Promoter recognition by a complex of Spx and the C-terminal domain of the RNA polymerase alpha subunit

Abstract

Background: Spx, an ArsC (arsenate reductase) family member, is a global transcriptional regulator of the microbial stress response and is highly conserved amongst Gram-positive bacteria. Bacillus subtilis Spx protein exerts positive and negative control of transcription through its interaction with the C-terminal domain of the RNA polymerase (RNAP) alpha subunit (alphaCTD). Spx activates trxA (thioredoxin) and trxB (thioredoxin reductase) in response to thiol stress, and bears an N-terminal C10XXC13 redox disulfide center that is oxidized in active Spx.

Methodology/principal findings: The structure of mutant Spx(C10S) showed a change in the conformation of helix alpha4. Amino acid substitutions R60E and K62E within and adjacent to helix alpha4 conferred defects in Spx-activated transcription but not Spx-dependent repression. Electrophoretic mobility-shift assays showed alphaCTD interaction with trxB promoter DNA, but addition of Spx generated a supershifted complex that was disrupted in the presence of reductant (DTT). Interaction of alphaCTD/Spx complex with promoter DNA required the cis-acting elements -45AGCA-42 and -34AGCG-31 of the trxB promoter. The Spx(G52R) mutant, defective in alphaCTD binding, did not interact with the alphaCTD-trxB complex. Spx(R60E) not only failed to complex with alphaCTD-trxB, but also disrupted alphaCTD-trxB DNA interaction.

Conclusions/significance: The results show that Spx and alphaCTD form a complex that recognizes the promoter DNA of an Spx-controlled gene. A conformational change during oxidation of Spx to the disulfide form likely alters the structure of Spx alpha helix alpha4, which contains residues that function in transcriptional activation and alphaCTD/Spx-promoter interaction. The results suggest that one of these residues, R60 of the alpha4 region of oxidized Spx, functions in alphaCTD/Spx-promoter contact but not in alphaCTD interaction.

Figure 1. Structure of reduced C10S Spx in complex with αCTD.

(A) Spx and αCTD are shown as teal and green ribbons, respectively, and their secondary structures are labelled. Helix α4, which is observed in oxidized Spx but has unraveled in the reduced form, is colored magenta. The residues mutated in this study, R60 and K62, are labelled and shown as sticks with carbon atoms colored white and nitrogen atoms either blue or magenta. Residues S10 and C13 are labelled and shown as sticks with carbon and sulphur atoms colored yellow and the γ-oxygen of S10, red. (B) Close up of the region surrounding helix α4 and residues C10/S10 and C13 after the superposition of the oxidized and reduced αCTD-Spx complex structures. Reduced Spx is shown as a magenta ribbon and oxidized Spx as a teal ribbon. The C10-C13 disulfide bond is shown in orange sticks and S10 and C13 from the reduced structure are shown as yellow sticks. In the reduced form residue R92 has moved 2.8 Å away from its position in the ammonium sulphate-containing oxidized form . The side chain of residue R60 beyond the Cβ atom is disordered in the sulphate-containing crystal form of oxidized Spx, which was used in the superposition visualized here .

Figure 2. Production of the wild-type and mutant Spx in B. subtilis.

B. subtilis cells expressing the wild-type and the mutant Spx^DD (the C-terminal two amino acids are substituted with aspartate residues, which renders the Spx protein insensitive to ClpXP protease) were grown in DS medium in the absence (lanes 1, 3, 5, and 7) and the presence of IPTG (lanes 2, 4, 6, and 8) as described in Materials and Methods. The lysate was prepared by the protoplast lysis method as described and 15 µg of total protein was resolved by SDS-polyacrylamide gel electrophoresis. Western blot analysis was carried out to detect Spx as shown previously. Lanes: M, molecular weight marker; 1 and 2, ORB6894 (Spx^DD), ORB6895 (Spx^DD-R60E), ORB6896 (Spx^DD-K62E), and ORB6897 (Spx^DD-K66E).

Figure 3. Effect of amino acid substitutions within and near helix α4 of Spx on the transcription of trxB (A) and srfA (B).

Strains carrying trxB-lacZ (A) or srfA-lacZ (B) were grown in DS medium. When the OD₆₀₀ was 0.4 to 0.5, each culture was divided into two flasks, and 1 mM IPTG was added to one flask to induce Spx^DD. Samples were taken at time intervals and β-galactosidase activities were measured. (A) Symbols: squares, ORB6894 with Spx^DD; circles, ORB6895 with Spx^DD-R60E; triangles, ORB6896 with Spx^DD-K62E; diamonds, ORB 6897 with Spx^DD-K66E. (B) Symbols: squares, ORB6129 with Spx^DD; circles, ORB6934 with Spx^DD-R60E; triangles, ORB6935 with Spx^DD-K62E; diamonds, ORB6936 with Spx^DD-K66E. Open symbols represent cells cultured without IPTG and closed symbols represent cells cultured with IPTG.

Figure 4. Effect of base substitutions of the trxB promoter on trxB expression.

Single base pair substitutions were generated in the trxB promoter (−115 to +47). The mutated promoters fused to lacZ were introduced in spx mutant strains expressing spx^DD from the IPTG-inducible Pspank-hy promoter. Expression of trxB-lacZ was determined in at least two independent isolates as described in Fig. 3. The effect of each base substitution is shown as a percentage of the peak trxB transcribed from the wild-type promoter, which was used as a control in each experiment. The peak expression was generally seen around 1.5 hr after the addition of IPTG.

Figure 5. Effect of the A-34T substitution on trxB transcription activated by the wild-type and mutant Spx.

Strains carrying trxB-lacZ fusions were grown in DS medium. When the OD₆₀₀ was 0.4 to 0.5, each culture was divided into two flasks, and 1 mM IPTG was added to one flask. Samples were taken at time intervals and β-galactosidase activities were measured. (A) Expression of the wild-type trxB-lacZ. Symbols: squares, ORB7276 with Spx^DD; triangles, ORB7282 with Spx^DD-R60E; diamonds, ORB7316 with Spx^DD-C10A; circles, ORB7337 with Spx^DD-G52R. (B) Expression of trxB(A-34T)-lacZ. Symbols: squares, ORB7342 with Spx^DD; triangles, ORB7343 with Spx^DD-R60E; diamonds, ORB7347 with Spx^DD-C10A; circles, ORB7348 with Spx^DD-G52R. Open symbols represent cells cultured without IPTG and closed symbols represent cells cultured with IPTG.

Figure 6. In vitro transcription from the wild-type and trxB(A-34T) promoters in the absence and presence of the wild-type and R60E Spx.

Either the wild-type or A-34T trxB template (1 nM) was incubated with 25 nM RNAP together with 25 nM σ^A in the presence of 7.5 nM Spx. The arrow shows the 66-base trxB transcript.

Figure 7. Interaction of αCTD and Spx variants with the regulatory region of the trxB promoter.

The trxB probe (−56 to −21) was generated by annealing of oligonucleotides followed by labeling of the 3′-end of the template strand using Klenow fragment and [³²P]dATP. Bands corresponding to the trxB/αCTD and trxB/Spx/αCTD complexes are marked with arrows. (A) EMSA analysis of αCTD and Spx binding to the trxB probe in reactions containing Spx variant or mixtures of mutant Spx proteins or mutant with the wild-type Spx (each at 5 µM). Abbreviations: W, wild-type Spx; G, Spx^G52R; R, Spx^R60E. (B) Redox-sensitive interaction was examined in the presence of DTT.

Figure 8. Effect of trxB base substitutions on interaction with αCTD and Spx.

The wild-type (W) and the trxB (G-44A/G-33A) mutant (M) probes were incubated with different concentrations of αCTD and Spx as described in Figure 7.

Figure 9. Interaction of αCTD and Spx with the spoVG promoter carrying AT-rich upstream sequences.

(A) The radiolabeled spoVG probe was generated as described in Figure 7. The spoVG-αCTD complex is marked with an arrow. (B) Competition of the trxB-αCTD and trxB-Spx-αCTD complexes with spoVG was examined by the addition of a 2- to 50-fold excess of cold spoVG probes.