Rapid Fluorescent-Based Detection of New Delhi Metallo-β-Lactamases by Photo-Cross-Linking Using Conjugates of Azidonaphthalimide and Zinc(II)-Chelating Motifs.

A method for rapid detection of metallo-β-lactamases NDM-5 and NDM-7 using conjugates of azidonaphthalimide and Zn(II) chelating motifs (like sulfonamides, hydroxamate, and terpyridine) is described. Incubation and irradiation, followed by gel electrophoresis, clearly show the presence of NDMs. The o-sulfonamide-based probe has the highest efficiency of detection for both the NDMs. The proteins are detectable at nM concentrations, and the method is also selective, works both in vitro and in vivo, as revealed by cellular imaging and also with clinical isolates.

Introduction

Antibiotic resistance has become a problem of immense concern in today’s world because of the appearance of so called “super bugs”.[1] These super bacteria harboring more than two antibiotic-resistant genes have become multidrug resistant, and infections caused by them are difficult to treat.[2] A major mechanism by which bacteria have acquired resistance is through production of metallo-β-lactamases (MBL)[3] which uses an active site zinc to hydrolyze the β-lactam rings of widely used penicillins and cephalosporins thus making them inactive.[4] As a result, carbapenems became one of the last lines of defense for treatment of such drug resistant infections. However, the emergence of new MBL-producing strains such as those producing New Delhi MBL (NDM BL) in recent years is conferring resistance[5] to almost all β-lactam antibiotics including carbapenems. NDM-1 was first identified in 2008 in a Klebsiella pneumoniae isolate and has become the focus of worldwide attention because of the rapid dissemination of the corresponding blaNDM-type gene mainly amongst the Enterobacteriaceae and Acinetobacter spp. Because of the first description of NDM-1, multiple (>15) NDM variants have been reported, mostly arising from mutations causing substitutions of one or two residues at different locations of NDM-1.[6] NDM-5 differs from NDM-1 by two substitutions at positions 88 (Val → Leu) and 154 (Met → Leu),[7] while NDM-7 differs at positions 130 (Asp → Asn) and 154 (Met → Leu).[8] Although the crystal structure of NDM-1 (Figure ) is known and not of NDM-5, it has been reported that the mutation at position 154 is proposed for inducing higher carbapenemase activity[7,8] and/or stabilization,[6d,9] which in turn affects the discovery of new antibiotics.[10] The mechanism of β-lactam resistance can be complex, involving, expression of β-lactamases and efflux pumps, porin loss, and alterations in penicillin-binding proteins (PBP). All these are associated with carbapenem resistance in gram negative bacteria.[11] One of the reasons[12,15,16] for NDM-harboring strains becoming a worldwide threat is certainly the lack of available antibiotics.[13,14] However, the problem has become more complicated because of the absence of a standardized phenotypic test for MBL detection. Rapid and early detection of MBL strains including NDM producers is crucial not only for the right treatment but also to prevent their dissemination and associated nosocomial infections.

Figure 1

Crystal structure of NDM-1. (A)NDM-1 monomer with highlighted motifs involved in protein dimerization and (B) proposed dimer form of NDM-1 with loop-10, loop-8, and alpha-3 residues involved in the interaction.

Results and Discussion

We became interested in developing an early detection method for NDM-producing micro-organisms. At present, laboratory detection of MBL (including NDM) producers employs different techniques, all of which depend upon assaying the carbapenemase activity,[13] like for example the modified Hodge test (MHT), UV spectrophotometry, MALDI TOF analysis, and Carba NP test. Amongst these, MHT is the only CLSI-recommended carbapenemase-screening method which, however, has a low sensitivity of ∼11% because of suppression of enzyme releases due to anchoring in the inner membrane.[17] The sensitivity of MHT can be improved somewhat by the addition of a nonionic surfactant (Triton X-100).[18] Other methods include PCR[19] and loop-mediated isothermal amplification[20] which, although more sensitive, put additional constrains like prior sequencing of the targeted gene and primer designing.

Recently, we have reported[21] an efficient detection protocol for HCA II and PBP-5 by a simple gel electrophoresis technique using an azidonaphthalimide template with a built-in fluorophore and a linker. This has simplified the previously reported template-based strategies.[22] Using a similar approach,[21] molecules 1–4 having Zn(II)-chelating motifs attached to the template were designed (Figure ) and synthesized in order to detect the presence of NDM-5 and NDM-7, in vitro and at a cellular level. Amongst all the probes, compound 2 with an o-sulphonamido moiety could detect the monomeric protein at ∼26 kDa with the most prominent visualizable fluorescent band under UV after photo-cross-linking followed by gel electrophoresis. The detection was also selective as demonstrated by preferential cross-linking of NDMs in the presence of other zinc enzymes like HCA II, carboxypeptidase, calf intestinal alkaline phosphatase, as well as nonmetallo enzymes like ovalbumin, class A serine-β-lactamase CTX-M-15[23] and also in cell lysate (both overexpressed and clinical isolates). Herein, we describe the synthesis of the probes and the results of subsequent SDS-polyacrylamide gel electrophoresis and microscopic studies in detail.

Figure 2

Target photoaffinity probes 1–4.

It may be pointed out that the design of the probes was inspired by a recent report by Cheng et al.[24] on the screening of sulfonamides as NDM-1 inhibitors. The strongest inhibitor has been shown to be a sulfonamide with ortho-ligating imino nitrogen. Moreover, molecular docking studies using the pdb structure of NDM-1 indicated the possibility of cation π-interaction as well as agostic interaction between the zinc ion and the aryl ring bearing the o-sulfonamide moiety of probe 2. For the p-sulfonamide, being much further away from the active site, the binding interactions are less. Interestingly with HCA II, which is a Zn(II)-containing enzyme, the probe 2 fails to show any binding, indicating possible selectivity of the probe (Figure ).

Figure 3

(a) Reported[24] NDM-1 inhibitor; docking poses of probe 2 (b) with NDM-1 and (c) with HCA II.

Backed up by the initial docking results, the synthesis started with the previously prepared GABA-derived azidonaphthalimide carboxylic acid 5. Like compound 1 whose synthesis has been reported earlier, the 2-sulphonamide[25] based compound 2 was similarly prepared by treating 5 with bromoacetyl sulfanilamide 6 in dimethylformamide (DMF) in the presence of K2CO3. The synthesis of compound 3 was accomplished from the sodium salt of 5 by reacting with bromomethyl terpyridine 7(26) in dry DMF at room temperature for 10 h. Working up of the reaction mixture with brine followed by drying over Na2SO4 and evaporation afforded a yellow solid from which the product was purified by repeated precipitation from acetone–hexane (for 1 and 2) and ethyl acetate-hexane (for 3). The fluorescent template 5 was converted to the corresponding hydroxamic acid 4 (a well-known bidentate Zn(II)–chelator for HDAC[27]) by sequential reaction, first with ethyl chloroformate and N-methyl morpholine in dry DCM followed by reaction with hydroxylamine hydrochloride in the presence of KOH (Figure ).

Figure 4

Synthesis of target compounds.

MBLs such as B1 subclass NDM and BcII are present in monomeric and, may be, dimeric forms in the solution. From the crystal structure (Figure ) of NDM-1, King and Strynadka[28] predicted that the variable loops around the active site lead to broader substrate specificity and hence higher resistance. It is also predicted that loop-10, loop-8, and alpha-3 residues of two monomers of NDM-1 proteins interact to form the dimer. Selevsek et al.[29] have shown that such dimers exist due to sharing of the same metal ions between two monomeric forms of the protein where the Zn ion acts as a bridge under low availability of Zn ions during protein folding.

The cross-linking was initially studied with purified NDM-5[30] under nonreducing conditions in which the protein exists both as monomeric and dimeric forms[28,29] in solution as revealed by the appearance of two distinct bands under CBB at ∼26 and ∼50 kDa corresponding to the mono and dimeric forms, respectively. All the probes were incubated with both NDM-5 and NDM-7 under irradiation with UV. The appearance of fluorescent bands visible under UV demonstrated successful cross-linking, the cross-linking being stronger (thus more selective) with the monomeric form as revealed by the relative fluorescent intensities of the bands vis-à-vis Coomassie blue stained-bands (Figure ). This observation is an indirect support to the hypothesis[28] that the Zn ions are shared between two monomeric proteins in the dimer thus making the metal ion less exposed. Gratifyingly, the cross-linking also works in the case of irradiation with visible light, (the gel picture which is included in the Supporting Information).

Figure 5

SDS-PAGE experiment[31] results show the binding efficiency of the synthesized probes with dimer and monomer forms of purified NDM-5. NDM-5 (final concentration 10 μM) was used for cross-linking with compound concentration kept at 20 μM. The gel was run under nonreduced conditions. Gel image captured under ultraviolet light has been denoted as UV and the gel image after Coomassie blue staining has been denoted as CBB. M denotes protein molecular weight marker.

The same experiment was then repeated with a fixed concentration of protein under reducing conditions to evaluate the relative efficiency of cross-linking of various probes with the monomeric form of NDM. In this case also, all the probe molecules showed cross-linking as observed by the appearance of fluorescent bands when visualized under UV after gel electrophoresis. ImageJ software analysis[32] revealed that amongst all the probes, compound 2 with the o-sulfonamido group showed the highest efficiency for visualization of NDM-5 (Figure ) and hence subsequent experiments regarding detection limit and cell lysate were carried out with probe 2.

Figure 6

(A) Screening of binding efficiency for synthesized fluorescent probes with purified NDM-5; M = protein mol wt marker; C = NDM-5 in 2% DMSO. The concentration of each compound was maintained at 10 μM. (B) ImageJ analysis of Figure A; the X-axis denotes the compound numbers. SDS-PAGE was run under reduced conditions.

For a protein concentration of 20 μM, the lowest concentration of compound 2 that can detect NDM-5 was 5 μM (figure in Supporting Information). With a compound 2 concentration of 20 μM, the protein (NDM-5) is detectable at as low as 500 nM concentration (Figure ).[33]

Figure 7

Protein concentration study with compound 2. Lane 1: 25 μM NDM-5 and DMSO (2%); lane 2: 25 μM NDM-5; lane 3: 20 μM NDM-5; lane 4: 15 μM NDM-5; lane 5: 10 μM NDM-5; lane 6: 5 μM NDM-5; lane 7: 2 μM NDM-5; lane 8: 1 μM NDM-5; lane 9: 0.5 μM NDM-5; lane 10: protein marker. Final concentration of compound 2 was kept 20 μM in each lane from lane 2–9. SDS-PAGE was run under reduced conditions.

For checking the selectivity of the probe molecule 2, incubation followed by UV irradiation was performed with different additional proteins that included other Zn-dependent enzymes like HCA II, carboxypeptidase A, and CI phosphatase. The cross-linking efficiency of probe 2 was also compared with ovalbumin and a serine β-lactamase CTX-M-15. The gel picture under UV vis-a-vis Coomassie stain failed to show any significant cross-linking with all these enzymes (Figure ). These results demonstrated the in vitro selectivity of compound 2 towards NDM-5. The probe 2 can also detect NDM-5 in cell lysate where the cells have been overexpressed with the NDM-5 gene. The method is also applicable to detect the presence of NDM-5 even in clinical isolates. The results were also very similar in the case of NDM-7. For example, probe 2, which selectively binds to the NDM-7 monomer, has higher efficiency compared to other probes and the detection limit in this case was 625 nM using 80 μM of 2 (figure in Supporting Information).

Figure 8

Results of cross-linking with different proteins and cell lysates. For (A) OV = ovalbumin (14 μM); M = protein mol wt marker; ND = purified NDM-5 (10 μM). For (B) HA = purified HCA II; M = protein mol wt marker. For (C) EC = E. coli cell lysate harboring CTX-M-15 overexpressed protein; C = E. coli cell lysate harboring NDM-5 overexpressed protein with 2% DMSO; ECN = E. coli cell lysate harboring NDM-5 overexpressed protein; M = protein mol wt marker; B = only 2% DMSO; CP = purified carboxypeptidase A (10 μM); CIP = purified calf intestine phosphatase (10 μM); ND = purified NDM-5 (2 μM); ECC = E. coli cell lysate from clinical isolate harboring NDM-5.[34]

Encouraged by the in vitro results, we then carried out in vivo studies using fluorescence cell imaging to detect NDM-5 and NDM-7. Upon expression of NDM-5 and NDM-7 in the Escherichia coli host (CS109) from an expressible plasmid (pBAD18-cam) under the control of an arabinose promoter and subsequent labeling of the cells with compound 2 revealed that only the CS109 cells expressing NDM-5 or NDM-7 were detectable under fluorescent microscopy (Figure ); no fluorescence was observed in the control CS109. The CS109 cells expressing NDM-5 as well as NDM-7 when labeled with compound 2 have clearly shown the localization of the MBL at the cell periphery. This is most likely at the periplasmic space of the host because the expressed NDM-5 or NDM-7 gene has the constitutive signal peptide for transporting across the cytoplasmic membrane (see Figure , column 2 and 3 fluorescent images). The observation further expands the cell permeability and precision of compound 2 in the detection of NDM-5 and NDM-7 at the cellular level.

Figure 9

Microscopic analysis of E. coli cells having overexpressed SHV-14 and NDM-5 proteins bound to compound 2.

Conclusions

In summary, the overall work provides an in-depth study on the utility of various protein binding fluorescent compounds in detection and localization of MBLs such as NDM-5 and NDM-7 both at the cellular level and in vitro. Moreover, the compound 2 showed selective binding towards both the NDMs over other proteins used in the study as revealed by the combination of in vitro binding studies with cell lysates and whole cell microscopy images. The method has been validated for use against clinical isolates. Future studies will be aimed towards other carbapenemase resistant MBLs like VIM and IMP and in vitro screening of possible inhibitors triggered by visible light.

Experimental Section

General Experimental Details

All reactions were performed under a nitrogen atmosphere unless otherwise stated. Glassware used in reactions was thoroughly oven-dried. All commercial grade reagents were used without further purification, and solvents were dried prior to use following the standard protocol. Thin-layer chromatography was carried out on precoated plates (silica gel 60 F254), and the spots were visualized by exposure to UV light and/or by staining with iodine. All crude products were purified by silica gel flash column chromatography (230–400 mesh) with petroleum ether/ethyl acetate as the eluent and characterized by NMR and mass spectrometry unless otherwise mentioned. Melting points were determined in open capillary tubes and are reported as uncorrected. 1H and 13C NMR spectra for all the compounds were recorded at 400/500/600 and 100/125/150 MHz (Bruker UltrashieldTM 400, AscendTM 500, Ascend 600), respectively. The spectra were recorded in deuterochloroform (CDCl3) and deuterated dimethyl sulfoxide (DMSO-d6) as solvents at room temperature. Chemical shifts are reported in ppm from tetramethylsilane with the solvent resonance as rgw internal standard (CDCl3: δH = 7.26, δC = 77.16 ppm). Data for 1H NMR are reported as follows: chemical shift (δ ppm), multiplicity (s = singlet, br s = broad singlet, d = doublet, t = triplet, q = quartet, m = multiplet, dd = doublet of doublet), coupling constant (Hz), and integration. Data for 13C NMR are reported as the chemical shift. HRMS spectra using ESI were recorded on an ESI–FTMS mass spectrometer.

General Experimental Procedure

Preparation of Carboxylic Acid Derivatives of 4-Azido-1,8-naphthalimide (5)

Procedure for preparation of the GABA–carboxylic acid derivative is the same as reported earlier (Cai, Y.; Zhan, J.; Shen, H.; Mao, D.; Ji, S.; Liu, R.; Yang, B.; Kong, D.; Wang, L.; Yang, Z. Anal Chem. 2016,88, 740).

Preparation of Bromoacetyl Derivatives of 2-Aminobenzene Sulfonamide (6)

To a solution of sulfanilamide 8 (0.35 g, 2.03 mmol) in dry THF (10 mL), K2CO3 (0.561 g, 4.06 mmol) was added and stirred for 30 min. Keeping the temperature at 0 °C, bromoacetyl chloride (0.2 mL, 2.44 mmol) was added dropwise to the reaction mixture which was stirred for a further period of 30 min at 0 °C. Water was added and the mixture was extracted with EtOAc (50 mL × 2), washed with brine, dried over Na2SO4, and concentrated in vacuo to get the product as a white crystalline solid (76% yield). Spectral data: 1H NMR (600 MHz, acetone-d6): δ 10.05 (s, 1 H), 8.40 (d, J = 8.4 Hz, 1 H), 8.03 (d, J = 8.0 Hz, 1 H), 7.69 (t, J = 7.9 Hz, 1 H), 7.39 (t, J = 7.7 Hz, 1 H), 6.94 (s, 2 H), 4.26 (s, 2 H). 13C NMR: δ (150 MHz, acetone-d6) 165.4, 135.7, 133.9, 132.4, 128.8, 125.1, 123.4, 30.4. HRMS: calcd for C8H9N2O3SBrNa (M + Na), 314.9415; found, 314.9424.

Synthesis of Final Compounds

Synthesis of 2-Oxo-2-((2-sulfamoylphenyl)amino)ethyl 4-(6-Azido-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)butanoate (2)

To a solution of GABA–carboxylic acid 5 (0.15 mmol) in dry DMF (5 mL) under N2, anhydrous K2CO3 (0.18 mmol) was added and stirred for 30 min at room temperature. A solution of bromoacetyl derivatives of sulphonamide 6 (0.18 mmol) in dry DMF (2 mL) was added and stirring was continued for 10 h at room temperature. The reaction was quenched by adding water (30 mL) and the aqueous layer was extracted with EtOAc (30 mL x 2). The combined organic layers were washed with brine, aq. NaHCO3 and water, dried over anhydrous Na2SO4, and concentrated in vacuo. The yellowish-brown gummy product was first precipitated from the acetone–hexane mixture and the precipitate was washed with hexane 2–3 times to furnish the target material as a yellow solid (70% yield). Spectral data: 1H NMR (400 MHz, DMSO-d6): δ 12.14 (s, 1H), 8.49 (d, J = 7.2 Hz, 1H), 8.42 (d, J = 8.0 Hz, 1H), 8.38 (d, J = 8.4 Hz, 1H), 7.82 (m, 2H), 7.68 (t, J = 7.5 Hz, 2H), 7.47 (t, J = 7.6 Hz, 1H), 7.37 (d, J = 8.2 Hz, 1H), 4.89 (s, 2H), 4.10 (t, J = 6.5 Hz, 2H), 2.58 (t, J = 7.2 Hz, 2H), 2.03–1.91 (m, 2H). 13C NMR (150 MHz, DMSO-d6): δ 172.1, 163.5, 163.1, 155.1, 142.9, 134.7, 133.3, 131.6, 131.6, 128.5, 128.4, 127.4, 126.7, 123.6, 122.3, 121.6, 118.3, 117.7, 116.0, 62.2, 30.9, 22.8. IR (cm–1): 2942, 2122, 1745, 1695, 1652, 1612, 1586, 1542, 1478, 1439, 1391, 1353, 1276, 1158, 1130, 996, 765, 588, 502. HRMS: calcd for C24H20N6O7SNa (M + Na), 559.1012; found, 559.1011.

Synthesis of 4-([2,2′:6′,2″-Terpyridin]-4′-yl)benzyl 4-(6-Azido-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)butanoate (3)

GABA–carboxylic acid 5 (0.11 g, 0.35 mmol) was dissolved in methanol and NaHCO3 (29 mg, 0.35 mmol) dissolved in a minimum amount of water was added. The mixture was stirred for 2 h at room temperature and then lyophilized to obtain the sodium salt of 5 as a yellow solid. The solid (0.14 g, 0.35 mmol) was dissolved in dry DMF (20 mL) and compound 7 (0.12 g, 0.35 mmol) was added. The reaction mixture was stirred for 12 h under a N2 atmosphere at room temperature. The reaction mixture was extracted with ethyl acetate (2 × 50 mL) and the combined organic layers were washed with brine to remove DMF. Removal of ethyl acetate furnished a yellow solid which was further reprecipitated with ethyl acetate and hexane (75% yield). Spectral data: 1H NMR (400 MHz, chloroform-d): δ 8.75–8.58 (m,7 H), 8.55 (d, J = 8.0 Hz, 1 H), 8.40 (d, J = 8.4 Hz, 1 H), 7.87 (d, J = 6.0 Hz, 4 H), 7.75–7.67 (m, 1 H), 7.48–7.41 (m, 3 H), 7.35 (t, J = 6.2 Hz, 2 H), 5.16 (s, 2 H), 4.26 (t, J = 7.0 Hz, 2 H), 2.54 (t, J = 7.4 Hz, 2 H), 2.15 (p, J = 7.1 Hz, 2 H). 13C NMR (125 MHz, chloroform-d): δ 172.9, 164.2, 163.7, 156.3, 156.1, 149.9, 149.3, 143.6, 138.4, 137.0, 132.4, 131.9, 129.3, 129.0, 128.8, 127.6, 127.0, 124.5, 124.0, 122.7, 121.5, 119.0, 114.8, 66.0, 39.7, 32.1, 23.6. IR (cm–1): 3404, 2931, 2123, 1708, 1352, 1272, 1111, 784, 726, 612. HRMS: calcd for C38H27N7O4H (M + H), 646.2203; found, 646.2201.

Synthesis of 4-(6-Azido-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)-N-hydroxybutanamide (4)

To a solution of the compound 5 (0.1 g, 0.31 mmol) in dichloromethane (20 mL) at 0 °C, ethyl chloroformate (35 μL, 0.37 mmol) and N-methyl morpholine (41 mg, 0.40 mmol) were added and the mixture was stirred for 10 min. Hydroxylamine (prepared by stirring a methanolic solution of hydroxylamine hydrochloride (32 mg, 0.47 mmol) with KOH in methanol, followed by filtration to remove KCl) was added and the mixture was stirred for 15 min. The solvent was removed under reduced pressure to obtain yellow gummy oil. After washing with dichloromethane, the hydroxamic acid 4 was obtained as a yellow solid (58% yield). Spectral data: 1H NMR (400 MHz, methanol-d4): δ 8.58 (d, J = 7.3 Hz, 1 H), 8.55 (d, J = 8.0 Hz, 1 H), 8.49 (d, J = 8.5 Hz, 1 H), 7.80 (t, J = 7.8 Hz, 1 H), 7.64 (d, J = 8.0 Hz, 1 H), 4.18 (t, J = 7.0 Hz, 2 H), 2.20 (t, J = 7.6 Hz, 2 H), 2.02 (p, J = 7.7 Hz, 2 H). 13C (125 MHz, methanol-d4): δ 172.2, 165.5, 165.1, 145.3, 133.1, 133.1, 130.4, 130.0, 128.1, 125.6, 123.7, 119.8, 116.3, 40.7, 31.5, 25.4. IR (cm–1): 3309, 2946, 2123, 1730, 1272, 1119, 733, 624. HRMS: calcd for, C16H13N5O4Na (M + Na) 362.0865; found, 362.0868.

Capture Experiment Protocol

The capture experiment was studied with overexpressed proteins at the cellular level and with purified proteins (in vitro) under different conditions of cross-linking under (UV light) and reducing and nonreducing conditions. In the absence of a reducing environment, the NDM protein exists as dimer and monomer forms in the solution. Moreover, under reducing conditions (i.e., in the presence of DTT or BME in the loading dye of SDS-PAGE), a single band of NDM protein was observed. However, under nonreducing condition, two distinct bands were observed (26 and 50 kDa approximately). In all the experiments purified proteins are used in the range 5–20 μM depending on the experimental requirement in 50 mM HEPES buffer (pH 7.2), each compound was used at 20 μM concentration (except in compound concentration study). The whole reaction was set in 50 μL reaction volume and crossing-linking was done under the UV light for 30 min. The image was recorded under Typhoon FLA7000, UV light and white light conditions to confirm the compound binding and protein stability. For more details about the buffer used and the capture experiment protocol see Basak et al. Chem. Commun., 2017,53, 13015.

SDS-PAGE Gel Experimental Details

Cellular Expression and Selective Binding of Compound to β-Lactamase Overexpressed Proteins in Whole Cell Lysates

Cellular expression of NDM-5 and NDM-7 was studied in E. coli CS109 cells under the control of an araC promoter. The cells were induced with 0.2% arabinose and incubated overnight at 37 °C. The protein expression was analyzed via 15% SDS-PAGE after sonication. Cell lysates used at 49 μL of 0.9 OD600nm, compound used at 20 μM concentration.

Overexpression of β-lactamase CTX-M-15 was studied in E. coli BL21 (DE3) under the control of a T7 promoter system. E. coli cell culture harboring recombinant CTX-M-15 was induced with 0.2 mM of IPTG at an OD600nm–0.6 and allowed to grow at 37 °C till OD600nm reached ∼1.0. The cells were then harvested and resuspended in 50 mM HEPES buffer and lysed by sonication. Prepared lysates were analyzed via 15% SDS-PAGE for expression. Cell lysates were used at a concentration of ∼1 μg/μL with compound used at 20 μM concentration.

Similarly, clinical isolate-harboring NDM-5 was grown under antibiotic stress (Ampicillin 200 μg/mL) till OD600nm reached ∼1.0. The cells were harvested, resuspended in 50 mM HEPES buffer, and lysed by sonication. Reaction conditions were kept similar to previous experiments with a final cell lysate protein concentration of ∼1.0 μg/μL.

Protein Purification and Compound Binding Study

The NDM-5, NDM-7, and CTX-M-15 proteins were overexpressed in the pET28a(+) vector under 0.5 mM IPTG induction, in BL21 (DE3) cells, at 20 °C overnight with constant shaking. The cells were sonicated and the cell lysate, thus obtained after removal of the cell debris, was loaded in the AKTA prime column (HisTrapTM HP, GE Healthcare, Uppsala, Sweden). The protein was washed initially with a low concentration of imidazole (25 mM) and subsequently eluted with 500 mM imidazole. Furthermore, excess imidazole was removed via dialysis and the concentration of purified protein was checked via the Bradford method. The SDS-PAGE analysis was done to assess the quality of purified proteins. The purity level of proteins after dialysis was more than 95%. Afterwards, the protein samples were used for further in vitro analysis.

Preferential Binding of the Compound to the NDM Monomer form than the Dimer Form

To know the preferential binding of fluorescent compounds to NDM monomer and dimer forms, 15% SDS-PAGE was done under nonreducing conditions. The NDM protein was used at 10 μM, the compound was used at 20 μM. The cross-linking was done under UV light for 30 min on ice, followed by heat denaturation (at 95 °C for 5 min), and 15% SDS-PAGE analysis.

Microscopic Analysis of Compound 2 Binding to Overexpressed NDM-5 and SHV-14 in E. coli Cells

Furthermore, to know the localization of the compound in E. coli cells harboring overexpressed NDM-5, NDM-7, and SHV-14 proteins, the bacterial cells were allowed to bind compound 2 (20 μM) for 15 min postinduction period of 16 h at 37 °C. Thereafter, the cells were allowed to stay under UV light for 15 min for cross-linking of compound 2 with overexpressed NDM-5, NDM-7, and SHV-14 within the E. coli cells. This step was followed by microscopy to record the images under both phase contrast and fluorescence (using DAPI filter). The microscopy slides were coated with 0.1% (w/v) poly-l-lysine prior to mounting of overexpressed bacterial cells. The experiment was also repeated with control E. coli cells devoid of NDM-5, NDM-7, and SHV-14.