Introduction

In bacteria, signal transduction in response to a wide variety of environmental stimuli is mediated by pairs of proteins that communicate with each other through a two-component signal transduction system (TCS, Figure 1) involving protein phosphorylation. TCS consists of a histidine kinase (HK) and a response regulator (RR) that have global roles in bacterial growth as well as in drug resistance, virulence, biofilm formation and quorum sensing.1 TCS inhibitors not only work as antibacterial agents but can also be developed as the inhibitors for drug resistance, biofilm formation, quorum sensing and virulence, as well as for plant-growth-regulating agents. In particular, TCSs are attractive as targets for antimicrobials for the following reasons. (1) Although many HK and RR genes are coded on the bacterial genome, few are found in lower eukaryotes. For example, only one HK and two RRs are present in the complete genome sequence of Saccharomyces cerevisiae, and none has been found from mammalian genomes. The HK/RR signal transduction system is distinct from serine/threonine and tyrosine phosphorylation in higher eukaryotes. (2) HKs and RRs possess a high degree of homology around their active sites. HKs are characterized by a conserved C-terminal catalytic domain containing a histidine residue, which is the site of autophosphorylation, and the N, G1, F and G2 boxes, which presumably form a nucleotide-binding surface. RRs contain a conserved N-terminal domain that includes the aspartate residue that is phosphorylated, a pair of aspartate residues preceding this site and a lysine residue, all of which contribute to the acidic pocket for the phosphorylation site. Such a high degree of structural homology in the catalytic domain of HKs and in the receiver domain of RRs suggests that multiple TCSs within a single bacterium could be inhibited simultaneously by a single inhibitor, thereby lowering the frequency of the appearance of drug-resistant strains. (3) Some TCSs are essential for bacterial viability: YycG/YycF in Bacillus subtilis, Staphylococcus aureus, Streptococcus pneumoniae, Enterococcus faecalis, Streptococcus pyogenes, Streptococcus mutans and Streptococcus epidermidis.2, 3 The YycG/YycF TCS, recently renamed WalK/WalR to reflect its function,2, 4 is highly conserved and specific to low-G+C Gram-positive bacteria, such as B. subtilis and S. aureus, in which it exerts an effect as a master regulatory system for cell wall metabolism and is essential for cell viability.4, 5, 6, 7, 8 As such, the WalK/WalR system constitutes an attractive target for the development of novel antimicrobial compounds, as inhibitors targeting WalK or WalR would be expected to have a bactericidal effect toward a broad range of major Gram-positive pathogens. Previously, a number of HK inhibitors have been identified,9, 10, 11, 12 but these studies did not conclusively show a direct relationship between cell death and HK inhibition rather than some other uncharacterized nonspecific effect. However, structure-based virtual screening of compounds potentially targeting the kinase activity of S. epidermidis WalK found several to be active against not only S. epidermidis but also S. aureus, S. aureus and S. mutans.13

Figure 1
figure 1

Two-component signal transduction system.

Full size image

WalR is a typical RR of the OmpR/PhoB subfamily with a characteristic winged helix-turn-helix DNA-binding domain.14, 15 These regulators control expression of their regulon genes by forming a head-to-head dimer of the receiver domains, using the conserved α4-β5-α5 face for intermolecular interactions, paired with a head-to-tail dimer of the winged helix-turn-helix motifs that bind to the tandem DNA repeats of their binding site.16 Inhibiting the cognate sensor kinase may not be sufficient to disrupt signaling, as RRs can often be phosphorylated by non-cognate sensor kinases17 or small-molecule phosphate donors,18 making them the more attractive target for inhibition.16, 19 Nevertheless, WalR-specific inhibitors have yet to be identified.

In this paper we describe a high-throughput screening system targeting the WalR RR, which uses a fusion between WalR and the DNA-binding domain of IclR to form a dimeric repressor and egfP as a reporter gene (Figure 2a). Using this system, we identified two novel inhibitors specifically targeting WalR in B. subtilis and S. aureus that were highly effective in killing these bacteria.

Figure 2
figure 2

Screening of WalR inhibitors. (a) High-throughput screening system. N100, N-terminal 100-a.a. DNA-binding domain of IclR; HD, homodimerization domain; IclR Box, −21 to +14 on iclR promoter region; RRs, response regulators (WalR, ArcA and OmpR). (b) Repression of EGFP intensity by response regulators and IclR. JM109 containing pFI003 (column 1), pFI014 (column 2), pFI001 (column 3), pArcA (column 4) or pOmpR (column 5) was cultured and fluorescence intensity was measured as described in Materials and methods, indicated as a value relative to that of JM109 containing pFI003. (c) Structural formulas of walrycins A and B. (d) Induction of EGFP by walrycins A or B. JM109 containing pFI014 (open circles), pFI001 (squares), pArcA (triangles) or pOmpR (diamonds) was cultured in the presence or absence of walrycins A or B. Fluorescence intensity is presented here as a value relative to that in the absence of walrycins A or B.

Full size image

Materials and methods

Bacteria strains and plasmids

Bacterial strains and plasmids used in this study are listed in Table 1. Antibiotics were used at the following concentrations: ampicillin: 50 μg ml–1 for B. subtilis and 100 μg ml–1 for E. coli; chloramphenicol: 5 μg ml–1 for B. subtilis and 10 μg ml for S. aureus; kanamycin: 5 μg ml–1.

Table 1 Bacterial strains and plasmids used in this study
Full size table

Plasmid construction

Primers used for plasmid construction are listed in Supplementary Table S1. To obtain pFI014, we PCR-amplified the DNA fragment containing the full length of walR of B. subtilis by using pBY33 as the template and YFIR4-399F and YFIR358-705R as the primers. The amplicons were digested with KpnI and ligated at the KpnI site of pFI00320 (Figure 2a). For pArcA and pOmpR, the DNA fragments were prepared by PCR using the genomic DNA from E. coli W3110 as the template and the following primers: pFI003-arcA-F and pFI003-arcA-R for pArcA and pFI003-ompR-F and pFI003-ompR-R for pOmpR. The amplicons were digested with KpnI and ligated at the KpnI site of pFI003.

B. subtilis walR was also amplified by PCR with pBY33 as the template and primers pANBF-1 and BsYycF-R. The PCR product was digested with KpnI and XbaI and then inserted into KpnI and XbaI sites of pAN1821 to obtain pABF. A DNA fragment containing the S. aureus walR was amplified by PCR using primers OSA211 and OSA250 and S. aureus chromosomal DNA as a template. The PCR product was digested with BamHI and PstI and then cloned into pMK4Pprot22 to obtain pSD1-7. For WalR protein purification, B. subtilis walR was amplified by PCR with pBY33 as the template and primers BsF-A-F and BsF-B-R. The PCR product was digested with NdeI and XhoI and then inserted into XhoI and NdeI sites of pET21a(+) (Novagen, Madison, WI, USA). The constructed plasmid was named pETBsFF. For WalKtru (C-terminal region, 204–611 a.a.) protein purification, B. subtilis walK was amplified by PCR with pBY33 as the template and primers YYCG(TRU)-F and YYCG(TRU)-R. The PCR product was digested with BamHI and NotI and then inserted into BamHI and NotI sites of pET21a(+). The constructed plasmid was named pETBsG-tru. A DNA fragment containing walRKyycH of S. aureus was amplified by PCR with S. aureus N31523 genomic DNA as the template and primers SAvicop-up and SAvicop-2dw. The PCR product was directly ligated to pSC-B (Stratagene, La Jolla, CA, USA) and was named pSCSawalRKH. S. aureus walR was sub-cloned from pSCSawalRKH to pET101/D-TOPO (Invitrogen, Carlsbad, CA, USA) by using primers TP-Sa315VR-F and TP-Sa315VR-3 and named pETSa315VR.

Chemical library

Two different kinds of chemical libraries were used for screening: one was produced by a new bioconversion system in recombinant E. coli cells;24 the other contained 10 000 compounds from ChemBridge Corporation (San Diego, CA, USA).

Screening method

A chemical library was screened using the E. coli-based high-throughput enhanced green fluorescent protein (EGFP) screening system. JM109 cells containing plasmids pFI001, pFI003, pFI014, pArcA or pOmpR were grown at 37 °C in LB (Luria-Bertani) medium containing ampicillin. Then, 1 μl of this culture was added to 96-titer wells containing 200 μl of LB–ampicillin medium and 1 μl of drugs, and the cultures were grown at 37 °C for 20 h. Fluorescence was measured using a Wallac 1420 ARVOsx (PerkinElmer Life Sciences, Boston, MA, USA) with excitation and emission wavelengths of 485 and 535 nm, respectively, and the optical density at 595 nm (OD595) was measured using a Model 3550 Microplate Reader (Bio-Rad Laboratories, Hercules, CA, USA). For normalization the fluorescence intensity of each sample was divided by its OD595 value.

Drug susceptibility assay

The MICs were examined by the serial agar dilution method using Mueller-Hinton agar (Difco Laboratories, Detroit, MI, USA).25 The MIC was determined as the lowest concentration at which growth was significantly inhibited.

Trans-phosphorylation of WalR by phosphorylated WalKtru

To prepare phosphorylated (P-) WalKtru, WalKtru (0.5 μM) was incubated at room temperature for 15 min in kinase buffer (50 mM Tris-HCl pH 8.5, 200 mM KCl, 100 mM NH4Cl, 25 mM MgCl2, 12.5 mM ATP and 0.4 μCi [γ-32P]ATP). After WalR (1 μM) was incubated at room temperature for 60 min in the presence of walrycins A or B, a P-WalKtru mixture was added and incubated for 15 min. The reaction was terminated by adding SDS sample buffer. The samples were analyzed using SDS-polyacrylamide gel electrophoresis. The gel was dried and exposed on an image plate and analyzed using FLA-7000 (Fuji Film, Tokyo, Japan).

RNA preparation and complementary DNA synthesis

For quantitative real-time PCRs (qRT-PCRs), the B. subtilis 168 and S. aureus N315 strains were grown with aeration to an OD 600 of 0.4, after which walrycins A or B was added and they were cultivated for another 5 min. The cells were harvested by centrifugation and immediately suspended in RNA-later (Ambion, Austin, TX, USA). RNA preparations were then performed with an SV total RNA Isolation System (Promega, Madison, WI, USA), followed by a DNase I treatment with the TURBO DNA-free Kit (Ambion) to eliminate residual contamination by genomic DNA. Reverse transcription reactions were performed with the High Capacity complementary DNA Reverse Transcription Kit (ABI, Foster City, CA, USA).

qRT-PCRs

Primers (Supplementary Table S2) were selected with Primer Express Software Version 3.0 (ABI). qRT-PCRs were performed with complementary DNA (0.1 ng), 250 nM gene-specific primer pair and PowerSYBR Green Master Mix (ABI). PCR amplification, detection and analyses were performed using StepOne Real Time PCR system (ABI). PCR conditions included an initial denaturation step at 95 °C for 10 min, followed by a 40-cycle amplification (95 °C for 15 s and 60 °C for 60 s). The specificity of the amplified product and the absence of primer dimer formation were verified by generating a melting curve. Melting curve analysis was performed by an initial denaturation step of 95 °C for 15 s followed by 60 °C for 1 min and again 95 °C for 15 s. Fluorescence was measured continually during the melting curve cycle (a stepwise 0.3 °C temperature), beginning at 60 °C. The absence of contaminating genomic DNA was verified by testing each sample in control reactions without a previous reverse transcription step. The critical threshold cycle (CT) was defined for each sample.

The expression levels were normalized using the 16S rRNA gene as an internal standard whose transcriptional level did not vary under our experimental conditions. Each assay was performed in triplicate and repeated using at least three independent RNA samples.

Preparation of 32P-end-labeled DNA probes

Reverse primer (10 pmol, 5′-AAGACGTCATTGATAAAGACGCACTCCGGT-3′) was phosphorylated with 10 μCi [γ-32P]ATP (5000 Ci mmol 1–1, PerkinElmer) by using T4 polynucleotide kinase (Takara Bio, Otsu, Japan). The promoter fragment (PyocH) was amplified by PCR using B. subtilis 168 genomic DNA as template, Ex Taq DNA polymerase (Takara Bio), and 32P-labeled reverse primer and forward primer (5′-AATACAGGGCTTATGCAAGGATGACAGACTA-3′). The labeled fragments were purified and the radioactivity was measured with a liquid scintillation counter.

Gel mobility shift assay

Gel mobility shift assays were carried out as described in Yamamoto et al.26 The 32P-labeled probe was incubated at 30 °C for 10 min with purified His6-protein in binding buffer. After addition of a DNA dye solution (40% (v v–1) glycerol, 0.025% (w v–1) bromophenol blue and 0.025% (w v–1) xylene cyanol, the mixtures were subjected to 6% polyacrylamide gel electrophoresis.

Fluorescent dyes and fluorescence microscopy

Exponentially growing cells were mounted on 2% agarose pads containing growth medium for microscopy use and stained with Synapto Green (membrane) and SYBR green I (DNA). Images ( × 1000) were acquired on a Radiance 2100 confocal laser scanning microscope (Bio-Rad Laboratories).

Protein purification

E. coli BL21(DE3) cells containing the pETBsFF, pETSa315VR, pKH55-21 or pKH43-21 plasmids27 were grown in 2 × YT medium at 30 °C. When the cell density reached an OD600 of 0.6, IPTG was added at the final concentration of 1 mM, and the cell suspension was cultivated for another 3 h before the cells were harvested by centrifugation, washed with lysis buffer (50 mM Tris-HCl (pH 8.0), 100 mM NaCl) and then stored at −80 °C until use. Frozen cells were suspended in lysis buffer containing 1 mM phenylmethylsulphonyl fluoride and then lysed by sonication. The lysates were cleared by centrifugation and filtered through a 0.22-μm filter. The filtrate was loaded onto a HisTrap HP column (ÄKTA prime system; GE Healthcare Bio-Science, Piscataway, NJ, USA) equilibrated with the lysis buffer, and His6-protein was eluted using an imidazole gradient from 0 to 500 mM. The fractions with the target protein were dialyzed against storage buffer (10 mM Tris-HCl (pH 8.0), 0.1 mM EDTA, 0.1 mM dithiothreitol and 10% (v v–1) glycerol), loaded onto a HiTrap Q FF (GE Healthcare Bio-Science) and eluted with a NaCl gradient of 0–500 mM in storage buffer. Fractions containing the His6-tagged proteins were collected and loaded onto a HiTrap Desalting column (GE Healthcare Bio-Science).

Analytical size-exclusion chromatography

Size-exclusion chromatography analysis was performed on a fast-performance liquid chromatography ÄKTA system using a Superdex 200 10/300 GL or Superdex 75 10/300 GL (all three devices from GE Healthcare Bio-Science). Experiments were carried out at 4 °C and a flow rate of 0.25 ml min–1 using 10 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.1 mM EDTA and 0.1 mM dithiothreitol as the mobile phase while monitoring absorbance at 280 nm was determined. Samples (300 μl) were prepared by incubating 2.5 nmol of protein in 10 mM Tris-HCl (pH 8.0), 0.1 mM EDTA and 0.1 mM dithiothreitol at 30 °C for 60 min in the presence or absence of 25 μg ml–1 of walrycin A. Before injection, samples were filtered through 0.22-μm filters.

Limited proteolysis with trypsin

After WalR proteins were digested with trypsin (Promega) as previously described,28 they were separated by SDS-polyacrylamide gel electrophoresis, transferred to a polyvinylidine difluoride membrane (Immobilon-P transfer membranes; Millipore, Bedford, MA, USA), and detected using anti-WalR antibody and enhanced chemiluminescence plus western blotting system (GE Healthcare Bio-Science).

Results and discussion

High-throughput screening of inhibitors targeting WalR

From the results of the crystal structure analysis of the Thermotoga maritime IclR protein,29 it was predicted that this protein is composed of two domains: the N-terminal DNA-binding domain containing the winged helix-turn-helix motif and a regulatory domain in the C-terminal of the protein, which is involved in its binding to a signal molecule. These two domains are linked by an α-helix.29 The binding of a signal molecule to the C-terminal domain of IclR is presumed to repress the transcription of target genes by modulating either DNA-binding activity or multimerization of IclR.29 Database homology searches have identified many IclR-like proteins, known as the IclR family, which show a conserved C-terminal region. E. coli IclR also forms a dimer and regulates both the aceBAK operon and its own expression.30, 31, 32 Furuta et al.20 previously reported that the region between 100 and 274 a.a. of IclR is involved in dimerization and that this region can be replaced with other functional homodimerization domains (Figure 1). To design a high-throughput screening system targeting the WalR RR, we constructed a chimeric repressor gene consisting of sequences encoding the 100 amino-terminal amino acid residues of IclR that constitute the DNA-binding domain (N100) fused to the walR, arcA or ompR RR genes (Figure 2a). To quantify the dimerization capacity of the resulting chimerical repressor proteins in terms of fluorescence intensity, the egfP reporter gene was placed directly downstream from the iclR promoter (Figure 2a) as described previously.20 In the presence of inhibitors targeting the RRs, the egfP gene is expressed and EGFP fluorescence intensity increases. WalR, ArcA and OmpR belong to the OmpR/PhoB subfamily, in which dimerization is essential for DNA-binding and transcriptional activation. Consequently, they are all capable of suppressing expression from the iclR promoter (Figure 2b). These results suggest that this high-throughput system could be used to screen inhibitors of RRs containing dimerization domains. To identify selective WalR inhibitors by measuring EGFP fluorescence intensity, we screened a chemical library against E. coli strains JM109/pFI014 (WalR) and JM109/pFI001 (IclR) in 96-titer wells. Through this screening process, we identified two compounds, compound A (4-methoxy-1-naphthol; CAS no. 84-85-5) and compound B (1,6-dimethyl-3-[4-(trifluoromethyl)phenyl]pyrimido[5,4-e][1,2,4]triazine-5,7-dione; CAS no. 878419-78-4) (Figure 2c) as WalR inhibitors. Both compounds strongly increased fluorescence intensity in a dose-dependent manner in strain JM109/pFI014 but did not significantly change fluorescence intensity of strain JM109/pFI001 (Figure 2d). Furthermore, to clarify that they actually target WalR, we analyzed their effect on phosphotransfer from autophosphorylated WalKtru (cytoplasmic HK region) to WalR (Figure 3). As a result, phosphotransfer from P-WalKtru to WalR was inhibited with increased concentrations of compounds A and B. These results strongly suggest that both compounds target WalR to inactivate the IclR–WalR chimera repressor for inducing GFP expression. Therefore, we named them walrycin A and walrycin B.

Figure 3
figure 3

Inhibition of phosphotransfer from P-WalKtru to WalR. Phosphotransfer from P-WalKtru to WalR was performed in the presence or absence of walrycin A (a, lanes 1–5) or walrycin B (b, lanes 6–10): lane 1, 0 μM; lane 2, 22.3 μM; lane 3, 44.8 μM; lane 4, 89.6 μM; lane 5, 179.2; lane 6, 0 μM; lane 7, 3.6 μM; lane 8, 14.5 μM; lane 9, 58.1 μM; lane 10, 232.4 μM.

Full size image

Regulation of WalR regulon genes

To establish that these compounds exert an effect specifically on the WalK/WalR system, the effects of walrycins on in vivo expression of WalR regulon genes4, 5, 8, 33, 34 were analyzed using qRT-PCR). In B. subtilis, WalR activates expression of ydjM (cell wall-associated protein), yocH (cell wall hydrolase) and yvcE (cwlO, cell wall endopeptidase), whereas it negatively regulates expression of yoeB (modulator of autolysin activity) and yjeA (peptideglycan deacetylase). In S. aureus, WalR activates expression of several cell wall metabolism genes including isaA (cell wall transglycosylase) and ssaA (cell wall amidase). As shown in Figure 4, addition of walrycins A and B to B. subtilis led to lowered expression of ydjM, yocH and yvcE and increased expression of yoeB and yjeA within 5 min (Figures 4a and b). Similarly, transcription of S. aureus WalR regulon genes isaA and ssaA stopped within 5 min of adding walrycins A and B to the culture (Figures 4a and b). Furthermore, both walrycin compounds had antimicrobial activity, with MICs of walrycins A and B of 64 and 0.39 μg ml–1 for B. subtilis 168 and 128 and 3.13 μg ml–1 for S. aureus N315, respectively. These results indicate that walrycins A and B target WalR and lead to cell death in both B. subtilis and S. aureus.

Figure 4
figure 4

Transcriptional regulation of WalR regulon genes by walrycins A or B. Total RNAs were extracted from B. subtilis or S.aureus cells, after which walrycin A (a; 0, 25 and 50 μg ml–1 for B. subtilis and 0, 50 and 100 μg ml–1 for S. aureus) or walrycin B (b; 0, 0.78 and 1.56 μg ml–1 for B. subtilis and 0, 3.13 and 6.25 μg ml–1 for S. aureus) was added, and they were cultivated for 5 min. After reverse transcription, specific complementary (c)DNAs were quantified using qRT-PCR. The results are expressed as the means of relative quantification to that of 0 μg ml–1 and s.d. of triplicate experiments using primers specific for WalR regulon genes, gyrA and 16S rRNA (normalizing gene).

Full size image

The MICs of walrycin B against B. subtilis and S. aureus were shown to be stronger than those of walrycin A, at which concentration the WalR regulon gene expressions were specifically effected to cause cell death. These results suggest that walrycins A and B actually target WalR in the B. subtilis and S. aureus. On the other hand, walrycins A and B inhibited phosphotransfer from WalK to WalR at similar concentrations in vitro. However, walrycin A induced EGFP at lower concentrations than did walrycin B in a high-throughput genetic system using E. coli (Figure 2). Walrycin A might enter E. coli cells easily, but not B. subtilis and S. aureus, whereas the reverse is true for walrycin B.

To confirm that walrycins target the WalR RR, the effects of WalR overproduction were analyzed. When WalR was overexpressed from a multi-copy plasmid (B. subtilis strain 168/pABF and S. aureus strain N315/pSD1-7), the MICs for walrycin A were increased twice over those of the control strains (168/pAN18 and N315/pMK4Pprot) to 128 and 256 μg ml–1, respectively. Under the same conditions, MICs for walrycin B were increased fourfold for B. subtilis and twofold for S. aureus. These results indicate that the walrycins specifically target WalR to cause cell death in both B. subtilis and S. aureus.

In this study we have shown that walrycins inhibit WalR activity in B. subtilis and S. aureus. These activities are involved in many crucial cell functions and influence cell development. Because autolysins are involved in the separation of daughter cells, one would expect WalR-depleted cells to be filamentous or aggregated. We used a confocal laser microscope to examine the effect of walrycin A on the morphology of B. subtilis and S. aureus cells. As shown in Figure 5, untreated B. subtilis and S. aureus cells had a typically short rod morphology or were well separated, whereas walrycins A and B-treated cells formed extremely long aseptate filaments or aggregates. Furthermore, after B. subtilis was incubated in the presence of walrycins A or B, cell lysis was observed (data not shown). These results are consistent with the previously reported phenotypes of WalR-depleted mutants.4, 6, 33

Figure 5
figure 5

Morphology changes induced by walrycins A or B. Confocal laser scanning microscopy of exponentially growing cells of B. subtilis 168 (a) and S. aureus N315 (b) in the presence or absence (Control) of walrycin A (50 μg ml–1 for B. subtilis and S. aureus) or walrycin B (0.19 μg ml–1 for B. subtilis and 0.78 μg ml–1 for S. aureus).

Full size image

A mode of action

B. subtilis and S. aureus WalRs, which were overproduced and purified from E. coli, revealed both monomeric and dimeric forms (Figure 6a-1, 2), whereas ArcA and OmpR formed the monomer as previously reported.35, 36, 37 After WalR was treated with walrycin A at 30 °C for 60 min followed by size-exclusion chromatography, a decrease in the concentration of WalR monomer and a concomitant increase in the dimeric form (Figure 6a-1, 2) were observed. This increase in the proportion of dimeric protein was not observed when purified ArcA or OmpR was incubated with walrycin A (Figure 6a-3, 4). As these results suggest that walrycin A can exert an effect specifically on monomeric WalR to increase dimer formation, we performed a size-exclusion chromatography on fractionated WalR-monomers with or without walrycin A treatment. As shown in Figure 6a-5, almost half of the monomer had transitioned into the dimeric form. When dimeric WalR was treated with walrycin A under the same conditions, however, no change was observed (Figure 6a-6). These results clearly show that walrycin A exerts an effect specifically on the monomeric form of WalR to increase the formation of WalR dimers. We designated the dimer that formed after the reaction of walrycin A and the WalR monomer as WalR(D′) and evaluated its structural characteristics (Figure 6b). WalR(D) and WalR(D′) were digested with trypsin, and the fragments were separated by SDS-polyacrylamide gel electrophoresis and analyzed using western blotting (Figure 6b). Results showed fragments arising from trypsin-digested WalR(D′), which were not detected in the WalR(D)-digested sample. These results indicate a possible altered conformation for WalR(D′) generated by the reaction with walrycin A when compared with endogenous WalR(D). The ability of WalR(D′) to bind to the promoter region of the B. subtilis yocH gene, which contains a WalR-binding site, was analyzed using gel mobility shift analysis (Figure 6c). Results showed that WalR(D) bound to the WalR box of PyocH and formed a DNA-WalR(D) complex that migrated at a slower rate than the free DNA. However, no protein/DNA complex was formed when the yocH probe was incubated with WalR(D′), indicating that no binding took place. These results show that walrycin A reacts with the WalR monomer to form a WalR dimer (WalR(D′)) with a different conformation from that of the endogenous dimeric WalR, and which lacks the ability to bind to WalR(D) operator sites (Figure 7).

Figure 6
figure 6

A mode of action of walrycin A targeting WalR. (a) Walrycin A interacts with monomeric WalR. After each protein (1–6) was incubated at 30 °C for 60 min without (solid line) or with (dotted line) 25 μg ml–1 of walrycin A, size-exclusion chromatography was performed as described in Materials and methods. WalRs (B. subtilis), ArcA and OmpR were separated using a Superdex 200 10/300 GL column, and S. aureus WalR was separated using a Superdex 75 10/300 GL column (GE Healthcare Bio-Science). 1, WalR (B. subtilis); 2, WalR (S. aureus); 3, ArcA (E. coli); 4, OmpR (E. coli); 5, the monomeric WalR (B. subtilis); 6, dimeric WalR (B. subtilis). M, monomer; D, dimer. (b) Conformational changes of WalR(D) and WalR(D′). WalR was digested with trypsin (protein to trypsin, 20:1). A total of 30 ng of WalR(D) (lanes 1–6) and WalR(D′) (lanes 7–12) were digested with 1.5 ng of trypsin for 0 (lanes 1 and 7), 15 (lanes 2 and 8), 30 (lanes 3 and 9), 60 (lanes 4 and 10), 90 (lanes 5 and 11) and 120 min (lanes 6 and 12). (c) Gel mobility shift analysis with WalR(D) and WalR(D′). A radiolabeled DNA fragment carrying the promoter of yocH was prepared as described in Materials and methods. The mobility of fragments without WalR addition is shown in lanes 1 and 5. Lanes 2–4 (WalR(D)) and 6–8 (WalR(D′)) show mobility after binding of 2.5 (lanes 2 and 6), 5 (lanes 3 and 7) or 10 pmol (lanes 4 and 8) of WalR.

Full size image
Figure 7
figure 7

A model of walrycin action.

Full size image

Concerning the difference in the mode of action between walrycins A and B, we conducted an in vitro phosphotransfer experiment from P-WalKtru to WalR in the presence or absence of walrycins (Figure 3). As a result, phosphotransfer was inhibited by both walrycins A and B. These results suggest that they likely inhibit WalR activity by a similar mechanism: by targeting WalR. Walrycin B is known as an analog of toxoflavin (a phytotoxin from Burkholderia glumae), which has been shown to have a strong MIC for B. subtilis and S. aureus but whose mode of action is not clear.38 The compound could also interact with WalR to cause bactericidal effects. Recently, Kim et al.39 reported that toxoflavin binds ToxR (an LysR-type transcriptional regulator) to activate synthetic gene expression of toxoflavin.

Walrycins are a new class of potent small-molecule compounds that kill bacterial cells by targeting the RR WalR and inhibiting this essential signal transduction pathway. They not only have therapeutic potential but will also prove to be useful reagents for the further study of the WalK/WalR TCS.