Structure-function comparisons of (p)ppApp vs (p)ppGpp for Escherichia coli RNA polymerase binding sites and for rrnB P1 promoter regulatory responses in vitro

Abstract

Precise regulation of gene expression is crucial for bacteria to respond to changing environmental conditions. In addition to protein factors affecting RNA polymerase (RNAP) activity, second messengers play an important role in transcription regulation, such as well-known effectors of the stringent response: guanosine 5'triphosphate-3'diphosphate and guanosine 3', 5'-bis(diphosphate) [(p)ppGpp]. Although much is known about importance of the 5' and 3' moieties of (p)ppGpp, the role of the guanine base remains somewhat cryptic. Here, we use (p)ppGpp's adenine analogs [(p)ppApp] to investigate how the nucleobase contributes to determine its binding site and transcriptional regulation. We determined X-ray crystal structure of Escherichia coli RNAP-(p)ppApp complex, which shows the analogs bind near the active site and switch regions of RNAP. We have also explored the regulatory effects of (p)ppApp on transcription initiating from the well-studied E. coli rrnB P1 promoter to assess and compare properties of (p)ppApp with (p)ppGpp. We demonstrate that contrary to (p)ppGpp, (p)ppApp activates transcription at this promoter and DksA hinders this effect. Moreover, pppApp exerts a stronger effect than ppApp. We also show that when ppGpp and pppApp are present together, the outcome depends on which one of them was pre-incubated with RNAP first. This behavior suggests a surprising Yin-Yang like reciprocal plasticity of RNAP responses at a single promoter, occasioned simply by pre-exposure to one or the other nucleotide. Our observations underscore the importance of the (p)ppNpp's purine nucleobase for interactions with RNAP, which may lead to a better fundamental understanding of (p)ppGpp regulation of RNAP activity.

Keywords: (p)ppApp; (p)ppGpp; RNAP; Transcription; ppGpp; rrnB P1.

Figure 1.. Structure of the ppApp-RNAP complex.

A. Electron density map of the ppApp and its binding site. Fo-Fc map (σ cut off = 4) is calculated by omitting ppApp for phase calculation and is shown as blue mesh with stick models of β (cyan) and β’ subunits (pink) of RNAP, as well as ppApp. Positions of adenine base and phosphate groups of ppApp are indicated. The catalytic Mg²⁺ ion is shown as a magenta sphere. Amino acid residues involved in the interaction with phosphate groups of ppApp are indicated. B. ppApp binding site. The β (cyan) and β’ (pink) subunits are shown as surface representation and ppApp is shown as a CPK/stick representation. Amino acid residues contact with adenine base via C2 position are highlighted and indicated (A426, red; Q465, yellow). C. The ppApp and ppGpp binding sites are shown in the E. coli RNAP σ⁷⁰ holoenzyme (RNAP: cartoon model with transparent surface; α, white; β, cyan; β’, pink; ω, gray; σ, orange; β flap and insert 11 domains are removed to see the ppApp binding site). ppApp and ppGpp are depicted as CPK representations. The clamp domain of β’ subunit is highlighted in green and the main channel of RNAP (MC) is indicated. Opening and closing of the clamp is indicated by a black arrow. D.magnified view of the ppApp binding site in (C). The DPBB domain (brown) and switch regions of RNAP involved in the ppApp binding are indicated. The σ_3.2 is highlighted in black. Myx bound at the switch regions is depicted in CPK/stick representation, which was constructed by superposing the RNAP-ppApp complex structure (this study) and the RNAP-Myx complex (PDB: 4YFX). E. model of open complex with ppApp was constructed by superposing the structures of the RNAP-ppApp complex (this study) and the transcription initiation complex (PDB: 4YLN). Template (t-DNA) and non-template DNA (nt-DNA) strands are depicted in white and black CPK/stick representation and the transcription start site (+1, green) is highlighted. A position of downstream DNA (dsDNA) is indicated. ppApp and active site Mg are depicted as CPK representation and magenta sphere. F. A magnified view of the ppApp binding site in (E). The transcription start site (+1, green) and the template DNA from −1 to −3 positions (brown) that occupy the ppApp binding site are indicated. Three positions of σ^3.2 in the apo-form, in the RNAP-ppApp complex and in the transcription initiation complex are depicted and indicated. Movement of σ^3.2 from the apo-form to the transcription initiation complex is indicated by a black arrow.

Figure 2.. Effect of (p)ppApp and ppGpp on the transcription initiating from the rrnB P1 promoter.

A. B. Transcription was performed at 90 mM KGlu and assessed in the presence of increasing concentrations of ADP, ppApp and pppApp (25–500 µM). C. Comparison of constant amount of ADP, a4p and a5p (250 µM) in the presence of increasing concentration of DksA (0–300 nM). D. Titration of a5p in an in vitro transcription initiating from the rrnB P1 promoter mutant (+1 transcription start site: A/G). The data are from at least 3 independent experiments, error bars represent S.D. Examples of autoradiograms displaying bands corresponding to rrnB P1 initiated transcripts resolved on sequencing gels are provided.

Figure 3.. Influence of pppApp and ppGpp on the transcription initiation stages.

A. Effect of 250μM pppApp (a5p) and 250 μM ppGpp (g4p) on promoter escape in the absence and presence of DksA (300 nM). B. Effect of 250 μM pppApp (a5p) and 250 μM ppGpp (g4p) on stability of open complexes at the rrnB P1 promoter. Half-lives of the open complexes were estimated by „run-off” transcription following heparin addition at time 0. The data are from at least 3 independent experiments, error bars represent S.D.

Figure 4.. RNA polymerase binding to the DNA fragment containing rrnB P1 promoter in the presence of pppApp.

A. Representative electrophoretic mobility shift assay (EMSA) gel of Cy5-labeled DNA fragment (5 nM) and indicated amounts of RNAP (0–160 nM). B. Quantifications of the percentage of RNAP–DNA complexes (shifted DNA) formed with RNAP only or with addition of 500 µM pppApp, normalized to the unbound DNA in each experiment. The data are from 3 independent experiments, error bars represent S.D.

Figure 5.. The effect of the order of addition of pppApp and ppGpp in in vitro transcription and EMSA experiments.

A. In vitro transcription carried out at different order of (p)ppNpp addition, as indicated. Details are provided in the text and Figure S1.B. EMSA experiments measuring stable DNA-RNAP complexes, carried out at different order of (p)ppNpp addition, as indicated. The amount of bound DNA is normalized to 20 nM RNAP in the presence of pppApp in panel I. Details are provided in the text and Figure S2. The data are from at least 3 independent experiments, error bars represent S.D.