lobe, indirectly, through fork loop 2, one of three flexible loops whose conformation is interconnected with the bridge helix and trigger loop/helices in a cap region (Hein and Landick, 2010; Seibold et al., 2010; Weinzierl, 2010). Bridging contacts of RfaH to the GL and clamp may stabilize the bridge helix-trigger loop/helices module through the cap and anchor regions, respectively, and suppress entry to the elemental pause. Thus, the contribution of the βGL-RfaH contact to AP could be either direct or indirect. Regardless, it is likely that RfaH binding to the clamp and βGL completes a proteinaceous encirclement of the DNA template, thereby increasing RNAP processivity.

Concluding remarks

Our results demonstrate that the βGL is essential for AP modification of E. coli RNAP by RfaH and NusG. Together with our previous data, which identified the β′ CH as the binding site for RfaH and NusG, and recent structural and functional analyses of archaeal Spt5 (Hirtreiter et al., 2010; Klein et al., 2011; Martinez-Rucobo et al., 2011), our observations suggest a model in which NusG-like proteins form a part of the clamp around the DNA, increasing RNAP processivity. This mechanism is likely ancient and ubiquitous: in all kingdoms, RNAP must bypass numerous barriers, such as sequences that induce pausing, DNA-bound regulators, or DNA-packaging proteins (e.g., nucleoid-associated proteins in Bacteria and nucleosomes in Eukarya), to get to the end of the transcribed region. It remains to be determined whether the NusG-like proteins restrict the clamp and other movements in their cognate RNAPs and how these effects are coupled to elongation, pausing, and termination. It is also possible that other, structurally unrelated, proteins inhibit pausing and termination by locking the two halves of RNAP together through binding to the same or different determinants on RNAP.

EXPERIMENTAL PROCEDURES

Proteins and reagents

Oligonucleotides were obtained from Integrated DNA Technologies (Coralville, IA) or Sigma Aldrich (St. Louis, MO), NTPs and [α32P]-NTPs were from Perkin Elmer (Boston, MA), restriction and modification enzymes – from NEB (Ipswich, MA), PCR reagents – from Roche (Indianapolis, IN), other chemicals - from Sigma (St. Louis, MO) and Fisher (Pittsburgh, PA). Plasmid DNAs and PCR products were purified using spin kits from Qiagen (Valencia, CA) and Promega (Madison, WI). Unless indicated otherwise, enzymes for RNA manipulations and qRT-PCR were from Epicentre (Madison, WI). ρ was purified as described in (Mooney et al., 2009b). The full-length RfaH variants, the RfaHN domain, and RNAP were purified as described in (Belogurov et al., 2007). NusG was purified as in (Artsimovitch and Landick, 2000). Plasmids are listed in Table S1.

Pause assays

Halted ECs were prepared in 50 μl of TGA buffer (20 mM Tris-HCl, 2 mM MgCl2, 20 mM NaCl, 5% glycerol, and 0.1 mM EDTA, pH 7.9) with E. coli RNAP (30 nM), ApU (100 μM), and starting NTPs (1 μM CTP, 5 μM ATP and GTP, 10 μCi [α32P]CTP, 3000 Ci/mmol). Transcription was restarted by the addition of CTP, ATP and UTP to 150 μM, GTP to 15 μM, and rifapentin to 25 μg/ml. Samples were removed at times indicated in figure legends and quenched by the addition of an equal volume of STOP buffer (10 M urea, 50 mM EDTA, 45 mM Tris-borate; pH 8.3, 0.1% bromophenol blue, 0.1% xylene cyanol). Samples were analyzed on denaturing urea-acrylamide gels. The RNA products were visualized and quantified using PhosphorImager and ImageQuant Software (GE Healthcare).

Sigma competition assay

The assay was performed on a linear DNA template (40nM) amplified from pIA807 (Sevostyanova et al., 2008). Halted ECs were prepared in 50 μl of TGA buffer with E. coli RNAP (30 nM), ApU (100 μM), and starting NTPs (1 μM CTP, 5 μM ATP and GTP, 10 μCi [α32P]CTP, 3000 Ci/mmol). Elongation factors (1 μM σ, 70 nM RfaH) were added followed by a 3-min incubation at 37°C. Transcription was restarted by the addition of GTP to 15 μM, CTP, ATP and UTP to 150 μM, and rifapentin to 25 μg/ml.

Gel mobility shift assay

TEC were reconstituted as described in (Artsimovitch and Landick, 2002). RfaH was labeled with [α-32P]ATP using the heart muscle kinase catalytic subunit (NEB); the unincorporated label was removed using a size exclusion G50 spin column (GE Healthcare). Reconstituted ECs were mixed with radiolabeled RfaH at 50 nM, incubated for 5 min at 37°C, and loaded onto 3% NuSieve agarose gels in 0.5X TBE. To obtain radiolabeled ECs, 10 μCi [α-32P]GTP (3000 Ci/mmol) were added instead of RfaH. After electrophoresis at room temperature at 5 V/cm for 4 hr, the gels were exposed to phosphorimager screens.

qRT-PCR

Total RNA was isolated from the WT or ΔrfaH stains grown under different conditions, and the amounts of RNA message in rfbB region was compared to that in the wbbI region. To test the effects of the βGL, overnight cultures of DH5α and IA149 (a ΔrfaH derivative of DH5α) E. coli strains transformed with pIA898 (ΔGL; D516V rpoB) or pIA183 (D516V rpoB) were diluted 1/100 into LB and grown for 3 h before addition of 1 mM IPTG. After 2 h of induction, rifapentin was added to 200 μg/ml, and cells were grown for 1 h. To test the effects of ρ, overnight cultures were diluted 1/100 into LB, grown for 3 h, and treated with BCM (20 μg/ml) for additional 30 minutes. Total RNA samples were isolated using the Nucleic Acid Isolation Kit (Epicentre). qRT-PCR analysis was performed using MiniOpticon cycler (BioRad; Hercules, CA) and MasterAmp™ GREEN Real-Time RT-PCR kit. For each sample, at least three repeats in two independent experiments (starting from cell growth and RNA isolation) were performed. To ensure an accurate quantification of RNA message in rfbB and wbbI regions, qRT-PCR assay was calibrated using dilutions of an in vitro synthesized RNA (64 pg, 16 pg, 4 pg, 1 pg, 0.25 pg and 0.0625 pg; see Supplemental Methods for the list of oligonucleotides and additional details). Tc (threshold cycle) from four independent runs were plotted against the concentration and fitted using Scientist 3.0 software (Micromath). The resulting equations were used to quantify the amount of rfbB and wbbI messages in total RNA samples.

Supplementary Material

01

Acknowledgments

We thank Ruth M. Saecker and Dmitri Svetlov for comments on the manuscript. This work was supported by the National Institutes of Health grants GM67153 (to IA) and GM38660 (to RL).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Artsimovitch I, Landick R. Pausing by bacterial RNA polymerase is mediated by mechanistically distinct classes of signals. Proc Natl Acad Sci U S A. 2000;97:7090–7095. doi: 10.1073/pnas.97.13.7090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Artsimovitch I, Landick R. The transcriptional regulator RfaH stimulates RNA chain synthesis after recruitment to elongation complexes by the exposed nontemplate DNA strand. Cell. 2002;109:193–203. doi: 10.1016/s0092-8674(02)00724-9. [DOI] [PubMed] [Google Scholar]
  3. Bailey MJ, Hughes C, Koronakis V. In vitro recruitment of the RfaH regulatory protein into a specialised transcription complex, directed by the nucleic acid ops element. Mol Gen Genet. 2000;262:1052–1059. doi: 10.1007/pl00008648. [DOI] [PubMed] [Google Scholar]
  4. Belogurov GA, Mooney RA, Svetlov V, Landick R, Artsimovitch I. Functional specialization of transcription elongation factors. EMBO J. 2009;28:112–122. doi: 10.1038/emboj.2008.268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Belogurov GA, Sevostyanova A, Svetlov V, Artsimovitch I. Functional regions of the N-terminal domain of the antiterminator RfaH. Mol Microbiol. 2010;76:286–301. doi: 10.1111/j.1365-2958.2010.07056.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Belogurov GA, Vassylyeva MN, Svetlov V, Klyuyev S, Grishin NV, Vassylyev DG, Artsimovitch I. Structural basis for converting a general transcription factor into an operon-specific virulence regulator. Mol Cell. 2007;26:117–129. doi: 10.1016/j.molcel.2007.02.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bies-Etheve N, Pontier D, Lahmy S, Picart C, Vega D, Cooke R, Lagrange T. RNA-directed DNA methylation requires an AGO4-interacting member of the SPT5 elongation factor family. EMBO Rep. 2009;10:649–654. doi: 10.1038/embor.2009.31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Burmann BM, Schweimer K, Luo X, Wahl MC, Stitt BL, Gottesman ME, Rosch P. A NusE:NusG complex links transcription and translation. Science. 2010;328:501–504. doi: 10.1126/science.1184953. [DOI] [PubMed] [Google Scholar]
  9. Cardinale CJ, Washburn RS, Tadigotla VR, Brown LM, Gottesman ME, Nudler E. Termination factor Rho and its cofactors NusA and NusG silence foreign DNA in E. coli. Science. 2008;320:935–938. doi: 10.1126/science.1152763. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Carter HD, Svetlov V, Artsimovitch I. Highly divergent RfaH orthologs from pathogenic proteobacteria can substitute for Escherichia coli RfaH both in vivo and in vitro. J Bacteriol. 2004;186:2829–2840. doi: 10.1128/JB.186.9.2829-2840.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Davis CA, Bingman CA, Landick R, Record MT, Jr, Saecker RM. Real-time footprinting of DNA in the first kinetically significant intermediate in open complex formation by Escherichia coli RNA polymerase. Proc Natl Acad Sci U S A. 2007;104:7833–7838. doi: 10.1073/pnas.0609888104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ederth J, Mooney RA, Isaksson LA, Landick R. Functional interplay between the jaw domain of bacterial RNA polymerase and allele-specific residues in the product RNA-binding pocket. J Mol Biol. 2006;356:1163–1179. doi: 10.1016/j.jmb.2005.11.080. [DOI] [PubMed] [Google Scholar]
  13. Epshtein V, Dutta D, Wade J, Nudler E. An allosteric mechanism of Rho-dependent transcription termination. Nature. 2010;463:245–249. doi: 10.1038/nature08669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hein PP, Landick R. The bridge helix coordinates movements of modules in RNA polymerase. BMC Biology. 2010;8:144. doi: 10.1186/1741-7007-8-141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hirtreiter A, Damsma GE, Cheung AC, Klose D, Grohmann D, Vojnic E, Martin AC, Cramer P, Werner F. Spt4/5 stimulates transcription elongation through the RNA polymerase clamp coiled-coil motif. Nucleic Acids Res. 2010;38:4040–4051. doi: 10.1093/nar/gkq135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Klein BJ, Bose D, Baker KJ, Yusoff ZM, Zhang X, Murakami KS. RNA polymerase and transcription elongation factor Spt4/5 complex structure. Proc Natl Acad Sci U S A. 2011;108:546–550. doi: 10.1073/pnas.1013828108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Landick R. RNA polymerase clamps down. Cell. 2001;105:567–570. doi: 10.1016/s0092-8674(01)00381-6. [DOI] [PubMed] [Google Scholar]
  18. Landick R, Wang D, Chan C. Quantitative analysis of transcriptional pausing by RNA polymerase: the his leader pause site as a paradigm. Meth Enzymol. 1996;274:334–352. doi: 10.1016/s0076-6879(96)74029-6. [DOI] [PubMed] [Google Scholar]
  19. Martinez-Rucobo FW, Sainsbury S, Cheung AC, Cramer P. Architecture of the RNA polymerase-Spt4/5 complex and basis of universal transcription processivity. EMBO J. 2011;30:1302–1310. doi: 10.1038/emboj.2011.64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Mooney RA, Davis SE, Peters JM, Rowland JL, Ansari AZ, Landick R. Regulator trafficking on bacterial transcription units in vivo. Mol Cell. 2009a;33:97–108. doi: 10.1016/j.molcel.2008.12.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Mooney RA, Schweimer K, Rosch P, Gottesman M, Landick R. Two structurally independent domains of E. coli NusG create regulatory plasticity via distinct interactions with RNA polymerase and regulators. J Mol Biol. 2009b;391:341–358. doi: 10.1016/j.jmb.2009.05.078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Morgan WD, Bear DG, von Hippel PH. Specificity of release by Escherichia coli transcription termination factor rho of nascent mRNA transcripts initiated at the lambda PR. J Biol Chem. 1984;259:8664–8671. [PubMed] [Google Scholar]
  23. Mukhopadhyay J, Das K, Ismail S, Koppstein D, Jang M, Hudson B, Sarafianos S, Tuske S, Patel J, Jansen R, et al. The RNA polymerase “switch region” is a target for inhibitors. Cell. 2008;135:295–307. doi: 10.1016/j.cell.2008.09.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Neuman KC, Abbondanzieri EA, Landick R, Gelles J, Block SM. Ubiquitous transcriptional pausing is independent of RNA polymerase backtracking. Cell. 2003;115:437–447. doi: 10.1016/s0092-8674(03)00845-6. [DOI] [PubMed] [Google Scholar]
  25. Nickels BE. Genetic assays to define and characterize protein-protein interactions involved in gene regulation. Methods. 2009;47:53–62. doi: 10.1016/j.ymeth.2008.10.011. [DOI] [PubMed] [Google Scholar]
  26. Peterlin BM, Price DH. Controlling the elongation phase of transcription with P-TEFb. Mol Cell. 2006;23:297–305. doi: 10.1016/j.molcel.2006.06.014. [DOI] [PubMed] [Google Scholar]
  27. Peters JM, Mooney RA, Kuan PF, Rowland JL, Keles S, Landick R. Rho directs widespread termination of intragenic and stable RNA transcription. Proc Natl Acad Sci U S A. 2009;106:15406–15411. doi: 10.1073/pnas.0903846106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Proshkin S, Rahmouni AR, Mironov A, Nudler E. Cooperation between translating ribosomes and RNA polymerase in transcription elongation. Science. 2010;328:504–508. doi: 10.1126/science.1184939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rees WA, Weitzel SE, Das A, von Hippel PH. Regulation of the elongation-termination decision at intrinsic terminators by antitermination protein N of phage lambda. J Mol Biol. 1997;273:797–813. doi: 10.1006/jmbi.1997.1327. [DOI] [PubMed] [Google Scholar]
  30. Richardson JP, Grimley C, Lowery C. Transcription Termination Factor Rho Activity Is Altered in Escherichia coli with suA Gene Mutations. Proc Natl Acad Sci U S A. 1975;72:1725–1728. doi: 10.1073/pnas.72.5.1725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Roberts JW, Shankar S, Filter JJ. RNA polymerase elongation factors. Annu Rev Microbiol. 2008;62:211–233. doi: 10.1146/annurev.micro.61.080706.093422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Schneider DA, French SL, Osheim YN, Bailey AO, Vu L, Dodd J, Yates JR, Beyer AL, Nomura M. RNA polymerase II elongation factors Spt4p and Spt5p play roles in transcription elongation by RNA polymerase I and rRNA processing. Proc Natl Acad Sci U S A. 2006;103:12707–12712. doi: 10.1073/pnas.0605686103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Seibold SA, Singh BN, Zhang C, Kireeva M, Domecq C, Bouchard A, Nazione AM, Feig M, Cukier RI, Coulombe B, et al. Conformational coupling, bridge helix dynamics and active site dehydration in catalysis by RNA polymerase. Biochim Biophys Acta. 2010;1799:575–587. doi: 10.1016/j.bbagrm.2010.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Sevostyanova A, Artsimovitch I. Functional analysis of Thermus thermophilus transcription factor NusG. Nucleic Acids Res. 2010;38:7432–45. doi: 10.1093/nar/gkq623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sevostyanova A, Svetlov V, Vassylyev DG, Artsimovitch I. The elongation factor RfaH and the initiation factor sigma bind to the same site on the transcription elongation complex. Proc Natl Acad Sci U S A. 2008;105:865–870. doi: 10.1073/pnas.0708432105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Squires CL, Greenblatt J, Li J, Condon C. Ribosomal RNA antitermination in vitro: requirement for Nus factors and one or more unidentified cellular components. Proc Natl Acad Sci U S A. 1993;90:970–974. doi: 10.1073/pnas.90.3.970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Stevens MP, Clarke BR, Roberts IS. Regulation of the Escherichia coli K5 capsule gene cluster by transcription antitermination. Mol Microbiol. 1997;24:1001–1012. doi: 10.1046/j.1365-2958.1997.4241780.x. [DOI] [PubMed] [Google Scholar]
  38. Svetlov V, Belogurov GA, Shabrova E, Vassylyev DG, Artsimovitch I. Allosteric control of the RNA polymerase by the elongation factor RfaH. Nucleic Acids Res. 2007;35:5694–5705. doi: 10.1093/nar/gkm600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Svetlov V, Nudler E. Clamping the clamp of RNA polymerase. Embo J. 2011;30:1190–1191. doi: 10.1038/emboj.2011.76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. 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. doi: 10.1016/j.molcel.2009.06.002. [DOI] [PubMed] [Google Scholar]
  41. Tagami S, Sekine S, Kumarevel T, Hino N, Murayama Y, Kamegamori S, Yamamoto M, Sakamoto K, Yokoyama S. Crystal structure of bacterial RNA polymerase bound with a transcription inhibitor protein. Nature. 2010;468:978–982. doi: 10.1038/nature09573. [DOI] [PubMed] [Google Scholar]
  42. 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. doi: 10.1016/j.molcel.2007.06.008. [DOI] [PubMed] [Google Scholar]
  43. Vassylyev DG, Sekine S, Laptenko O, Lee J, Vassylyeva MN, Borukhov S, Yokoyama S. Crystal structure of a bacterial RNA polymerase holoenzyme at 2.6 A resolution. Nature. 2002;417:712–719. doi: 10.1038/nature752. [DOI] [PubMed] [Google Scholar]
  44. 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. doi: 10.1038/nature05931. [DOI] [PubMed] [Google Scholar]
  45. Wang MB, Dennis ES. SPT5-like, a new component in plant RdDM. EMBO Rep. 2009;10:573–575. doi: 10.1038/embor.2009.101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Weinzierl RO. The nucleotide addition cycle of RNA polymerase is controlled by two molecular hinges in the Bridge Helix domain. BMC Biol. 2010;8:134. doi: 10.1186/1741-7007-8-134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yarnell WS, Roberts JW. Mechanism of intrinsic transcription termination and antitermination. Science. 1999;284:611–615. doi: 10.1126/science.284.5414.611. [DOI] [PubMed] [Google Scholar]
  48. Zhang G, Campbell EA, Minakhin L, Richter C, Severinov K, Darst SA. Crystal structure of Thermus aquaticus core RNA polymerase at 3.3 A resolution. Cell. 1999;98:811–824. doi: 10.1016/s0092-8674(00)81515-9. [DOI] [PubMed] [Google Scholar]
  49. Zhang J, Palangat M, Landick R. Role of the RNA polymerase trigger loop in catalysis and pausing. Nat Struct Mol Biol. 2010;17:99–104. doi: 10.1038/nsmb.1732. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS306005-supplement-01.doc (391.5KB, doc)

RESOURCES