Potent inhibition of bacterial DNA gyrase by digallic acid and other gallate derivatives

Abstract

Bacterial DNA gyrase, an essential enzyme, is a validated target for discovering and developing new antibiotics. Here we screened a pool of polyphenols and discovered that digallic acid is a potent DNA gyrase inhibitor. We also found that several food additives based on gallate, such as dodecyl gallate, potently inhibit bacterial DNA gyrase. Interestingly, the IC₅₀ of these gallate derivatives against DNA gyrase is correlated with the length of hydrocarbon chain connecting to the gallate. These new bacterial DNA gyrase inhibitors are the ATP competitive inhibitors of DNA gyrase. Our results also show that digallic acid and certain gallate derivatives potently inhibit E. coli DNA topoisomerase IV. Several gallate derivatives have strong antimicrobial activities against Staphylococcus aureus and methicillin-resistant Staphylococcus aureus (MRSA). This study provides a solid foundation for the design and synthesis of gallate-based DNA gyrase inhibitors that may be used to combat antibacterial resistance.

Graphical Abstract

Several simple derivatives of gallic acid, such as digallic acid and dodecyl gallate, are potent bacterial DNA gyrase inhibitors although gallic acid per se, a polyphenol found in many plants, does not inhibit DNA gyrase. Modeling studies show that a hydrophobic tail or a second gallic acid can fill a cavity of gyrase B subunit near the ATP binding site and block ATP binding.

Introduction

Prokaryotic DNA gyrase is a type IIA DNA topoisomerase essential to bacterial survival.^[1,2] Due to its importance, bacterial DNA gyrase is a validated and highly valuable target for discovering new antibiotics to treat bacterial infection.^[3–5] Fluoroquinolones (FQs), which target bacterial DNA gyrase, are among the most prescribed and successful antibiotics.^[6–8] The mechanism of antibacterial activities of FQs is to stabilize the gyrase-DNA cleavage-complex,^[9–11] which is ultimately responsible for cell death.^[12] Since certain bacteria have another type IIA DNA topoisomerase, topoisomerase IV,^[13] FQs also target topoisomerase IV and stabilize DNA–topoisomerase IV complexes which leads to generate DNA double breaks during DNA replication.^[14,15] This gyrase and topoisomerase IV dual poisoning mechanism makes FQs among the most effective antibiotics.^[12] Unfortunately, FQ resistance has emerged.^[16–20] New and more effective antibiotics are needed to treat FQ-resistant bacterial infections.^[12,21,22] Additionally, FQs have been explored extensively.^[12,23] It is likely that the limits and potential of FQs have been reached.^[12,23] Another concern for the use of FQs is that these compounds can cause serious side effects including tendonitis and tendon rupture, peripheral neuropathy, hyperglycemia, and aortic dissections and aortic aneurysm.^[24–28] As a result, US Food and Drug Administration (FDA) issued several warnings for the use of FQs and added black box warnings on all FQs.^[29] Therefore, there is a need to develop and identify new types of compounds targeting bacterial DNA gyrases for treating bacterial infections.

Natural products, such as polyphenols, are rich sources to identify antibiotics or antibacterial agents.^[30–32] For instance, ellagic acid, found in a large variety of foods including pecan, walnuts, cranberries, strawberries, and pomegranates,^[33] is a potent bacterial DNA gyrase inhibitor^[34] and has antibacterial activities.^[35,36] Antibiotic novobiocin, produced by Streptomyces niveus,^[37] is also a polyphenol^[38] and has strong antibacterial activities especially against Gram-positive bacteria.^[39] It potently inhibits bacterial DNA gyrase.^[39] Mechanistically, novobiocin is a competitive inhibitor of the ATPase of bacterial DNA gyrase with the inhibition dissociation constant K_i of ~10 nM.^[40,41] Structural studies showed that novobiocin competes with ATP for the ATP-binding site of gyrB.^[42] However, due to its toxicity, novobiocin has been withdrawn from clinical use.^[39] Other polyphenols, such as epigallocatechin gallate (EGCG),^[43] haloemodin,^[44] and quercetin^[45] also have anti-bacterial DNA gyrase activity. Nevertheless, chemically, these polyphenols are not easy to be modified or derivatized, which is required for improving their potency. This prompted us to identify polyphenol-based gyrase inhibitors that can be easily modified/derivatized.

In this paper, we report the discovery of a new potent bacterial DNA gyrase inhibitor, digallic acid after we screened a pool of naturally-occurring polyphenols. We also find that several food additives based on gallic acid or gallate potently inhibit bacterial DNA gyrase. These gallate derivative inhibit bacterial DNA topoisomerase IV as well. Two derivatives octyl gallate and dodecyl gallate have strong anti-bacterial activities against Staphylococcus aureus and methicillin-resistant Staphylococcus aureus (MRSA). These simple gallate-based gyrase inhibitors can be easily modified/derivatized and should have potential to be developed into anti-bacterial agents.

Results and Discussion

Since several naturally occurring polyphenols, such as ellagic acid and EGCG, are potent bacterial DNA gyrase inhibitors,^[34,43] we decided to examine the inhibition activities of a group of polyphenols collected in our laboratory against E. coli DNA gyrase (Figure S1). As expected, ellagic acid, EGCG, and 3,5-dicaffeoylquinic acid strongly inhibit E. coli DNA gyrase (Figures 1, S1, and S2). We found that tannic acid also potently inhibits E. coli DNA gyrase with an IC₅₀ of 1 μM (Figure 1e). Since tannic acid is composed of several digallic acids covalently attached to a glucose (Figure 1a), we reasoned that digallic acid, found in Pistacia lentiscus and other plants and fruits,^[46] should also strongly inhibit E. coli DNA gyrase. Indeed, our results showed that digallic acid potently inhibit E. coli DNA gyrase with an estimated IC₅₀ of 2 μM (Figure 1f). In contrast, gallic acid does not inhibit E. coli DNA gyrase up to 500 μM (Figure 1g, Figure S1, and Table 1).

Figure 1.

(a-d) Chemical structure of (a) tannic acid, (b) digallic acid, (c) gallic acid and (d) egallic acid. (e-h) Inhibition of E. coli DNA gyrase supercoiling activities by tannic acid, digallic acid, gallic acid, and ellagic acid. Novobiocin was used as a positive control (Figure S3). Symbols: Rx, relaxed plasmid pAB1; Sc, supercoiled pAB1.

Table 1.

The IC50 values (M) of gallate derivatives against E. coli DNA gyrase and topoisomerase IV

	Gyrase		Topoisomerase IV
Compound	Gel-based	SDFQ	Gel-based	SDFQ
Gallic acid	>500	N/Aa	>500	N/Aa
Digallic acid	2	1.9	8	7.3
Butyl Gallate	>100	>100	>100	>100
Octyl Gallate	50	25.96	50	41.2
Dodecyl Gallate	15	13.77	50	36.9
Phenyl Gallate	>100	>100	>100	>100
Biphenyl Gallate	20	18.8	25	23.6
Novobiocin	0.5	0.45	10	4.3

^aNot available

Next, we obtained several compounds with a gallic acid or gallate attached to a hydrophobic moiety through an ester bond (Figure 2a–e) and tested their inhibition activities against E. coli DNA gyrase. Our results are shown in Figure 2f and g. All these gallate derivatives inhibit E. coli DNA gyrase (Figure 2 and Table 1). Interestingly, a long hydrophobic chain or group significantly increase the inhibition activities of these gallate derivatives against E. coli DNA gyrase. For example, the IC₅₀ value of dodecyl gallate is much lower than that of butyl gallate (Figure 2; and Table 1). Likewise, the IC₅₀ value of biphenyl gallate is also lower than that of phenyl gallate (Figure 2 and Table 1).

Figure 2.

(a-e) Chemical structure of gallate derivatives. (a) butyl gallate (BG), (b) octyl gallate (OG), (c) dodecyl gallate (DG), (d) phenyl gallate (PG), and (e) biphenyl gallate (BPG). (f and g) Inhibition of E. coli DNA gyrase by gallate derivatives. Symbols: Rx, relaxed plasmid pAB1; Sc, supercoiled pAB1. (f) Lanes 1–6 contain 0, 5, 10, 20, 50 and 100 μM of butyl gallate, respectively. Lanes 7–11 indicate 5, 10, 20, 50 and 100 μM of octyl gallate, respectively. Lanes 12–16 indicate 5, 10, 20, 50 and 100 μM of dodecyl gallate, respectively. (g) Lanes 1–6 contain 0, 5, 10, 20, 50 and 100 μM of phenyl gallate, respectively. Lanes 7–11 indicate 5, 10, 20, 50 and 100 μM of biphenyl gallate, respectively. Lanes 12 and 13 represent supercoiled and relaxed plasmid pAB1, respectively.

The Inhibition of E. coli DNA gyrase by these gallate derivatives was also confirmed using a supercoiling-dependent fluorescence quenching (SDFQ) assay (Figure 3a). As shown in Fig. 3a and our previous publications,^[41,47] supercoiling of the fluorescently-labeled plasmid pAB1_FL905 greatly decreases the fluorescence intensity of the plasmid. In the presence of a gyrase inhibitor, the fluorescence intensity of pAB1_FL905 should not change. The SDFQ assay is an excellent assay to quantitatively analyze the inhibition of DNA gyrase by inhibitors. Indeed, the IC₅₀ values obtained through the SDFQ assays are almost identical to those obtained by gel-based gyrase inhibiton assays (Table 1).

Figure 3.

(a and b) SDFQ-based DNA gyrase supercoiling assays. (a) An experimental strategy to study DNA gyrase inhibitors. The IC₅₀ values can be determined by a titration experiment. (b) Titration experiments to determine the IC₅₀ of digallic acid (red up triangles), octyl gallate (green down triangles), dodecyl gallate (blue squares), biphenyl gallate (black circles), and novobiocin (stars). (c and d) Digallic acid and other gallate derivatives are competitive inhibitors of the ATPase activities of E. coli DNA gyrase. The ATPase assays were performed as described in Material and Methods. (c) Lineweaver–Burk plot or double-reciprocal plot of E. coli DNA gyrase in the absence (open triangles) or presence (closed squares) of 500 nM digallic acid. (d) ATP hydrolysis activities of E. coli DNA gyrase in presence of different gallate derivatives (100 μM). Symbol: DNA gyrase only, open down triangles; digallic acid, red up triangles; octyl gallate, green down triangles; dodecyl gallate, blue squares; biphenyl gallate, black circles; and novobiocin (open circles).

Previous studies showed that ellagic acid (a condensed dimer of gallic acid) and epigallocatechin gallate (a gallate derivative) are competitive inhibitors of bacterial DNA gyrase’s ATPase.^[43] In this study, we also conducted the ATPase kinetic studies in the absence or presence of digallic acid or other gallate derivatives. Our results are shown in Figure 3c and d. As expected, digallic acid and other gallate derivatives greatly inhibit the ATPase activities of E. coli DNA gyrase (Figure 3c and d). Fitting of these kinetic results to the Michaelis–Menten equation yielded the following kinetic parameters: K_M of 0.59 ± 0.22 mM and V_max of 16.3 ± 2.5 nM/s in the absence of an inhibitor and K_M of 1.44 ± 0.73 mM and V_max of 17.8 ± 5.0 nM/s in the presence of 500 nM of digallic acid. These results demonstrate that digallic acid is a competitive inhibitor of bacterial DNA gyrase’s ATPase (Figure 3c): the intercept on the 1/V₀ axis of the Lineweaver–Burk or double-reciprocal plot is the same in the presence or absence of digallic acid showing that digallic acid competes with ATP for its binding sites of E. coli gyrase. The K_i was calculated to be 347 nM.

We also tested the inhibition activities of digallic acid and other gallate derivatives against E. coli DNA topoisomerase IV. Our results are shown in Figure 4 and Figure S3b for relaxation and also in Figure S4 for decatenation. Novobiocin is a known topoisomerase IV inhibitor with an IC₅₀ of ~10 μM, which is consistent with previously published results (PMID: 21693461).^[48] Digallic acid potently inhibits E. coli topoisomerase IV with an estimated IC₅₀ ~8 μM for relaxation (Figure 4a and d) and ~4 μM for decatenation (Figure S4a and c). Octyl gallate, dodecyl gallate, and biphenyl gallate also strongly inhibit E. coli DNA topoisomerase IV (Figure 4b, c and d; Figure S4b and c; Table 1). As expected, gallic acid does not inhibit E. coli DNA topoisomerase IV up to 500 μM (Figure S3c and Table 1). Interestingly, digallic acid and these gallate derivatives also inhibit the ATPase activities of bacterial DNA topoisomerase IV (Figure 4e). In contrast, at 100 μM, digallic acid and other gallate derivatives do not inhibit E. coli DNA topoisomerase I (Figure 5a), human DNA topoisomerase I (Figure 5b), and IIα (Figure 5c). These results suggest that these gallic acid derivatives only target bacterial DNA gyrase and topoisomerase IV. Additionally, our DNA cleavage assays showed that digallic acid and other gallate derivatives are not bacterial DNA gyrase poisons, specific DNA gyrase inhibitors that stabilize and stimulate gyrase-mediated double-stranded breaks and single-stranded nicks^[49] (Figure 5d).

Figure 4.

Inhibition of E. coli DNA topoisomerase IV by digallic acid and other gallate derivatives. Relaxation assays by E. coli DNA topoisomerase IV were performed as described under Material and Methods. The ATPase assays were performed as described in Material and Methods as well. (a) Digallic acid potently inhibits E. coli DNA topoisomerase IV. Lanes 2–8 contain 0, 6.25, 8, 10, 12.5, 25, and 50 μM of digallic acid, respectively. Lane 1 is supercoiled plasmid pAB1. IC₅₀ of digallic acid against E. coli DNA topoisomerase IV was estimated to be ~8 μM. (b) Inhibition of gallate derivatives against E. coli DNA topoisomerase IV at 100 μM. (c) Titration experiments to determine IC₅₀ of dodecyl gallate, octyl gallate, and biphenyl gallate against E. coli DNA topoisomerase IV. Symbols: Digallic acid, DA; butyl gallate, BG; octyl gallate, OG; dodecyl gallate, DG; phenyl gallate, PG; and biphenyl gallate, BPG; Ln, linear DNA, Nk, nicked DNA, Rx, relaxed DNA, Sc, supercoiled DNA. (d) SDFQ-based DNA relaxation assays to determined IC₅₀ of digallic acid (red up triangles), octyl gallate (green down triangles), dodecyl gallate (blue squares), biphenyl gallate (black circles), and novobiocin (stars). The supercoiled, fluorescently-labeled plasmid pAB1_FL905 was used. Fluorescence measurements were performed in a Biotek Synergy H1 Hybrid Plate Reader using a wavelength of 494 nm and 521 nm for excitation and emission, respectively. (e) ATP hydrolysis activities of E. coli DNA topoisomerase IV in presence of different gallate derivatives (100 μM). Symbol: DNA gyrase only, open down triangles; digallic acid, red up triangles; octyl gallate, green down triangles; dodecyl gallate, blue squares; biphenyl gallate, black circles; and novobiocin (open circles).

Figure 5.

Digallic acid and gallate derivatively did not strongly inhibit E. coli DNA topoisomerase I (a), human DNA topoisomerase I (b), and human DNA topoisomerase IIα (c). 100 μM of compounds were used. (d) DNA gyrase-mediated DNA cleavage assays were performed as described under Material and Methods. Lanes 2–6 contain 0, 6.25, 12.5, 50, 100, 150, and 200 μM of digallic acid, respectively. Lane 1 contains 10 μM of ciprofloxacin. Symbols: Digallic acid, DA; butyl gallate, BG; octyl gallate, OG; dodecyl gallate, DG; phenyl gallate, PG; and biphenyl gallate, BPG; Ln, linear DNA, Nk, nicked DNA, Rx, relaxed DNA, Sc, supercoiled DNA.

Figure 6 shows molecular models for digallic acid, dodecyl gallate, and biphenyl gallate binding to E. coli DNA gyrase B subunit. These models were generated by molecular docking with the crystal structure of gyrase B subunit (PDB ID: 1EI1), followed by 100 ns molecular dynamics simulations as described in Material and Methods. Our method is validated by docking ADPNP to E. coli DNA gyrase subunit B followed by molecular dynamic simulation (Figure S5), which resulted in a very similar complex to the experimentally obtained ADPNP-E coli DNA gyrase complex (PDB ID: 1EI1). Similar to novobiocin, digallic acid, dodecyl gallate, and biphenyl gallate also bind to the ATP binding site of the gyrase B subunit.^[42,50] Digallic acid, dodecyl gallate or biphenyl gallate form a hydrogen bond with Asp73 (Figure 6). Previous studies have shown that hydrophobic interactions significantly contribute to the binding of novobiocin to the protein.^[42,50] We also find that the hydrophobic groups of gallate derivatives are in contact with the hydrophobic amino acids, which may greatly enhance the inhibition against E. coli DNA gyrase.

Figure 6.

Molecular models of digallic acid (a-c), dodecyl gallate (d-f), and biphenyl gallate (g-i) binding to the ATP binding pocket of E. coli DNA Gyrase B. Molecular models were constructed as described in Materials and Methods. A hydrogen bond is formed between all gallate derivatives (digallic acid, dodecyl gallate, and biphenyl gallate) and residue Asp73. c, f, and i shows schematic 2-D representations of E. coli DNA gyrase-ligand complexes. Hydrogens bonds and distances are highlighted as green dashed lines. Hydrophobic contacts are shown as an arc with spokes radiating towards the ligand atoms.

Several gallate derivative, such as butyl gallate, octyl gallate, and dodecyl gallate, are widely used as food additives and were shown to have antibacterial activities.^[51] In this study, we also tested the antibacterial activities of digallic acids and these gallate derivatives and determined their MIC against several bacterial strains. Our results are summarized in Table 2. Consistent with previously published results,^[51] octyl and dodecyl gallate showed antibacterial activities against B. subtilis, S. aureus, and MRSA. Digallic acid also showed some antibacterial activities against these two gram-positive bacteria. Interestingly, the antibacterial activities of these gallate derivatives are coincided with their inhibition against bacterial DNA gyrase, suggesting that DNA gyrase is a potential target for these compounds for their antibacterial activities. Biphenyl gallate showed some antibacterial activities against E. coli. Nevertheless, these gallate derivatives do not have strong antibacterial activities against Gram-negative E. coli strains ATCC 25922 and BAS3023. A likely scenario is that these compounds could not enter E. coli cells by crossing bacterial cell wall and membrane to inhibit DNA gyrase in vivo. Another possibility is that these gallate derivatives are ester and may be hydrolyzed in the MIC assays. Additional studies are needed to confirm these possibilities.

Table 2.

Antibacterial activity of gallate derivatives

Bacterial strains	MIC (μM)
	Digallic acid	Butyl gallate	Octyl gallate	Dodecyl gallate	Phenyl gallate	Biphenyl gallate	Ciprofloxacin	Novobiocin
S. aureus ATCC 14775	200	>200	200	200	200	50	1.25	<6.25
MRSA ATCC 33591	200	>200	100	200	200	50	40	<6.25
B. subtilis ATCC 6633	>200	>200	12.5–25	25–50	>200	50–100	0.4	<6.25
E. coli ATCC 25922	>200	>200	>200	>200	>200	>200	0.0125	157.5
E. coli imp BAS3023	>200	>200	50	200	>200	200	0.0019	6.25

One advantage of these new gallate-based gyrase inhibitors is that they can be easily modified/derivatized to enhance their potency. For instance, new functional groups can be attached to the gallate through an ester or amide bond. Using a combined effort of biochemistry, molecular modeling, medicinal chemistry, and antibacterial testing, new antibacterial agents targeting bacterial DNA gyrase can be developed/synthesized.

Conclusion

In summary, we found that digallic acid and several gallate derivatives are potent bacterial DNA gyrase inhibitors. Our results show that a long hydrophobic chain or group attached to the gallate moiety significantly increases the gyrase inhibition activities of these gallate derivatives although gallic acid per se does not inhibit E. coli DNA gyrase. These new gyrase inhibitors are competitive inhibitor of bacterial DNA gyrase’s ATPase and do not cause gyrase-mediated single-stranded nicks and double-stranded breaks. They also inhibit E. coli DNA topoisomerase IV and do not inhibit E. coli DNA topoisomerase I, human DNA topoisomerase I, and human DNA topoisomerase IIα.

Experimental Section

Materials

Human topoisomerase I, IIα, and kinetoplast DNA (kDNA) were purchased from TopoGen, Inc. (Buena Vista, CO). E. coli DNA gyrase and topoisomerase I were purified as described previously.^[41,52] E. coli DNA topoisomerase IV was kindly provided by Keir C. Neuman at the National Heart, Lung, and Blood Institute. Ethidium bromide, buffer-saturated phenol, and ethanol were purchased from Fisher Scientific, Inc. Gallic acid, ellagic acid, butyl gallate, octyl gallate, and dodecyl gallate were bought from Fisher Scientific, Inc. Digalic acid was purchased from Santa Cruz Biotechnology, Inc. Phenyl gallate (NSC333571) and biphenyl gallate (NSC406820) were obtained through NCI Developmental Therapeutics Program. Vendors of other polyphenols analyzed in this study are provided in Table S1. Plasmid pAB1 and fluorescently-labeled plasmid pAB1_FL905 were described previously.^[47]

Gel- and supercoiling dependent fluorescence quenching (SDFQ)-based gyrase DNA supercoiling and inhibition assays

DNA supercoiling assays were carried out in 30 μL of 1×gyrase buffer (35 mM Tris-HCl, 24 mM KCl, 4 mM MgCl₂, 2 mM DTT, 0.1 mg/mL BSA, 6.5% glycerol, and 1.75 mM ATP, pH 7.5) containing 8.9 nM of E. coli DNA gyrase and 400 ng of relaxed plasmid pAB1 or pAB1_FL905. The reaction mixtures were incubated at 37°C for 30 minutes. For gel-based assays, the reactions were stopped by the addition of 1 μL of stop solution (2% SDS and 200 mM EDTA). DNA samples were analyzed by using 1% agarose gels in 1×TAE buffer followed by ethidium bromide staining and photographed under UV light. For SDFQ-based assays, DNA samples were transferred to a 384-well plate. Fluorescence measurements were performed in a Biotek Synergy H1 Hybrid Plate Reader using a wavelength of 494 nm and 521 nm for excitation and emission, respectively. For DNA gyrase inhibition assays, each compound was mixed with E. coli DNA gyrase and incubated on ice for 5 min. Then, relaxed pAB1 was added to initiate the supercoiling reaction.

DNA relaxation assays by different DNA topoisomerases

DNA relaxation assays by E. coli DNA topoisomerase I, human DNA topoisomerase I, and human DNA topoisomerase IIα was described previously.^[41] DNA relaxation assays by E. coli DNA topoisomerase IV was performed in 20 μL of 1×topoisomerase IV buffer (35 mM Tris-HCl, 24 mM KCl, 4 mM MgCl₂, 2 mM DTT, 0.1 mg/mL BSA, 6.5% glycerol, 100 mM potassium glutamate, and 1.75 mM ATP, pH 7.5) containing 2 nM of DNA topoisomerase IV and 200 ng of supercoiled plasmid pAB1. The reaction mixtures were incubated at 37°C for 30 minutes and stopped by the addition of 1 μL of stop solution (2% SDS and 200 mM EDTA). DNA samples were analyzed by using 1% agarose gels in 1×TAE buffer followed by ethidium bromide staining and photographed under UV light.

DNA decatenation assays by E. coli DNA topoisomerase IV

DNA decatenation assays by E. coli DNA topoisomerase IV was performed in 30 μL of 1×topoisomerase IV buffer (35 mM Tris-HCl, 24 mM KCl, 4 mM MgCl₂, 2 mM DTT, 0.1 mg/mL BSA, 6.5% glycerol, 100 mM potassium glutamate, and 1.75 mM ATP, pH 7.5) containing 1 nM of DNA topoisomerase IV and 200 ng of kDNA. The reaction mixtures were incubated at 37°C for 30 minutes and stopped by the addition of 1 μL of stop solution (2% SDS and 200 mM EDTA). DNA samples were analyzed by using 1% agarose gels in 1×TAE buffer followed by ethidium bromide staining and photographed under UV light.

E. coli DNA gyrase cleavage assays

250 ng of plasmid pBR322 and 20 nM of E. coli DNA gyrase were mixed and incubated in 1×gyrase buffer at 37°C for 15 minutes in the presence of an inhibitor or compound. After the incubation, 0.2% SDS and 0.1 mg/ml proteinase K were added to the reaction mixtures to trap the gyrase-inhibitor-DNA complex and digest the gyrase, respectively, and incubated for an additional 30 min at 37 °C. DNA samples were analyzed in 1% agarose gel containing 0.5 μg/mL ethidium bromide in 1×TAE buffer and photographed under UV light.

E. coli DNA Gyrase and topoisomerase IV ATPase Linked Assay

A linked assay that couples the ATP hydrolysis with the conversion of NADH to NAD⁺ was used to determine the ATPase activity of E. coli DNA gyrase. Briefly, the assays were conducted in 60 μL of 1×ATPase buffer (10 mM Tris.HCl (pH 7.5), 0.2 mM EDTA, 1 mM magnesium chloride, 1 mM DTT, and 2% (w/v) glycerol) containing 50 nM of E. coli DNA gyrase or gyrB, 200 ng of relaxed pAB1, 0.8 mM of phosphoenol pyruvate, 1.2 units of pyruvate kinase, 1.7 units of lactate dehydrogenase, and 0.4 mM of NADH at 37 °C. After the reaction mixtures were incubated at 37 °C for 5 minutes, 2 mM of ATP was added to initiate the reaction. Absorbance at 340 nm was used to measure the ATPase activities at 37 °C.

Minimum inhibitory concentrations assays

Antibacterial minimum inhibitory concentrations (MICs) were obtained from three independent experiments using broth microdilution methods in 96-well plates according to Clinical and Laboratory Standards Institute guidelines.⁸ Cells were growth from a singles colony in Müller-Hinton II broth (MHIIB) (Becton Dickinson) medium for 18 h at 37 °C in agitation. The cultures were then diluted to OD₆₀₀ of 0.1 using MHIIB medium. 50 μL of the diluted cultures were added to 50 μL of serially diluted compounds in MHIIB in a 96-well plate and incubated at 37 °C for another 18–20 h. After OD₆₀₀ was measured to monitor the inhibition, 0.02% resazurin was added to each well and incubated for 4 hours at 37 °C. After the incubation, fluorescence with a wavelength of 494 nm and 521 nm for excitation and emission, respectively, was measured using a plate reader and used to calculate the MIC values. Bacterial strains S. aureus ATCC 14775, MRSA ATCC BAA44, B. subtilis ATCC 6633, E. coli ATCC 25922, and E. coli BAS3023 that carries an imp4213 mutation^[53,54] conferring permeability to small molecules were used to determine the MICs.

Molecular dynamics simulation and docking

To perform docking against a flexible receptor, conformations of the E. coli DNA gyrase subunit B were generated using molecular dynamics (MD) simulations. Charmm-Gui web interface,^[55] was used to prepare the simulation system. Specifically, gyrase subunit B was solvated in a cubic water box with TIP3 water and the system was neutralized by adding 0.15M of MgCl₂ resulting in a system containing ~ 31,000 atoms. The particle mesh Ewald (PME) method^[56] was used for calculating the long-range ionic interactions. The systems were minimized for 100,000 steps, followed by a 250 ps equilibration at 310 K with 1 fs time step. 200-ns production simulation with 2-fs time step were performed at a constant pressure of 1 atm. and T = 310 K. The Nose–Hoover Langevin-piston method^[57] was used for pressure coupling, with a piston period of 50 fs and a decay of 25 fs, and the Langevin temperature coupling with a friction coefficient of 1 ps⁻¹ was used for maintaining the temperature. From the simulation trajectory, 1,000 protein pdb frames of gyrase subunit B conformations were extracted using Visual Molecular Dynamics (VMD)^[58] for molecular docking.

Digallic acid, dodecyl gallate, and biphenyl gallate were docked to 1,000 MD-generated gyrase subunit B conformations. The protein pdb files and the small molecule structures were first converted to pdbqt format for docking. AutoDock Vina 1.1.269 was used to perform molecular docking. The compounds were screened against the protein conformations and the resulting scores of the complexes were sorted and ranked according to their binding affinities using custom scripts. Another 100-ns molecular dynamics simulation was performed for each top-ranked protein-drug complex with the same input settings described above.