Ebsulfur as a potent scaffold for inhibition and labelling of New Delhi metallo-β-lactamase-1 in vitro and in vivo

Article abstract of DOI:10.1016/j.bioorg.2018.11.035

The superbug infection caused by New Delhi metallo-β-lactamase (NDM-1) has grown into an emerging threat, labelling and inhibition of NDM-1 has proven challenging due to its shuttling between pathogenic bacteria. Here, we report a potent covalent scaffold, ebsulfur, for targeting the protein in vitro and in vivo. Enzymatic kinetic study indicated that eighteen ebsulfurs gained except 1a–b and 1f inhibited NDM-1, exhibiting an IC₅₀ value ranging of 0.16–9 μM, and 1g was found to be the best, dose- and time-dependent inhibitor with an IC₅₀ of 0.16 μM. Also, these ebsulfurs effectively restored the antibacterial activity of cefazolin against E. coli expressing NDM-1, and the best effect was observed to be from 1g, 1i and 1n, resulting in an 256-fold reduction in MIC of the antibiotic at a dose of 16 μg/mL. The equilibrium dialysis study implied that the ebsulfur disrupted the coordination of one Zn(II) ion at active site of NDM-1. Labelling of NDM-1 using a constructed fluorescent ebsulfur Ebs-R suggested that the inhibitor covalently bound to the target through SDS-PAGE analysis in vitro. Also, labelling NDM-1 in living E. coli cells with Ebs-R by confocal microscopic imaging showed the real-time distribution change process of intracellular recombinant protein NDM-1. Moreover, the cytotoxicity of these ebsulfurs against L929 mouse fibroblastic cells was tested, and their capability to restore antibacterial activity of antibiotic against clinical strains E. coli EC08 producing NDM-1 was determined. The ebsulfur scaffold proposed here is valuable for development of the covalent irreversible inhibitors of NDM-1, and also for labelling the target in vitro and in vivo.

Full text of DOI:10.1016/j.bioorg.2018.11.035

Accepted Manuscript
Ebsulfur as a potent scaffold for inhibition and labelling of New Delhi Metallo-
β-lactamase-1 in vitro and in vivo
Jianpeng Su, Jiayun Liu, Cheng Chen, Yuejuan Zhang, Kewu Yang
PII:
S0045-2068(18)31037-X
https://doi.org/10.1016/j.bioorg.2018.11.035
YBIOO 2643
DOI:
Reference:
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
16 September 2018
19 November 2018
20 November 2018
Please cite this article as: J. Su, J. Liu, C. Chen, Y. Zhang, K. Yang, Ebsulfur as a potent scaffold for inhibition and
labelling of New Delhi Metallo-β-lactamase-1 in vitro and in vivo, Bioorganic Chemistry (2018), doi: https://doi.org/
10.1016/j.bioorg.2018.11.035
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Ebsulfur as a potent scaffold for inhibition and labelling
of New Delhi Metallo-β-lactamase-1 in vitro and in vivo
Jianpeng Su1, Jiayun Liu2, Cheng Chen1, Yuejuan Zhang1, Kewu Yang1*
1Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of
Education, Chemical Biology Innovation Laboratory, College of Chemistry and Materials Science,
Northwest University, 1 Xuefu Avenue, Xi’an 710127, P. R. China.
2Department of Clinical Laboratory, Xijing Hospital, Medical University of the Air Force, Xi'an 710032, P.
R. China
ABSTRACT: The superbug infection caused by New Delhi metallo-β-lactamase (NDM-1) has
grown into an emerging threat, labelling and inhibition of NDM-1 has proven challenging due to
its shuttling between pathogenic bacteria. Here, we report a potent covalent scaffold, ebsulfur,
for targeting the protein in vitro and in vivo. Enzymatic kinetic study indicated that eighteen
ebsulfurs gained except 1a-b and 1f inhibited NDM-1, exhibiting an IC50 value ranging of 0.16-9
μM, and 1g was found to be the best, dose- and time-dependent inhibitor with an IC50 of 0.16 μM.
Also, these ebsulfurs effectively restored the antibacterial activity of cefazolin against E. coli
expressing NDM-1, and the best effect was observed to be from 1g, 1i and 1n, resulting in an
256-fold reduction in MIC of the antibiotic at a dose of 16 μg/mL. The equilibrium dialysis study
implied that the ebsulfur disrupted the coordination of one Zn(II) ion at active site of NDM-1.
Labelling of NDM-1 using a constructed fluorescent ebsulfur Ebs-R suggested that the inhibitor
covalently bound to the target through SDS-PAGE analysis in vitro. Also, labelling NDM-1 in living
E. coli cells with Ebs-R by confocal microscopic imaging showed the real-time distribution change
process of intracellular recombinant protein NDM-1. Moreover, the cytotoxicity of these
ebsulfurs against L929 mouse fibroblastic cells was tested, and their capability to restore
antibacterial activity of antibiotic against clinical strains E.coli EC08 producing NDM-1 was
determined. The ebsulfur scaffold proposed here is valuable for development of the covalent
irreversible inhibitors of NDM-1, and also for labelling the target in vitro and in vivo.
KEYWORDS: antibiotic resistance, metallo-β-lactamase, NDM-1, inhibitor, ebsulfurs
1. Introduction
β-Lactam antibiotics, including penicillins, cephalosporins, and carbapenems, remain the
most important and frequently used antimicrobial agents, constituting more than 50% of the
antibiotics prescribed worldwide [1]. However, the immoderate use of β-lactams has resulted in a
large number of bacteria that are resistant to most all antibiotics. Most commonly, bacteria
become resistant to β-lactams by producing β-lactamases, which catalyze the cleavage of
β-lactam rings of the antibiotics [2, 3]. β-Lactamases have been categorized into four classes: A, B,
C, and D, depending on their amino acid sequence homologies [4]. Classes A, C, and D are
referred to as serine-β-lactamases (SβLs) which utilize an active site serine to hydrolyze the

β-lactam ring via an enzyme-acyl intermediate [5]. The class B enzymes, also called
metallo-β-lactamases (MβLs), use either 1 or 2 Zn(II) ions to catalyze β-lactams hydrolysis [6].
MβLs are further divided into three subclasses B1, B2, and B3, based on their amino acid
sequence and metal occupancies [1, 7].
Antibiotic resistance has become a concerning global health problem. The development of
β-lactamase inhibitors is an essential strategy for maintaining the usefulness of the existing
β-lactam antibiotics. Although SβL inhibitors, such as clavulanic acid, tazobactam, and sulbactam,
have been used clinically in combination with β-lactam antibiotics [1], there are no MβL inhibitors
for clinic purpose to date [8]. Accordingly, the development of MβL inhibitor is urgently needed.
New Delhi metallo-β-lactamase-1 (NDM-1), a B1 subclass MβL, hydrolyzes almost all
clinically available β-lactam antibiotics, including carbapenems which are considered “last resort
antibiotics” [9-11]. The plasmid-encoded NDM-1 gene has been shown to horizontally transfer to
other pathogenic bacteria and spread rapidly [12], therefore it is a daunting challenge to
real-time monitor it in the dynamic cellular environment. Even worse, there is no simple and
practical method for labelling/tracking NDM-1 in vitro and in vivo.
Given the enormous clinical importance of NDM-1, a large number of NDM-1 inhibitors have
been reported, such as D- and L-captopril [13, 14], aspergillomarasmine A (AMA) [15],
pyridine-2-carboxylic acid and its derivatives [16, 17], and cyclic boronates [18]. Recently, ebselen
was reported to be a covalent inhibitor of NDM-1 by forming a S–Se bond with the Cys221
residue at active site of the enzyme [19, 20]. The crystal structures of NDM-1 have revealed the
Zn(II) ion at the Zn1 site is coordinated by His116, His118, His196, and a bridging water/hydroxide
molecule (Wat1), while the Zn(II) ion at the Zn2 site is coordinated by Asp120, Cys221, His263, the
bridging water and an apical water molecule (Wat2) in NDM-1 [21]. Based on the residue Cys221
at active site of NDM-1, which was expected to be bound by the sulfur-containing molecules
through forming disulfide bond in inhibiting the enzyme as proposed in Fig. 1, in this work, we
developed a series of ebsulfur derivatives as previously reported methods [22-25], and also
constructed a fluorescent labelling agent Ebs-R through attaching rhodamine B to an ebsulfur
moiety, which was shown to label and visualize NDM-1 in vitro and in vivo. Moreover, the
cytotoxicity of these ebsulfurs against L929 mouse fibroblastic cells was tested, and their
capability to restore antibacterial activity of antibiotic against E. coli cells producing NDM-1 was
determined.
2. Results and discussion
2.1 Chemistry
Eighteen ebsulfurs 1a−o, 2a−c and a fluorescent inhibitor Ebs-R were synthesized by a
synthetic route as shown in Scheme S1 as previously reported methods [24, 25]. Briefly,
2,2'-disulfanediyldibenzoic acid (a) react with SOCl2 by refluxing in SOCl2 to give
2,2'-disulfanediyldibenzoyl chloride (b) in a 80% yield, and b was treated with the corresponding
aniline and Et3N in DCM to afford compounds 1a-o in 45–81% yield [25]. The intermediate
isocyanates (d) was obtained by reaction of primary amine (c), triphosgene and Et3N, and d
reacted with 1a to offer desire compounds 2a-c [24]. Fluorescent inhibitor Ebs-R was synthesized
by esterification of 1f with Rhodamine B. The structures of ebsulfurs and Ebs-R are shown in Fig.
1
1. All compounds synthesized were characterized by H and 13C NMR and confirmed by HRMS
(see Supporting Information).

Fig. 1. The active site structures and proposed interaction mechanism of NDM-1 and the
synthesized ebsulfurs with various substitutes.
2.2 Activity evaluation of ebsulfur on NDM-1 in vitro
To test whether these ebsulfurs were NDM-1 inhibitors, NDM-1 was over-expressed and
purified as previously described [26], and detailed in the Experimental Section. The enzyme
inhibition experiments with these compounds were conducted on an Agilent UV8453 UV-Vis
spectrophotometer using 40 μM cefazolin as substrate, and the concentrations of inhibitors were
varied between 0-100 μM. Enzyme and inhibitor were pre-incubated for 20 h before adding
cefazolin, which was then monitored at 262 nm, to determine the initial reaction rate in the
absence and presence of inhibitor in triplicate, and the average value was recorded.
Percent inhibition, defined as enzyme activity without inhibitor (100%) minus residual
activity with 10 μM inhibitor, is showed in Fig. S1. It is observed that all of these ebsulfurs except
1a, 1b and 1f exhibited more than 50% inhibition on NDM-1 at a concentration of 50 nM. Also,
the inhibitor concentrations causing 50% decrease of enzyme activity (IC50) were determined
using cefazolin as substrate. The IC50 data (Table 1) indicated that all compounds except 1a and 1f

inhibited NDM-1 with an IC50 value range from 0.16 to 24 μM, and 1g was found to be the best
inhibitor (IC50=0.16 μM), which is consistent with its high percent inhibition at a concentration of
10 μM (Fig. S1).
Table 1
IC50 values of ebsulfurs against metallo-β-lactamases NDM-1a
Compd
1a
IC50 (μM)
98.6±0.02
24.1±0.04
8.17±0.02
2.5±0.03
Compd
1k
IC50 (μM)
0.79±0.04
0.77±0.05
1.05±0.04
0.66±0.02
0.2±0.02
1b
1c
1l
1m
1n
1d
1e
2.32±0.02
85.3±0.03
0.16±0.02
0.48±0.03
0.59±0.01
0.52±0.01
1o
1f
2a
0.26±0.03
0.72±0.02
0.4±0.04
1g
2b
1h
1i
2c
Ebs-R
0.74±0.02
1j
aThe substrate used was cefazolin and inhibitor concentration was varied between 0.1 and 100
μM.
2.3 Irreversible and time-dependent inhibition
In order to identify the inhibition mode of ebsulfurs, the hydrolysis of cefazolin (80 μM)
inhibited by 1g was real-time monitored at 262 nm on an Agilent UV8453 spectrometer (Fig. 2A).
The enzyme concentration was 50 nM, and the hydrolysis is monitored within 20 min. It is clearly
observed that the cefazolin was completely hydrolyzed within 200 s in the absence of inhibitor,
while the residual concentration of cefazolin increased gradually with the increase of 1g dose.
When the concentration of the inhibitor was 4 μM, only approximately twenty-five percent
antibiotics was hydrolyzed, implying that 1g is an irreversible inactivator of NDM-1. Also, we
monitored the residual activity of NDM-1 after different term of pre-incubate with 1g for 24 h
(Fig. 2B). It shows the activity of NDM-1 gradually decreases with pre-incubate time, and its
activity reached the lowest level after pre-incubate for about 20 h, indicating that the ebsulfur is
time-dependent inhibitor.
Fig. 2. Inhibition of cefazolin hydrolysis in the presence of NDM-1 enzyme by ebsulfur 1g at
various concentrations (A), the residual activity of NDM-1 (50 nM) after different term of
pre-incubation with 4 μM 1g (B), and the metal content of NDM-1 (8 μM) after dialysis versus

various concentrations of EDTA or 1g in 50 mM HEPES, pH 7.5 (C).
2.4 Equilibrium Dialysis
To further understand the interaction of ebsulfur and NDM-1, equilibrium dialysis studies were
conducted [17]. The metal content of protein samples was determined by using an IRIS
Advantage Inductively Coupled Plasma spectrometer with atomic emission spectroscopy
detection (ICP-AES). The protein samples were diluted to 8 μM with 50 mM Tris, pH 7.5. A
calibration curve with four standards and a correlation coefficient of greater than 0.999 was
generated using Zn reference solutions from Fisher Scientific. The following emission wavelengths
were chosen to ensure the lowest detection limits possible: Zn, 213.856 nm [26]. As expected,
the Zn (II) ion content of NDM-1 was significantly reduced when incubated with different
concentrations of 1g (Fig. 2C), when the concentration of 1g was 25 μM, about an equivalent
Zn(II) ions were removed from NDM-1, while when the inhibitor concentration continues to
increase, the Zn(II) ion content in enzyme did not decrease, revealing that one Zn(II) ion from
active site of each NDM-1 molecule was removed by the ebsulfur. In contrast, the Zn (II) ions of
NDM-1 were almost completely removed when incubated with 200 μM EDTA. These results are
consistent with the interaction mechanism of NDM-1 and ebsulfur that we proposed (Fig. 1).
2.5 MIC determination
The ability of ebsulfur to restore antimicrobial activity of cefazolin against E. coli producing
NDM-1 was investigated by determining the minimum inhibitory concentrations (MICs) in the
absence and presence of ebsulfurs 1a-o, 2a-c and Ebs-R as previously reported method [27]. E.
coli BL21 (DE3) harboring plasmids pET26b-NDM-1 was used to assess these inhibitors. The MIC
data (Table 2A) indicated that the ebsulfurs gradually increased antimicrobial effect of cefazolin
against the E. coli cells with an increasing inhibitor dose, and the highest dose of 1g, 1i and 1n
tested (16 μg/mL) resulted in MICs of cefazolin 256-fold decreased, respectively. However, the
inhibitors alone (16 μg/mL) did not inhibit cell growth, revealing that the efficacy of ebsulfurs to
restore antibiotic activity is due to their inhibition to the NDM-1 harbored in bacteria. Further, to
identify if ebsulfur exhibits activity against the bacteria expressing wild-type NDM-1, the capacity
of ebsulfurs to restore the antimicrobial activity of cefazolin against clinical strain E. coli EC08
producing NDM-1 was investigated. The clinical strain was from the Health Science Center at Xi’an
Jiaotong University (Xian, China). The MIC data (Table 2B) show that 1g-I, 1k, 1n and 2a-b
resulted in 2–32-fold reduction in MIC of cefazolin, revealing that ebsulfur also restored activity
of the antibiotic against clinical isolates producing NDM-1.
Table 2
Antibacterial activities (MICs, μg/mL) of cefazolin against E. coli BL21 strain expressing NDM-1 in
the absence and presence of ebsulfur at a concentration ranging from 2 to 16 μg/mL (A) , and
cefazolin against clinical strain E.coli EC08 producing NDM-1 in the absence and presence of
ebsulfur at a concentration ranging from 16 to 128 μg/mL (B).
A
Compd\conc cefazolin
256
+2
+4
+8
+16 Compd\conc cefazolin
128 256
+2
+4
32
+8
8
+16
2
256
256
128
128
1a
1k

256
256
256
256
256
256
256
256
256
256
256
128
128
256
64
128
128
64
64
256
8
128
64
32
16
128
4
64
64
16
16
64
1
256
256
256
256
256
256
256
256
128
128
128
256
128
128
128
256
32
64
8
8
4
8
1b
1c
1d
1e
1f
1l
1m
1n
32
2
1
128
64
64
64
128
64
8
64
2
1o
2a
16
64
32
2
1g
1h
1i
2b
128
64
8
8
2
16
32
2c
32
64
8
1
Ebs-R
256
16
4
1j
B
Compd\conc cefazolin
+16
+32
+64 +128 Compd\conc cefazolin
+16
+32
+64
+128
10000
10000
10000
10000
10000 5000 5000 625
10000 10000 5000 5000
10000 5000 5000 1250
10000 5000 1250 1250
10000
10000
10000
5000
5000
1250 312.5
5000
10000 10000 10000 5000
1g
1h
1i
1n
2a
2b
10000 10000 5000
1k
2.6 Real-time activity monitoring of NDM-1 in living E. coli cells
A simple and straightforward UV-Vis approach was employed for real-time activity monitoring
of NDM-1 in living bacterial cells in the presence and absence of 1g [28]. E. coli BL21 cells that
produce NDM-1 were used to assess the ebsulfur inhibitor. Inhibition of cefazolin hydrolysis in E.
coli cells harboring NDM-1 by ebsulfur 1g is shown in Fig. 3A. It is clearly observed that 1g
effectively inhibits NDM-1 in living cells, exhibiting an IC50 value of 35.1 μM on the cells (Fig. 3B),
which is much larger than the data (0.16 μM) from above assays in vitro using the same antibiotic
as enzyme substrate. This is probably due to the poor transport through porin or efflux pump,
which limits the inhibitor’s access to NDM-1 in the native environment of bacteria and
consequently results in reduced inhibition of cefazolin hydrolysis [28].
Fig. 3. Inhibition of cefazolin hydrolysis in NDM-1 harbored E. coli cells by 1g at various
concentrations (A), and IC50 measurement for E. coli BL21 cells (B). For each experiment, the E.
coli cells harboring NDM-1 (OD600 = 0.2) were first incubated with 1g for 10 min and 100 μM
cefazolin was subsequently added.

2.7 Fluorescent labelling NDM-1 with Ebs-R
To identify whether the binding of ebsulfurs to NDM-1 occurred, we constructed a fluorescent
ebsulfur Ebs-R by introducing Rhodamine B onto ebsulfur molecule and identified the binding of
protein-small molecule by SDS-PAGE. The scheme of labelling NDM-1 with Ebs-R is illustrated in
Fig. 4A. The purified NDM-1 was incubated with Ebs-R and the resulting sample was assessed by
SDS-PAGE. The labelled protein was observed by photographing with iPhone 6s under UV
illumination (365 nm) followed by staining with Coomassie Brilliant Blue (CBB). Comparison of
the gel pictures gave an indication of protein labelling [29]. When NDM-1 sample was mixed with
different concentrations of Ebs-R, a series of yellow protein bands with different fluorescence
intensities were observed, and the CBB staining suggested that the bands are corresponded to
NDM-1 (Fig. 4B, Lanes 6-11). In contrast, when NDM-1 was treated with Rhodamine B, no yellow
protein bands were observed (Fig. 4B, Lane 3), indicating that the Rhodamine B could not label
NDM-1. Besides, when adding 5% mercaptoethanol to the gel loading buffer, incubation of
NDM-1 with Ebs-R did not result in a surveyable yellow band (Fig. 4B, Lane 4), implying that the
disulfide bond formed between NDM-1 and Ebs-R was reduced by mercaptoethanol. Also, Ebs-R
exhibited a dose- and time-dependent labelling to NDM-1 (Fig. 4B and 4C), which are consistent
with those of the above irreversible and time-dependent inhibition studies. These
characterizations suggested that Ebs-R covalently bound to the cysteine at active site of NDM-1.
In vivo experiments were also carried out with cells grown in the presence of Ebs-R (Fig. 4D).
E. coli BL21 cells harboring NDM-1 (OD600 = 0.5) were incubated with 40 μM Ebs-R at 37 °C for 2 h,
the cells were washed with PBS through centrifugation for five times (10,400 x g, 10 min) to
remove any adsorbed Ebs-R on the cell surface, and then resuspended in PBS, lysed by
ultrasonication and the resulting supernatant was analyzed by SDS-PAGE. It was clearly observed
that Ebs-R can penetrate through the cell membrane and label the NDM-1 Inside cells. Also, to
assay whether the ebsulfur selectively label MβLs, Ebs-R (20 μM) was employed for labelling the
MβLs from B1-B3 subclasses. As shown in Fig. 4E, the NDM-1 and IMP-1 (B1) and ImiS (B2) were
labelled by Ebs-R in varying degrees, but L1 was not, suggesting that the ebsulfur, like ebselen
[20], targets the MβLs with cysteine residue at the active site(s), while the most prominent
fluorescent band to NDM-1 suggesting that the ebsulfur could be for labelling the protein that is
over-expressed in vivo [30].

Fig. 4. The gel electrophoresis analysis of NDM-1-labelled by the fluorescent reagent Ebs-R. (A)
Structure and mechanism of labelling NDM-1 with Ebs-R. (B) Lanes 1-3: 80 μM Ebs-R, 20 μM
NDM-1, 20 μM NDM-1 treated with 80 μM Rhodamine B; Lane 4: 20 μM NDM-1 and 80 μM
Ebs-R(containing 5% β-ME); Lanes 6-11: 20 μM NDM-1 was treated with different concentrations
of Ebs-R (final concentrations were 80, 40, 20, 10, 5, 2.5 μM, respectively). (C) Lane 1: control, 20
μM NDM-1; Lanes 3-10: 20 μM NDM-1 and 20 μM Ebs-R incubated for different time. (D) Left
Lane: 40 μM purified NDM-1 and 40 μM Ebs-R; middle Lane: 40 μM Ebs-R; right Lane: cell lysate
was treated with 40 μM Ebs-R. (E) Lanes 1-4: 20 μM MβLs (from left to right: NDM-1, IMP-1, ImiS
and L1) was treated with 20 μM Ebs-R.
2.8 Tracking NDM-1 and Real-time imaging in living bacterial cells with Ebs-R
To track the NDM-1 inside living bacterial cells, fluorescence imaging of the E. coli BL21 cells
harboring NDM-1 with Ebs-R was performed using Andor Dragonfly Confocal Microscope (Andor,
UK) [31]. Fig. 5 shows the images of NDM-1 E. coli cells incubated with Ebs-R in MHB for different
time. The interaction of Ebs-R with bacteria was imaged using microscopy technique, where a
clear visualization of the interation of ebsulfur and intracellular enzyme (NDM-1) has been
intuitively demonstrated. Ebs-R was found to associate at certain sites on and in the bacteria
depending on incubation time. Incubation with Ebs-R for 8 h, a high density of bacterial cells with
either membrane-associated or uniform localization of inhibitor throughout the cytosol of the
bacterial cells was observed (Fig. 5A). Incubation for 16 or 24 h, we observed that enzyme NDM-1
accumulated at the cell poles (Fig. 5B-C), possibly reflecting the formation of inclusion bodies [32].
Furthermore, incubation for 48 h, we observed that the enzyme finally looked like diffused
probably due to the lysed of bacterial cells (Fig. 5D). The real-time distribution change process of
intracellular recombinant protein NDM-1 is shown in Fig. 5E-H.

Fig. 5. Confocal microscopic real-time imaging of E. coli BL21 cells expressing NDM-1 after treated
with Ebs-R (40 μM) for different time (A-D), and the real-time distribution change process of the
recombinant protein NDM-1 inside living E. coli BL21 cells (E-H).
2.9 Uptake of ebsulfurs by bacterial cells
Flow cytometer would be more appreciate technique for measurement of fluorescence
intensity of only those labelled inhibitors that are internalized into the cells [33]. NDM-1 E. coli
cells were incubated with Ebs-R at concentration of 0, 10, 20 and 40 μM, respectively. After a
certain period of incubation, the cells were repeatedly washed with PBS (to remove any Ebs-R
adsorbed on the surface of cells) and investigated by analysis using flow cytometer. The results of
cellular uptake of inhibitor into NDM-1 E. coli cells are shown in Fig. 6. It shows that the ratio of
detectable labelled E. coli cells increase from 9.8 to 76.6% with the increase of Ebs-R dose from
10 to 40 μM, exhibiting a dose-dependent cellular uptake, in comparison with the controls,
revealing that the ebsulfur molecules can bond stably with bacteria.

Fig. 6. Flow cytometer analysis of NDM-1 E. coli cells incubation with Ebs-R at different
concentrations. NDM-1 E. coli cells were incubated with 0 (A), 10 (B), 20 (C), and 40 µM Ebs-R (D).
2.10 Cytotoxicity assay
For biomedical applications, the potential toxicity of ebsulfur inhibitors is a major concern. As
the representatives of ebsulfurs, 1g, 2a and Ebs-R were subjected to a cytotoxicity assay using
mouse fibroblast cells (L929) treated with different inhibitor concentrations (2.5, 5, 10, 20, 40,
and 80 μM). As shown in Fig. 7, no compounds except 1g tested affected viability of the L929
mouse fibroblastic cells at a concentration up to 40 μM, indicating that these compounds have
low cytotoxicity and may be used clinical purposes.
Fig. 7. Cytotoxicity assay of ebsulfurs 1g, 2a and Ebs-R on L929 cells at an inhibitor concentration
ranging from 2.5 to 80 µM. DMSO was used as a negative control, and cell viability was measured

by MTT assay.
3. Conclusion
Based on the residue Cys221 at active site of NDM-1, which was expected to be bound by the
sulfur-containing molecules through forming a disulfide bond in inhibiting the enzyme as
1
13
proposed in Fig. 1, eighteen ebsulfur derivatives were synthesized, characterized by H, C NMR
and HRMS, and evaluated in vitro and in in vivo. All of these ebsulfurs gained inhibited NDM-1,
especially 1g, exhibiting an IC50 value of 0.16 μM. Also, these ebsulfurs effectively restored the
antibacterial activities of cefazolin against E. coli expressing NDM-1, and the best effect was
observed to be from 1g, 1i and 1n, resulting in an 256-fold reduction in MIC of the antibiotic at a
concentration of 16 μg/mL. Inhibitory and equilibrium dialysis studies suggested that the ebsulfur
time-dependently and covalently binds to NDM-1, which has been examined with fluorescent
Ebs-R constructed through SDS-PAGE analysis. Real-time activity monitoring of the NDM-1 inside
living E. coli BL21 cells indicated that 1g effectively inhibited the enzyme with an IC50 value of
35.1 μM. Tracking the NDM-1 in living E. coli cells through labelling it with Ebs-R by confocal
microscopic imaging showed that the target initially distributed uniformly in the cytosol of
bacteria cells, then accumulated in formation of inclusion bodies at the cell poles, and finally
dispersed probably due to the lysed of bacterial cells. The ebsulfur scaffold proposed here is
valuable for development of the covalent NDM-1 inhibitors and also for labelling the target in
vitro and in vivo.
4. Experimental
4.1 Materials
The reagents were purchased from Aladdin (Shanghai) Trading Co., Ltd. All other starting
materials were purchased from commercial sources and purified using standard methods.
Analytical thin layer chromatography (TLC) was carried out on silica gel F254 plates with
visualization by ultraviolet radiation. 1H and 13C NMR spectra were recorded on a Bruker 400 MHz
NMR spectrometer. Chemical shifts are given in parts per million (ppm) on the delta scale. The
peak patterns are reported as singlet (s), doublet (d), triplet (t), quartet (q), and multiplet (m).
The spectra were recorded with TMS as internal standard. Coupling constants (J) were reported
in Hertz (Hz). Mass spectra were obtained on a micro TOF-Q (BRUKER) mass spectrometer.
Activity evaluation of inhibitors was performed on an Agilent 8453 UV-Vis spectrometer. Confocal
fluorescence images were collected on an Andor Dragonfly Confocal Microscope (Andor, UK).
4.2 Chemistry
4.2.1 Preparation of ebsulfurs 1a-o following procedure A:
2-(Chlorosulfanyl)benzoyl chloride (b) has been obtained by 2,2'-disulfanediyldibenzoic acid
(a) react with SOCl2 according to the procedure described [22]. To a stirred and ice-cooled
mixture of primary amine (3 mmol) and Et3N (0.45 mL, 3.2 mmol ) in anhydrous CH2Cl2 (40 mL), 2,
2'-dithiobenzoyl chloride (b, 0.50 g, 1.5 mmol ) in anhydrous CH2Cl2 (20 mL) was added dropwise
over 20 min. The reaction mixture was warmed-up to RT, stirred for 12 h, and washed with aq. sat.
NaHCO3 (30 mL) and H2O (30 mL). The organic layer was dried over anhydrous Na2SO4,
concentrated in vacuo, and the resulting residue was purified by flash column chromatography

(SiO2, EtOAc:hexane/1:2) to afford the final compounds 1a-o [25].
4.2.1.1. 1,2-benzisothiazoline-3-one (1a)
Compound 1a was prepared according to standard procedure A by using b and ammonia
aqueous solution, obtained a white solid in 80% yield. 1H NMR (400 MHz, CDCl3) δ 8.16–8.08 (m,
13
1H), 7.74–7.65 (m, 2H), 7.52–7.44 (m, 1H). C NMR (101 MHz, DMSO) δ 165.26, 147.86, 130.51,
125.31, 125.17, 124.59, 121.93. HRMS (ESI) m/z: 152.0167 (Calcd. for [M+H+]+: 152.0165 m/z).
4.2.1.2. 2-methylbenzo[d]isothiazol-3(2H)-one (1b)
Compound 1b was prepared according to standard procedure A by using b and Methylamine
aqueous solution, obtained a white solid in 75% yield. 1H NMR (400 MHz, CDCl3) δ 8.02 (d, J = 7.9
13
Hz, 1H), 7.64–7.50 (m, 2H), 7.43–7.35 (m, 1H), 3.43 (s, 3H). C NMR (101 MHz, CDCl3) δ 165.52,
139.95, 131.70, 126.51, 125.47, 124.35, 120.25, 30.44. HRMS (ESI) m/z: 188.0132 (Calcd. for
[M+Na+]+: 188.0141 m/z).
4.2.1.3. 2-propylbenzo[d]isothiazol-3(2H)-one (1c)
Compound 1c was prepared according to standard procedure A by using b and Propylamine,
obtained a yellow solid in 60% yield. 1H NMR (400 MHz, CDCl3) δ 7.98 (d, J = 7.9 Hz, 1H), 7.52 (dt,
J = 10.2, 4.3 Hz, 2H), 7.34 (ddd, J = 8.0, 6.7, 1.4 Hz, 1H), 3.88–3.76 (m, 2H), 1.82–1.67 (m, 2H),
13
0.98–0.89 (m, 3H). C NMR (101 MHz, CDCl3) δ 165.29, 140.12, 131.59, 126.50, 125.34, 124.73,
120.33, 45.44, 22.86, 11.10. HRMS (ESI) m/z: 194.0640 (Calcd. for [M+H+]+: 194.0634 m/z).
4.2.1.4. 2-allylbenzo[d]isothiazol-3(2H)-one (1d).
Compound 1d was prepared according to standard procedure A by using b and Allylamine,
1
obtained a yellow solid in 66% yield. H NMR (400 MHz, CDCl3) δ 8.05 (d, J = 7.9 Hz, 1H),
7.66–7.51 (m, 2H), 7.40 (t, J = 7.4 Hz, 1H), 5.94 (ddt, J = 16.4, 10.1, 6.1 Hz, 1H), 5.40–5.26 (m, 2H),
4.51 (d, J = 6.1 Hz, 2H). 13C NMR (101 MHz, CDCl3) δ 165.15, 140.40, 132.36, 131.81, 126.74,
125.48, 124.67, 120.39, 119.32, 46.20. HRMS (ESI) m/z: 192.0470 (Calcd. for [M+H+]+: 192.0477
m/z).
4.2.1.5. 2-isobutylbenzo[d]isothiazol-3(2H)-one (1e).
Compound 1e was prepared according to standard procedure A by using b and isobutylamine,
1
obtained a yellow solid in 65% yield. H NMR (400 MHz, CDCl3) δ 8.00 (d, J = 7.9 Hz, 1H), 7.55
(ddd, J = 14.1, 10.5, 4.4 Hz, 2H), 7.35 (ddd, J = 8.0, 6.9, 1.3 Hz, 1H), 3.68 (d, J = 7.4 Hz, 2H),
2.19–2.00 (m, 1H), 0.95 (d, J = 6.7 Hz, 6H). 13C NMR (101 MHz, CDCl3) δ 165.49, 140.18, 131.62,
126.59, 125.33, 124.61, 120.25, 51.09, 28.98, 19.87. HRMS (ESI) m/z: 230.0618 (Calcd. for
[M+Na+]+: 230.0610 m/z).
4.2.1.6. 2-(2-hydroxyethyl)benzo[d]isothiazol-3(2H)-one (1f)
Compound 1f was prepared according to standard procedure A by using b and 2-Aminoethanol,
obtained a yellow solid in 55% yield. 1H NMR (400 MHz, CDCl3) δ 8.08–7.99 (m, 1H), 7.62 (ddd, J =
8.2, 7.1, 1.2 Hz, 1H), 7.55 (d, J = 8.1 Hz, 1H), 7.41 (ddd, J = 8.0, 7.1, 1.0 Hz, 1H), 4.05 (dt, J = 6.4,
1.7 Hz, 2H), 4.01–3.93 (m, 2H). 13C NMR (101 MHz, CDCl3) δ 166.41, 140.72, 131.97, 126.63,
125.60, 124.22, 120.23, 62.00, 47.49. HRMS (ESI) m/z: 218.0239 (Calcd. for [M+Na+]+:218.0246

m/z).
4.2.1.7. 2-phenylbenzo[d]isothiazol-3(2H)-one (1g)
Compound 1g was prepared according to standard procedure A by using b and Aniline,
1
obtained a yellow solid in 85% yield. H NMR (400 MHz, DMSO) δ 8.06 (d, J = 8.1 Hz, 1H),
13
8.00–7.93 (m, 1H), 7.82–7.67 (m, 3H), 7.60–7.46 (m, 3H), 7.42–7.32 (m, 1H). C NMR (101 MHz,
DMSO) δ 163.75, 140.55, 137.64, 133.05, 129.92, 127.44, 126.64, 126.48, 124.91, 124.80, 122.40.
HRMS (ESI) m/z: 228.0472 (Calcd. for [M+H+]+: 228.0477 m/z).
4.2.1.8. 2-(p-tolyl)benzo[d]isothiazol-3(2H)-one (1h)
Compound 1h was prepared according to standard procedure A by using b and p-Toluidine,
1
obtained a yellow solid in 83% yield. H NMR (400 MHz, CDCl3) δ 8.09 (d, J = 7.9 Hz, 1H), 7.64
(ddd, J = 8.2, 7.2, 1.2 Hz, 1H), 7.59–7.51 (m, 3H), 7.47–7.39 (m, 1H), 7.31–7.21 (m, 2H), 2.38 (s,
3H). 13C NMR (101 MHz, CDCl3) δ 164.07, 139.95, 137.23, 134.52, 132.25, 129.97, 127.12, 125.76,
124.83, 124.75, 120.13, 21.17. HRMS (ESI) m/z: 242.0645(Calcd. for [M+H+]+: 242.0634 m/z).
4.2.1.9. 2-(4-chlorophenyl)benzo[d]isothiazol-3(2H)-one (1i)
Compound 1i was prepared according to standard procedure A by using b and
p-Chlorophenylamine, obtained a yellow solid in 82% yield. 1H NMR (400 MHz, CDCl3) δ 8.10 (d, J
13
= 7.9 Hz, 1H), 7.73–7.64 (m, 3H), 7.59 (d, J = 8.1 Hz, 1H), 7.51–7.39 (m, 3H). C NMR (101 MHz,
CDCl3) δ 164.13, 139.64, 135.82, 132.61, 132.52, 129.49, 127.28, 126.01, 125.65, 124.60, 120.14.
HRMS (ESI) m/z: 283.9899 (Calcd. for [M+Na+]+: 283.9907 m/z).
4.2.1.10. Methyl 4-(3-oxobenzo[d]isothiazol-2(3H)-yl)benzoate (1j)
Compound 1j was prepared according to standard procedure A by using b and 4-aminobenzoic
acid methyl ester, obtained a yellow solid in 78% yield. 1H NMR (400 MHz, CDCl3) δ 8.19–8.07 (m,
3H), 7.93–7.85 (m, 2H), 7.68 (dd, J = 7.1, 1.1 Hz, 1H), 7.60 (d, J = 8.1 Hz, 1H), 7.50–7.42 (m, 1H),
3.94 (s, 3H). 13C NMR (101 MHz, CDCl3) δ 166.28, 164.18, 141.65, 139.47, 132.85, 130.85, 127.78,
127.37, 126.09, 124.84, 122.90, 120.13, 52.27. HRMS (ESI) m/z: 308.0349 (Calcd. for [M+Na+]+:
308.0351 m/z).
4.2.1.11. 2-(furan-2-ylmethyl)benzo[d]isothiazol-3(2H)-one (1k)
Compound 1k was prepared according to standard procedure A by using b and
1
furan-2-ylmethanamine, obtained a yellow solid in 75% yield. H NMR (400 MHz, CDCl3) δ
8.10–8.01 (m, 1H), 7.59 (ddd, J = 8.3, 7.1, 1.2 Hz, 1H), 7.54–7.48 (m, 1H), 7.45–7.34 (m, 2H), 6.43
(dd, J = 3.2, 0.5 Hz, 1H), 6.37 (dd, J = 3.2, 1.9 Hz, 1H), 5.06 (s, 2H). 13C NMR (101 MHz, CDCl3) δ
165.07, 149.28, 143.20, 140.46, 131.93, 126.83, 125.51, 124.42, 120.39, 110.64, 109.82, 40.08.
HRMS (ESI) m/z: 232.0418 (Calcd. for [M+H+]+: 232.0426 m/z).
4.2.1.12. 2-benzylbenzo[d]isothiazol-3(2H)-one (1l)
Compound 1l was prepared according to standard procedure A by using b and Benzylamine,
1
obtained a yellow solid in 50% yield. H NMR (400 MHz, DMSO) δ 7.93 (dd, J = 13.5, 8.0 Hz, 2H),
7.73–7.65 (m, 1H), 7.46 (t, J = 7.5 Hz, 1H), 7.41–7.27 (m, 5H), 5.03 (s, 2H). 13C NMR (101 MHz,
DMSO) δ 164.86, 141.03, 137.40, 132.42, 129.17, 128.40, 126.15, 122.47, 46.79. HRMS (ESI) m/z:

264.0458 (Calcd. for [M+Na+]+: 264.0459 m/z).
4.2.1.13. 2-(3-phenylpropyl)benzo[d]isothiazol-3(2H)-one (1m)
Compound 1m was prepared according to standard procedure A by using b and
1
3-phenylpropan-1-amine, obtained a yellow solid in 52% yield. H NMR (400 MHz, CDCl3) δ 8.09
(d, J = 7.8 Hz, 1H), 7.70–7.55 (m, 2H), 7.44 (t, J = 7.2 Hz, 1H), 7.34 (t, J = 7.4 Hz, 2H), 7.30–7.20 (m,
3H), 3.98 (t, J = 7.0 Hz, 2H), 2.75 (t, J = 7.7 Hz, 2H), 2.21–2.07 (m, 2H). 13C NMR (101 MHz, CDCl3)
δ 165.40, 140.92, 140.14, 131.74, 128.50, 128.42, 126.63, 126.13, 125.50, 124.74, 120.39, 43.49,
32.80, 31.18. HRMS (ESI) m/z: 270.0955 (Calcd. for [M+H+]+: 270.0953 m/z).
4.2.1.14. 2-(2-aminophenyl)benzo[d]isothiazol-3(2H)-one (1n)
Compound 1n was prepared according to standard procedure A by using b and
1,2-diaminobenzene, obtained a yellow solid in 56% yield. 1H NMR (400MHz, DMSO) δ 8.06–7.98
(m, 1H), 7.98–7.88 (m, 1H), 7.80–7.69 (m, 1H), 7.58–7.44 (m, 1H), 7.24–7.09 (m, 2H), 6.92–6.80
(m, 1H), 6.72–6.58 (m, 1H), 5.24 (s, 2H). 13C NMR (101 MHz, DMSO) δ 164.25, 146.59, 142.21,
132.41, 130.27, 130.11, 126.53, 125.89, 124.61, 122.32, 120.74, 116.56. HRMS (ESI) m/z:
243.0579 (Calcd. for [M+H+]+: 243.0586 m/z).
4.2.1.15. 2,2'-(ethane-1,2-diyl)bis(benzo[d]isothiazol-3(2H)-one) (1o)
Compound 1o was prepared according to standard procedure A by using b and
1
ethylenediamine, obtained a yellow solid in 75% yield. H NMR (400 MHz, DMSO) δ 7.93 (d, J =
8.1 Hz, 2H), 7.84 (d, J = 7.8 Hz, 2H), 7.66 (dd, J = 11.2, 4.0 Hz, 2H), 7.42 (t, J = 7.4 Hz, 2H), 4.16 (s,
4H). 13C NMR (101 MHz, DMSO) δ 165.10, 141.37, 132.33, 126.10, 125.96, 124.23, 122.47, 42.94.
HRMS (ESI) m/z: 351.0225 (Calcd. for [M+Na+]+: 351.0232 m/z).
4.2.2 Preparation of ebsulfurs 2a-c following procedure B:
To a solution of triphosgene (2.96 g, 10 mmol) in DCM (20 ml) was added dropwise to
primary amine c (10 mmol) in DCM (20 ml) followed by dropwise addition of triethylamine (3 ml)
in DCM (10 ml). The solvent was removed on a rotary evaporator. The resulting residue (d) was
dissolved in DCM (20 ml), and 1a (1.51 g, 10 mmol) in THF (20 ml) was added. After the mixture
was refluxed for 30 min, the solvent was removed on a rotary evaporator, the residue was
dissolved in acetone (30 ml) and mixed with water (30 ml), the resulting precipitate was collected
on a funnel by vacuum filtration and washed with water–acetone (1:1, 4×5 ml) to afford the final
compounds 2a-c [24].
4.2.2.1. 3-oxo-N-propylbenzo[d]isothiazole-2(3H)-carboxamide (2a)
Compound 2a was prepared according to standard procedure B by using 1a and
Propylamine, obtained a white solid in 76% yield. 1H NMR (400 MHz, CDCl3) δ 8.90 (s, 1H), 8.02 (d,
J = 7.9 Hz, 1H), 7.77–7.64 (m, 1H), 7.58 (d, J = 8.1 Hz, 1H), 7.49–7.36 (m, 1H), 3.42 (dd, J = 12.9,
7.0 Hz, 2H), 1.73–1.61 (m, 2H), 1.01 (t, J = 7.4 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ 165.10,
151.13, 140.84, 133.75, 127.18, 125.76, 125.13, 120.48, 42.35, 22.84, 11.36. HRMS (ESI) m/z:
259.0518 (Calcd. for [M+Na+]+: 259.0517 m/z).
4.2.2.2. N-butyl-3-oxobenzo[d]isothiazole-2(3H)-carboxamide (2b)

Compound 2b was prepared according to standard procedure B by using 1a and Butylamine,
1
obtained a white solid in 78% yield. H NMR (400 MHz, CDCl3) δ 8.88 (s, 1H), 8.02 (d, J = 7.9 Hz,
1H), 7.75–7.65 (m, 1H), 7.58 (d, J = 8.1 Hz, 1H), 7.42 (t, J = 7.5 Hz, 1H), 3.45 (dd, J = 12.8, 7.0 Hz,
2H), 1.63 (dt, J = 19.7, 7.2 Hz, 2H), 1.43 (dq, J = 14.6, 7.3 Hz, 2H), 0.96 (t, J = 7.3 Hz, 3H). 13C NMR
(101 MHz, CDCl3) δ 165.10, 151.10, 140.86, 133.75, 127.18, 125.76, 125.15, 120.49, 40.38, 31.58,
20.03, 13.70. HRMS (ESI) m/z: 273.0698 (Calcd. for [M+Na+]+:273.0702 m/z).
4.2.2.3. N-(furan-2-ylmethyl)-3-oxobenzo[d]isothiazole-2(3H)-carboxamide (2c)
Compound 2c was prepared according to standard procedure B by using 1a and
1
furan-2-ylmethanamine, obtained a white solid in 69% yield. H NMR (400 MHz, CDCl3) δ 9.20 (s,
1H), 8.00 (d, J = 7.9 Hz, 1H), 7.74–7.64 (m, 1H), 7.57 (d, J = 8.1 Hz, 1H), 7.47–7.33 (m, 2H), 6.33 (dt,
13
J = 6.1, 2.5 Hz, 2H), 4.63 (d, J = 5.7 Hz, 2H). C NMR (101 MHz, CDCl3) δ 165.10, 151.01, 150.53,
142.49, 140.89, 133.90, 127.28, 125.83, 124.88, 120.49, 110.45, 107.86, 37.46. HRMS (ESI) m/z:
297.0296 (Calcd. for [M+Na+]+:297.0304 m/z).
4.2.3 Preparation of ebsulfurs Ebs-R following procedure C:
To a stirred solution of 1f (3.0 equiv) in DCM was added DCC (1.3 equiv), DMAP (0.13 equiv)
and Rhodamine B (1.0 equiv) at 0 °C. The reaction mixture was stirred for 10 min at 0 °C, and
then overnight at room temperature, the solvent was removed in vacuum and extracted with
DCM (3 × 10 mL). The combined organic extracts were dried over anhydrous Na2SO4, the solvent
removed was under reduced pressure, and the crude product was purified by flash column
chromatography on silica gel (CHCl3/CH3OH, 9:1) to afford the final compound Ebs-R, as dark
purple crystals with metallic glass, yield 50%. 1H NMR (400 MHz, CDCl3) δ 8.38 (dd, J = 7.7, 1.2 Hz,
1H), 8.00 (d, J = 7.8 Hz, 1H), 7.87–7.75 (m, 3H), 7.68–7.62 (m, 1H), 7.41 (t, J = 7.5 Hz, 1H), 7.30 (dd,
J = 7.5, 1.1 Hz, 1H), 7.08 (d, J = 9.5 Hz, 2H), 6.92 (dd, J = 9.5, 2.4 Hz, 2H), 6.76 (d, J = 2.4 Hz, 2H),
4.36 (t, J = 5.1 Hz, 2H), 4.12 (t, J = 5.1 Hz, 2H), 3.63 (q, J = 7.1 Hz, 8H), 1.32 (t, J = 7.1 Hz, 12H). 13C
NMR (101 MHz, CDCl3) δ 165.58, 164.53, 158.67, 157.67, 155.48, 140.78, 133.87, 133.44, 132.10,
131.67, 131.28, 130.73, 130.26, 129.09, 126.35, 125.52, 123.85, 121.15, 114.32, 113.44, 96.23,
63.74, 46.13, 42.65, 12.67. HRMS (ESI) m/z: 620.2569 (Calcd. for [M-Cl- ]+:620.2577 m/z).
4.3 Over-expression and purification of NDM-1
NDM-1 was overexpressed and purified as previously described [26]. Truncated wild type
NDM-1 (Q37-R270) was cloned into pET-26b vector to facilitate protein folding and purification.
The recombinant plasmid was transformed in E. coli BL21 (DE3) and expressed with a molecular
mass of approximately 25.1 kDa. The cells were plated on LB-agar plates containing 25 μg/mL
kanamycin, a single colony was used to inoculate 50 mL of LB containing 25 μg/mL kanamycin.
After the preculture grew overnight at 37 C , 10 mL of overnight culture cells in LB was used to
inoculate 4 × 1 L of LB containing 25 μg/mL kanamycin. The cells were allowed to grow at 37 °C
with shaking until the cells reached an optical density at 600 nm of 0.6-0.8. Protein production
was induced with 1 mM IPTG, and the cells were shaken at 25 °C for 3 h. The cells were collected
by centrifugation (30 min at 8,275 × g) and resuspended in 25 mL of 30 mM Tris, pH 8.0,
containing 500 mM NaCl. The cells were lysed by ultrasonication, and the cell debris was
separated by centrifugation (30 min at 32,583 × g). The cleared supernatant was dialyzed versus
30 mM Tris, pH 8.0, containing 100 μM ZnCl2 for 36 h at 4 °C, centrifuged (25 min at 32,583 × g)

to remove insoluble matter, and loaded onto an equilibrated Q-Sepharose column. Bound
proteins were eluted with a 0-500 mM NaCl gradient in 30 mM Tris, pH 8.0, containing 100 μM
ZnCl2 at 2 mL/min. Fractions (2 mL) containing NDM-1 were pooled and concentrated with an
Amicon ultrafiltration cell equipped with an YM-10 membrane. The crude protein NDM-1 was run
through a G75 column and eluted with 30 mM Tris, pH 8.0, containing 200 mM NaCl. Protein
purity was ascertained by SDS-PAGE and protein concentration was determined using Beer’s law
and an extinction coefficient of 27,960 M−1cm−1 at 280 nm.
4.4 Determination of IC50 values
The inhibitor concentration causing 50% decrease of enzyme activity (IC50) was determined at
37 °C using cefazolin as substrate of NDM-1. Inhibitors were dissolved in a small volume of DMSO
and then diluted with 30 mM Tris, pH 7.5 for NDM-1. The final concentrations of DMSO in
inhibition experiments were below 0.1%, control experiments verified that the 0.1% DMSO had
no inhibitory activity against the MβLs tested. All inhibitors were assayed at six different
concentrations ranging from 0 to 25 μM, and substrate concentration was 50 μM. Enzyme and
inhibitor were pre-incubated for 20 h before adding cefazolin. The hydrolysis of cefazolin was
monitored at 262 nm on an Agilent UV8453 spectrometer, and the hydrolytic rates were
determined in triplicate. The IC50 values for all analyzed compounds were calculated by plotting
the average percentage inhibition against inhibitor concentration and fitting of the data using
GraphPad Prism 5.0 in a logistic dose-response model (see Supporting Information for IC50
measurements).
4.5 Equilibrium Dialysis
NDM-1 (final concentration 8 μM) in 5 mL of 50 mM HEPES at pH 7.5 was mixed with 1g at a
concentration between 0 and 200 μM. After incubation for 2 h, the solutions were dialyzed
versus 500 mL of metal-free 50 mM HEPES at pH 7.5 for 3×4 h (dialysis tubing MWCO 8000-14000,
Bioreagent). The Zn(II) ion content in the resulting NDM-1 samples was determined using
inductively coupled plasma with atomic emission spectroscopy (ICP-AES, IRIS Advantage). A
calibration curve with four standards and a correlation coefficient of greater than 0.999 was
generated using Zn reference solutions from Fisher Scientific. The emission wavelength was set to
213.856 nm, as previously described [17, 26].
4.6 Determination of MIC
Single colony of E. coli BL21 (DE3) containing plasmids pET26b-NDM-1 on LB agar plates was
transferred to 5 mL of Mueller-Hinton (MH) liquid medium and grown at 37 °C overnight. The
bacterial cells were collected by centrifugation (4,150 x g, 10 min). After discarding the
supernatant, the pelleted cells were re-suspended in MH medium and diluted to an OD600 of 0.5.
MIC values were determined by using the Clinical and Laboratory Standards Institute (CLSI)
macrodilution (tube) broth method [27]. The MIC was interpreted as the lowest concentration of
the drug that completely inhibited the visible growth of bacteria after incubating plates for at
least 16 h at 37 °C. Each inhibitor was tested in triplicate in at least two independent experiments
and the highest MIC value was reported.

4.7 Identification of labelled protein by SDS-PAGE
Equal volumes of NDM-1 and Ebs-R were pre-incubated in 10 mM Tris-HCl, pH 7.5, at 25 °C for
2 h. labelled protein was solubilized in 2 × SDS gel loading buffer (100 mM Tris-HCl buffer (pH 6.8),
2.5% SDS, 20% glycerol or 100 mM Tris-HCl buffer (pH 6.8), 2.5% SDS, 20% glycerol and 10%
mercaptoethanol ) and resolved by SDS-PAGE [34]. The gel was visualized under UV light at 365
nm, and then was stained with Coomassie Brilliant Blue. All images of the gels were captured
using an iPhone 6s.
4.8 Fluorescence imaging and flow cytometer analysis
NDM-1 harbored E.coli cells (OD600 = 0.5) were incubated with compound Ebs-R (0-40 μM) in
tubes at 37 °C for different time. After incubation, the cells were repeatedly washed with PBS for
five times (10,400 x g, 5 min), to remove any adsorbed Ebs-R on surface of the cell. The cells were
re-suspended in PBS to give the final samples for confocal microscopic imaging and flow
cytometer analysis [20].
4.9 Cytotoxicity assays
A cytotoxicity of inhibitors 1g, 2a and Ebs-R to mouse fibro-blast cells (L929) was assayed as
pervious described [35]. The cells with a density of 1.0 × 104 cells/well in 100 μL of culture
medium were seeded into 96-well plates and maintained for 24 h. Then solutions of inhibitors 1g,
2a and Ebs-R with work concentrations (2.5, 5, 10, 20, 40, and 80 μM) were added to 96-well
plates, respectively, and incubated for another 48 h. Six wells containing only cells suspended in a
mixture of 99 μL of complete medium and 1 μL of DMSO were used as the control for
investigating cell-viability. Three wells containing only the complete medium were used as the
blank control. Following that, the medium was removed, and 100 μL of fresh culture medium and
10 μL of CCK8 were added to each well. After incubation for 4 h, the 96-well plates were then
vigorously shaken to solubilize the formed product and the absorbance at a wavelength of 490
nm was read on a Microplate Reader and analyzed. All experiments were conducted in triplicate.
Acknowledgments
This work was supported by grants 81361138018 and 21572179 (to K. W. Y.) from the
National Natural Science Foundation of China.
Conflicts of interest
None
References
[1] S. M. Drawz, R. A. Bonomo, Three decades of β-lactamase inhibitors, Clin. Microbiol. Rev. 23 (2010)
160-201.
[2] A. L. Demain, R. P. Elander, The beta-lactam antibiotics: Past, present, and future, Antonie Van
Leeuwenhoek. 75 (1999) 5-19.
[3] A. A. Medeiros, Evolution and dissemination of beta-lactamases accelerated by generations of
beta-lactam antibiotics, Clin. Infect. Dis. 24 Suppl 1 (1997) S19-45.
[4] R. P. Ambler, The structure of β-lactamases, Philosophical Transactions of the Royal Society of

London:series B, Biological Sciences. 289 (1980) 321-331.
[5] H. Z, P. G, B. B, C. MW, Role of the Zn1 and Zn2 sites in metallo-beta-lactamase L1, JACS. 130 (2008)
14207-14216.
[6] M. W. Crowder, J. Spencer, A. J. Vila, Metallo-beta-lactamases: Novel weaponry for antibiotic
resistance in bacteria, Acc. Chem. Res. 39 (2006) 721-728.
[7] G. G, G.-S. I, B. C, et al., Update of the standard numbering scheme for class B β-lactamases,
Antimicrob. Agents Chemother. 48 (2004) 2347-2349.
[8] M. R. Meini, L. I. Llarrull, A. J. Vila, Overcoming differences: The catalytic mechanism of
metallo‐β‐lactamases, FEBS Lett. 589 (2016) 3419-3432.
[9] P. W. Groundwater, S. Xu, F. Lai, et al., New delhi metallo-β-lactamase-1: Structure, inhibitors and
detection of producers, Future Med. Chem. 8 (2016) 993-1012.
[10] A. U. Khan, L. Maryam, R. Zarrilli, Structure, genetics and worldwide spread of new delhi
metallo-β-lactamase (NDM-1): A threat to public health, BMC Microbiol. 17 (2017) 101.
[11] D. Yong, M. A. Toleman, C. G. Giske, et al., Characterization of a new metallo-β-lactamase gene,
blandm-1, and a novel erythromycin esterase gene carried on a unique genetic structure in
klebsiella pneumoniae sequence type 14 from india, Antimicrob. Agents Chemother. 53 (2009)
5046-5054.
[12] T. R. Walsh, Emerging carbapenemases: A global perspective, Int. J. Antimicrob. Agents. 36 (2010)
S8-S14.
[13] D. T. King, L. J. Worrall, R. Gruninger, N. C. J. Strynadka, New delhi metallo-β-lactamase: Structural
insights into β-lactam recognition and inhibition, JACS. 134 (2012) 11362-11365.
[14] Y. Guo, J. Wang, G. Niu, et al., A structural view of the antibiotic degradation enzyme NDM-1 from
a superbug, Protein & Cell. 2 (2011) 384.
[15] K. AM, R. SA, W. W, et al., Aspergillomarasmine a overcomes metallo-β-lactamase antibiotic
resistance, Nature. 510 (2014) 503-506.
[16] L. E. Horsfall, G. Garau, B. M. R. Liénard, et al., Competitive inhibitors of the cpha
metallo-β-lactamase from aeromonas hydrophila, Antimicrob. Agents Chemother. 51 (2007)
2136-2142.
[17] A. Y. Chen, P. W. Thomas, A. C. Stewart, et al., Dipicolinic acid derivatives as inhibitors of new delhi
metallo-β-lactamase-1, J. Med. Chem. 60 (2017) 7267-7283.
[18] J. Brem, R. Cain, S. Cahill, et al., Structural basis of metallo-β-lactamase, serine-β-lactamase and
penicillin-binding protein inhibition by cyclic boronates, Nature Communications. 7 (2016)
12406.
[19] J. Chiou, S. Wan, K. F. Chan, et al., Ebselen as a potent covalent inhibitor of new delhi
metallo-β-lactamase (NDM-1), Chem. Commun. 51 (2015) 9543-9546.
[20] C. Chen, Y. Xiang, K. W. Yang, et al., A protein structure-guided covalent scaffold selectively targets
the B1 and B2 subclass metallo-β-lactamases, Chem. Commun. 54 (2018) 4802-4805.
[21] H. M. Zhang, Q. Hao, Crystal structure of NDM-1 reveals a common β-lactam hydrolysis
mechanism, Faseb Journal Official Publication of the Federation of American Societies for
Experimental Biology. 25 (2011) 2574-2582.
[22] J. Lu, S. K. Vodnala, A. L. Gustavsson, et al., Ebsulfur is a benzisothiazolone cytocidal inhibitor
targeting the trypanothione reductase of trypanosoma brucei, J. Biol. Chem. 288 (2013)
27456-27468.
[23] H. M. Gordhan, S. L. Patrick, M. I. Swasy, et al., Evaluation of substituted ebselen derivatives as

potential trypanocidal agents, Biorg. Med. Chem. Lett. 27 (2016) 537-541.
[24] D. Liu, Z. Tian, Z. Yan, et al., Design, synthesis and evaluation of 1,2-benzisothiazol-3-one
derivatives as potent caspase-3 inhibitors, Biorg. Med. Chem. 21 (2013) 2960-2967.
[25] H. X. Ngo, S. K. Shrestha, K. D. Green, S. Garneau-Tsodikova, Development of ebsulfur analogues
as potent antibacterials against methicillin-resistant staphylococcus aureus, Biorg. Med. Chem.
24 (2016) 6298-6306.
[26] H. Yang, M. Aitha, A. M. Hetrick, et al., Mechanistic and spectroscopic studies of
metallo-β-lactamase NDM-1, Biochemistry. 51 (2012) 3839-3847.
[27] F. R. Cockerill, Methods for dilution antimicrobial susceptibility tests for bacteria that grow
aerobically : Approved standard. Clinical and Laboratory Standards Institute: 2000.
[28] K. W. Yang, Y. Zhou, Y. Ge, Y. Zhang, Real-time activity monitoring of new delhi
metallo-β-lactamase-1 in living bacterial cells by uv-vis spectroscopy, Chem. Commun. 53 (2017)
8014-8017.
[29] S. Mizukami, S. Watanabe, Y. Hori, K. Kikuchi, Covalent protein labeling based on noncatalytic
β-lactamase and a designed fret substrate, JACS. 131 (2009) 5016-5017.
[30] M. Singha, S. Roy, S. D. Pandey, et al., Use of azidonaphthalimide carboxylic acids as fluorescent
templates with a built-in photoreactive group and a flexible linker simplifies protein labeling
studies: Applications in selective tagging of hcaii and penicillin binding proteins, Chem. Commun.
53 (2017) 13015-13018.
[31] J. M. Xiao, L. Feng, L. S. Zhou, et al., Novel fluorescent cephalosporins: Synthesis, antimicrobial
activity and photodynamic inactivation of antibiotic resistant bacteria, Eur. J. Med. Chem. 59
(2013) 150-159.
[32] A. Rokney, M. Shagan, M. Kessel, et al., E. Coli transports aggregated proteins to the poles by a
specific and energy-dependent process, J. Mol. Biol. 392 (2009) 589-601.
[33] X. Yong, J. G. Wilkes, T. J. Moskal, et al., Development of a flow cytometry-based method for rapid
detection of escherichia coli and shigella spp. Using an oligonucleotide probe, PLoS One. 11
(2016) e0150038.
[34] U. K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage
t4, Nature. 227 (1970) 680-685.
[35] D. Gao, A. Li, L. Guan, et al., Solvent-dependent ratiometric fluorescent merocyanine dyes:
Spectral properties, interaction with bsa as well as biological applications, Dyes & Pigments. 129
(2016) 163-173.

Table of content