TraR directly regulates transcription initiation by mimicking the combined effects of the global regulators DksA and ppGpp

Saumya Gopalkrishnan; Wilma Ross; Albert Y Chen; Richard L Gourse

doi:10.1073/pnas.1704105114

. 2017 Jun 26;114(28):E5539–E5548. doi: 10.1073/pnas.1704105114

TraR directly regulates transcription initiation by mimicking the combined effects of the global regulators DksA and ppGpp

Saumya Gopalkrishnan ^a, Wilma Ross ^a, Albert Y Chen ^a, Richard L Gourse ^a,¹

PMCID: PMC5514744 PMID: 28652326

Significance

TraR is a 73-amino acid protein encoded in the transfer region operon (tra) of the Escherichia coli conjugative F plasmid. Here we describe evidence for a model in which TraR mimics the combined regulatory activities of the transcription factor DksA and the second messenger ppGpp. By interacting with the residues in RNA polymerase that form a ppGpp binding pocket, we suggest that TraR provides a means for regulating transcription initiation by targeting the secondary channel even under conditions that do not result in induction of ppGpp. TraR-like proteins appear to be ubiquitous in bacteria even in phyla distant from the proteobacteriaceae, in support of the accumulating evidence that very small proteins can have a large impact on bacterial biology.

Keywords: TraR, bacterial transcription initiation, RNA polymerase, ppGpp/DksA, F element

Abstract

The Escherichia coli F element-encoded protein TraR is a distant homolog of the chromosome-encoded transcription factor DksA. Here we address the mechanism by which TraR acts as a global regulator, inhibiting some promoters and activating others. We show that TraR regulates transcription directly in vitro by binding to the secondary channel of RNA polymerase (RNAP) using interactions similar, but not identical, to those of DksA. Even though it binds to RNAP with only slightly higher affinity than DksA and is only half the size of DksA, TraR by itself inhibits transcription as strongly as DksA and ppGpp combined and much more than DksA alone. Furthermore, unlike DksA, TraR activates transcription even in the absence of ppGpp. TraR lacks the residues that interact with ppGpp in DksA, and TraR binding to RNAP uses the residues in the β′ rim helices that contribute to the ppGpp binding site in the DksA–ppGpp–RNAP complex. Thus, unlike DksA, TraR does not bind ppGpp. We propose a model in which TraR mimics the effects of DksA and ppGpp together by binding directly to the region of the RNAP secondary channel that otherwise binds ppGpp, and its N-terminal region, like the coiled-coil tip of DksA, engages the active-site region of the enzyme and affects transcription allosterically. These data provide insights into the function not only of TraR but also of an evolutionarily widespread and diverse family of TraR-like proteins encoded by bacteria, as well as bacteriophages and other extrachromosomal elements.

It is becoming increasingly clear that small proteins can play important regulatory roles in bacteriophage and bacterial biology (1, 2). A case in point is TraR, a 73-amino acid protein encoded in the transfer region operon (tra) of the Escherichia coli conjugative F plasmid (3). TraR shares 29% sequence identity with the C-terminal half of DksA (4), a 151-residue protein that binds to and regulates transcription initiation by E. coli RNA polymerase (RNAP) at specific promoters (5–7). TraR also shares homology with predicted proteins of similar length elsewhere in the bacterial sequence database that are encoded by conjugative plasmids and bacteriophages (e.g., phages 186, P2, and lambda) (8, 9).

TraR is expressed from the Py promoter as part of the major tra operon transcript (10). However, insertion mutations in traR did not alter the efficiency of F element DNA transfer, making its role in conjugation unclear (3). Herman and colleagues showed that TraR complements a ∆dksA mutant strain when expressed from its natural tra operon promoter on a conjugative plasmid or ectopically from a trp-lac promoter on a standard expression plasmid (4), and it activates transcription by RNAP containing the alternative σ factor, σ^E (11). Because Eσ^E activity is up-regulated by perturbations of the outer membrane and transcribes genes whose products (chaperones and proteases) alleviate membrane stress, it was suggested that TraR helps mediate a response to disruptions of the cell membrane caused by assembly of the conjugation pilus and DNA transfer apparatus (11).

TraR also affects transcription initiation by the primary holoenzyme, Eσ⁷⁰. Like the transcription factor DksA, TraR inhibits transcription from the rRNA promoter rrnB P1 in vivo and in vitro, and it activates transcription from the Eσ⁷⁰-dependent amino acid biosynthesis promoter plivJ in vivo (4). However, it was unclear whether activation of transcription by TraR was direct or indirect.

DksA directly inhibits transcription in the absence of the “second messenger” ppGpp in vitro, but it functions synergistically with ppGpp to inhibit transcription to a much greater extent (5). The effects of ppGpp on transcription inhibition are mediated by two binding sites ∼60 Å apart on RNAP, one at the interface of the β′ and ω subunits (site 1) (12, 13), and one at the interface of β′ and DksA in the RNAP secondary channel (site 2) (14). Only site 2 plays a role in activation of transcription, explaining why both DksA and ppGpp are required for positive control (7, 14). Because ppGpp concentrations change dramatically in response to a variety of nutritional stresses, DksA and ppGpp together coordinate transcription with the translational status of the cell, even though DksA concentrations remain relatively constant with the growth phase (15).

In this study, we provide information about the mechanism of TraR function. We show that TraR is much more active in modulating transcription than DksA alone, even though it binds to the RNAP secondary channel with only slightly higher affinity. We also show that TraR inhibits or activates a number of Eσ⁷⁰-dependent promoters in vitro, its activity is not affected by ppGpp binding at site 1, and unlike DksA, it does not create a binding site for ppGpp analogous to site 2. We describe a model for TraR action in which TraR mimics the combined effects of ppGpp and DksA by using the residues in RNAP that help form ppGpp site 2. TraR adds to the growing repertoire of known factors—including DksA, GreA, GreB, and Rnk (15–20)—that target the β′ rim helices of RNAP to modulate transcriptional output. Members of the TraR class of regulators are present in diverse bacterial species and are encoded by a large family of phages and extrachromosomal elements.

Results

TraR Is More Active than DksA in Direct Inhibition of Transcription in Vitro, but It Has a Similar Affinity for RNAP.

TraR inhibits transcription from rrnB P1 (Fig. 1A). The IC₅₀ for inhibition by purified TraR in vitro was ∼50 nM, compared with ∼1.3 µM for inhibition by DksA alone, a difference of 26-fold (Fig. 1B). TraR was also more active than DksA for inhibition of the rpsT P2 promoter (for ribosomal protein S20), another promoter previously shown to be inhibited by DksA (21) (Fig. S1 A and B). The effect of TraR was promoter-specific; TraR did not inhibit transcription from the plasmid-encoded RNA-I promoter or the lacUV5 promoter (Fig. 1 A and B and Fig. S2A).

Fig. S1. — TraR is more active than DksA for inhibition of the ribosomal protein S20 promoter *rpsT* P2 and shifts occupancy of *rpsT* P2–RNAP complexes to a closed-like form. (A) Representative gels showing the *rpsT* P2 and *RNA*-1 transcripts from plasmid pRLG14658 in the absence of factor or with 4–4,000 nM TraR or DksA. (B) Transcripts were quantified by phosphorimaging (n = 3). IC₅₀ for TraR = ∼80 nM and for DksA = ∼300 nM. (C) Representative gel showing DNase I footprints of RNAP bound to the *rpsT* P2 promoter from plasmid RLG11272, ³²P-labeled in the nontemplate strand, in the presence or absence of 5 µM TraR or 5 µM DksA. DNase I digested fragment with no proteins (lanes 1 and 2), with RNAP alone (lanes 3 and 4), with RNAP and 5 µM DksA (lanes 5 and 6), or with RNAP and 5 µM TraR (lanes 7 and 8). A+G sequence ladder is at *Left* and undigested fragment is in lane 9. Superimposed traces of the gel lanes are shown at *Right*. Colored dots indicate the downstream boundary of DNase I protection in the absence of DksA or TraR (green dot; +20) or in the presence of DksA or TraR (red or blue dots; +5). The upstream boundary of protection in lanes 3–8 is ∼−40 (n = 3).

Fig. S2. — (A) Activation of phisG, plivJ, or *RNA-1* by TraR (0–500 nM). (B) Activation shown reflects maximum activation observed. Transcription of *iraP* P1 is activated to a different extent by TraR vs. DksA/ppGpp. Values plotted with TraR (1 µM), DksA (1 µM), or DksA (2 µM) and ppGpp (50 µM) together, relative to transcription without factors (set to 1).

To determine whether TraR binds to the secondary channel of RNAP, like DksA, we used a cross-linking approach. TraR-³²P-HMK or ³²P-HMK-DksA were incubated with RNAPs containing the cross-linkable amino acid Bpa substituted for β′ trigger loop residues Q929 or R933, or for β′ R1148, a position outside of the secondary channel, as a control. Previously, we found that β′ R933-Bpa cross-linked to DksA (16). β′ R933-Bpa cross-linked to both DksA and TraR, but Fig. 1C shows that β′ Q929-Bpa cross-linked more efficiently to DksA than to TraR. β′ R1148-Bpa did not cross-link to either DksA or TraR. These results suggest that TraR binds in close proximity to the trigger loop of RNAP but its precise position in the channel may be slightly different from DksA.

We next used a competition assay to determine whether TraR’s greater effect on transcription inhibition compared with DksA resulted from a higher affinity for RNAP. Binding of ³²P-labeled DksA to RNAP was assessed in the presence or absence of competing unlabeled DksA or TraR using an Fe²⁺ cleavage assay in which ferrous iron was substituted for the active site Mg²⁺ in RNAP to generate hydroxyl radicals upon addition of DTT. The hydroxyl radicals cleave bound full-length ³²P-labeled DksA (151 residues) in the tip region, producing an N-terminal product of ∼73 residues. The 1 µM unlabeled DksA inhibited cleavage of 1 µM ³²P-DksA by 50%, as expected (22), whereas 0.6 µM TraR was sufficient to inhibit cleavage of ³²P-DksA by 50% (Fig. 1D). These data suggest that the apparent affinity of TraR for RNAP is only slightly greater than that of DksA, insufficient to account for the ∼26-fold lower IC₅₀ for transcription inhibition of rrnB P1 by TraR (Fig. 1B).

TraR Shifts Occupancy of the RNAP–Promoter Complex to an Earlier Kinetic Intermediate and Decreases Complex Lifetime.

To determine whether TraR, like DksA, inhibits transcription by reducing open complex occupancy, we compared the effects of TraR and DksA on DNase I footprints of RNAP on rrnB P1 (Fig. 1E). RNAP protected promoter DNA to ∼+12 with respect to the transcription start site (+1). Addition of DksA shifted the downstream boundary of protection back to ∼+1, consistent with our previous observations (23). TraR also shifted the downstream boundary of the complex from ∼+12 to ∼+1. At the ribosomal protein promoter, rpsT P2, DksA and TraR each shifted the downstream boundary of the complex from ∼+20 to ∼+5 (Fig. S1C). We infer that TraR and DksA have similar effects on promoter complex formation, shifting occupancy from a kinetic intermediate in which there is protection well downstream of the transcription start site to an earlier intermediate with a boundary of protection much closer to the transcription start site.

DksA reduces the half-lives of open complexes formed by all promoters, and it inhibits transcription from those promoters that make intrinsically short-lived complexes (5, 23). To facilitate comparing rate measurements of complexes with DksA vs. TraR, we used the rrnB P1(dis) promoter, a variant that forms more stable complexes than rrnB P1. At a concentration of 15 nM of each protein, the effect of TraR on promoter complex lifetime was much greater than for DksA (Fig. 1F). Over 30-fold more DksA (500 nM) than TraR (15 nM) was required to reduce promoter complex half-life to the same extent (Fig. 1F). These results are consistent with the much greater effects of TraR than DksA on transcription inhibition (Fig. 1 A and B).

TraR Directly Activates Transcription by Eσ⁷⁰ in Vitro.

Although it was shown previously that TraR stimulates transcription from a positively regulated promoter in vivo, even in a strain lacking ppGpp (4), this activation could have been indirect, resulting from direct inhibition of rRNA transcription by TraR, leading to a subsequent increase in the availability of RNAP and thus increased occupancy of promoters subsaturated for RNAP (24). To address whether the effect of TraR was direct, we measured whether it increased the activities of several amino acid biosynthesis or transport promoters in vitro that were shown previously to be activated by DksA and ppGpp together (7). TraR increased transcription from the argI, thrABC, hisG, and livJ promoters three- to fourfold; half-maximal activation was achieved at ∼50 nM TraR (Fig. 2 A and B and Fig. S2A), similar to the concentration required for inhibition of rrnB P1 (Fig. 1 A and B). Activation by TraR was promoter-specific: Under the same conditions, TraR had no effect on the lacUV5 promoter (Figs. 1 A and B and 2B) and only a small (∼twofold) effect on the RNA-1 promoter (Fig. S2A). In contrast to its requirement for activation by DksA, ppGpp was not required for activation by TraR (Fig. 2 C and D). TraR also activates the σ^E-dependent rpoH P3 promoter without ppGpp (11).

Fig. 2. — TraR activates transcription by Eσ⁷⁰ at DksA/ppGpp-regulated promoters. (A) Representative gel images showing multiple round transcription from the *thrABC* or *argI* promoters in the presence of increasing concentrations of TraR (0–2 µM) indicated by a wedge. (B) Quantification of data as in A (n = 3). Transcription was normalized to that in the absence of TraR. *lacUV*5 promoter is shown for comparison (gel image is in Fig. 1A). (C) Representative gel image showing transcription from pthrABC with TraR (500 nM), ppGpp (1.5–400 µM), or TraR (500 nM) and ppGpp (1.5–400 µM). (D) Quantification of data for ppGpp alone (200 µM), TraR alone (500 nM), or TraR (500 nM) + ppGpp (200 µM) expressed relative to values in the absence of factors (n = 3). Error bars indicate SD. (E) Transcription from the *dsrA* promoter, at the indicated concentrations of DksA alone, TraR alone, or DksA + 200 µM ppGpp, relative to transcription without factor (n = 3).

The degree of activation by TraR and DksA/ppGpp was not always the same. On the promoter for the small RNA, DsrA (25), and on the promoter for the antiadapter protein, IraP (26), TraR had a smaller effect than ppGpp/DksA (Fig. 2E and Fig. S2B).

TraR Functions Independently of ppGpp.

ppGpp binds to two sites on RNAP, one at the interface of the β′ and ω subunits (site 1) and one at the interface of DksA and the rim helices of β′ (site 2) (14). We used a DRaCALA (differential radial capillary action of ligand) assay (27), which measures binding of ³²P-ppGpp directly (Materials and Methods), to ask whether TraR creates a ppGpp binding site analogous to site 2, the site created by DksA (Fig. 3 A and B). In the absence of DksA, ppGpp bound to WT RNAP (Fig. 3 A, i), and it did not bind to RNAP lacking the ω subunit (∆ω RNAP; i.e., lacking site 1) (Fig. 3 A, ii). However, ppGpp bound to ∆ω RNAP when DksA was included, creating site 2 (Fig. 3 A, iii). As expected, addition of 1 mM nonradioactive ppGpp competed with binding of the radio-labeled ppGpp (Fig. 3 A, iv). ³²P-ppGpp did not bind to ∆ω RNAP when as much as 10 µM TraR was included (Fig. 3 A, v–vii), nor did ³²P-ppGpp bind to purified TraR without RNAP (Fig. 3 A, ix). We conclude that TraR does not create a ppGpp binding site on RNAP analogous to site 2, and it does not bind ppGpp by itself.

Consistent with the lack of a TraR-dependent ppGpp binding site (Fig. 3 A and B), inhibition of transcription by TraR did not require ppGpp and was only slightly stronger in the presence of ppGpp. With WT RNAP, TraR (60 nM) inhibited rrnB P1 ∼threefold in the absence of ppGpp, and ppGpp by itself inhibited rrnB P1 ∼two- to threefold (Fig. 3C), consistent with our previous results (12). When both TraR and ppGpp were added to WT RNAP together, rrnB P1 was inhibited ∼fivefold, reflecting an additive effect of the two factors. The additional effect from ppGpp was not observed with ∆ω RNAP, indicating that it derived from ppGpp binding to site 1.

The extent of inhibition of rrnB P1 transcription varied with TraR concentration (Figs. 1B and 3D), and was the same, relative to reactions without TraR, in either the presence or the absence of ppGpp (Fig. 3D) (IC₅₀ ∼50–55 nM). In contrast, the extent of inhibition by DksA was greatly amplified by ppGpp, consistent with our previous results (5, 14) (Fig. 3D). TraR inhibited rrnB P1 much more than DksA alone, and at least as much as DksA and ppGpp together. Thus, TraR is insensitive to the presence of ppGpp and functionally mimics the effects of DksA and ppGpp together.

Strains lacking either ppGpp or DksA or both are unable to grow on medium lacking amino acids (4, 5). Uninduced levels of TraR expressed from pTrc99a were sufficient to complement strains lacking DksA or both DksA and ppGpp (Fig. 3E, sectors 1 and 3), but longer incubation times (Fig. 3F, sector 2) or induced (higher) levels of TraR (Fig. 3G, sector 2) were required to complement the strain lacking only ppGpp. It is possible that DksA competes with TraR for the secondary channel in the ∆relA∆spoT strain, resulting in a requirement for higher TraR levels to complement the ppGpp defect. Taken together, our data suggest that TraR mimics the effect of ppGpp and DksA together in vitro and when supplied ectopically in vivo, and that ppGpp does not affect TraR function (Figs. 2 C and D and 3 D–G).

TraR and DksA Interact Differently with the Same Part of RNAP.

TraR’s higher activity than DksA and its ppGpp-independence (Figs. 1–3) and the observation that β′ Q929-Bpa cross-linked much more efficiently to DksA than to TraR (Fig. 1C) suggest that there are differences in the way TraR and DksA interact with RNAP. No high-resolution structural information is available for DksA bound to RNAP, but genetic and biochemical studies support a model in which several different parts of RNAP interact directly with DksA. These include: the β′ secondary channel rim, including residue E677, which interacts with the DksA globular domain (14, 17, 28); the RNAP active site region at the base of the secondary channel and the trigger loop, which interact with the DksA coiled-coil tip (16, 17, 29) (Fig. 1C); and the sequence insertion 1 (SI1) subdomain in the nearby β-subunit, which binds to the C-terminal helix of DksA (17).

When RNAP contained the β′ E677A substitution or either of two deletions in β SI1, TraR, like DksA, failed to inhibit rrnB P1 (Fig. 4A) or to activate the thrABC promoter (Fig. 4B). However, other substitutions in the rim helices (β′ N680A, K681A, or the double substitution, β′ N680A/K681A) had different effects on responses to TraR vs. DksA. The three variants were defective in responding to TraR, either for inhibition of rrnB P1 (Fig. 4C) or for activation of thrABC (Fig. 4D), whereas the same substitutions did not interfere at all with DksA function. In fact, DksA functioned slightly better with the mutant RNAPs than the WT RNAP in the absence of ppGpp. The IC₅₀ ratios (mutant/WT) for inhibition of rrnB P1 by TraR were 4.7 for K681A RNAP, 2.4 for N680A RNAP, and 3.4 for N680A/K681A RNAP (Fig. 4C legend for IC₅₀ values). In contrast, the IC₅₀ ratios (mutant/WT) for inhibition by DksA were 0.75 for β′ N680A, 0.83 for K681A, and 0.5 for β′ N680A/K681A RNAP (14).

Because nearly full TraR function with the β′ N680A and K681A RNAPs was observed at high TraR concentrations, it is likely that a reduction in TraR affinity was responsible for the defect in function of the rim helix variants. We suggest that DksA and TraR both bind to the rim helices, and TraR mimics the effect of DksA and ppGpp together by interacting directly with the same or nearby residues in the rim helices that help form ppGpp binding site 2.

Alignment of TraR with DksA and Analysis of Critical TraR Residues.

Alignment of TraR with DksA (Fig. 5A) indicated some common features. First, like DksA, TraR contains two aspartates and an alanine residue near its N terminus, D3, D6, and A8, which could correspond to D71, D74, and A76, the residues near the tip of the DksA coiled-coil. Second, TraR also contains four cysteine residues, C37A, C40A, C58A, and C61A, which could correspond to the four cysteines, C114, C117, C135, and C138, that form a zinc-binding motif in the DksA globular domain (6). Consistent with this proposal, we found that zinc was present at a 1:1 molar ratio in purified WT TraR when examined by inductively coupled plasma mass spectrometry (Materials and Methods). Third, TraR contains an isoleucine at residue 20 (I20) that could correspond to N88 in DksA, where an N-to-I substitution strongly increased DksA binding to RNAP (30). Fourth, the C-terminal helix of DksA is critical for its interactions with RNAP, especially E143 (17). In our alignment, E66 of TraR corresponded to E143 in DksA.

To determine whether these residues in TraR were important for function, single substitutions were constructed and the variants were screened for complementation of a ΔdksA strain for growth on minimal medium when expressed ectopically from the pTrc99a plasmid (Table S1). We then overexpressed the variants from the T7 promoter on pET28a and purified them to analyze their RNAP binding properties and their effects on transcription in vitro (see TraR expression and purification section in SI Materials and Methods).

Table S1.

Effects of TraR substitutions on ∆dksA complementation and on TraR levels

Position of TraR substitution	TraR variant	Growth on MM (no IPTG)	Growth on MM (+0.1 mM IPTG)	TraR levels in vivo
N-terminal α-helix	empty vector	−	−
	WT TraR	+	+	+
	S2A	+	+	±
	D3A	−	−	±
	E4A	+	+	+
	D6A	−	−	−
	E7A	+	+	+
	A8T	−	−	−
	Y9A	+	+	+
	S10A	+	+	+
	V11A	+	+	+
	T12A	+	+	+
	E13A	+	+	+
	I20A	−	−	+
C4-type zinc-finger globular domain	C37A	−	−	−
	C40A	−	−	−
	C58A	−	−	+
	C61A	−	−	−
	P33A	+	+	±
	V34A	+	±	+
	G41A	+	+	+
	P43A	+	+	+
	I44A	−	−	−
	P45A	+	+	±
	E46A	+	+	+
	A47T	±	+	+
	R48A	−	−	−
	R49A	+	+	+
	I51A	−	±	+
	F52A	−	+	+
	P53A	+	+	+
	G54A	+	+	+
	V55A	+	+	+
	V59A	−	+	+
C-terminal α-helix	Q62A	±	±	+
	Y64A	+	+	±
	Q65A	+	+	+
	E66A	−	−	+
	R67A	+	+	+
	Q68A	+	+	+
	R69A	+	+	+
	K70A	+	+	+
	H71A	+	+	±
	Y72A	+	+	+

Open in a new tab

Variants with single substitutions at the 42 positions indicated in Fig. 7A were constructed in pTrc99a (41), expressed in a ∆dksA strain with and without IPTG induction (0.1 mM), and analyzed by Western blots. Complementation was assessed by growth on agar plates containing minimal medium (MM) lacking all 20 amino acids. Low-level expression of WT TraR from the pTrc99a promoter in the absence of induction (leaky expression, column 3) is sufficient to compensate for the ∆dksA mutation. (+) indicates growth on MM plates. −, indicates no growth; ±, indicates partial complementation (slow growth compared with WT TraR). Western blots with TraR-specific antibody were measured on extracts from cells in which the pTrc99a-traR constructs were induced with 0.1 mM IPTG. (+) indicates TraR levels similar to WT TraR. −, indicates no TraR detected; ±, indicates TraR levels lower than WT TraR.

TraR variants with substitutions for D3, D6, or A8, the positions proposed to correspond to D71, D74, and A76 in DksA, were unable to complement a ∆dksA strain (Table S1). As measured by the competition assay described in Fig. 1D, purified D3A, D6A, and A8T TraR were only partially defective for binding to RNAP in vitro (Fig. 5B), yet all three variants were almost completely defective in inhibiting rrnB P1 and activating thrABC. Regulation of transcription was severely defective even at high concentrations of D3A and D6A TraR, where binding to RNAP was almost indistinguishable from WT TraR (Fig. 5 C and D; see legend for IC₅₀ values). Thus, the N-terminal residues in TraR are required for function subsequent to RNAP binding. A substitution for D3 had a more severe defect in TraR function than a substitution for the analogous residue in DksA, D71 (29), indicating that TraR and DksA do not interact with RNAP identically. The defects in function of the N-terminal variants are consistent with previous reports that TraR D6N is defective for inhibition of rrnB P1 and rpoH P3-lacZ fusions in vivo (4, 11).

Substitutions for I20 and E66, the positions corresponding to N88 and E143 in DksA (see above), eliminated complementation of a ∆dksA mutant (Table S1) and were defective for inhibition and activation of transcription in vitro, even though the purified proteins retained substantial RNAP binding activity (Fig. 5). We suggest that these TraR variants bind to RNAP but they might be positioned incorrectly, accounting for the loss of function.

TraR variants with alanine substitutions at C37, C40, C58, or C61, the residues in our alignment corresponding to the Zn-binding motif in DksA (Fig. 5A), were unable to complement the ΔdksA strain for growth on minimal medium after ectopic expression from pTrc99a (Table S1). Because we were unable to purify the cysteine variants after overexpression from pET28a, we could not distinguish whether these variants had specific defects vs. defects in overall structure.

Models for TraR and the TraR–RNAP Complex.

The alignment of TraR and DksA and the genetic and biochemical data on the roles of specific residues in RNAP or TraR on TraR function were used to create a model for the TraR–RNAP complex. No structural information is available for TraR, so a computational model for the structure of TraR was generated using the RaptorX web server (31) (raptorx.uchicago.edu/StructurePrediction/predict/). RaptorX chose the known structure of DksA (1TJL) (6) as a template for prediction of the TraR structure. In this model, TraR contains a long N-terminal α-helix (residues 1–29), a globular domain containing a C4-type zinc finger (residues 30–61), and a C-terminal α-helix (residues 62–73) (Fig. 6A). These features are similar to features in the C-terminal half of the DksA structure (Fig. 6B).

Our model for the TraR–RNAP complex is shown in Fig. 6C. The position of TraR in complex with RNAP was based on the position of DksA in our model of the DksA–RNAP complex (14). We considered two possible alignments of TraR with DksA. In one, the globular domains and C-terminal helices of the two proteins were aligned in PyMol, but with this alignment, TraR D3, D6, and A8 and the coiled-coil tip of DksA (D71, D74, and A76) did not superimpose, because the lengths of the α-helices of TraR and DksA differed by eight to nine amino acids. In the alternative alignment, which we favor because of the importance of D3, D6, and A8 to TraR function, these N-terminal residues and the coiled-coil tip residues of DksA, as well as a portion of the adjoining α-helices, were superimposed, but the globular domains were offset because of the length difference in the α-helices (Fig. 6B). This alignment of TraR and DksA was used to create the model of the TraR–RNAP complex. The model of the complex is consistent with cross-linking of both DksA and TraR to Q933 in the trigger loop (16) (Fig. 1C) and positions the N-terminal region of TraR to make a close approach to the active site region of RNAP, like DksA (14, 16, 17).

β′ E677, N680, and K681 interact with TraR in our model of the complex, in support of the hypothesis that TraR mimics the effect of DksA and ppGpp together by interacting directly with the same β′ residues that help form ppGpp binding site 2 (Fig. 6 D and E). Consistent with the inability of ppGpp to bind to the TraR–RNAP complex (Fig. 3A), the N680 and K681 interaction with TraR would compete with the availability of these residues for interaction with ppGpp. Furthermore, TraR does not contain the residues corresponding to those in DksA that interact with ppGpp in site 2 (DksA L95, K98, R129, K139) (14) (Fig. 5A), consistent with the absence of an effect of ppGpp on TraR function.

In our model of the complex, TraR I20 is in a distorted segment of the N-terminal α-helix but it does not make direct contacts to the RNAP rim (Fig. 6 A and D). Explanations for the strong effect of the I20A variant on TraR function and for the effect of DksA N88I on binding to RNAP await further investigation. In contrast, E66 interacts directly with the rim helices (Fig. 6A) in the model, suggesting that effects of the E66A variant result from changes in direct contacts to RNAP.

Some parts of RNAP that likely interact with TraR are mobile, including βSI1 and the trigger loop, and thus these interactions are not illustrated. We emphasize that the lack of an experimentally determined TraR structure and the absence of some modules of RNAP with which TraR interacts makes our model of the complex speculative.

Conservation of Residues Among TraR Homologs.

The degree of conservation of each TraR residue in a representative group of 100 TraR homologs was investigated using the ConSurf web server (consurf.tau.ac.il/2016/) (32). We limited analysis to ORFs of 69–75 residues (i.e., similar to the length of TraR) from a variety of bacterial, plasmid, and bacteriophage sources (Fig. 7A and Table S2). The five residues tested in Fig. 5 B–D and the four cysteines forming the putative Zn finger were among the most highly conserved residues in TraR homologs (Fig. 7A).

Fig. 7. — Conservation of TraR and effects of globular domain substitutions on TraR function. (A) ConSurf Server (consurf.tau.ac.il/2016/) analysis of TraR residues conserved in a group of 100 TraR homologs of similar length from bacteriophage, plasmid, and bacterial sources (Table S2). Color key (degree of conservation) is at right. Dots indicate positions tested for complementation of a ∆*dksA* mutation in vivo (Table S1). (B) Binding by WT TraR and TraR variants to WT RNAP by competition with ³²P-DksA by Fe²⁺ -mediated cleavage as in Fig. 5B; 0.6 µM WT TraR, 0.7 µM A47T, 1.8 µM I51A TraR, or 9.8 µM I44A TraR reduced cleavage of bound ³²P-DksA by ∼50%. R48A TraR did not compete at any of the concentrations tested (n = 2). (C) Inhibition of *rrnB* P1 transcription by WT TraR and TraR variants as in Fig. 5C. IC₅₀ = ∼55 nM for WT TraR, >4 µM for I44A and R48A, ∼310 nM for A47T, ∼545 nM for I51A. (D) Activation of the *thrABC* promoter as in Fig. 5D with WT TraR and TraR variants (n = 2).

Table S2.

Sequences of TraR-like proteins used for ConSurf analysis

Bacterial group or phage family	Species	Sequence
Gammaproteobacteria
1.	Escherichia coli (F element)	`MSDEADEAYSVTEQLTMTGINRIRQKINSHGIPVYLCEACGNPIPEARRKIFPGVTLCVECQAYQERQRKHYA`
2.	Salmonella enterica	`MSDEADEAYSVTEQMTMTGINRIRQKINAHGIPVYLCEACGNPVPEARRKIFPGVTLCVECQAYQERQRKHYA`
3.	Trabulsiella odontotermitis	`MTDEADRAQQQTDQMTKAGIDRAREMLTRPGSAWCEDCGAVIPAARRSCLPGAVTCVDCQSGREASARHRRPTG`
4.	Morganella morganii	`MSDEIDRANDHAALVLESQIAAARITAAGVSAFECEGCGKPIPEPRRRAVIGCTMCIDCQAIDELKNKHYRSV`
5.	Escherichia albertii	`MADEIERFQSHQAMLEADRIQRIRQQLTGISTLFCDDCGAPVPAERRRLIPGVRKCVGCQEVQEIQARHYRRWQ`
6.	Xenorhabdus bovienii	`MSKALDLAIQHVDEMLERRIAAHVNRPVGVSAFKCESCGNPIPEQRRMIIAGVTLCAPCQNVFELKQKHYRSE`
7.	Dickeya solani	`MADSMDISQEQQAMLLDAQIAHARKAPTAHAAHVCEDCDAPIPEARRQAIPGVTCCVACQEIREQKQRHYRPPS`
8.	Pectobacterium atrosepticum	`MPDEIDRDQEFNEQRLEEMIEQNRFKPAITPSLFHCQFCGNPIPEKRRETLPGVSTCTECQSILERRRR`
9.	Citrobacter freundii	`MADTIDLAQQREQEDRERYINKARSRIAAPSRFICEKCDAPIPEARRIAIPGVDLCVTCQQIDELKSKHYRGV`
10.	Cronobacter dublinensis	`MADAMDLVQQRVEEERERHINKARSRQAVPSRFLCESCSAPIPEARRAALPGVELCVTCQEITELKSAHYRGAV`
11.	Methylobacter luteus	`MSSAWAGGTDAIMERMEQMTSLFIEETRKSMYAGESAIECEECGEPIPESRRKAIACKYCLACQTALEKNSR`
12.	Yersinia intermedia	`MPDEIDRDQEFNEQRLEEMIEKSRFKPAVTPSLFYCQFCGKPIPEKRRQTLPGVTTCTDCQSILERRRR`
13.	Edwardsiella tarda	`MSDAIDMAQRRAEEELARNLQRVTARAVHPSASFCEECGDPIPEARRRAVPGVLCCITCQEVLELKSKHYQGV`
14.	Kluyvera intermedia	`MADAMDLAQQRELEDRERHISNVRSRIAAPSRFLCEECDIPIPEARRIAIPGVTFCVTCQEVTELKSKHYRGV`
15.	Providencia rustigianii	`MSDVIDRANDHAALVLEQHIEAARKPVNSVSAFECENCDHPIPEARRQAVMGCTLCIDCQILSELKDKHYRSV`
16.	Photorhabdus luminescens	`MSDSVDIAAEHIAATLERQIKAVTDRGISVSAFECECCGNSIPEERRKAVIGVTCCITCQSIIELKNKHYRSV`
17.	Rahnella aquatilis	`MADAMDLAQLREQEDRERHISNARSRIAAPSRFLCEECDAPIPEARRSAIPGVAFCVTCQQIAELKSKHYRGV`
18.	Gallibacterium anatis	`MTDNLDRAQQIEQMQREIALKKHRTFKGVSALYCQDCDEPIPEARRKALPGCTRCTECQTVFEQQKRNFSR`
19.	Mannheima haemolytica	`MSDQIDRANELAEKAREAALAKILQNQTACTSLFECEDCGEPIPEKRREMVIGCTRCIECQTIYEHKQKGYRR`
20.	Gallibacterium genomosp. 1	`MTDNLDRAQQIEQMQREIALKKHRTFKGVSALYCQDCDEPIPEARRKALPGCTRCTECQTIFEQQKRNFSR`
21.	Aeromonas hydrophila	`MDLIDRATQQAERMLAAQLDNQLGRSHHQGESLYHCEECGDPIPEARRLHVPGVRLCVSCKSRAERRGQ`
22.	Serratia marcescens	`MPDLLDLVQQRQQEILTGQINAARHQGGVSSSICEECDQPIPAARRAAFPGVIRCVSCQTIHEQKAKHFRG`
23.	Arsenophonus nasoniae	`MSDPIDKANELAQQQLIDQIKQVTERKKGIARFHCEDCEKAIPQARRIASPGCIRCVDCQSFFESKQQHYR`
24.	Pluralibacter gergoviae	`MTPEIIDAAQEYIERNTAAGIDALRIDHSAVSATECRDCGYPIPEARRKFVPGVQTCIECQELAEEKDRIRGGI`
25.	Yokenella regensburgei	`MADAMDFAQQREQEDRERHIHNARSRITAPSRFFCEECDAPIPEARRIAIHGVTLCVTCQQIAELKSKHYRGV`
26.	Haemophilus parainfluenzae	`MTDQFDRAQELEQMTRDIALQKHRTFKAISAFYCEDCDIPIPEKRRKLIQGVTRCVDCQQKYEMQQRNFRK`
27.	Brenneria goodwinii	`MPDEIDRDQEFNEQRLEEMIEQNRFKPAPTPSLLHCRFCGKPIPEKRRQTLPGVTTCTQCQAKLERRQR`
28.	Necropsobacter rosorum	`MTDILDRAQQVEEMAREIALKKHRTFKAVSRLYCEECGALIPEKRRQLIHGVTRCVTCQEIHEKQRRNYRR`
29.	Cardiobacterium hominis	`MDIADKAAEIEEMNRAQALAARRRIPAAPGCAECEDCGDPIPEARRRAYPSATRCTECQARHERHAKNR`
30.	Pasteurella bettyae	`MTDIFDRAQAIEEKQRELSLQNRFKAVHRISLQECQDCGEPIPEQRRKMVQGCTRCVDCQQVYEQRLKGFRR`
31.	Acinetobacter baumannii	`MADFADVASTLSEQDLDHALANIKNFDQVSNYECEDCGAEIPERRRALGNVKLCIDCQTAVESKSKHFRGGL`
32.	Pseudomonadaceae bacterium D2441	`MADVIDATNDRIQAELDAAIAACVGGVPLNESALYCEECDSEIPEARRVAVKGCRFCVHCQGMYETNQKRYRAR`
33.	Mangrovibacter sp. MFB070	`MAELIDEANERAEFNISIAIANARINHDAVSATHCKACGEPIPEKRRVAVPGCTMCAYCKSDAELKRKQVGGL`
34.	Vibrio cholerae O395	`MDVIDDAAKTEAKFQQMALANHRARAMQTAYLPSRTHCLECDDPIPKERQEKVKGCQYCTPCQAAKEQR`
35.	Kosakonia sacchari	`MQDEIDRDQEFNEQRLEEMIEQSRFKPGATPSLHYCRLCGNPIPEKRRETLPGVTTCTECQAFIERKKR`
36.	Necropsobacter rosorum	`MTDQFDRAQQLEEMQREIALKKHRTFKAVSRLYCEDCDAPIPEKRRQLIHGVTRCVICQEREEKRQRNFRK`
37.	Pseudoalteromonas haloplanktis	`MSKIDDAQKIDQHLLDAALSIQREQTNKPGATFMHCQECGIDIPKQRRDAVKNCSTCVDCQSLIEIQQQHYRH`
38.	Gilvimarinus agarilyticus	`MDIADRAQIEQDRAMARFEQARAAAPVPVSAVDCESCGYEIPEARRAGVPGVQTCIECQTLIERGVLWL`
39.	Gilliamella apicola	`MKDIIDQANDLAQLELDNLLANRQTFIGESVIDCIECGEPIPEKRRRLIKGCHLCIDCQSLKELRTKR`
40.	Kluyvera cryocrescens	`MDIIDSASEVEELQRNAALAAHRINRLAVSAMRCSDCDEELPQARRIAYPGCTMCVSCQADAERRNRRM`
41.	Enterobacter aerogenes	`MTAEIIDQANELAQQRIDMAIAAHRINRNAVSAEHCSECGEDIPAPRRAAVPGCQTCAECQSVIELKNKQRGL`
42.	Methylomonas denitrificans	`MSTAWAGGTDAILERMDKMTELFVAETRKAIYTGESATHCEECGEPIPEARRQAIACKYCLSCQTELEKG`
43.	Agarivorans gilvus	`MCDKAEALEQQQRQQALNNALNRPLEHGRIIDGTKVCICCDEPIPATRLQAAPNAVRCIDCQALEEAYGNG`
44.	Moraxella bovoculi	`MTDIIDQAQHQTELILAQHIKRARITQTSLPSATHCIDCDEPIPLLRQQKVVGCQRCVACQSDFENQKARR`
45.	Dichelobacter nodosus	`MDLNDYAAELEEYYRQESITKILNYSKVQNQSKTGRKSCIDCGEEIAVARRKVYPHAQRCVTCQIVAEKKNK`
46.	Siccibacter colletis	`MTAEIIDQASELEEMLREQAIAAHRIDRNAVSAEHCVECGEDIPELRRLKVPGCQRCASCQADEELRMKQGRG`
47.	Psychromonas ossibalaenae	`MDDLDRASEHEMAQTQRLINRHQQSSKNTFITSAHNCTLCDEAIPAARRFAVPGCQLCVSCQSLSEQGI`
48.	Erwinia piriflorinigrans	`MADAMDRVQQRVEEELQRHLRKALTRPATVTREQCVVCCAPIPPARQNAIPGVQCCVTCQEIGELKSKHYQSKAL`
49.	Cedecea neteri	`MADSMDLVQQRVEEDRLLHINNARAKTPGASRVLCIDCDAPIPPARRRAVPGVQCCVTCQEISELKGKHYNGGAV`
50.	Chania multitudinisentens	`MDELDYLQAEVDARLHRQIEQARWLPARPVSAFVCEDCGVPIPDKRRLALPGVQCCVDCQTINEAMSGRYARRGR`
51.	Leclercia adecarboxylata	`MADSMDLVQKRVEENLQRHIHNARTRKPGIARVLCIDCDAPIPTARRQAIPGVQCCVTCQEIAELKGKHYTRGAL`
Alphaproteobacteria
52.	Rhodopseudomonas palustris	`MTDQLDEAQALEERERDDCIARAAGQIKGHGATHCVVCGEPIDEARRKAMPSATRCVDCQESRERWSRSHRGR`
53.	Tistrella mobilis	`MADDADIAADRAAQAAADAIARHLRRPRPVGRPTCRVCGDVIPEDRRAALPWARTCIDCARGEEREAACRTG`
54.	Magnetospirillum magneticum	`MADLADLAKEREEANREAALAAFRARARPTGESAYFCRSCGERIPDERRAAVPGTNHCTFCAQQITAASPHSIR`
55.	Agrobacterium sp. SUL3	`MNFGGNAAFEQAELRAEQEREAAIADASRTLRGPGTMQCEDCGGDIARERRLALPSATRCIVCQTMLERSRV`
56.	Ochrobactrum anthropic	`MFGNDKQLELAEERADQAREAGIASVRLALSQIGADFCVSCGAEIEPERRKALPSARRCVECQTVFEKAKRK`
57.	Pleomorphomonas koreensis	`MNAEEKLRLAEERVEADRQRRIDAAIAAVAAPGTDSCIDCGDAIPEERRIAAPWAKRCIECQEFYEAERLSK`
58.	Phaeospirillum fulvum	`MDAYDDAQALEQRQRDQAVTTAAALARARQQGIGSDICVVCEEPIPEGRRRAAPHANTCLACQAELEATTRWP`
59.	Polymorphum gilvum	`MSQASNFDVELAEARVEREREAGVNAVRARLQGAGAADCVDCGEEIPPARRRAMPSARRCAGCQTRIEGRHR`
60.	Martelella mediterranea	`MNREDRLEAVTEGHDAARQRRIESARRALAASGAKECADCGELIPEARRKAAPFARRCAACQSEREREFYCR`
61.	Pelagibacterium halotolerans	`MHFSNRDYDLAEARSEQERAAGIKAAQAALCGEGVQACVDCLDPIPEARRKAVPSAKRCAGCQTKRERGR`
62.	Azospirillum thiophilum	`MDFGDEGEARSAFLTAGAISAVRGRLMQAGSTACIDCDTAIPPARRAALPSATRCVVCQERHERGL`
63.	Methylobacterium extorquens	`MADDVDMIQHLDEMAMDGKLAAIRSRTETTDLAPEDCDDCGGDIPMERRRAAPWTRRCVRCQTLAERAFGGRA`
64.	Rhizobium leguminosarum	`MSGCDEKDFDLAERMADRQREAAIRLARQALAIDGTFDCQDCTNEIEPARREAVPSARRCIACQRRYEREKGKRR`
Betaproteobacteria
65.	Neisseria weaver	`MTDIADKASESEDLFLAEALYKAGRPSEYQGASNYECDDCGDPIPEARRQAVPGCTRCVYCQEYFEHGYP`
66.	Chitinimonas koreensis	`MSDEVDLANDAVQNRLDGAIAMQRAALHRQGTPECEDCGDDIPLDRRAVYPAATRCVACQGRRETLNRQGVRA`
67.	Comamonas sp. E6	`MTDFFDRAQARELQLREDALRDQTRRAGLVGKTMTDSATECIDCDAQIPEARRWAMPGCQRCVKCQSTFEQQKP`
68.	Sideroxydans lithotrophicus	`MRPEDRAQELELNEWEARQQAAIQPEPTRESAKWCKAAGCGERIPEDRRNAVPGVQFCIECQERNEFMGKER`
69.	Gulbenkiania mobilis	`MTDFFDRAQELEMRQRDEALARQADAARQQGASLSHCSDCGEEIAALRREKVPGCTRCVDCQTAAEKALRR`
70.	Brachymonas chironomi	`MGDIADRASAREAELLQEALYQQRRRSAGTPGQTSATDCVECGAAIPLARQQAVPGVQTCIECQTALEAAARLA`
71.	Stenoxybacter acetivorans	`MDWAERAALNEAVFRETALARVGVNAPPANATGLCLECGTPIPAARLQAIPNAAYCVACQVLFEAHPPHLP`
72.	Acidovorax caeni	`MTDDIDRAQAREAELLADALRDHARRAGLAGKTVADSAEFCQACAEDIPDARRRAVPGVQFCVACQARRERKGNL`
Deltaproteobacteria
73.	Desulfatibacillum alkenivorans	`MPDPTDRAQEAAADYTRFSIEAARSRGSQSTLADCQDCGDPIPEARRKAVPGCTRCLDCQETYEQRRGR`
74.	Desulfonatronovibrio hydrogenovorans	`MADIVDRAQQSQNFFMDQAQAEMQRRLSQARAAATNNTPECIECGDMIPKARRAAVPGCLRCIECQREVEDGLT`
75.	Desulfovibrio sp. J2	`MADECDMAQDMEALHRRIALESLHSATTGRNSLYYCEECGDPIPEARRQAVPGVRLCLECQEEADACHR`
76.	Desulfobulbaceae bacterium BRH_c16a	`MDQFDRAQELDAYYRDQAIELHRKRMEVGGDSLTHCLECGGEIPEARRTILPGCTHCVDCVEEFERKRKRL`
77.	Geobacter sp. M18	`MADEIDLAQEISERHLEAALAGHLDHKPAAESLTHCEECGRPIPEGRRQAQPGCTRCVRCQSGIEADILTHWRN`
78.	Desulfuromonas sp. TF	`MDDIDRAQGINEQLQADALEAHSRRRTVGESLTECEDCEEEIPAARRKAVPGCTRCIDCQRKFEWIQGRR`
79.	Desulfovibrio sp. A2	`MADAADIATERECILRQEALARLHAARGAQRTSLASRATCEECGDPIPQARRAAVPGVRLCITCQKELEA`
80.	Desulfobulbus elongates	`MMDDVDIAQELREAELRHLLAARAYALPHGESAEECDDCGEPIPEARRQAAPGCTRCVFCQQRFERRMREGR`
81.	Desulfovibrio aminophilus	`MADEVDMAQQVHEREVAAAISLIVHRREDEGPEYVDGVPCCRDCGDPIPARRLAALPGIGRCVACQELADRAA`
Bacilli
82.	Bacillus thuringiensis Sbt003	`MADIIDSASEIEELQRNTAIKMRRLNHQAISATHCCECGDPIDERRRLVVQGCRTCASCQEDLELISKQRGSK`
Nitrospira
83.	Leptospirillum ferriphilum	`MDIVDLTQIRFEHEDNLILKSRKKAERESALFCEDCGIRIPDQRRKYVLGVRTCVSCQSVRERDNAQDF`
Actinobacteria
84.	Mycobacterium tuberculosis	`MSDEIDRANDHAALVLESQIAAARITAAGVSAFECEGCGKQIPEPRRRAVIGCTMCIDCQAIDELKNKHYRSV`
85.	Rhodococcus qingshengii	`MSDEADEAYSVTEQLTMTGINRIRQKINAHGIPVYLCEACGNPIPEARRKIFPGVTLCVECQAYQERQRKHYA`
Bacteriophages (Siphoviridae family)
86.	Escherichia coli bacteriophage Lambda	`MADIIDSASEIEELQRNTAIKMRRLNHQAISATHCCECGDPIDERRRLVVQGCRTCASCQEDLELISKQRGSK`
87.	Enterobacteria bacteriophage BP-4795	`MADIIDSASEIEELQRNTAIKMRRLNYQTVSATHCCECGDPIDERRRLAVQGCRTCASCQEELELISKQRGSK`
88.	Pseudomonas phage D3	`MDIVDIANDYAERELAERLYSRVKYVGESLYECEDCGEEIPVARRELVPGVRKCLSCQEYLEEINGR`
89.	Mycobacterium bacteriophage Hawkeye	`MTAAADTGMCKMCGRVVLKAELKLHWLHGYLCQLCNNGIKKIQEKVFPGGASQSRSEGGMNVGNHSD`
Bacteriophages (Myoviridae family)
90.	Escherichia bacteriophage 186	`MADAMDLAQLREQEDRERHISNARSRRHEVSAFICEECDAPIPEARRRAIPGVQCCVTCQEILELKSKHYNGGAL`
91.	Enterobacteria bacteriophage P2	`MPDNVDFIQEQQAELLERQINAARVKHCGVSALVCEECDAPIPAVRRAAYPSATRCVSCQSVFEAKNKHYRRMA`
92.	Vibrio bacteriophage VP882	`MSDIADQAQDVIEQHLTASLANRKHNINPAIPSAKHCDDCESEIPEARRRSLPGVRLCVDCASLQEIKGRHQR`
93.	Vibrio bacteriophage VHML	`MTDCSADPLDRAAALSQAHLEVSLSRIKKFEGVSAHECVECGSEIPKKRRELLQGVTDCVDCAAIKETLSKNYMR`
94.	Aeromonas bacteriophage phiO18P	`MDDIDRATRHAARMLAVQLANQVGKGHYQGESLHQCEECGDDIPEGRRRHVPGVRLCVPCQTRLERLAR`
95.	Mannheimia bacteriophage vB_MhM_1127AP1	`MSDQIDRANELAEKAREAALAKILQNQTACTSLFECEDCGEPIPEKRREMVIGCTSCIECQTIYEHKQKGYRR`
96.	Salmonella bacteriophage RE-2010	`MADAMDLVQQRVEEERQRHIRAARAKTPGVSRVLCIECEAPIPPARRRAIPGVQLCITCQEIAELKGKHYNGGAV`
97.	Salmonella bacteriophage Fels-2	`MADSMDLVQQRVEEERQRHIHTARNKTPGVSRVLCIDCDAPIPPARRRAIPGVQCCITCQEIAELKGKHYNGGAV`
98.	Yersinia bacteriophage L-413C	`MPDNVDFIQEQQAELLERQINAARVKHCCASALVCEECDTPMPAARRAAYPSATRCVSCQSVFEAKNKHYRRMA`
Bacteriophages (Podoviridae family)
99.	Shigella bacteriophage 75/02 Stx	`MADIIDNAAEIEELQRNLSLQKYKSESNAPSATHCCECGDPIDERRRQAVRGCRTCASCQQDIELINKQRGVK`
100.	Enterobacteria phage 2851	`MADIIDSASEIEELQRNTAIKMRRLNYQTVSATHCCECGDPIDERRRLAVQGCRTCASCQEDLELISKQRGSK`

Open in a new tab

Sequences of 100 TraR homologs between 69 and 75 residues in length obtained from an NCBI Blast search of E. coli TraR homologs. Bacterial groups and phage families are indicated. A Clustal Omega alignment of these sequences was entered into the ConSurf Server (consurf.tau.ac.il/2016/) for determination of conserved residues (Fig. 7A).

We chose eight additional residues in the globular domain of TraR for investigation in vitro based primarily on their conservation or proximity to conserved residues: P43, I44, P45, E46, A47, R48, R49, and I51 (Fig. 7 and Fig. S3). Two variants with substitutions in the putative Zn-binding domain, A47T and I51A, retained a substantial capacity to bind RNAP (Fig. 7B), suggesting they did not have global structural defects and yet were still functionally compromised (Fig. 7 C and D), consistent with the TraR–RNAP model (Fig. 6) in which the globular domain is proposed to interact with the β′ rim helices. Two other variants in the putative Zn-binding domain, I44A and R48A, had severe defects on RNAP binding, so without further data we cannot attribute their defects to specific rather than global effects on structure. Four other TraR globular domain variants—P43A, P45A, E46A, and R49A—functioned similarly to WT TraR both in vivo and in vitro, despite the fact that two of these residues, P45 and R49, were among the most highly conserved residues in our alignment of TraR homologs (Fig. 7A and Fig. S3). Twenty-five additional variants were tested by complementation analysis in vivo (Fig. S4 and Table S1). Many variants complemented, indicating that the identity of the WT residue was not essential for function.

Fig. S3. — Globular domain substitutions P43A, P45A, E46A, or R49A in TraR do not alter inhibition in vitro. Transcription from *rrnB* P1 in the presence of WT or TraR variants was as described in Fig. 1 and plotted relative to transcription without factors. The IC₅₀ for P43A TraR was ∼100 nM, for P45A ∼140 nM, for E46A ∼110 nM, and for R49A ∼105 nM.

Fig. S4. — Levels of TraR variant proteins in vivo. (A) TraR variants were analyzed with Western blots with an anti-TraR polyclonal antibody after induction from pTrc99a with 0.1 mM IPTG. One-microgram of lysate was loaded in each lane. Purified WT TraR was loaded as a size marker in lane 1. Levels of the D3A, D6A, and A8T variants and three of the four cysteine variants were significantly reduced relative to WT TraR. TraR E66A and C58A levels were similar to WT TraR. (B) Purified TraR variants (0.5 µg in each lane) were analyzed with Western blots after overexpression from pET28a and Ni-affinity purification. D3A and D6A were just as reactive as WT TraR. Thus, differences in apparent concentrations in A are likely to result from protein instability rather than from differential antibody reactivity.

TraR Appears to Function as a Monomer.

Because TraR is only half the length of DksA, and it was reported previously that DksA variants with a shorter N-terminal tail were more active than WT DksA (33), it seemed plausible that a DksA variant consisting only of its C-terminal half (residues 69–151) (Fig. S5), the half corresponding to TraR, might be functional. When expressed from the pTrc99a vector in vivo, this “half-DksA” variant was easily detected in a Western blot (Fig. S5A), but it did not complement a strain lacking DksA for growth on minimal medium (Fig. S5 B and C).

Fig. S5. — A half-DksA variant, similar in length to WT TraR, does not complement a ∆*dksA* mutant. (A) Representative Western blot from cell lysates made from a ∆*dksA* strain carrying either the WT gene or the *half*-*dksA* gene fused to an IPTG-inducible promoter. Cells were harvested 1, 2, or 3 h after induction with 0.5 mM IPTG. One-microgram of cell lysate was loaded in each lane, and purified DksA-HMK was loaded in lane 1 for comparison. Somewhat lower amounts of the half-DksA variant were observed than of WT DksA, which could be attributable to lower stability of the half-DksA variant or to reduced ability of the DksA antibody to recognize the half-DksA peptide. (B and C) Growth on plates containing defined medium without amino acids and (B) 0.1 mM IPTG or (C) 1 mM IPTG. Sector 1: empty vector control. Sector 2: plasmid containing *half*-*dksA* gene. Sector 3: plasmid containing the full-length *traR* gene. Sector 4: plasmid containing the full-length *dksA* gene. Even if the half-DksA concentration was lower than the WT DksA concentration, it is likely that the half-DksA would have still resulted in at least partial complementation, because we showed previously that even 50% of the WT concentration supplied from a plasmid was sufficient to complement a ∆*dksA* strain fully (5, 15).

Because a DksA variant consisting only of its C-terminal half was not functional, we also tested whether TraR might form a dimer, accounting for its ability to function like DksA. We measured TraR’s oligomeric state directly by sedimentation equilibrium analytical ultracentrifugation (SE-AUC) and its molecular mass by electrospray ionization (ESI)-mass spectroscopy (Fig. S6A). In neither case was there evidence for TraR dimers. The molecular mass of purified TraR was 8,648.2 Da, consistent with a monomer lacking the N-terminal methionine and containing four extra residues (LVPR) at the C terminus remaining from the cleaved thrombin site (Materials and Methods).

Finally, we mixed together TraR samples with and without a His-tag. The two proteins were distinguishable by size by PAGE (Fig. S6B), but no untagged TraR coeluted from the nickel resin with the His-tagged protein, suggesting that TraR monomers do not associate to form dimers (Fig. S6B). Taken together, the SE-AUC, the ESI-mass spectrometry data, and the mixed heterodimer analysis all suggest that TraR does not form dimers in solution.

SI Materials and Methods

Bacterial Strains, Primers, and Growth Conditions.

Strains and plasmids are listed in Table S3. Primers for mutagenesis or sequencing are listed in Table S4. Bacteria were grown in LB Lennox media, or on LB agar plates or M9 Minimal medium agar plates lacking all amino acids. When required, media were supplemented with ampicillin (100 μg/mL) or kanamycin (30 μg/mL). IPTG (0.1 mM, 0.5 mM, or 1 mM, as indicated) was used to induce expression of products cloned in pTrc99a or pET28a. Plates for the minimal medium complementation assay were incubated at 30 °C.

Table S3.

Strains and plasmids

Strain no. or plasmid	Genotype or description	Source
Strain no.
RLG6348	dksA::tet VH1000 = MG1655 pyrE + lacI lacZ	(5)
RLG7075	BL21 (DE3) dksA::tet	(5)
RLG9962	DH10B pIA299 rpoZ::kan rpoB ∆SI1–1.2 (Δ240–284 ΩAAA)	Present study
RLG11415	XL10Gold pIA299 rpoC-929BPA	(16)
RLG11416	XL10Gold pIA299 rpoC-933BPA	(16)
RLG12121	BL21 (DE3) rpoZ::kan pIA299	(12)
RLG12586	XL10Gold pIA900 rpoB ∆225–343ΩGG	Present study
RLG12648	BL21 (DE3) rpoZ::kan pIA299 rpoC-K681A	(incorrectly called RLG13840 in ref. 14)
RLG12773	XL10Gold pIA299 rpoC-R1148BPA	(16)
RLG13788	BL21 (DE3) rpoZ::kan pIA299 rpoC-E677A	Present study
RLG13084	RLG 7075 pET28a-traR-His6, Kan^R	Present study
RLG13085	RLG 7075 pET28a-traR-Kinase-His6, Kan^R	Present study
RLG13807	BL21DE3 rpoZ::kan pIA299 rpoC-N680A	(14)
RLG13840	BL21DE3 rpoZ::kan pIA299 rpoC-N680A K681A	(14)
RLG14304	RLG 6348 pTrc99a-Half dksA, Amp^R	Present study
RLG14718	RLG 6348 pTrc99a-empty vector, Amp^R	Present study
RLG14719	RLG 6348 pTrc99a-WT-traR, Amp^R	Present study
RLG14720	RLG 6348 pTrc99a-S2A traR, Amp^R	Present study
RLG14721	RLG 6348 pTrc99a-D3AtraR, Amp^R	Present study
RLG14722	RLG 6348 pTrc99a-E4AtraR, Amp^R	Present study
RLG14723	RLG 6348 pTrc99a-D6A traR, Amp^R	Present study
RLG14724	RLG 6348 pTrc99a-E7A traR, Amp^R	Present study
RLG14725	RLG 6348 pTrc99a-Y9A traR, Amp^R	Present study
RLG14726	RLG 6348 pTrc99a-S10A traR, Amp^R	Present study
RLG14727	RLG 6348 pTrc99a-V11A traR, Amp^R	Present study
RLG14728	RLG 6348 pTrc99a-T12A traR, Amp^R	Present study
RLG14729	RLG 6348 pTrc99a-E13A traR, Amp^R	Present study
RLG14740	RLG 6348 pTrc99a-C37A traR, Amp^R	Present study
RLG14741	RLG 6348 pTrc99a-C40A traR, Amp^R	Present study
RLG14742	RLG 6348 pTrc99a-C58A traR, Amp^R	Present study
RLG14743	RLG 6348 pTrc99a-C61A traR, Amp^R	Present study
RLG14730	RLG 6348 pTrc99a-Q62A traR, Amp^R	Present study
RLG14731	RLG 6348 pTrc99a-Y64A traR, Amp^R	Present study
RLG14732	RLG 6348 pTrc99a-Q65A traR, Amp^R	Present study
RLG14733	RLG 6348 pTrc99a-E66A traR, Amp^R	Present study
RLG14734	RLG 6348 pTrc99a-R67A traR, Amp^R	Present study
RLG14735	RLG 6348 pTrc99a-Q68A traR, Amp^R	Present study
RLG14736	RLG 6348 pTrc99a-R69A traR, Amp^R	Present study
RLG14737	RLG 6348 pTrc99a-K70A traR, Amp^R	Present study
RLG14738	RLG 6348 pTrc99a-H71A traR, Amp^R	Present study
RLG14739	RLG 6348 pTrc99a-Y72A traR, Amp^R	Present study
RLG14834	RLG 6348 pINIII-dksA, Amp^R	Present study
RLG14855	RLG 6348 pTrc99a-P43A traR, Amp^R	Present study
RLG14856	RLG 6348 pTrc99a-I44A traR, Amp^R	Present study
RLG14857	RLG 6348 pTrc99a-P45A traR, Amp^R	Present study
RLG14858	RLG 6348 pTrc99a-E46A traR, Amp^R	Present study
RLG14859	RLG 6348 pTrc99a-A47T traR, Amp^R	Present study
RLG14860	RLG 6348 pTrc99a-R48A traR, Amp^R	Present study
RLG14861	RLG 6348 pTrc99a-R49A traR, Amp^R	Present study
RLG14862	RLG 6348 pTrc99a-I51A traR, Amp^R	Present study
RLG15073	MG1655 dksA::tet pTrc99a-empty vector, Amp^R	Present study
RLG15074	MG1655 dksA::tet pTrc99a-traR, Amp^R	Present study
RLG15075	MG1655 relA251::kan spoT207::cat pTrc99a-empty vector, Amp^R	Present study
RLG15076	MG1655 relA251::kan spoT207::cat pTrc99a-traR, Amp^R	Present study
RLG15077	MG1655 dksA::tet relA251::kan spoT207::cat pTrc99a-empty vector, Amp^R	Present study
RLG15078	MG1655 dksA::tet relA251::kan spoT207::cat pTrc99a-traR, Amp^R	Present study
RLG15104	RLG6348 pTrc99a-A8T traR, Amp^R	Present study
RLG15105	RLG6348 pTrc99a-I20A traR, Amp^R	Present study
RLG15106	RLG6348 pTrc99a-P33A traR, Amp^R	Present study
RLG15107	RLG6348 pTrc99a-G41A traR, Amp^R	Present study
RLG15108	RLG6348 pTrc99a-V34A traR, Amp^R	Present study
RLG15109	RLG6348 pTrc99a-P53A traR, Amp^R	Present study
RLG15110	RLG6348 pTrc99a-F52A traR, Amp^R	Present study
RLG15111	RLG6348 pTrc99a-G54A traR, Amp^R	Present study
RLG15112	RLG6348 pTrc99a-V55A traR, Amp^R	Present study
RLG15113	RLG6348 pTrc99a-V59A traR, Amp^R	Present study
Plasmid
pRLG770	In vitro transcription vector, Amp^R	(14)
pRLG5073	pRLG770 with thrABC (−72 to +16)	(24)
pRLG7340	pSL6 with rrnB P1 (−73 to +50)	(42)
pRLG11272	pSL6 with rpsT P2 (−68 to +50)	Present study
pRLG11350	pRLG770 with iraP1 (−207 to +21)	(14)
pRLG13065	pRLG770 with rrnBP1 (−88 to +50)	(14)
pRLG6120	pRLG770 with rrnB P1(dis) (−66 to +9)	(5)
pRLG13067	pRLG770 with lacUV5 (−59 to +38)	(24)
pRLG13070	pRLG770 with plivJ (−60 to +13)	(24)
pRLG13072	pRLG770 with dsrAp205	(25)
pRLG13084	pET28a-WT traR-His6, Kan^R	Present study
pRLG13085	pET28a-WT traR-Kinase-His6, Kan^R	Present study
pRLG13098	pRLG770 with argI (−45 to +32)	(24)
pRLG13099	pRLG770 with hisG (−60 to +1)	(7)
pRLG14658	pRLG770 with rpsT P2 (−89 to +50)	(21)
pRLG14852	pET28a-D3A traR-His6	Present study
pRLG14853	pET28a-D6A traR-His6	Present study
pRLG14854	pET28a-E66A traR-His6	Present study
pRLG14844	pET28a-P43A traR-His6	Present study
pRLG14845	pET28a-I44A traR-His6	Present study
pRLG14846	pET28a-P45A traR-His6	Present study
pRLG14847	pET28a-E47A traR-His6	Present study
pRLG14848	pET28a-A47T traR-His6	Present study
pRLG14849	pET28a-R48A traR-His6	Present study
pRLG14850	pET28a-R49A traR-His6	Present study
pRLG14851	pET28a-I51A traR-His6	Present study
pRLG15135	pET28a-A8T traR-His6	Present study
pRLG15136	pET28a-I20A traR-His6	Present study

Open in a new tab

Table S4.

Oligonucleotides

Primer name or no.	Sequence
6897	5′ `CATCCGGCTCGTATAATGTGTGG` 3′
6898	5′ `CCGCTTCTGCGTTCTGATTTAATC` 3′
6899	5′ `CATGGTCTGTTTCCTGTGTGAAAT` 3′
6900	5′ `GGCTGTTTTGGCGGATGAGAG` 3′
6893	5′ `CTAGAGGGGAATTGTTATCCGCTCAC` 3′
6984	5′ `TGAGATCCGGCTGCTAACAAAGC` 3′
7349	5′ `ATGCGTCCGGCGTAGAGGATCGAG` 3′
7861	5′ `CTAGTTATTGCTCAGCGG` 3′
S2A traR	5′ `GGGAGGAATATTCCGTGGCTGATGAAGCCGATG` 3′
pTrc-D3A traR	5′ `GGAATATTCCGTGAGTGCGGAAGCCGATGAAG` 3′
pET28a-D3A traR-His6	5′ `TTAACTTTAAGAAGGAGATATACCATGAGTGCGGAAGCTGATGAGGCATATTC` 3′
E4A traR	5′ `GGAATATTCCGTGAGTGATGCAGCCGATGAAGC` 3′
pTrc-D6A traR	5′ `GAATATTCCGTGAGTGATGAAGCCGCGGAGGCATATTCAGTGACAGAACAACTG` 3′
pET28a-D6A traR-His6	5′ `GAAGGAGATATACCATGAGTGATGAAGCTGCGGAGGCATATTCAGTGACAGAAC` 3′
E7A traR	5′ `AGTGATGAAGCCGATGCAGCATATTCAGTGAC` 3′
Y9A traR	5′ `GCCGATGAAGCAGCTTCAGTGACAGAACAAC` 3′
S10A traR	5′ `GCCGATGAAGCATATGCAGTGACAGAACAACTG` 3′
V11A traR	5′ `CGATGAAGCATATTCAGCGACAGAACAACTGAC` 3′
T12A traR	5′ `CGATGAAGCATATTCAGTGGCAGAACAACTGACC` 3′
E13A traR	5′ `GAAGCATATTCAGTGACAGCACAACTGACCATG` 3′
C37A traR	5′ `CCTGTTTATCTCGCTGAAGCATGCGGAAATC` 3′
C40A traR	5′ `CTCTGTGAAGCAGCCGGAAATCCTATTCCGG` 3′
C58A traR	5′ `GGTGTGACGTTGGCCGTTGAATGTCAGGCG` 3′
C61A traR	5′ `GTTGTGCGTTGAAGCTCAGGCGTATCAGG` 3′
Q62A traR	5′ `GTGCGTTGAATGTGCGGCGTATCAGGAAAG` 3′
Y64A traR	5′ `GTTGAATGTCAGGCGGCTCAGGAAAGACAG` 3′
Q65A traR	5′ `GAATGTCAGGCGTATGCGGAAAGACAGAGAAAAC` 3′
E66A traR	5′ `GTCAGGCGTATCAGGCAAGACAGAGAAAAC` 3′
R67A traR	5′ `GGCGTATCAGGAAGCACAGAGAAAACATTATG` 3′
Q68A traR	5′ `GCGTATCAGGAAAGAGCGAGAAAACATTATGC` 3′
R69A traR	5′ `TCAGGAAAGACAGGCAAAACATTATGCATAAG` 3′
K70A traR	5′ `CAGGAAAGACAGAGAGCACATTATGCATAAGTC` 3′
H71A traR	5′ `GAAAGACAGAGAAAAGCTTATGCATAAGTCAG` 3′
Y72A traR	5′ `CAGAGAAAACATGCTGCATAAGTCAGTCGCAG` 3′
P43A traR	5′ `GAAGCATGCGGAAATGCTATTCCGGAAGCC` 3′
I44A traR	5′ `TGCGGAAATCCTGCTCCGGAAGCCCGGCGG` 3′
P45A traR	5′ `GGAAATCCTATTGCGGAAGCCCGGCGG` 3′
E46A traR	5′ `GGAAATCCTATTCCGGCAGCCCGGCGGAAAATA` 3′
A47T traR	5′ `AATCCTATTCCGGAAACGCGGCGGAAAATATTTCC` 3′
R48A traR	5′ `CCTATTCCGGAAGCCGCGCGGAAAATATTTCCC` 3′
R49A traR	5′ `ATTCCGGAAGCCCGGGCGAAAATATTTCCCGGT` 3′
I51A traR	5′ `GAAGCCCGGCGGAAAGCATTTCCCGGTGTGAC` 3′

Open in a new tab

Plasmid Construction.

TraR expression plasmids for in vivo analysis (pTrc99a; traR expressed from the IPTG-inducible tac promoter) or for purification (pET28a; traR expressed from the T7 promoter) were constructed by Gibson Assembly (New England Biolabs) using synthetic gene fragments specific for cloning into each vector (gBlock; IDT DNA) (Table S5). The gBlock sequence for pTrc99a cloning contained traR sequence from the F element flanked by 29 bp of tra operon sequence upstream, including the GTG start codon and 21 bp downstream of the traR gene. The gBlock sequence for pET28a cloning had the native traR coding sequence, but with an ATG start codon, a C-terminal His6 tag, and contained or lacked an HMK site (Table S5). Each gBlock was flanked at each end by ∼20 bp of sequence homologous to the insertion site in the appropriate plasmid. For assembly reactions, plasmids were linearized by amplification with Q5 high-fidelity polymerase (New England Biolabs) and primers 6899 and 6900 (for pTrc99a) (Table S4), or primers 6893 and 6984 (for pET28a). Assembly reactions were transformed into NEB5α competent cells, plated on LB ampicillin or kanamycin agar, and correct inserts were verified by sequencing using primers 6897 and 6898 (for pTrc99a) or 7349 and 7861 (for pET28a). For assay of TraR function in vivo, pTrc99a-traR was transformed into RLG6348 and for overexpression and purification of TraR, pET28a-traR-His6 or pET28a-traR-HMK-His6 was transformed into RLG7075.

Table S5.

Geneblocks

gBlock	Sequence
pTrc99a-traR	`ATTTCACACAGGAAACAGACCATGGAATTCCGTGTCAGAACATGTCGGGAGGAATATTCCGTGAGTGATGAAGCCGATGAAGCATATTCAGTGACAGAACAACTGACCATGACAGGAATAAACCGGATACGCCAGAAAATAAATGCTCATGGTATTCCTGTTTATCTCTGTGAAGCATGCGGAAATCCTATTCCGGAAGCCCGGCGGAAAATATTTCCCGGTGTGACGTTGTGCGTTGAATGTCAGGCGTATCAGGAAAGACAGAGAAAACATTATGCATAAGTCAGTCGCAGAACATAGTGATTTAATGGATCCTCTAGAGTCGACCTG`
pET28a- traR-His6	`GTGAGCGGATAACAATTCCCCTCTAGAAATAATTTTGTTTAACTTTAAGAAGGAGATATACCATGAGTGATGAAGCCGATGAAGCATATTCAGTGACAGAACAACTGACCATGACAGGAATAAACCGGATACGCCAGAAAATAAATGCTCATGGTATTCCTGTTTATCTCTGTGAAGCATGCGGAAATCCTATTCCGGAAGCCCGGCGGAAAATATTTCCCGGTGTGACGTTGTGCGTTGAATGTCAGGCGTATCAGGAAAGACAGAGAAAACATTATGCACTGGTGCCGCGTGGCAGCAGCAGCGGCCACCATCACCATCACCATTAAGATCCGGCTGCTAACAAAGC`
pET28a- traR-Kinase-His6	`GTGAGCGGATAACAATTCCCCTCTAGAAATAATTTTGTTTAACTTTAAGAAGGAGATATACCATGAGTGATGAAGCCGATGAAGCATATTCAGTGACAGAACAACTGACCATGACAGGAATAAACCGGATACGCCAGAAAATAAATGCTCATGGTATTCCTGTTTATCTCTGTGAAGCATGCGGAAATCCTATTCCGGAAGCCCGGCGGAAAATATTTCCCGGTGTGACGTTGTGCGTTGAATGTCAGGCGTATCAGGAAAGACAGAGAAAACATTATGCACGTCGTGCATCTGTTCTGGTGCCGCGTGGCAGCAGCAGCGGCCACCATCACCATCACCATTAAGATCCGGCTGCTAACAAAGC`
pTrc99a-half-dksA	`ATAACAATTTCACACAGGAAACAGACCATGGAATTCGTTAAGGAGAAGCAACATGCCGGACCCGGTAGACCGTGCAGCCCAGGAAGAAGAGTTCAGCCTCGAACTGCGTAACCGCGATCGCGAGCGTAAGCTGATCAAAAAGATCGAGAAGACGCTGTGCGAATCCTGCGGTGTTGAAATTGGTATTCGCCGTCTGGAAGCGCGCCCGACAGCCGATCTGTGCATCGACTGCAAAACGCTGGCTGAAATTCGCGAAAAACAGATGGCTTAATTACAGCCGTTCCATCACGTTTACCACAAAGCTTGGCTGTTTTGGCGGATGAGAGAAGATTTTC`

Open in a new tab

Start and stop codons are in bold.

Overexpression and Purification of TraR and TraR Variants.

LB Kan cultures (1 L), inoculated from overnight LB Kan cultures at a starting OD₆₀₀ of 0.025, were grown to an OD₆₀₀ of 0.5. Expression of TraR was induced by addition of IPTG (1 mM final) with continued growth at 37 °C for 3 h. Cells were pelleted, resuspended in 50 mM NaPO₄ (pH 8.0), 0.25 M NaCl, 5 mM imidazole, and 10% glycerol, lysed by sonication, and TraR was purified by Ni-NTA chromatography (Qiagen), essentially as described previously (4). The lysate was passed twice through a Ni-NTA column (0.75 mL bed volume), washed with ∼20 mL of column buffer with 25 mM imidazole, then ∼20 mL of column buffer with 50 mM imidazole. TraR-His6 was eluted with buffer containing 300 mM imidazole, fractions containing TraR-His6 were pooled and dialyzed against thrombin cleavage buffer at 4 °C [10 mM Tris⋅Cl (pH 8.0), 0.1 mM EDTA, 2 mM DTT, 250 mM NaCl, 10% glycerol]. The histidine tag was removed by cleavage with biotinylated thrombin (1 unit/mg of protein; EMD Millipore) for ∼3 h at 22 °C. Free His6 tag and uncleaved protein were removed by passage through a combined streptavidin agarose (0.35 mL) and Ni-NTA (0.25 mL) column, and the flow-through containing TraR without the His6 tag was dialyzed overnight against storage buffer [10 mM Tris⋅Cl (pH 8.0), 0.1 mM EDTA, 2 mM DTT, 250 mM NaCl, 50% glycerol]. Stocks were stored at −80 °C, and the working aliquots were stored at −20 °C. Protein concentrations were determined with the Bradford assay reagent (Bio-Rad), using BSA as a standard. TraR was diluted into storage buffer before use. The purified proteins were typically »95% pure.

Some TraR variants produced weak or no signals in Western blots of protein extracts when they were expressed ectopically from pTrc99a (Fig. S4A and Table S1). As a result, we cannot rule out that their failure to complement the ∆dksA mutant in vivo resulted from protein instability. Nevertheless, in some cases we were able to purify the proteins after overexpression from pET28a (e.g., the D3A, D6A, and A8T variants). These purified variant proteins were as reactive to our TraR antibody as the WT protein, indicating that the low signals in the Western blots when the variants were produced from pTrc99a did not reflect a defect in antibody recognition (Fig. S4B). The high expression afforded by the T7-BL21 expression system is probably sufficient to explain the success with purification of mutant proteins that were not stable when expressed from pTrc99a.

In Vitro Transcription.

Reactions were carried out as previously described (14, 15, 23). Supercoiled plasmid templates for transcription were derivatives of pRLG770 (Table S3). Multiple round transcription reactions (25 µL) contained 50 ng supercoiled plasmid DNA in 10 mM Tris⋅HCl, pH 7.9, 170 mM NaCl, 10 mM MgCl₂, 1 mM DTT and 0.1 μg/μL BSA, 500 μM ATP, 200 μM GTP, 200 μM CTP, 10 μM UTP, 1 μCi [α-³²P] UTP), and the indicated concentrations of TraR, DksA, or ppGpp (TriLink Biotechnologies). Reactions were initiated by addition of WT or variant RNAPs (10 nM), incubated for 15 min at 22 °C, and terminated by addition of an equal volume of stop solution (95% formamide, 20 mM EDTA, 0.05% bromophenol blue, 0.05% xylene cyanol). Effects of TraR and DksA on RNAP are qualitatively the same at 22 °C and 37 °C. Transcripts were separated on 6% acrylamide–7 M urea denaturing gels and analyzed by phosphorImaging (ImageQuant 5.2 software). IC₅₀ values were quantified by fitting the data points from reactions containing up to 4 µM TraR to a single exponential decay curve using SigmaPlot 10.0. Activation was quantified by fitting the points to a single exponential rise to max curve.

RNAP–promoter complex lifetime was determined as described previously (5). Complexes were formed by incubation of the plasmid containing the rrnB P1(dis) promoter at room temperature with TraR or DksA as indicated. Reactions contained 10 mM Tris⋅HCl, pH 7.9, 170 mM NaCl, 10 mM MgCl₂, 1 mM DTT and 0.1 μg/μL BSA. Heparin (10 µg/mL) was used as a competitor to sequester free RNAP, and the fraction of complexes remaining at different times was determined by transcription after addition of aliquots to an NTP mix.

DNaseI Footprinting.

DNaseI footprints were performed as described previously (35). Plasmid pRLG7340 (rrnB P1, −73 and +50) was digested with NcoI-HF (New England Biolabs), the template strand was 3′ end-labeled with ³²P-dCTP (Perkin-Elmer) using Sequenase (USB), and an rrnB P1 promoter fragment was produced by digestion with NheI-HF (New England Biolabs). The rrnB P1 promoter fragment was excised from a 5% native polyacrylamide gel, diffused overnight into buffer (200 mM NaCl, 20 mM Tris⋅Cl pH 7.5, 1 mM EDTA), purified using a Qiagen PCR purification kit, and stored in 10 mM Tris⋅Cl pH 8.0. The nontemplate strand of an rpsT P2 fragment was labeled similarly by digesting plasmid pRLG11272 (containing an rpsT P2 fragment with endpoints −68 and +50), digesting first with NheI-HF, labeling with ³²P, and then digesting with NcoI-HF.

For DNaseI footprinting reactions, 5 nM RNAP was added to end-labeled DNA in buffer containing 30 mM KCl, 10 mM Tris⋅Cl, pH 8.0, 10 mM MgCl₂, 1 mM DTT, 0.1 mg/mL BSA, in the presence or absence of 5 μM TraR or DksA, and incubated at 37 °C for 10 min. DNaseI (Worthington DPFF) was added to a concentration of 10 μg/mL for 30 s, and the reactions were stopped by addition of 10 mM EDTA, 0.3 M sodium acetate, and an equal volume of phenol. Glycogen (20 µg) was added to the aqueous phase, and the DNA was precipitated with ethanol, washed with 100% ethanol, dried, resuspended in 6 μL loading buffer (8 M urea, 0.5× TBE, 0.05% bromophenol blue, 0.05% xylene cyanol), and electrophoresed on 9.5% acrylamide–7 M urea gels. Gels were dried, visualized using phosphorimaging, and quantified using ImageQuant 5.2.

³²P-ppGpp Binding Assay.

The DRaCALA assay, adapted from ref. 27, was used to measure ³²P-ppGpp binding to proteins. ³²P-ppGpp was synthesized as described previously using γ-[³²P]ATP and GDP (Amresco) and was purified by PEI cellulose thin-layer chromatography (TLC) in 1.5 M KH₂PO₄ pH 3.4, eluted from the TLC plate in 4 M LiCl, precipitated, and then stored in aliquots at −80 °C, as described previously (12, 14). ³²P-ppGpp (∼10 nM) was incubated for 10 min at 22 °C with 2 µM RNAP (WT or Δω RNAP), in the presence or absence of either TraR (0.5 µM, 2 µM or 10 µM) or DksA (20 µM), or with TraR alone (10 µM) in 10-µL reactions containing 15 mM Tris⋅Cl pH 8, 30 mM KCl, 10 mM MgCl₂, 1 mM DTT, 5% glycerol. The DksA concentration used (20 µM) was saturating under these conditions. No binding to DksA alone was observed in previous experiments (14). Where indicated, cold competitor ppGpp (1 mM) was added to reactions. Duplicate 4-µL aliquots were spotted onto dry Protran BA85 nitrocellulose filters, and dried filters were quantified by phosphorimaging. The fraction of total counts bound to the central protein spot was determined by correction of protein-bound counts for background of unbound ³²P-ppGpp, and was expressed as the percent of total counts in the entire spot (27). Radioactivity in the outer ring of the spot was not included when determining background correction values.

TraR–RNAP Binding (Fe²⁺ Cleavage Competition Assay).

TraR (or DksA) binding to RNAP was determined by competition with ³²P-labeled DksA binding to RNAP. ³²P-labeled DksA binding to RNAP was detected by cleavage with hydroxyl radicals generated by Fe²⁺ in the RNAP active site (22). Briefly, Mg²⁺ in the active site of RNAP was removed by spin filtration, and ³²P-labeled HMK-DksA (1 µM) was preincubated with core RNAP (0.1 µM) for 10 min at 22 °C, as described. Next, 0–16 µM unlabeled TraR (or DksA) was then added, and after 10 min (NH₄)₂Fe(SO₄)₂ and DTT were added simultaneously to initiate DksA cleavage by the Fe²⁺ that replaced Mg²⁺ in the RNAP active site. After 10 min, samples were run on 12% NuPAGE Bis-Tris gels in MES buffer, and dried gels were analyzed by phosphorimaging. Binding was calculated after normalization to the fraction of ³²P-DksA cleaved in the absence of unlabeled competitor.

Western Blotting.

Twenty-five–milliliter cultures of RLG6348 with pTrc99a-traR (Table S3) were inoculated at an OD₆₀₀ of 0.025 and grown at 30 °C to an OD₆₀₀ of 0.5, and 0.1 mM IPTG was then added to induce TraR (WT or variant) expression in vivo for 1 h. Cells (OD₆₀₀ equivalent of 10.0) were pelleted, resuspended in 0.6 mL of 20 mM Tris⋅HCl, pH 8.0, 150 mM KCl, 1 mM MgCl₂, 1 mM DTT, and lysed by vortexing with 0.1-mm glass beads. Cell debris was removed, and total cellular protein concentration was determined using the Bradford Assay Reagent (Bio-Rad) with BSA as a protein standard. Whole-cell protein lysates (1 µg/lane) were separated on 12% acrylamide SDS gels in MES buffer (Invitrogen), and transferred to a nitrocellulose membrane (GE Healthcare). The membrane was incubated for 1.5 h with primary antibody (1:8,000 dilution of rabbit polyclonal anti-TraR; Covance Research Product), washed four times with PBST (PBS with 0.1% Tween-20), and incubated for 1 h with secondary antibody (1:10,000 peroxidase conjugated goat anti-rabbit IgG; Thermo Scientific) for 1 h. After four washes with PBST, bands were detected by chemiluminescent substrate (Clarity Western ELC Substrate; Bio-Rad).

Bpa Cross-Linking.

The nonnatural, UV cross-linkable amino acid Bpa was incorporated at individual RNAP residues by overexpression of RNAP subunits from a multisubunit plasmid containing amber codons at the residues to be substituted using a strain expressing an evolved tRNA and tRNA synthetase pair (40). Cross-linking was performed as described previously (16, 35). ³²P-labeled HMK-DksA or ³²P-labeled TraR-HMK (50 nM) was incubated for 10 min at room temperature with Bpa-RNAP (500 nM) in buffer containing 30 mM KCl, 10 mM Tris pH 7.9 and 10 mM MgCl₂. Tubes containing the samples were placed directly on a 365 nm UV source for 3 min, and samples were run on 4–12% Bis-Tris NuPAGE gels with Mops gel running buffer (Invitrogen). Cross-linked products with the mobility predicted for β′ cross-linked to DksA or TraR on SDS PAGE were detected by phosphorimaging.

Cloning, Expression, and Western Blotting of the Half-DksA Variant.

A 335-nt gene fragment (pTrc99a-half-dksA gBlock) (Table S5), encoding only the part of DksA that aligns by sequence to full-length TraR, was cloned into pTrc99a (Fig. 5A for sequence alignment). The half-DksA variant was constructed as described above for pTrc99a-traR and transformed into RLG6348, resulting in strain RLG14304.

To determine the stability of the half-DksA variant, total protein was extracted as described for TraR after expression from pTrc99a in the ΔdksA strain. Cells were grown to an OD₆₀₀ of 0.5, and 0.5 mM IPTG was added to induce expression. Samples were harvested at 1, 2, or 3 h after IPTG addition, and Western blots were performed as described above using anti-DksA primary antibody (1:8,000) and goat anti-rabbit secondary antibody (1:10,000).

Discussion

What Accounts for the Greater Activity of TraR Relative to DksA Without ppGpp?

We found that the activity of TraR was much higher than the activity of DksA in the absence of ppGpp; it inhibited transcription much more than DksA alone and it activated transcription even in the absence of ppGpp. The high activity of TraR was similar to that of DksA and ppGpp together. Because the TraR–β′ interface did not create a binding site for ppGpp analogous to the site that forms at the DksA–β′ interface, site 2, TraR was completely insensitive to ppGpp. Rather, we propose that TraR excludes binding of ppGpp to RNAP by occupying the same surface on the β′ rim helices as used by DksA to form site 2.

The extraordinary activity of TraR was not attributable to a much higher affinity for RNAP. Rather, we suggest that TraR allosterically alters the DNA binding surface in the main channel of RNAP, as proposed previously for DksA (21). For DksA, this conformational change (in the presence of ppGpp) was proposed to involve allosteric effects of an interaction of the C-terminal helix with the SI1 region the β-subunit (14). These interactions could impact interactions of the DksA coiled-coil tip aspartate residues with the trigger loop/bridge helix of RNAP, which could in turn allosterically affect promoter DNA interactions in the main channel (23, 34). Alternatively, β-subunit–SI1 interactions could affect promoter binding by altering the downstream DNA binding interface in β (34, 35). Interactions of TraR with the trigger loop/active site region and/or β SI1 would thus mimic the allosteric changes caused by ppGpp binding to site 2 in the RNAP–DksA complex.

Multiple Secondary Channel Binding Proteins: Why both DksA and TraR?

DksA and the Gre factors modulate E. coli RNAP activity by targeting the RNAP secondary channel. DksA and the Gre factors share some structural features, including a coiled-coil with conserved aspartic acid residues at the tip, but they have very little sequence similarity, very different globular domains, and perform distinct functions. GreA and GreB are transcription elongation factors that stimulate RNAP’s intrinsic RNA cleavage activity to modulate escape from pauses and arrest sites during the elongation phase (18, 19). In contrast, DksA does not promote RNA cleavage (6) but rather acts to regulate transcription initiation in conjunction with ppGpp (5, 7). Although high concentrations of Gre factors can inhibit transcription initiation in vitro or when overexpressed in vivo, there is little evidence that they regulate open complex formation under physiological conditions in vivo (15). Because the Gre factors do not contain the amino acid sequences that contribute to site 2, ppGpp and the Gre factors do not synergize (15).

Unlike the functions of DksA and the Gre factors, the functions of DksA and TraR do overlap. Why then does the F element encode TraR when ppGpp/DksA already can provide the same function? Although TraR is not essential DNA transfer during conjugation (3, 10), previous studies established that it is coexpressed with other tra operon products from the F element Py promoter (10). Conjugation efficiency is highest in exponential growth (36), but ppGpp levels are very low during these conditions (37). We speculate that TraR’s function is to regulate host promoters in the absence of ppGpp that are normally targeted by ppGpp/DksA, including rRNA promoters and amino acid biosynthesis-related promoters, but also some promoters needed for helping to mitigate membrane damage during conjugation (4, 11). Finally, because TraR and DksA/ppGpp can regulate the same promoters to different extents (Fig. 2 and Fig. S2B), in theory TraR could alter expression to favor conjugation, but it remains to be determined whether either ppGpp/DksA or TraR directly regulates production of specific tra region transcripts or whether conjugation efficiency is altered in strains lacking DksA/ppGpp.

TraR suppresses the amino acid auxotrophy of a ∆dksA strain when supplied from a mini-F plasmid, and we found that it regulates transcription, both negatively and positively, when expressed from the trp-lac promoter on pTrc99a, a standard expression plasmid, even without induction (4) (Fig. 3E and Table S1). Thus, a very low concentration of TraR appears to be sufficient to regulate transcription in vivo. In fact, we were unable to detect TraR in Western blots under this condition. Because 1 ng was the lowest amount of purified TraR that we could detect in Western blots, and we could not detect a comparable TraR signal even with as much as 15 µg of cell lysate loaded per lane, we estimate that there is <1 ng of TraR in 15 µg of total cellular protein in uninduced cells. Additional studies will be needed to determine the TraR concentration and its physiological consequences when TraR is made from the Py promoter and DksA is present. Whatever the TraR concentration is during conjugation, we note that single-molecule studies suggest that the short residence time of secondary channel binding factors on RNAP helps them cooperate to regulate transcription while minimizing mutual interference (38).

The conservation of TraR homologs across major bacteriophage and bacterial groups suggests that its regulatory functions are strongly selected for in evolution. In general, ppGpp does not target RNAP directly in species distant from the proteobacteria (14). Thus, it is conceivable that TraR-like proteins have taken on some of the functions of ppGpp/DksA in the nonproteobacteria. Future characterization of the function of TraR homologs encoded by other conjugal plasmids, mobile elements, or bacteriophage may shed light on the roles of these factors in horizontal gene transfer.

Materials and Methods

Additional details for all procedures are in SI Materials and Methods.

Strains, Plasmids, Oligonucleotides, and Geneblock Sequences.

Strains and plasmids are listed in Table S3, oligonucleotide sequences are in Table S4, and Geneblock (gBlock) sequences are in Table S5.

Purification of TraR, DksA, and RNAP.

TraR and variants were purified from BL21DE3 dksA::Tn10 containing pET28a-traR-His6 plasmids by Ni-NTA chromatography (Qiagen) essentially as described previously (11). The histidine tag was cleaved off, and TraR without the His6 tag was dialyzed against storage buffer containing 10 mM Tris⋅Cl (pH 8.0), 0.1 mM EDTA, 2 mM DTT, 250 mM NaCl, and 50% glycerol. Protein concentrations were determined with the Bradford assay reagent (Bio-Rad) using BSA as a standard. HMK-DksA was purified as described previously (5). WT RNAP (core and holoenzyme), Bpa-containing RNAPs, and RNAPs not containing Bpa were purified as described previously (14, 35, 39).

In Vitro Transcription.

Single- or multiple-round in vitro transcription reactions were carried out as described previously (5, 14). Reactions contained TraR, DksA, or ppGpp (TriLink Biotechnologies) at the concentrations indicated in the figure legends.

DNaseI Footprinting.

DNaseI footprints were performed as described previously (35). Five-nanomolars of RNAP were added to ³²P-dCTP (Perkin-Elmer) end-labeled DNA in buffer containing 30 mM KCl, 10 mM Tris⋅Cl, pH 8.0, 10 mM MgCl₂, 1 mM DTT, and 0.1 mg/mL BSA.

³²P-ppGpp Binding Assay.

The DRaCALA assay, adapted from ref. 27, was used to measure ³²P-ppGpp binding to proteins essentially as described previously (14), with WT or Δω RNAP, ±TraR or DksA.

Site-Directed Mutagenesis.

Substitutions were introduced into pTrc99a-traR and/or pET28a-traR using the QuikChange Lightning Multi Site-Directed Mutagenesis Kit (Agilent) with mutagenic primers listed in Table S4. Mutations were confirmed by DNA sequencing.

Bpa Cross-Linking.

Cross-linking was performed as described previously (16, 35) using ³²P-labeled HMK-DksA and ³²P-labeled TraR-HMK.

TraR–RNAP Binding (Fe²⁺ Cleavage Competition Assay).

Western Blotting.

Western blots were performed using standard procedures using a rabbit polyclonal anti-TraR antibody (Covance Research Product).

TraR Conservation Analysis.

TraR-like proteins were identified by National Center for Biotechnology Information (NCBI) Blast using E. coli F plasmid-encoded TraR as a query. One-hundred different TraR homologs between 69 and 75 amino acids in length, representing a range of bacteria and bacteriophages, were chosen from ∼500 TraR-like sequences, aligned using Clustal Omega, and analyzed for conservation of individual residues using the ConSurf Server (consurf.tau.ac.il/2016/) (Table S2). Degree of conservation is indicated in Fig. 7A. TraR homologs were identified in Proteobacteria, as well as distantly related species in Nitrospira, Bacilli, and Actinobacteria. The database annotations suggest that many of these sequences are in conjugal plasmids, prophage, or mobile elements. Among bacteriophages, TraR-like proteins were found in the Myoviridae, Siphoviridae, and Podoviridae families.

Modeling of TraR Bound to RNAP.

A model for the structure of TraR was generated using RaptorX (raptorx.uchicago.edu/StructurePrediction/predict/). All 73 TraR residues were modeled. Six (8%) were predicted as disordered, and the secondary structure was predicted to contain 61% helix, 2% β-sheet, and 35% loop (Fig. 6A). The TraR–RNAP model was generated in Pymol based on genetic and biochemical data presented here and on models for how DksA binds in the secondary channel of RNAP (14, 17). The N-terminal region of TraR (including residues D3, D6, and A8) was aligned in PyMol with the coiled-coil tip region of DksA (Fig. 6B). TraR was positioned manually, because limited sequence similarity in the TraR N-terminal region precluded a Pymol-generated alignment.

Inductively Coupled Plasma-Mass Spectrometry Analysis.

Elemental analysis of a purified TraR preparation for determination of zinc content was performed at the University of Wisconsin–Madison Soil & Plant Analysis Laboratory, Department of Soil Science.

Sedimentation Equilibrium Analytical Ultracentrifugation.

The oligomeric state of purified TraR was determined by the University of Wisconsin–Madison Biophysics Instrumentation Facility.

ESI-Mass Spectrometry.

ESI-mass spectrometry was performed by the Mass Spectrometry/Proteomics Facility in the Biotechnology Center, University of Wisconsin–Madison to determine the molecular mass of TraR and its oligomeric state.

Acknowledgments

We thank Karen Wasserman for help with antibody production; Navneet Singh for preliminary Western blot analysis of TraR variants; Suzanne Treml for strain constructions; and Christophe Herman for helpful discussions. This work was supported by NIH Grant R37 GM37048 (to R.L.G.).

Footnotes

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1704105114/-/DCSupplemental.

References

1.Liu X, Jiang H, Gu Z, Roberts JW. High-resolution view of bacteriophage lambda gene expression by ribosome profiling. Proc Natl Acad Sci USA. 2013;110:11928–11933. doi: 10.1073/pnas.1309739110. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Storz G, Wolf YI, Ramamurthi KS. Small proteins can no longer be ignored. Annu Rev Biochem. 2014;83:753–777. doi: 10.1146/annurev-biochem-070611-102400. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Maneewannakul K, Ippen-Ihler K. Construction and analysis of F plasmid traR, trbJ, and trbH mutants. J Bacteriol. 1993;175:1528–1531. doi: 10.1128/jb.175.5.1528-1531.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Blankschien MD, et al. TraR, a homolog of a RNAP secondary channel interactor, modulates transcription. PLoS Genet. 2009;5:e1000345. doi: 10.1371/journal.pgen.1000345. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Paul BJ, et al. DksA: A critical component of the transcription initiation machinery that potentiates the regulation of rRNA promoters by ppGpp and the initiating NTP. Cell. 2004;118:311–322. doi: 10.1016/j.cell.2004.07.009. [DOI] [PubMed] [Google Scholar]
6.Perederina A, et al. Regulation through the secondary channel—Structural framework for ppGpp-DksA synergism during transcription. Cell. 2004;118:297–309. doi: 10.1016/j.cell.2004.06.030. [DOI] [PubMed] [Google Scholar]
7.Paul BJ, Berkmen MB, Gourse RL. DksA potentiates direct activation of amino acid promoters by ppGpp. Proc Natl Acad Sci USA. 2005;102:7823–7828. doi: 10.1073/pnas.0501170102. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Doran TJ, Loh SM, Firth N, Skurray RA. Molecular analysis of the F plasmid traVR region: traV encodes a lipoprotein. J Bacteriol. 1994;176:4182–4186. doi: 10.1128/jb.176.13.4182-4186.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Sergueev K, Court D, Reaves L, Austin S. E. coli cell-cycle regulation by bacteriophage lambda. J Mol Biol. 2002;324:297–307. doi: 10.1016/s0022-2836(02)01037-9. [DOI] [PubMed] [Google Scholar]
10.Frost LS, Ippen-Ihler K, Skurray RA. Analysis of the sequence and gene products of the transfer region of the F sex factor. Microbiol Rev. 1994;58:162–210. doi: 10.1128/mr.58.2.162-210.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Grace ED, et al. Activation of the σE-dependent stress pathway by conjugative TraR may anticipate conjugational stress. J Bacteriol. 2015;197:924–931. doi: 10.1128/JB.02279-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Ross W, Vrentas CE, Sanchez-Vazquez P, Gaal T, Gourse RL. The magic spot: A ppGpp binding site on E. coli RNA polymerase responsible for regulation of transcription initiation. Mol Cell. 2013;50:420–429. doi: 10.1016/j.molcel.2013.03.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Zuo Y, Wang Y, Steitz TA. The mechanism of E. coli RNA polymerase regulation by ppGpp is suggested by the structure of their complex. Mol Cell. 2013;50:430–436. doi: 10.1016/j.molcel.2013.03.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Ross W, et al. ppGpp binding to a site at the RNAP-DksA interface accounts for its dramatic effects on transcription initiation during the stringent response. Mol Cell. 2016;62:811–823. doi: 10.1016/j.molcel.2016.04.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Rutherford ST, et al. Effects of DksA, GreA, and GreB on transcription initiation: Insights into the mechanisms of factors that bind in the secondary channel of RNA polymerase. J Mol Biol. 2007;366:1243–1257. doi: 10.1016/j.jmb.2006.12.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Lennon CW, et al. Direct interactions between the coiled-coil tip of DksA and the trigger loop of RNA polymerase mediate transcriptional regulation. Genes Dev. 2012;26:2634–2646. doi: 10.1101/gad.204693.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Parshin A, et al. DksA regulates RNA polymerase in Escherichia coli through a network of interactions in the secondary channel that includes sequence insertion 1. Proc Natl Acad Sci USA. 2015;112:E6862–E6871. doi: 10.1073/pnas.1521365112. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Laptenko O, Lee J, Lomakin I, Borukhov S. Transcript cleavage factors GreA and GreB act as transient catalytic components of RNA polymerase. EMBO J. 2003;22:6322–6334. doi: 10.1093/emboj/cdg610. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Opalka N, et al. Structure and function of the transcription elongation factor GreB bound to bacterial RNA polymerase. Cell. 2003;114:335–345. doi: 10.1016/s0092-8674(03)00600-7. [DOI] [PubMed] [Google Scholar]
20.Lamour V, et al. Crystal structure of Escherichia coli Rnk, a new RNA polymerase-interacting protein. J Mol Biol. 2008;383:367–379. doi: 10.1016/j.jmb.2008.08.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Lemke JJ, et al. Direct regulation of Escherichia coli ribosomal protein promoters by the transcription factors ppGpp and DksA. Proc Natl Acad Sci USA. 2011;108:5712–5717. doi: 10.1073/pnas.1019383108. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Lennon CW, Gaal T, Ross W, Gourse RL. Escherichia coli DksA binds to free RNA polymerase with higher affinity than to RNA polymerase in an open complex. J Bacteriol. 2009;191:5854–5858. doi: 10.1128/JB.00621-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Rutherford ST, Villers CL, Lee JH, Ross W, Gourse RL. Allosteric control of Escherichia coli rRNA promoter complexes by DksA. Genes Dev. 2009;23:236–248. doi: 10.1101/gad.1745409. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Barker MM, Gaal T, Gourse RL. Mechanism of regulation of transcription initiation by ppGpp. II. Models for positive control based on properties of RNAP mutants and competition for RNAP. J Mol Biol. 2001;305:689–702. doi: 10.1006/jmbi.2000.4328. [DOI] [PubMed] [Google Scholar]
25.Repoila F, Gottesman S. Signal transduction cascade for regulation of RpoS: Temperature regulation of DsrA. J Bacteriol. 2001;183:4012–4023. doi: 10.1128/JB.183.13.4012-4023.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Bougdour A, Gottesman S. ppGpp regulation of RpoS degradation via anti-adaptor protein IraP. Proc Natl Acad Sci USA. 2007;104:12896–12901. doi: 10.1073/pnas.0705561104. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Roelofs KG, Wang J, Sintim HO, Lee VT. Differential radial capillary action of ligand assay for high-throughput detection of protein-metabolite interactions. Proc Natl Acad Sci USA. 2011;108:15528–15533. doi: 10.1073/pnas.1018949108. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Satory D, Halliday JA, Sivaramakrishnan P, Lua RC, Herman C. Characterization of a novel RNA polymerase mutant that alters DksA activity. J Bacteriol. 2013;195:4187–4194. doi: 10.1128/JB.00382-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Lee JH, Lennon CW, Ross W, Gourse RL. Role of the coiled-coil tip of Escherichia coli DksA in promoter control. J Mol Biol. 2012;416:503–517. doi: 10.1016/j.jmb.2011.12.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Blankschien MD, et al. Super DksAs: Substitutions in DksA enhancing its effects on transcription initiation. EMBO J. 2009;28:1720–1731. doi: 10.1038/emboj.2009.126. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Källberg M, et al. Template-based protein structure modeling using the RaptorX web server. Nat Protoc. 2012;7:1511–1522. doi: 10.1038/nprot.2012.085. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Ashkenazy H, Erez E, Martz E, Pupko T, Ben-Tal N. ConSurf 2010: Calculating evolutionary conservation in sequence and structure of proteins and nucleic acids. Nucleic Acids Res. 2010;38:W529–W533. doi: 10.1093/nar/gkq399. [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Furman R, Tsodikov OV, Wolf YI, Artsimovitch I. An insertion in the catalytic trigger loop gates the secondary channel of RNA polymerase. J Mol Biol. 2013;425:82–93. doi: 10.1016/j.jmb.2012.11.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Mekler V, Minakhin L, Borukhov S, Mustaev A, Severinov K. Coupling of downstream RNA polymerase-promoter interactions with formation of catalytically competent transcription initiation complex. J Mol Biol. 2014;426:3973–3984. doi: 10.1016/j.jmb.2014.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Winkelman JT, et al. Crosslink mapping at amino acid-base resolution reveals the path of DNA in a scrunched initial transcribing complex. Mol Cell. 2015;59:768–780. doi: 10.1016/j.molcel.2015.06.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Frost LS, Manchak J. F-phenocopies: Characterization of expression of the F transfer region in stationary phase. Microbiology. 1998;144:2579–2587. doi: 10.1099/00221287-144-9-2579. [DOI] [PubMed] [Google Scholar]
37.Murray HD, Schneider DA, Gourse RL. Control of rRNA expression by small molecules is dynamic and nonredundant. Mol Cell. 2003;12:125–134. doi: 10.1016/s1097-2765(03)00266-1. [DOI] [PubMed] [Google Scholar]
38.Tetone LE, et al. Dynamics of GreB-RNA polymerase interaction allow a proofreading accessory protein to patrol for transcription complexes needing rescue. Proc Natl Acad Sci USA. 2017;114:E1081–E1090. doi: 10.1073/pnas.1616525114. [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Burgess RR, Jendrisak JJ. A procedure for the rapid, large-scall purification of Escherichia coli DNA-dependent RNA polymerase involving polymin P precipitation and DNA-cellulose chromatography. Biochemistry. 1975;14:4634–4638. doi: 10.1021/bi00692a011. [DOI] [PubMed] [Google Scholar]
40.Chin JW, Martin AB, King DS, Wang L, Schultz PG. Addition of a photocrosslinking amino acid to the genetic code of Escherichia coli. Proc Natl Acad Sci USA. 2002;99:11020–11024. doi: 10.1073/pnas.172226299. [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Amann E, Ochs B, Abel KJ. Tightly regulated tac promoter vectors useful for the expression of unfused and fused proteins in Escherichia coli. Gene. 1988;69:301–315. doi: 10.1016/0378-1119(88)90440-4. [DOI] [PubMed] [Google Scholar]
42.Winkelman JT, Chandrangsu P, Ross W, Gourse RL. Open complex scrunching before nucleotide addition accounts for the unusual transcription start site of E. coli ribosomal RNA promoters. Proc Natl Acad Sci USA. 2016;113:E1787–E1795. doi: 10.1073/pnas.1522159113. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r1] 1.Liu X, Jiang H, Gu Z, Roberts JW. High-resolution view of bacteriophage lambda gene expression by ribosome profiling. Proc Natl Acad Sci USA. 2013;110:11928–11933. doi: 10.1073/pnas.1309739110. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r2] 2.Storz G, Wolf YI, Ramamurthi KS. Small proteins can no longer be ignored. Annu Rev Biochem. 2014;83:753–777. doi: 10.1146/annurev-biochem-070611-102400. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r3] 3.Maneewannakul K, Ippen-Ihler K. Construction and analysis of F plasmid traR, trbJ, and trbH mutants. J Bacteriol. 1993;175:1528–1531. doi: 10.1128/jb.175.5.1528-1531.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r4] 4.Blankschien MD, et al. TraR, a homolog of a RNAP secondary channel interactor, modulates transcription. PLoS Genet. 2009;5:e1000345. doi: 10.1371/journal.pgen.1000345. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r5] 5.Paul BJ, et al. DksA: A critical component of the transcription initiation machinery that potentiates the regulation of rRNA promoters by ppGpp and the initiating NTP. Cell. 2004;118:311–322. doi: 10.1016/j.cell.2004.07.009. [DOI] [PubMed] [Google Scholar]

[r6] 6.Perederina A, et al. Regulation through the secondary channel—Structural framework for ppGpp-DksA synergism during transcription. Cell. 2004;118:297–309. doi: 10.1016/j.cell.2004.06.030. [DOI] [PubMed] [Google Scholar]

[r7] 7.Paul BJ, Berkmen MB, Gourse RL. DksA potentiates direct activation of amino acid promoters by ppGpp. Proc Natl Acad Sci USA. 2005;102:7823–7828. doi: 10.1073/pnas.0501170102. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r8] 8.Doran TJ, Loh SM, Firth N, Skurray RA. Molecular analysis of the F plasmid traVR region: traV encodes a lipoprotein. J Bacteriol. 1994;176:4182–4186. doi: 10.1128/jb.176.13.4182-4186.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r9] 9.Sergueev K, Court D, Reaves L, Austin S. E. coli cell-cycle regulation by bacteriophage lambda. J Mol Biol. 2002;324:297–307. doi: 10.1016/s0022-2836(02)01037-9. [DOI] [PubMed] [Google Scholar]

[r10] 10.Frost LS, Ippen-Ihler K, Skurray RA. Analysis of the sequence and gene products of the transfer region of the F sex factor. Microbiol Rev. 1994;58:162–210. doi: 10.1128/mr.58.2.162-210.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r11] 11.Grace ED, et al. Activation of the σE-dependent stress pathway by conjugative TraR may anticipate conjugational stress. J Bacteriol. 2015;197:924–931. doi: 10.1128/JB.02279-14. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r12] 12.Ross W, Vrentas CE, Sanchez-Vazquez P, Gaal T, Gourse RL. The magic spot: A ppGpp binding site on E. coli RNA polymerase responsible for regulation of transcription initiation. Mol Cell. 2013;50:420–429. doi: 10.1016/j.molcel.2013.03.021. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r13] 13.Zuo Y, Wang Y, Steitz TA. The mechanism of E. coli RNA polymerase regulation by ppGpp is suggested by the structure of their complex. Mol Cell. 2013;50:430–436. doi: 10.1016/j.molcel.2013.03.020. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r14] 14.Ross W, et al. ppGpp binding to a site at the RNAP-DksA interface accounts for its dramatic effects on transcription initiation during the stringent response. Mol Cell. 2016;62:811–823. doi: 10.1016/j.molcel.2016.04.029. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r15] 15.Rutherford ST, et al. Effects of DksA, GreA, and GreB on transcription initiation: Insights into the mechanisms of factors that bind in the secondary channel of RNA polymerase. J Mol Biol. 2007;366:1243–1257. doi: 10.1016/j.jmb.2006.12.013. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r16] 16.Lennon CW, et al. Direct interactions between the coiled-coil tip of DksA and the trigger loop of RNA polymerase mediate transcriptional regulation. Genes Dev. 2012;26:2634–2646. doi: 10.1101/gad.204693.112. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r17] 17.Parshin A, et al. DksA regulates RNA polymerase in Escherichia coli through a network of interactions in the secondary channel that includes sequence insertion 1. Proc Natl Acad Sci USA. 2015;112:E6862–E6871. doi: 10.1073/pnas.1521365112. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r18] 18.Laptenko O, Lee J, Lomakin I, Borukhov S. Transcript cleavage factors GreA and GreB act as transient catalytic components of RNA polymerase. EMBO J. 2003;22:6322–6334. doi: 10.1093/emboj/cdg610. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r19] 19.Opalka N, et al. Structure and function of the transcription elongation factor GreB bound to bacterial RNA polymerase. Cell. 2003;114:335–345. doi: 10.1016/s0092-8674(03)00600-7. [DOI] [PubMed] [Google Scholar]

[r20] 20.Lamour V, et al. Crystal structure of Escherichia coli Rnk, a new RNA polymerase-interacting protein. J Mol Biol. 2008;383:367–379. doi: 10.1016/j.jmb.2008.08.011. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r21] 21.Lemke JJ, et al. Direct regulation of Escherichia coli ribosomal protein promoters by the transcription factors ppGpp and DksA. Proc Natl Acad Sci USA. 2011;108:5712–5717. doi: 10.1073/pnas.1019383108. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r22] 22.Lennon CW, Gaal T, Ross W, Gourse RL. Escherichia coli DksA binds to free RNA polymerase with higher affinity than to RNA polymerase in an open complex. J Bacteriol. 2009;191:5854–5858. doi: 10.1128/JB.00621-09. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r23] 23.Rutherford ST, Villers CL, Lee JH, Ross W, Gourse RL. Allosteric control of Escherichia coli rRNA promoter complexes by DksA. Genes Dev. 2009;23:236–248. doi: 10.1101/gad.1745409. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r24] 24.Barker MM, Gaal T, Gourse RL. Mechanism of regulation of transcription initiation by ppGpp. II. Models for positive control based on properties of RNAP mutants and competition for RNAP. J Mol Biol. 2001;305:689–702. doi: 10.1006/jmbi.2000.4328. [DOI] [PubMed] [Google Scholar]

[r25] 25.Repoila F, Gottesman S. Signal transduction cascade for regulation of RpoS: Temperature regulation of DsrA. J Bacteriol. 2001;183:4012–4023. doi: 10.1128/JB.183.13.4012-4023.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r26] 26.Bougdour A, Gottesman S. ppGpp regulation of RpoS degradation via anti-adaptor protein IraP. Proc Natl Acad Sci USA. 2007;104:12896–12901. doi: 10.1073/pnas.0705561104. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r27] 27.Roelofs KG, Wang J, Sintim HO, Lee VT. Differential radial capillary action of ligand assay for high-throughput detection of protein-metabolite interactions. Proc Natl Acad Sci USA. 2011;108:15528–15533. doi: 10.1073/pnas.1018949108. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r28] 28.Satory D, Halliday JA, Sivaramakrishnan P, Lua RC, Herman C. Characterization of a novel RNA polymerase mutant that alters DksA activity. J Bacteriol. 2013;195:4187–4194. doi: 10.1128/JB.00382-13. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r29] 29.Lee JH, Lennon CW, Ross W, Gourse RL. Role of the coiled-coil tip of Escherichia coli DksA in promoter control. J Mol Biol. 2012;416:503–517. doi: 10.1016/j.jmb.2011.12.028. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r30] 30.Blankschien MD, et al. Super DksAs: Substitutions in DksA enhancing its effects on transcription initiation. EMBO J. 2009;28:1720–1731. doi: 10.1038/emboj.2009.126. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r31] 31.Källberg M, et al. Template-based protein structure modeling using the RaptorX web server. Nat Protoc. 2012;7:1511–1522. doi: 10.1038/nprot.2012.085. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r32] 32.Ashkenazy H, Erez E, Martz E, Pupko T, Ben-Tal N. ConSurf 2010: Calculating evolutionary conservation in sequence and structure of proteins and nucleic acids. Nucleic Acids Res. 2010;38:W529–W533. doi: 10.1093/nar/gkq399. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r33] 33.Furman R, Tsodikov OV, Wolf YI, Artsimovitch I. An insertion in the catalytic trigger loop gates the secondary channel of RNA polymerase. J Mol Biol. 2013;425:82–93. doi: 10.1016/j.jmb.2012.11.008. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r34] 34.Mekler V, Minakhin L, Borukhov S, Mustaev A, Severinov K. Coupling of downstream RNA polymerase-promoter interactions with formation of catalytically competent transcription initiation complex. J Mol Biol. 2014;426:3973–3984. doi: 10.1016/j.jmb.2014.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r35] 35.Winkelman JT, et al. Crosslink mapping at amino acid-base resolution reveals the path of DNA in a scrunched initial transcribing complex. Mol Cell. 2015;59:768–780. doi: 10.1016/j.molcel.2015.06.037. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r36] 36.Frost LS, Manchak J. F-phenocopies: Characterization of expression of the F transfer region in stationary phase. Microbiology. 1998;144:2579–2587. doi: 10.1099/00221287-144-9-2579. [DOI] [PubMed] [Google Scholar]

[r37] 37.Murray HD, Schneider DA, Gourse RL. Control of rRNA expression by small molecules is dynamic and nonredundant. Mol Cell. 2003;12:125–134. doi: 10.1016/s1097-2765(03)00266-1. [DOI] [PubMed] [Google Scholar]

[r38] 38.Tetone LE, et al. Dynamics of GreB-RNA polymerase interaction allow a proofreading accessory protein to patrol for transcription complexes needing rescue. Proc Natl Acad Sci USA. 2017;114:E1081–E1090. doi: 10.1073/pnas.1616525114. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r39] 39.Burgess RR, Jendrisak JJ. A procedure for the rapid, large-scall purification of Escherichia coli DNA-dependent RNA polymerase involving polymin P precipitation and DNA-cellulose chromatography. Biochemistry. 1975;14:4634–4638. doi: 10.1021/bi00692a011. [DOI] [PubMed] [Google Scholar]

[r40] 40.Chin JW, Martin AB, King DS, Wang L, Schultz PG. Addition of a photocrosslinking amino acid to the genetic code of Escherichia coli. Proc Natl Acad Sci USA. 2002;99:11020–11024. doi: 10.1073/pnas.172226299. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r41] 41.Amann E, Ochs B, Abel KJ. Tightly regulated tac promoter vectors useful for the expression of unfused and fused proteins in Escherichia coli. Gene. 1988;69:301–315. doi: 10.1016/0378-1119(88)90440-4. [DOI] [PubMed] [Google Scholar]

[r42] 42.Winkelman JT, Chandrangsu P, Ross W, Gourse RL. Open complex scrunching before nucleotide addition accounts for the unusual transcription start site of E. coli ribosomal RNA promoters. Proc Natl Acad Sci USA. 2016;113:E1787–E1795. doi: 10.1073/pnas.1522159113. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

TraR directly regulates transcription initiation by mimicking the combined effects of the global regulators DksA and ppGpp

Saumya Gopalkrishnan

Wilma Ross

Albert Y Chen

Richard L Gourse

Series information

Significance

Abstract

Results

TraR Is More Active than DksA in Direct Inhibition of Transcription in Vitro, but It Has a Similar Affinity for RNAP.

Fig. 1.

Fig. S1.

Fig. S2.

TraR Shifts Occupancy of the RNAP–Promoter Complex to an Earlier Kinetic Intermediate and Decreases Complex Lifetime.

TraR Directly Activates Transcription by Eσ70 in Vitro.

Fig. 2.

TraR Functions Independently of ppGpp.

Fig. 3.

TraR and DksA Interact Differently with the Same Part of RNAP.

Fig. 4.

Alignment of TraR with DksA and Analysis of Critical TraR Residues.

Fig. 5.

Table S1.

Models for TraR and the TraR–RNAP Complex.

Fig. 6.

Conservation of Residues Among TraR Homologs.

Fig. 7.

Table S2.

Fig. S3.

Fig. S4.

TraR Appears to Function as a Monomer.

Fig. S5.

Fig. S6.

SI Materials and Methods

Bacterial Strains, Primers, and Growth Conditions.

Table S3.

Table S4.

Plasmid Construction.

Table S5.

Overexpression and Purification of TraR and TraR Variants.

In Vitro Transcription.

DNaseI Footprinting.

32P-ppGpp Binding Assay.

TraR–RNAP Binding (Fe2+ Cleavage Competition Assay).

Western Blotting.

Bpa Cross-Linking.

Cloning, Expression, and Western Blotting of the Half-DksA Variant.

Discussion

What Accounts for the Greater Activity of TraR Relative to DksA Without ppGpp?

Multiple Secondary Channel Binding Proteins: Why both DksA and TraR?

Materials and Methods

Strains, Plasmids, Oligonucleotides, and Geneblock Sequences.

Purification of TraR, DksA, and RNAP.

In Vitro Transcription.

DNaseI Footprinting.

32P-ppGpp Binding Assay.

Site-Directed Mutagenesis.

Bpa Cross-Linking.

TraR–RNAP Binding (Fe2+ Cleavage Competition Assay).

Western Blotting.

TraR Conservation Analysis.

Modeling of TraR Bound to RNAP.

Inductively Coupled Plasma-Mass Spectrometry Analysis.

Sedimentation Equilibrium Analytical Ultracentrifugation.

ESI-Mass Spectrometry.

Acknowledgments

Footnotes

References

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

TraR Directly Activates Transcription by Eσ⁷⁰ in Vitro.

³²P-ppGpp Binding Assay.

TraR–RNAP Binding (Fe²⁺ Cleavage Competition Assay).

³²P-ppGpp Binding Assay.

TraR–RNAP Binding (Fe²⁺ Cleavage Competition Assay).