Biochimica et Biophysica Acta 1804 (2010) 1369–1375
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b b a p a p
Evaluation of substituted triazol-1-yl-pyrimidines as inhibitors of Bacillus anthracis acetohydroxyacid synthase Vinayakumar Gedi a, Kumaresan Jayaraman b, Satish Kalme a,1, Hye-Yeon Park a, Hae-Chul Park a, Im-Joung La a, Hoh-Gyu Hahn c, Moon-Young Yoon a,⁎ a b c
Department of Chemistry and Institute of Natural Sciences, Hanyang University, Seoul 133-791, Republic of Korea Centre for Biotechnology, Anna University, Chennai-25, India Division of Life Sciences, Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea
a r t i c l e
i n f o
Article history: Received 13 November 2009 Received in revised form 21 January 2010 Accepted 3 February 2010 Available online 17 February 2010 Keywords: Acetohydroxyacid synthase Bacillus anthracis Docking Reconstitution Triazol-1-yl-pyrimidines
a b s t r a c t Acetohydroxyacid synthase (AHAS), a potential target for antimicrobial agents, catalyzes the first common step in the biosynthesis of the branched-chain amino acids. The genes of both catalytic and regulatory subunits of AHAS from Bacillus anthracis (Bantx), a causative agent of anthrax, were cloned, overexpressed in Escherichia coli, and purified to homogeneity. To develop novel anti-anthracis drugs that inhibit AHAS, a chemical library was screened, and four chemicals, AVS2087, AVS2093, AVS2387, and AVS2236, were identified as potent inhibitors of catalytic subunit with IC50 values of 1.0 ± 0.02, 1.0 ± 0.04, 2.1 ± 0.12, and 2.0 ± 0.08 µM, respectively. Further, these four chemicals also showed strong inhibition against reconstituted AHAS with IC50 values of 0.05 ± 0.002, 0.153 ± 0.004, 1.30 ± 0.10, and 1.29 ± 0.40 µM, respectively. The basic scaffold of the AVS group consists of 1-pyrimidine-2-yl-1H-[1,2,4]triazole-3-sulfonamide. The potent inhibitor, AVS2093 showed the lowest binding energy, − 8.52 kcal/mol and formed a single hydrogen bond with a distance of 1.973 . As the need for novel antibiotic classes to combat bacterial drug resistance increases, the screening of new compounds that act against Bantx-AHAS shows that AHAS is a good target for new anti-anthracis drugs. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The branched-chain amino acids (BCAAs) are synthesized by plants, algae, fungi, bacteria, and archaeans but not by animals. Therefore, the enzymes of the BCAA biosynthetic pathway are potential targets for the development of herbicides, fungicides, and antimicrobial compounds. Some of the most popular herbicides used worldwide over the past 20 years act by inhibiting the first common enzyme in the BCAA biosynthetic pathway, acetohydroxyacid synthase (AHAS, EC 2.2.1.6). AHAS is a ThDP and FAD-dependent enzyme and is capable of catalyzing the synthesis of (2S)-acetolactate (AL) from two molecules of pyruvate or (2S)-2-aceto-2-hydroxybutyrate (AHB) from pyruvate and 2-ketobutyrate [1]. Bacterial AHASs are composed of large catalytic (∼ 60 kDa) and small regulatory (9– 18 kDa) subunits. Isolated catalytic subunits are similar to holoenzymes in their cofactor dependence and specificity toward substrates
Abbreviations: Bantx, B. anthracis; HE, reconstituted AHAS; BCAAs, branched-chain amino acids; CSU, catalytic subunit; RSU, regulatory subunit; SUs, sulfonylureas; IMs, imidazolinones; TPs, triazolopyrimidines ⁎ Corresponding author. Tel.: +82 2 2220 0946; fax: +82 2 2299 0763. E-mail address:
[email protected] (M.-Y. Yoon). 1 Present address: National Center for Biomedical Engineering Sciences, National University of Ireland, Galway, Ireland. 1570-9639/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2010.02.002
[2]. Reconstitution of AHAS from purified subunits has previously been demonstrated not to be significantly different from the native protein [3–5]. Bantx is a gram-positive bacterium, and the causative agent of anthrax. The Centers for Disease Control and Prevention categorize Bantx as a critical high-priority category A biothreat agent, based on its potential for causing mass casualties in the event of a bioterror attack [6]. Serious limitations in approved therapeutics, as well as the existence of resistance, underscore a compelling need for new therapeutic agents against this agent. However, previous studies of leucine and isoleucine-valine-leucine auxotrophs of Mycobacterium tuberculosisH37Rv [7], leucine auxotrophs of Mycobacterium bovis BCG [8], and the AHAS mutant of Burkholderia pseudomallei [9] have shown that the AHAS of pathogenic microorganisms could be a target for antimicrobial agents. Further interest in targeting AHAS has been raised since several structurally unrelated classes of herbicides, including the sulfonylureas (SUs), imidazolinones (IMs), triazolopyrimidines (TPs), and pyrimidyl-oxybenzoates, have been shown to specifically inhibit the enzyme [10–12]. A hit compound that inhibits the growth of various strains of M. tuberculosis, including MDR strains, has been identified by screening a chemical library against recombinant M. tuberculosis AHAS [13]. Recently, we showed the inhibitory activities of several herbicides (SUs and IMs) against the purified catalytic subunit (CSU) of Bantx-AHAS and evaluated the binding
1370
V. Gedi et al. / Biochimica et Biophysica Acta 1804 (2010) 1369–1375
modes of the potent inhibitors (CE) by molecular docking [11]. Here, we screened the substituted triazol-1-yl-pyrimidines analogues to measure inhibition potency and to discover new inhibitors of AHAS. In this study, we have individually cloned, purified, and reconstituted the CSU and regulatory subunit (RSU) of Bantx-AHAS. We describe the high-throughput screening (HTS) of a novel chemical library for in vitro inhibition of Bantx-AHAS CSU and inhibition potency against the reconstituted AHAS (HE). In addition, the binding modes and in vivo inhibition of Bantx-AHAS activity of the selected compounds were analyzed. 2. Materials and methods 2.1. Materials Brain heart infusion (BHI) broth was purchased from Becton Dickinson (Sparks, NV, USA). Sodium pyruvate, FAD, ThDP, MgCl2, IPTG (isopropylβ-D-thiogalactopyranoside), creatine, β-ME (β-mercaptoethanol), and αnapthol were obtained from Sigma Aldrich (St. Louis, MO, USA). All other chemicals were obtained from commercial sources and were of the highest purity grade available. KSW chemicals were obtained from the Korean Institute of Science and Technology (KIST, Seoul, Korea). AVS series compounds were obtained from the Korean Chemical Bank (KRICT, Daejeon, Korea). 2.2. Cloning and construction of expression plasmids The DNA fragments encoding the open reading frames of the catalytic (ilvB: TIGR locus NT05BA1436) and the RSU (ilvN: TIGR locus NT05BA1437) subunits of Bantx-AHAS were amplified by PCR, using genomic DNA as a template. The cloning and construction of the expression plasmid of CSU (pET-ilvB) were carried out as previously described [14]. The RSU coding region was amplified by PCR, using primers that introduced an EcoRI site immediately upstream of the initial methionine codon and a HindIII site immediately downstream of the RSU gene. The EcoRI–HindIII fragment of the PCR product was inserted into the IPTG-inducible plasmid pET28, yielding pET-ilvN, which produces a fusion protein with an N-terminal hexahistidine tag. 2.3. Expression and purification of AHAS subunits The expression plasmids pET-ilvB and pET-ilvN, encoding the BantxAHAS catalytic and RSU genes, were transformed into Escherichia coli BL21 (DE3). The induction and purification of the CSU were carried out as described previously [14]. The purification of the RSU was carried out similarly to the CSU, with a few modifications. The frozen cell pellet from the 0.5-L expression culture was resuspended in 25 ml of pH 8.0 buffer containing 20 mM sodium phosphate, 0.5 M NaCl, 10 mM imidazole, 1 mM PMSF, 0.5 mg/ml lysozyme, and 0.5% Triton X-100 and incubated at room temperature for 30 min, followed by ultrasonication. Cell debris was removed by centrifugation at 14,000 rpm for 30 min, and the clear supernatant was loaded on a Ni2+-charged Hi-Trap chelating column which was pre-equilibrated with a pH 8.0 20 mM sodium phosphate, 0.5 M NaCl, and 10 mM imidazole solution. The column was washed with 100 ml of pH 8.0 washing buffer containing 20 mM sodium phosphate, 0.5 M NaCl, and 50 mM imidazole. The bound RSU was eluted by running a linear gradient of imidazole (50–500 mM). Protein fractions were concentrated and desalted against a pH 8.0 solution of 20 mM Tris–HCl, 5 mM EDTA, and 5 mM β-mercaptoethanol using a HiTrap desalting column.
37 °C in a 100 mM potassium phosphate buffer (pH 8.0), containing 10 mM MgCl2, 1 mM ThDP, and 50 μM FAD. The reaction was initiated by the addition of 100 mM pyruvate, and the AHAS activity was assayed by colorimetric assay, as described previously [11]. 2.5. Log P and screening of Bantx-AHAS inhibitors The ACD/log P program was used to calculate the log P (octanol/ water partition coefficient) value of the chemical library. The ACD/log P algorithm is based on the log P contributions of separate atoms, structural fragments, and intramolecular interactions between fragments [15]. To identify novel inhibitors of Bantx-AHAS, a HTS of a chemical library was performed against CSU, as described previously [13]. For the inhibition study, all chemicals were dissolved in dimethyl sulfoxide (DMSO). The inhibition activity of selected compounds was further analyzed against the HE activity. Unless otherwise noted, all HE experiments were performed with a 15-min incubation of 8.6 pM of CSU and 600 pM of RSU, in standard assay buffer. All kinetic studies were performed using a discontinuous colorimetric assay, with the inhibitors in various concentrations, as appropriate for the particular experiment. 2.6. In vivo inhibition of Bantx-AHAS The in vivo inhibition effect of selected AVS group compounds was observed in a glucose–glutamate–glycine–salt (GGGS) synthetic medium [16]. A 1% inoculum of E. coli BL21 (carrying the BantxAHAS gene) stationary growth phase cells were added to the 4 ml of GGGS medium, which had been preincubated with varied concentrations of the AVS compounds, and incubated at 37 °C for 12 h. The control sample was made by using only DMSO for pre-incubation. Growth of the microorganisms was determined at OD600 over regular time intervals. After 12 h of incubation, the cells were collected by centrifugation, and the cell lysis and AHAS activity were determined, as reported previously [14]. 2.7. Preparation of the macromolecules As in the previous report [11], the crystal structure of Arabidopsis thaliana AHAS in complex with CE (PDB code: 1YBH) was used as a template to model the Bantx-AHAS. The modeling was carried out using Modeller software. The cofactors were included in the initial model, and the model was improved by energy minimization. The modeled structure was validated using PROSA and analyzed using Ramachandran plot techniques. 2.8. Molecular docking AutoDock 4.0 [17] combined with the Lamarckian genetic algorithm (LGA) was used for the docking study. Appropriate charges and hydrogens were added to the modeled protein. The ligands were modeled by making all the torsional bonds free in the structure. The docking area was calculated with a three-dimensional grid box with grid points of 100 × 124 × 100 Å, with a line spacing of 0.375 Å. Each docking experiment consisted of 50 runs. Based on the ligand torsion angles, calculations were done and grouped on the basis of root mean square deviation (RMSD) over 50 runs. The docking results were analyzed based on the binding energy and hydrogen bond distance. 3. Results
2.4. Reconstitution and enzyme assay
3.1. Expression and purification of AHAS
The initial reconstitution experiments were conducted at a fixed concentration of CSU (8.6 pM) and varied concentrations of RSU (0– 900 pM). Purified CSU and RSU solutions were incubated for 15 min at
The cloning of PCR-amplified genes of Bantx-AHAS CSU and RSU into the pET28a (+) vector yielded the expression plasmids pET-ilvB and pET-ilvN, respectively. Both expression plasmids were transformed into
V. Gedi et al. / Biochimica et Biophysica Acta 1804 (2010) 1369–1375
1371
E. coli BL21 (DE3) cells and induced with 0.4 M IPTG. The induced proteins were purified by Ni+ affinity chromatography, and the molecular weights were estimated to be 62 kDa and 19 kDa, near the expected size of CSU and RSU, respectively (Fig. 1). However, the purified RSU was found to be aggregated at concentrations higher than 0.3 mg/ml and during dialysis. A Hi-trap desalting column was used to prevent aggregation, and the protein concentrations were kept at less than 0.3 mg/ml. 3.2. Reconstitution The specific activity of the CSU was calculated to be 1.07 U/mg. Although the CSU alone exhibited AHAS activity, reconstitution with the RSU yielded a 3.5-fold increase in specific activity. Activity as a function of RSU concentration is shown in Fig. 2. The substrate saturation kinetics of HE were determined, and the Km value was estimated to be 6.1 mM. However, the pyruvate saturation curves of Bantx-AHAS HE do not follow Michaelis–Menten kinetics; they demonstrate positive cooperativity toward the substrate, with a Hill coefficient of 1.47 (Fig. 3). The HE was weakly inhibited by valine and was not inhibited by leucine and isoleucine (data not shown).
In this equation, V0 is the reaction rate without inhibitor, Vf is the rate at saturating inhibition, and [I] is the inhibitor concentration.
3.3. Log P and screening of AHAS inhibitors The ACD/log P program was used to calculate log P values of the compounds in the chemical library, then compounds with a predicted log P between 0.5 and 5.0 were selected (Table 2). Based on the HTS, five potent inhibitors of Bantx-AHAS CSU were identified by inhibiting ≥ 90% activity at 100 µM concentrations (Fig. 4A): AVS2387 (1-(4,6dimethoxypyrimidin-2-yl)-5-methylsulfanyl-1H-[1,2,4]triazole-3- sulfonicacid (6-chloro-3-fluoro-2-nitro-phenyl)-amide), AVS2093 (1(4,6-dimethoxypyrimidin-2-yl)-N-(2-isopropyl-6-nitrophenyl)- 5(methoxymethyl)-1H-1,2,4-triazole-3-sulfonamide), AVS2236 (1-(4chloro-6-dimethoxypyrimidin-2-yl)-5-methoxy-N-(2-methyl-6nitrophenyl)-1H-1,2,4-triazole-3-sulfonamide), AVS2087 (1-(4,6dimethoxypyrimidin-2-yl)-N-(2-isopropyl-6-nitrophenyl)-5-methyl-1H-1,2,4-triazole-3-sulfonamide), and KSW30191 (2-(2-chloro-phenoxy)-3-fluoro-[1,4] naphthoquinone). The inhibition potency of the selected compounds was further tested using HE. All four AVS chemicals were found to inhibit HE in a similar fashion, by ≥90% activity (Fig. 4B), whereas the KSW30191 was showed very weak inhibition, inhibiting ∼12% activity at 100 µM concentrations (data not shown). The 50% inhibition concentrations (IC50) for selected compounds were analyzed by fitting the data to Eq. (1). The values are shown in Table 1.
v=
ðV0 −Vf Þ·IC50 + Vf IC50 + ½I
Fig. 2. Reconstitution of AHAS HE by increasing concentrations of the RSU protein. Assays contained 8.6 pM of CSU and varying concentrations of the RSU, as shown. Specific activity is expressed as units per milligram of CSU protein.
3.4. Inhibition kinetics The inhibition mechanisms of the most potent inhibitors were selected for further study. Initial rates in the presence of inhibitors were measured as a function of pyruvate concentration at fixed inhibitor concentrations (data not shown). The Kii and Kis values were determined by fitting the data to Eq. (2) or (3), for noncompetitive and uncompetitive inhibition, respectively. v = Vmax × ½S = fKm ð1 + ½I = Kis + ½Sð1 + ½I = Kii Þg
ð2Þ
v = Vmax × ½S = fKm + ½Sð1 + ½I = Kii Þg
ð3Þ
In these equations, [S] and [I] are the concentrations of substrate (pyruvate) and inhibitor. Kis is the inhibition constant derived from the slope, whereas Kii is the constant derived from the intercepts of 1/ V vs 1/S plots. The initial kinetics indicates the AVS compounds were
ð1Þ
Fig. 1. 15% SDS–PAGE analysis of the purification stages of B. anthracis-AHAS RSU and purified CSU. Lane MK, Sigma wide range marker; lane 1, uninduced BL21 (DE3); lane 2, induced BL21 (DE3); lane 3, soluble fraction; lane 4, purified RSU (19 kDa, indicated by arrow); lane 5, purified CSU (62 kDa, indicated by arrow).
Fig. 3. Substrate saturation curve of reconstituted B. anthracis HE. The reaction mixtures containing 8.6 pM CSU and 600 pM of RSU in 100 mM potassium phosphate buffer (pH 7.4) containing 10 mM MgCl2, 1 mM ThDP, and 50 μM FAD were preincubated at 37 °C for 15 min. The reaction was initiated by the addition of various concentrations of pyruvate, as shown.
1372
V. Gedi et al. / Biochimica et Biophysica Acta 1804 (2010) 1369–1375
Fig. 5. The stereochemical spatial arrangement of amino acid residues in the modeled 3D structure of Bantx-AHAS in favored region of the Ramachandran plot. The disallowed residues were labeled.
3.5. In vivo inhibition of Bantx-AHAS
Fig. 4. The relative activity of the Bantx-AHAS CSU (A) and HE (B) as a function of inhibitor concentration. The IC50 values were mentioned at Table 1.
noncompetitive inhibitors of both Bantx-AHAS CSU and HE (data not shown). At any inhibitor concentration, when noncompetitive inhibition is occurring, a portion of the enzyme will remain nonproductive in the enzyme–substrate–inhibitor complex form, as we observed the Vmax in the presence of inhibitor is less than the Vmax in the absence of inhibitor. In contrast to the AVS chemicals, KSW30191 was found to be an uncompetitive inhibitor of BantxAHAS CSU (data not shown). However, the KSW30191 inhibition kinetics were not present against HE; instead, there was very weak inhibition. The resulting Kii values indicate the inhibition potency of AVS and KSW chemicals toward Bantx-AHAS (Table 1).
In an attempt to determine the in vivo inhibition effect of the AVS group chemicals, the inhibition of E. coli BL21 (Bantx-AHAS) growth was measured at varied concentrations of the inhibitors. When the cells were incubated for 12 h (stationary growth phase) at a 1.5 mM concentration, AVS2093 and AVS2387 showed ≥85% growth inhibition, whereas AVS2236 and AVS2087 showed ≥ 21% and 52% inhibition, respectively (data not shown). Next, we determined the AHAS activity of the cells grown in the presence of each AVS chemical in the GGGS medium. Cells grown in the presence of 1.5 mM of AVS2093 and AVS2387 for 12 h showed ≥96% inhibition of AHAS activity, as compared to the control cells (E. coli BL21 cells containing Bantx-AHAS and grown without inhibitor), whereas the cells grown in the presence of 1.5 mM of AVS2236 and AVS2087 showed ∼45% and 60% inhibition of AHAS activity, respectively (data not shown). As another control, the AHAS activity was determined in E. coli BL21 cells without the plasmids carrying the Bantx-AHAS. These cells showed 4.7-fold lower AHAS activity than the cells carrying Bantx-AHAS plasmid. AVS2093, AVS2387, AVS2236, and AVS2087 showed 35%, 39%, 22%, and 25% inhibition of E. coli BL21 AHAS activity, respectively (data not shown). The amount of AHAS activity inhibited in E. coli BL21 cells carrying Bantx-AHAS plasmid is significantly higher (P b 0.001) as compared to inhibition in E. coli BL21 cells without Bantx-AHAS plasmid.
Table 1 Inhibition kinetics of potent chemicals against Bantx-AHAS CSU and HE. Inhibitor
AVS 2087 AVS 2093 AVS 2387 AVS 2236 KSW 30191 a
Not determined.
IC50 (µM)
Kii (µM)
Kis (µM)
CSU
HE
CSU
HE
CSU
HE
1.0 ± 0.02 1.0 ± 0.04 2.1 ± 0.12 2.0 ± 0.08 4.0 ± 0.25
0.05 ± 0.002 0.153 ± 0.004 1.30 ± 0.10 1.29 ± 0.40 NDa
0.24 ± 0.034 1.9 ± 0.5 4.0 ± 1.54 26.2 ± 4.6 3.8 ± 0.49
0.03 ± 0.41 1.03 ± 0.13 2.91 ± 0.09 11.09 ± 0.24 NDa
0.14 ± 0.028 0.88 ± 0.07 1.19 ± 0.20 0.84 ± 0.12 —
0.01 ± 0.006 0.34 ± 0.64 0.61 ± 0.19 0.49 ± 0.32 NDa
V. Gedi et al. / Biochimica et Biophysica Acta 1804 (2010) 1369–1375
3.6. Homology modeling and docking analysis The CSU of Bantx-AHAS has about 44% amino acid sequence identity and 60% sequence similarity to A. thaliana-AHAS (data not shown). The model of Bantx-AHAS was obtained based on the X-ray crystallography coordinates of A. thaliana-AHAS (1YBH). The model was deemed reliable as ∼ 99.6% of the residues (glycine and proline were ignored) were located in the allowed regions of the Ramachandran plot (Fig. 5). After the addition of appropriate charges and hydrogens, the docking was carried out with a proper grid. The lowest-energy conformer was chosen for further analysis from among the 50 runs, and a careful analysis of hydrogen bonds was performed. The most potent inhibitor of CSU, AVS2093, showed the strongest binding energy (− 8.52), of all the compounds, and the other three ligands had much lower binding energies (Table 2).
1373
which is a structurally similar ligand with an additional functional group, methoxymethane, forms a hydrogen bond with the backbone amino group HN of ARG158. This additional functional group caused the hydrogen bond to have a length of 1.973 Å (Fig. 6B). AVS2387 has different functional groups from the above two ligands, including F, Cl, and SCH3. These changes made the binding modes likely to appear at different sites than those of the other two ligands. A single hydrogen bond was found between the HN (main chain amino group) of MET260 and the O atom of 4-methoxy-pyrimidine, with a distance of 2.194 Å (Fig. 6D). Another ligand, AVS2236, was found to form two hydrogen bonds. An O atom of NO2 in AVS2236 interacts with GLU320 and VAL303, forming hydrogen bonds with lengths of 1.632 and 1.842 Å, respectively (Fig. 6C). These interactions establish the stable binding mode and orientation of the ligand and protein complex. 4. Discussion
3.7. Binding modes of the ligands As showed in Fig. 6A, two hydrogen bonds were found; between AVS2087 and LYS132, and AVS2087 and ASP309, with distances of 2.059 Å and 2.008 Å, respectively. The Oδ2 atom of ASP309 is hydrogen bonded with the H10 atom of SO2, whereas a hydrogen bond is formed between the O (in the carbonyl of the main chain) atom of LYS132 and the O5 of AVS2087 where H14 is shared. AVS2093,
To characterize the new inhibitors of Bantx-AHAS, the individual subunits of Bantx-AHAS were cloned, expressed in E. coli, and purified to homogeneity. The purified individual subunits were reconstituted to form a HE, which had a specific activity higher than the purified CSU. As in other AHASs, the Bantx-AHAS HE exhibited positive cooperativity toward the substrate, pyruvate [2,13]. The biochemical characteristics, as well as the substrate and cofactor requirements of
Table 2 Calculated binding energies and hydrogen bonding interactions for AVS chemicals in Bantx-AHAS CSU. Ligand name
Log P
Binding energy (kcal/mol)
Hydrogen bonding and length (Å)
AVS2087
2.21 ± 1.24
Structure
− 7.6
H10- OD2:ASP309 (2.006) H14- O:LYS132 (2.059)
AVS2093
2.06 ± 1.26
− 8.52
O7- HN:ARG158 (1.973)
AVS2236
2.64 ± 1.30
− 7.75
H12- OE1:GLU320 (1.632) O6- HN:VAL303 (1.842)
AVS2387
2.82 ± 1.03
− 7.91
O4- HN:MET260 (2.194)
1374
V. Gedi et al. / Biochimica et Biophysica Acta 1804 (2010) 1369–1375
Fig. 6. Binding modes of AVS chemicals in B. anthracis-AHAS CSU. (A) AVS2087; (B) AVS2093; (C) AVS2236; (D) AVS2387.
CSU, have been described previously [14]. Feedback inhibition of the Bantx-AHAS HE by valine was weak, with a maximum inhibition of 30% (data not shown), which is similar to a few other bacterial AHASs [18–20]. SUs, IMs and TPs have been the most widely used herbicidal AHAS inhibitors; however, their structures were not chosen on the basis of their interactions with binding sites in bacterial AHAS [2]. Among the three herbicide groups, TPs were found to be the weakest inhibitors of Bantx-AHAS, whereas SUs and IMs showed potent inhibition [11]. In the past, few modifications of TPs were used to successfully increase the potency of inhibition [21,22]. In this study, a HTS of substituted triazol-1-yl-pyrimidine derivatives revealed four compounds (AVS2387, AVS2093, AVS2236, and AVS2087) that acted as potential inhibitors of Bantx-AHAS CSU. The basic scaffold of AVS chemicals are similar to triazolopyrimidines, consisting of a di- or trisubstituted aromatic ring, linked by a short bridge to a substituted 2[1,2,4]-triazol-1-yl-pyrimidine (Table 2). Subsequently, related compounds with additional substituent groups in the scaffold, such as electron withdrawing groups (NO2, F, Cl, or CH3) on benzene, CH3, OCH3, or SCH3 groups on triazole, and OCH3 or Cl on pyrimidine, were found to be 150-fold more potent inhibitors of B. anthracis-AHAS than the TP herbicides. Moreover, the inhibition effects of AVS2093 and AVS2087 against HE were demonstrated to be greater than the CSU, with approximately 6- and 20-fold increases in the IC50 values: 0.15
and 0.05 µM, respectively. However, the AVS2236 and AVS2387 showed similar IC50 values against either HE or CSU (Table 1). All the AVS compounds noncompetitively inhibited both Bantx-AHAS CSU and HE. The IC50 and inhibition kinetics strongly suggest that the AVS compounds are the most potent inhibitors of Bantx-AHAS of any herbicides reported to date. Along with the AVS compounds, KSW30191, a known AHAS inhibitor [23], inhibited Bantx-AHAS CSU activity with an IC50 of 4.0 μM. The main scaffold of KSW30191 is 2-phenoxy-[1,4]naphthoquinone, a quinone moiety [23]. However, this chemical showed very weak inhibition against HE, suggesting that conformational changes during reconstitution may affect the inhibition ability. The polarity of a substance is an important factor in determining whether a given pharmaceutical compound can effectively reach its site of action. In this study, the log P values were better for the four AVS chemicals, as compared to other tested chemicals (Table 2). This finding is in agreement with the findings of Lipinski et al. [24], that most commercial herbicides have a log P value ≤ 5.0. Molecular docking experiments were carried out to understand the molecular mechanisms of compounds in contact with AHAS molecules. The substituted triazol-1-yl-pyrimidine derivatives had significantly different binding site arrangements than those of the other herbicides. As in previous reports [22], the substitution of a methoxymethane group had a significant effect on inhibition potency and on the
V. Gedi et al. / Biochimica et Biophysica Acta 1804 (2010) 1369–1375
structure of the binding region in triazol-1-yl-pyrimidine derivatives. The binding modes’ elucidation for the most active inhibitors includes several features that can be exploited in further development of this lead series. In summary, based on HTS results, a new series of substituted triazol-1-yl-pyrimidine derivatives were identified and characterized as Bantx-AHAS CSU inhibitors, and their inhibitory activities were also evaluated against the HE. The results indicate that substituted triazol1-yl-pyrimidine derivatives are potent AHAS inhibitors, and the molecular docking simulations of four AVS compounds with AHAS CSU revealed the details of the interactions. We conclude that a triazol-1-yl-pyrimidine skeleton is a promising structural template for the development of novel AHAS inhibitors and that molecular modeling study results may shed further light on this in a future study. Acknowledgements
[10]
[11]
[12]
[13]
[14]
[15]
[16]
This work was supported by the Korea Research Foundation grant funded by the Korea government (KRF-2008-313-C00530).
[17]
References
[18]
[1] H.E. Umbarger, B. Brown, Isoleucine and valine metabolism in Escherichia coli: VIII. The formation of acetolactate, J. Biol. Chem. 233 (1958) 1156–1160. [2] R.G. Duggleby, S.S. Pang, Acetohydroxyacid synthase, J. Biochem. Mol. Biol. 33 (2000) 1–36. [3] V. Vinogradov, M. Vyazmensky, S. Engel, I. Belenky, A. Kaplun, O. Kryukov, Z. Barak, D.M. Chipman, Acetohydroxyacid synthase isozyme I from Escherichia coli has unique catalytic and regulatory properties, Biochim. Biophys. Acta 1760 (2006) 356–363. [4] M. Vyazmensky, C. Sella, Z. Barak, D.M. Chipman, Isolation and characterization of subunits of acetohydroxy acid synthase isozyme III and reconstitution of the holoenzyme, Biochemistry 35 (1996) 10339–10346. [5] C.M. Hill, S.S. Pang, R.G. Duggleby, Purification of Escherichia coli acetohydroxyacid synthase isoenzyme II and reconstitution of active enzyme from its individual pure subunits, Biochem. J. 327 (1997) 891–898. [6] R.M. Atlas, Bioterrorism: from threat to reality, Annu. Rev. Microbiol. 56 (2002) 167–185. [7] I. Guleria, R. Teitelbaum, R.A. McAdam, G. Kalpana, W.R. Jacobs Jr., B.R. Bloom, Auxotropic vaccines for tuberculosis, Nat. Med. 2 (1996) 334–337. [8] F.C. Bange, A.M. Brown, W.R. Jacobs Jr., Leucine auxotrophy restricts growth of Mycobacterium bovis BCG in macrophages, Infect. Immun. 64 (1996) 1794–1799. [9] T. Atkins, R.G. Prior, K. Mack, P. Russell, M. Nelson, P.C.F. Oyston, G. Dougan, R.W. Titball, A mutant of Burkholderia pseudomallei, auxotrophic in the branched chain
[19]
[20]
[21]
[22]
[23]
[24]
1375
amino acid biosynthetic pathway, is attenuated and protective in a murine model of melioidosis, Infect. Immun. 70 (2002) 5290–5294. J.A. McCourt, S.S. Pang, L.W. Guddat, R.G. Duggleby, Elucidating the specificity of binding of sulfonylurea herbicides to acetohydroxyacid synthase, Biochemistry 44 (2005) 2330–2338. S. Kalme, C.N. Pham, V. Gedi, D.T. Le, J.D. Choi, S.K. Kim, M.Y. Yoon, Inhibitors of Bacillus anthracis acetohydroxyacid synthase, Enzym. Microb. Tech. 43 (2008) 270–275. J.A. McCourt, R.G. Duggleby, Acetohydroxyacid synthase and its role in the biosynthetic pathway for branched-chain amino acids, Amino Acids 31 (2006) 173–210. K.J. Choi, Y.G. Yu, H.G. Hahn, J.D. Choi, M.Y. Yoon, Characterization of acetohydroxyacid synthase from Mycobacterium tuberculosis and the identification of its new inhibitor from the screening of a chemical library, FEBS Lett. 579 (2005) 4903–4910. K.J. Choi, C.N. Pham, H. Jung, S.H. Han, J.D. Choi, J. Kim, M.Y. Yoon, Expression of acetohydroxyacid synthase from Bacillus anthracis and its potent inhibitors, Bull. Kor. Chem. Soc. 28 (2007) 1109–1113. A. Williams, E. Kolovanov, A rebuttal regarding recent comparisons of SciLogP ultra (Scivision Inc.) with ACD/log P prediction software, Procedings of the log P 2000 symposium, 2000, pp. 1–23. F. Buono, R. Testa, D.G. Lundgren, Physiology of growth and sporulation in Bacillus cereus: I. Effect of glutamic and other amino acids, J. Bacteriol. 91 (1966) 2291–2299. G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J. Olson, Automated docking using a Lamarckian Genetic Algorithm and an empirical binding free energy function, J. Comput. Chem. 19 (1998) 1639–1662. Y. Zohar, M. Einav, D.M. Chipman, Z. Barak, Acetohydroxyacid synthase from Mycobacterium avium and its inhibition by sulfonylureas and imidazolinones, Biochim. Biophys. Acta 1649 (2003) 97–105. I. Porat, M. Vinogradov, M. Vyazmensky, C.D. Lu, D.M. Chipman, A.T. Abdelal, Z. Barak, Cloning and characterization of acetohydroxyacid synthase from Bacillus stearothermophilus, J. Bacteriol. 186 (2004) 570–574. M. Vyazmensky, S. Engel, O. Kryukov, D. Berkovich-Berger, L. Kaplun, Construction of an active acetohydroxyacid synthase I with a flexible linker connecting the catalytic and the regulatory subunits, Biochim. Biophys. Acta 1764 (2006) 955–960. S.K. Namgoong, H.J. Lee, Y.S. Kim, J.H. Shin, J.K. Che, D.Y. Jang, G.S. Kim, J.W. Yoo, M.K. Kang, M.W. Kil, J.D. Choi, S.I. Chang, Synthesis of the quinoline-linked triazolopyrimidine analogues and their interactions with the recombinant tobacco acetolactate synthase, Biochem. Biophys. Res. Commun. 258 (1999) 797–801. C.N. Chen, L.L. Lv, F.Q. Ji, Q. Chen, H. Xu, C.W. Niu, Z. Xi, G.F. Yang, Design and synthesis of N-2, 6-difluorophenyl-5-methoxyl-1, 2, 4-triazolo[1, 5-a]- pyrimidine-2-sulfonamide as acetohydroxyacid synthase inhibitor, Bioorg. Med. Chem. 17 (2009) 3011–3017. K.J. Choi, K.M. Noh, D.E. Kim, B.H. Ha, E.E. Kim, M.Y. Yoon, Identification of the catalytic subunit of acetohydroxyacid synthase in Haemophilus influenzae and its potent inhibitors, Arch. Biochem. Biophys. 466 (2007) 24–30. C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings, Adv. Drug. Deliv. Rev. 46 (2001) 3–26.