Dithiocarbamate as a valuable scaffold for the inhibition of metallo-β-lactmases

Article abstract of DOI:10.3390/biom9110699

The ‘superbug’ infection caused by metallo-β-lactamases (MβLs) has grown into an emergent health threat. Given the clinical importance of MβLs, a novel scaffold, dithiocarbamate, was constructed. The obtained molecules, DC1, DC8 and DC10, inhibited MβLs NDM-1, VIM-2, IMP-1, ImiS and L1 from all three subclasses, exhibiting an IC₅₀ < 26 μM. DC1 was found to be the best inhibitor of ImiS (IC₅₀ < 0.22 μM). DC1-2, DC4, DC8 and DC10 restored antimicrobial effects of cefazolin and imipenem against E. coli-BL21, producing NDM-1, ImiS or L1, and DC1 showed the best inhibition of E. coli cells, expressing the three MβLs, resulting in a 2-16-fold reduction in the minimum inhibitory concentrations (MICs) of both antibiotics. Kinetics and isothermal titration calorimetry (ITC) assays showed that DC1 exhibited a reversible, and partially mixed inhibition, of NDM-1, ImiS and L1, with Ki values of 0.29, 0.14 and 5.06 μM, respectively. Docking studies suggest that the hydroxyl and carbonyl groups of DC1 form coordinate bonds with the Zn (II) ions, in the active center of NDM-1, ImiS and L1, thereby inhibiting the activity of the enzymes. Cytotoxicity assays showed that DC1, DC3, DC7 and DC9 have low toxicity in L929 mouse fibroblastic cells, at a dose of up to 250 μM. These studies revealed that the dithiocarbamate is a valuable scaffold for the development of MβLs inhibitors.

Full text of DOI:10.3390/biom9110699

biomolecules
Communication
Dithiocarbamate as a Valuable Scaffold for the
Inhibition of Metallo-β-Lactmases
Ying Ge †, Li-Wei Xu †, Ya Liu, Le-Yun Sun, Han Gao, Jia-Qi Li and Kewu Yang *
Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of Education, the College
of Chemistry and Materials Science, Northwest University, Xi’an 710127, China; geyingxgs@163.com (Y.G.);
lwxu717@163.com (L.-W.X.); liuya951214@163.com (Y.L.); marcolsun@163.com (L.-Y.S.);
gaohanyfz@163.com (H.G.); lijiaqi01120@163.com (J.-Q.L.)
* Correspondence: kwyang@nwu.edu.cn; Tel./Fax: +86-29-8153-5035
† Those authors contributed equally to this work.
ꢀꢁꢂꢀꢃꢄꢅꢆꢇ
ꢀꢁꢂꢃꢄꢅꢆ
Received: 11 October 2019; Accepted: 4 November 2019; Published: 5 November 2019
Abstract: The ‘superbug’ infection caused by metallo-
emergent health threat. Given the clinical importance of M
was constructed. The obtained molecules, DC1, DC8 and DC10, inhibited M
IMP-1, ImiS and L1 from all three subclasses, exhibiting an IC50 < 26 M. DC1 was found to be the
best inhibitor of ImiS (IC50 < 0.22 M). DC1-2, DC4, DC8 and DC10 restored antimicrobial effects of
cefazolin and imipenem against E. coli-BL21, producing NDM-1, ImiS or L1, and DC1 showed the best
inhibition of E. coli cells, expressing the three M Ls, resulting in a 2-16-fold reduction in the minimum
β
-lactamases (M
Ls, a novel scaffold, dithiocarbamate,
Ls NDM-1, VIM-2,
βLs) has grown into an
β
β
µ
µ
β
inhibitory concentrations (MICs) of both antibiotics. Kinetics and isothermal titration calorimetry
(ITC) assays showed that DC1 exhibited a reversible, and partially mixed inhibition, of NDM-1, ImiS
and L1, with Ki values of 0.29, 0.14 and 5.06 µM, respectively. Docking studies suggest that the
hydroxyl and carbonyl groups of DC1 form coordinate bonds with the Zn (II) ions, in the active center
of NDM-1, ImiS and L1, thereby inhibiting the activity of the enzymes. Cytotoxicity assays showed
that DC1, DC3, DC7 and DC9 have low toxicity in L929 mouse fibroblastic cells, at a dose of up to
250 µM. These studies revealed that the dithiocarbamate is a valuable scaffold for the development of
MβLs inhibitors.
Keywords: antibiotic resistance; metallo-β-lactamase; inhibitor; dithiocarbamate; activity assay
1. Introduction
Dating from the discovery of penicillin,
antibacterial drugs for the treatment of bacterial infection. Currently, about 60% of antibacterial
drugs used are -lactam antibiotics [ ]. However, the overuse of -lactams has resulted in many
bacteria that produce -lactamases, and these bacteria are resistant to the most commonly used
antibiotics. -Lactamases hydrolyse -lactam antibiotics by breaking the -lactam ring, rending the
drugs ineffective [ ]. So far, more than 2000 kinds of -lactamases have been discovered [ ]. They are
categorized into four classes:—A, B, C, and D—depending on their amino acid sequence homologies [ ].
Class A, C and D enzymes are collectively referred to as serine β-lactamases (SβLs), and catalyze the
hydrolysis of -lactams through a serine residue as an active group. Class B enzymes, also called
metallo- -lactamase (M Ls), are usually dependent on Zn(II) ions [ ]. M Ls are further divided into
subclasses B1, B2, and B3, based on amino acid sequence homology and Zn(II) content [6].
The “Superbugs” resistance, mediated by M Ls, has resulted in the failure of almost all antibiotics
used in clinics, leading to a global medical crisis. Therefore, the development of -lactamase inhibitors
β-lactam antibiotics have been developed into the main
β
1
β
β
β
β
β
2
β
3
4
β
β
β
5
β
β
β
is an essential strategy for maintaining the usefulness of the existing
β-lactam antibiotics. Currently, the
Biomolecules 2019, 9, 699; doi:10.3390/biom9110699
www.mdpi.com/journal/biomolecules

Biomolecules 2019, 9, 699
2 of 13
clavulanic and other S
βLs inhibitors, such as sulbactam and tazobactam, have targeted the inactivation
of the drug-resistant bacteria that produce S
these S L inhibitors are ineffective against the drug-resistant bacteria harboring M
no M L inhibitors for available clinical purposes to date. Due to the biomedical importance of M
a large number of small molecule M Ls inhibitors have been reported [ ], such as thiols [ ], dicarboxylic
acids [10 11], sulfates [12], tetrazoles [13], aspergillomarasmine A (AMA) [14], hydroxamates [15],
and ebselen [16].
Dithiocarbamates (DCs) have been reported to be inhibitors of carbonic anhydrase (CA) [17].
CA, like M Ls, are zinc-containing metalloenzymes [18]. Initially, DCs were used to treat crops,
to protect against insects and fungi [19]. Later, they were used in anticancer [20 21], antibacterial [22 23],
antifungal [24 25], and antiglaucoma [17] applications. Wang and co-workers reported that the DC
βLs in combination with
β
-lactam antibiotics [
Ls [ ], and there are
Ls,
1]. However,
β
β
7
β
β
β
8
9
,
β
,
,
,
derivatives with azacycloalkane had an inhibitory effect on drugresistant bacteria containing NDM-1
or IMP-4 genes [26]. Further mechanism studies indicated that these compounds may chelate Zn(II)
ions, thereby restoring the antibacterial qualites of meropenem. However, these reported compounds
are mostly limited to the aliphatic-substituted dithiocarbamates. In order to develop the structurally
diverse dithiocarbamates necessary to screen MβLs inhibitors, we designed and synthesized aromatic
acylaminodithiocarbamates and assayed their biological activities.
To develop the aromatic acylaminodithiocarbamates as inhibitors of M
βLs, and to use them in
combination with β-lactams to combat antibiotic-resistant bacteria-producing MβLs, in this work,
the dithiocarbamates (DC1-10) were synthesized (Figure 1). The positions (ortho, meta, para position)
of the aromatic substituents were adjusted, to determine the optimal position for the compound to bind
to the active sites of the target enzymes, which confer the best inhibitory effects. The biological activities
of these DCs were tested as inhibitors against the purified MβLs VIM-2, IMP-1, NDM-1, ImiS and L1,
which are representatives of the B1, B2 and B3 subclasses of M
βLs, respectively, and their inhibition
mode was evaluated thermodynamically by isothermal titration calorimetry (ITC). Furthermore,
the capability of these inhibitors to restore the antimicrobial activity of antibiotics against E. coli cells
expressing M
βLs was determined. In addition, molecular docking was performed, exploring the
binding mode between inhibitors and MβLs.
Figure 1. Structures of the synthesized dithiocarbamate derivatives.

Biomolecules 2019, 9, 699
3 of 13
2. Materials and Methods
2.1. General Information
The substituted benzoic acids were purchased from Aladdin (Shanghai) Trading Co., Ltd. and
all other 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 (Bruker Daltonics Inc., Billerica, MA, USA). 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 an internal standard. Coupling constants
(J) were reported in Hertz (Hz). Mass spectra were obtained on a micro TOF-Q (BRUKER) mass
spectrometer. Inhibition studies were performed on an Agilent 8453 UV-Vis spectrometer (Santa Clara,
CA, USA).
2.2. Synthesis of Inhibitors
The methyl arylcarboxylates and benzoylhydrazines were prepared using previously reported
methods [27]. The intermediate benzoylhydrazine was dissolved in sufficient amount of ethanol and
stirred until clarified at room temperature, and the solution was mixed with potassium hydroxide
at a molar ratio of 1:1.5. Following this, an equimolar ratio of carbon disulfide was added, and the
resulting mixture was stirred at room temperature for 3 h. Insoluble matter was removed by
filtration, and the obtained filtrate was distilled under reduced pressure and dried to give the
dithiocarbamates. The spectrogram information for the dithiocarbamates (DC1-10) is supplied in the
Supporting Information.
2.3. Inhibition Studies
The inhibitor concentration causing a 50% decrease in enzyme activity (IC50) was determined on
an Agilent UV8453 spectrometer at 25 ◦C using cefazolin as a substrate, for NDM-1, VIM-2, IMP-1
and L1, and imipenem for ImiS, respectively. The dithiocarbamates, M
were prepared with 30 mM Tris, pH 7.0. The final concentrations of NDM-1, VIM-2, IMP-1, ImiS,
and L1 were 13, 100, 70, 50 and 58 nM, and the concentrations of cefazolin and imipenem were 40 M,
respectively. In the test conditions, the Km for cefazolin (NDM-1, VIM-2, IMP-1 and L1) and imipenem
(ImiS) were determined to be 15, 97, 73, 60 and 52 M, respectively. Enzymes and inhibitors were
βLs and substrates samples
µ
µ
pre-incubated for 5 min before adding substrate. The hydrolyses of cefazolin and imipenem were
monitored at 262 and 300 nm, respectively. The IC50 values were nonlinearly fitted by plotting the
average percentage inhibition against inhibitor concentration. The inhibition modes of DC1 to ImiS,
NDM-1 and L1 were assayed by generating Lineweaver–Burk plots, and Ki values were obtained by
fitting initial velocity versus substrate concentration at each inhibitor concentration, using SigmaPlot
12.0 (Systat Software Inc., San Jose, CA, USA). Concentrations of DC1 were varied between 0 and 20
µM and substrate (cefazolin and imipenem) were varied between 20 and 120 µM.
2.4. Isothermal Titration Calorimetry (ITC) Assays
Isothermal titration calorimetry (ITC) assays were performed on a Malvern MicroCal iTC 200
instrument (Malvern Instruments, Malvern, UK) at 25 ◦C, by a single injection mode. During ITC
experiments the reference cell was filled with degassed water. The MβLs, substrates, and inhibitor DC1
were prepared with 50 mM Tris, pH 7.0. The concentration of M Ls used was in the range 100–200 nM,
β
concentration of substrate was 1 mM, and concentrations of DC1 were varied between 0 and 200 µM.
Before starting the measurement, the enzyme was pre-incubated with different concentrations of DC1
for 5 min, and then delivered into the sample cell (210 µL). The hydrolysis reaction in the absence
of inhibitors was also determined as a control. The stirring rate of the syringe was set to 750 rpm.
The system was equilibrated at the desired 25 ◦C. Substrate at a concentration of 1 mM (cefazolin for

Biomolecules 2019, 9, 699
4 of 13
NDM-1 and L1, imipenem for ImiS), in the syringe was titrated into the sample cell, in the mode of a
single 38 L injection. Heat flow was recorded as a function of time. Data were collected every 1 s
µ
until the signal reached the baseline and continued to be recorded for the length of time necessary to
generate the final baseline.
2.5. MIC Determination
The minimum inhibitory concentration (MIC) of antibiotics, both alone and in the presence of
enzyme inhibitors, was determined using the broth micro-dilution method. Single colonies of E. coli
BL21, containing plasmid, pET26b-NDM-1, pET26b-ImiS or pET26b-L1 in lysogeny broth agar plates,
were transferred into 5 mL of Mueller–Hinton (MH) liquid medium. Strains grown in MH medium
to OD600 = 0.45 were used as inocula, after an 84-fold dilution to 1
Cefazolin and imipenem were dissolved in MH medium to prepare 1024, 512, 256, 128, 64, 32, 16, 8, 4
and 2 g/mL stock solutions, respectively. Compound DCs as inhibitors were dissolved in MH media,
to prepare 1024, 512, 256, 128, 64 and 32 g/mL stock solutions. The prepared solutions with different
antibiotic concentrations (50 L) were diluted to 100 L, with different concentrations of inhibitor
solution (50 L), then 100 L inoculum was added sequentially into the prepared solutions, to yield a
×
105 CFU/mL in MH medium.
µ
µ
µ
µ
µ
µ
quarter of the initial concentrations. The mixtures were incubated at 37 ◦C for 16 h. The MIC results
were taken as the lowest concentration that completely inhibited visible growth. Each measurement
was performed in duplicate.
2.6. Docking Studies
Docking studies of compound DC1 into the active site of NDM-1 (PDB: 4EYL [28]), CphA (PDB:
2QDS [29], which was used as a closely related relative of ImiS) and L1 (PDB: 2AIO [30]) were performed
by AutoDock 4.2 [31]. The carboxyl groups were deprotonated, resulting in an overall charge of -1e for
the compounds. A charge of +1.4e was assigned to the two Zn(II) in the active site, while +0.2e was
added to each of its ligands [31]. The grid and docking parameter files were prepared using Zn (II) van
der Waals parameters = 0.25 kcal/mol and r0 = 1.95 Å [32]. Enzymes were treated as a rigid receptor,
while ligands were treated as flexible. The grid box was centered between the two active-site Zn (II)
ions, with dimensions of 70
×
70
×
70 grid points, with grid points spaced at 0.375 Å. The mutation
rate and crossover rates were set at 0.02 and 0.8, respectively, while the maximum energy evaluations
and generations’ numbers were set at 2,500,000 and 27,000, respectively. Default values were kept for
all other parameters, and no constraints were used. Fifty conformations were generated according to
the Lamarckian genetic algorithm, and grouped into clusters based on a root mean square deviation
(RMSD) with a tolerance of 2.0 Å. The lowest–energy (highest ranked) clusters were closely examined.
2.7. Cytotoxicity Evaluation
The cytotoxicity assays were performed to evaluate the toxicity of DC1, DC3, DC7 and DC9
in mouse fibroblast cells (L-929). The cells were seeded into 96-well plates at a cell density of
1.0
×
104 cells/well, in 100
µL of culture medium and maintained for 24 h. Then, the inhibitors with
working concentrations (15.6, 31.25, 62.5, 125, 250, and 500 µM) were added to the 96-well plates and
incubated for another 48 h. Three wells containing only cells were suspended in a mixture solution
of 98
only the complete medium were used as blank controls. Following that, the medium was removed,
and 100 L of fresh culture medium and 10 L MTT were added to each well. After incubation for
µL complete medium, as controls for cell viability without inhibitors. Three wells containing
µ
µ
4 h, the trays 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, and the data are expressed as mean ± standard deviation.

Biomolecules 2019, 9, 699
5 of 13
3. Results and Discussion
3.1. Synthesis
Ten dithiocarbamates (DC1-10) were synthesized by a synthetic route, shown in Scheme A
(Supporting Information). Briefly, the intermediate methyl arylcarboxylates were made by esterifying
the corresponding benzoic acids, and then were converted into benzoylhydrazines by condensing
with hydrazine [27]. The resulting benzoylhydrazines reacted with carbon disulfide in KOH aqueous
solution at room temperature for 3 h, and the crude products gained were distilled to offer the
dithiocarbamates (DCs) [33]. DCs were characterized by 1H and 13C NMR and HRMS (Supporting
Information).
3.2. Inhibition Studies
To evaluate the inhibitory activity of the dithiocarbamates, we over-expressed and purified
M
β
Ls NDM-1, VIM-2 and IMP-1 (B1), ImiS (B2), and L1 (B3) from three subclasses as previously
reported [34 37]. The inhibition experiments were performed with an Agilent UV8453 spectrometer.
Cefazolin (40 M) was used as substrate for NDM-1, VIM-2, IMP-1 and L1, and imipenem (40 M)
for ImiS. Hydrolysis of substrates was monitored at 262 (cefazolin) and 300 nm (imipenem). The
concentrations of inhibitors were varied between 0 and 200 M. The initial reaction rates were
–
µ
µ
µ
determined in the absence and presence of inhibitors in triplicate, and the generated average values
were recorded.
The collected concentrations of dithiocarbamates DC1-10 causing a 50% decrease in enzyme
activity (IC50) are listed in Table 1. The results indicate that DC1, DC8 and DC10 were potent
broad-spectrum inhibitors of all five M
to be the most potent inhibitor of ImiS (IC50 < 0.22
showed inhibitory potency with an IC50 value range of 0.38–15.25
VIM-2 and IMP-1, with an IC50 value range of 0.85–9.31 M, and DC8 was the most potent inhibitor
(IC50 = 0.85 M). ImiS was inhibited by DC1-2, DC4, DC6, DC8 and DC10, with an IC50 value range of
0.21–2.56 M. Analysis of IC50 data of these DCs reveals an interesting structure–activity trend—the
β
Ls tested, exhibiting IC50 values < 26
M). For NDM-1, DC1-2, DC4, DC6-8, and DC10
M. DC1, DC8, and DC10 inhibited
µM, and DC1 was found
µ
µ
µ
µ
µ
inhibitory activity of a single benzene ring and its ortho-substituted products is better than the avtivity
of para- and meta-substituted products. Comparison of the IC50 values of DC1 and DC8 indicates
that the hydroxyl group had little contribution to the inhibitory activity of DC1. While the inhibition
activity of DCs regarding NDM-1 is better than VIM-2, this is probably because the NDM-1 contains a
shallower and wider substrate–binding groove, which more easily accommodates substrate or inhibitor
molecules [38
curves are shown in Figure 2. It can be observed that the compound had more than 90% inhibition
against NDM-1, IMP-1, ImiS, and L1 at concentrations of 20, 10, 40 and 8 M, respectively. Also, while
,39]. The inhibition of DC1 in four MβLs was analyzed, and the generated inhibition
µ
DC1, DC8 and DC10 were incubated with NDM-1 for 8 h, respectively, the residual activity assays
showed that the compounds had no time-dependent inhibition of the enzyme (Figure S1).
To further identify the inhibition mode of the DCs against M
of NDM-1, ImiS and L1 by DC1, at different concentrations, with different substrate concentrations.
The concentrations of the inhibitor were varied between 0 and 20 M, and substrate (imipenem
or cefazolin) concentrations were varied between 20 and 120 M. Enzyme and inhibitors were
pre-incubated for 5 min before starting the kinetic assays. All experimental hydrolytic rates of
substrates were determined in triplicate. The inhibition mode of dithiocarbamates on M Ls was
βLs, we performed inhibition kinetics
µ
µ
β
identified by generating Lineweaver–Burk plots, and Ki values were gained by fitting the initial
velocity versus substrate concentration at each inhibitor concentration, using SigmaPlot 12.0. The
Lineweaver–Burk plots of imipenem and cefazolin hydrolysis by NDM-1, ImiS and L1 in the absence
and presence of DC1 are shown in Figure 3. DC1 is shown to exhibit a partially mixed inhibition
against NDM-1, ImiS and L1, with Ki values of 0.29, 0.14 and 5.06 µM, respectively, which were slightly
smaller than the corresponding IC50 values.

Biomolecules 2019, 9, 699
6 of 13
Table 1. IC50 values of dithiocarbamate derivatives (DC1-10) against MβLs (µM).
B1
VIM-2 a
B2
B3
L1 a
Compd.
NDM-1 a
IMP-1 a
ImiS b
DC1
DC2
DC3
DC4
DC5
DC6
DC7
DC8
DC9
DC10
0.38 ± 0.01
5.64 ± 0.10
67.12 ± 0.90
11.39 ± 0.40
91.31 ± 1.70
15.25 ± 0.90
0.52± 0.01
9.31 ± 0.20
3.16 ± 0.10
166.22 ± 2.70
>200.00
0.21 ± 0.01
2.50 ± 0.10
>200.00
1.08 ± 0.05
128.82 ± 2.10
2.56 ± 0.05
>200.00
0.67 ± 0.01
156.23 ± 2.20
0.82 ± 0.01
7.17 ± 0.10
86.81 ± 1.6
>200.00
>200.00
82.76 ± 1.10
139.07 ± 2.05
>200.00
102.78 ± 1.45
176.54 ± 2.80
>200.00
>200.00
>200.00
>200.00
154.37 ± 3.20
190.79 ± 4.00
25.81 ± 0.12
>200.00
63.46 ± 1.30
3.07 ± 0.10
115.20 ± 3.10
5.25 ± 0.20
96.38 ± 1.20
0.85 ± 0.01
86.16 ± 0.90
2.16 ± 0.05
0.53 ± 0.01
62.66 ± 1.05
1.52 ± 0.05
11.68 ± 0.6
The antibiotics used were cefazolin a and imipenem b, respectively. The inhibitor concentration used was up to
200 µM.
Figure 2. Inhibition curves of DC1 against NDM-1, IMP-1, ImiS and L1.
Figure 3. Lineweaver–Burk plots with NDM-1 (a), ImiS (b), and L1 (c) using cefazolin and imipenem as
substrate, respectively, in the absence and presence of DC1, at a concentration in the range of 0–20 µM.
3.3. Inhibition Mode Assays
Isothermal titration calorimetry (ITC) was also employed, to identify the inhibition of the
dithiocarbamates on M Ls and study their thermodynamic behaviour. The progress of M L-catalyzed
substrate hydrolysis in the absence and presence of inhibitors, was investigated by measuring thermal
power (
cal/sec), which is a function of time, on a Malvern MicroCal ITC 200 instrument at 25 ◦C.
β
β
µ
The evaluation was performed in 50 mM Tris, pH 7.0. The concentrations of the enzymes NDM-1,
ImiS, and L1 were 150, 100 and 100 nM, respectively. The substrate concentration was 1 mM, and its
inhibitor was varied between 0 and 200 µM.
The heat–flow curves of DC1’s inhibition of the antibiotic hydrolysis catalyzed by NDM-1, ImiS
and L1, at different concentrations, are shown in Figure 4a–c. The negative thermal compensation
power means that the hydrolysis of substrate was an exothermic reaction. It can be observed that, in the

Biomolecules 2019, 9, 699
7 of 13
absence of inhibitors, the enzyme activity was higher, the catalytic hydrolysis of substrate emitted more
heat per unit time, the corresponding compensation power was higher, the reaction time was shorter,
and the heat–flow curve was more concentrated. As the inhibitor concentration increases, the enzyme
activity decreases, the heat liberated of substrate per unit time gradually decreases, the time required
to hydrolyze the substrate gradually increases, and the heat–flow curve is elongated. ITC assays
confirmed the inhibitory effects of the dithiocarbamate on NDM-1, ImiS and L1. However, under the
conditions of a certain concentration of enzyme and substrate, the total heat Q released (corresponding
to the integral area of the heat–flow curve) was a basically fixed value (Figure 4d–f).
Recently, our studies revealed that the Q seems to be an important indicator of the inhibitors’
mode of action [40]. When Q remained constant, the inhibitor molecule was a reversible inhibitor, but
the inhibitor molecule that reduced Q was an irreversible inhibitor. It can be observed in Figure 4d–f
that the Q value remains almost unchanged in the presence or absence of inhibitors, likely because the
formation of the enzyme-inhibitor complex (EI), that slowed down substrate hydrolysis, was reversible.
ITC assays indicated that the dithiocarbamate which maintains a constant Q is a reversible inhibitor.
We assayed the binding reversibility of DC1 to M
after a large, rapid dilution of the M L-DC1 complex [41]. The results (Figure S2) suggest that DC1 is a
reversible inhibitor; this is consistent with the results from the thermodynamic evaluation.
βLs by measuring the enzymatic activity recovery
β
Figure 4. Overlaid heat flow curves of NDM-1- (
ImiS-catalysed hydrolysis of imipenem ( ) in the absence and presence of DC1 at a concentration
range of 1–200 µM, using a single injection mode ITC assay at 25 ◦C.
a,d) and L1- (c,f) catalysed hydrolysis of cefazolin and
b,e
3.4. MIC Assays
The ability of the dithiocarbamate derivatives to inhibit M
βLs, and to restore the antimicrobial
activity of antibiotics, was investigated by determining the minimum inhibitory concentrations (MICs)
of existing antibiotics in the presence and absence of DC1-10, as previously reported [42]. E. coli BL21
(DE3) containing plasmids pET26b-NDM-1, pET26b-ImiS and pET26b-L1 were used to assess these

Biomolecules 2019, 9, 699
8 of 13
inhibitors. The collected MIC data are listed in Table 2A–C. The MIC data show that DC1-2, DC4, DC8
and DC10 gradually restored the antimicrobial effects of the -lactam antibiotics against NDM-1, ImiS
and L1-producing bacteria E. coli-BL21, with an increased inhibitor dose. A dose of 64 g/mL inhibitor
β
µ
resulted in a 2-16-fold decrease in MICs of cefazolin and imipenem, and the DC1 was found to show
the best inhibitory effects against the three bacteria tested, resulting in a 16-fold reduction in MICs of
both cefazolin and imipenem, indicating that the dithiocarbamate restored activity of the
antibiotics against the antibiotic–resistant bacteria–producing M Ls. However, the DC3 and DC5-6
had no inhibitory effects on the three bacteria tested. Besides, the inhibitor alone did not inhibit cell
growth at a dose up to 512 g/mL, revealing that the efficacy of dithiocarbamate to restore -lactam’s
antibiotic activity lies in their inhibition of the MβLs harbored in bacteria.
β-lactam
β
µ
β
Table 2. Antibacterial activities (MICs,
µg/mL) of imipenem on E. coli BL21 expressing ImiS (
A),
cefazolin against E. coli BL21 expressing NDM-1 (
B
) and L1 ( ) in the presence of dithiocarbamates at a
C
concentration ranging from 8 to 128 µg/mL and 16 to 256 µg/mL.
A a
Compd.\conc.
DC1
8
8
16
8
32
4
64
4
128
1
DC2
DC3
DC4
16
16
8
8
16
8
8
16
8
8
16
8
4
16
4
DC5
DC6
DC7
DC8
DC9
DC10
16
16
16
16
16
16
16
16
16
8
16
8
16
16
16
8
16
8
16
16
16
8
16
8
16
16
16
8
16
4
B a
Compd.\conc.
DC1
8
32
16
16
32
16
64
8
128
8
DC2
DC3
DC4
DC5
DC6
DC7
DC8
DC9
64
128
64
128
128
128
64
128
128
64
128
64
128
128
32
64
128
128
64
128
32
128
128
16
32
128
128
32
128
32
128
128
8
32
128
32
32
128
16
128
128
8
16
64
16
DC10
C a
Compd.\conc.
DC1
16
32
64
64
128
32
256
8
64
128
32
128
128
128
32
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
DC2
DC3
DC4
DC5
DC6
DC7
DC8
DC9
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
64
128
128
128
64
DC10
a
The MICs of cefazolin alone on E. coli-BL21 with NDM-1 or L1 were 128
on E. coli-BL21-ImiS was 16 µg/mL.
µg/mL, and the MIC of imipenem alone

Biomolecules 2019, 9, 699
9 of 13
3.5. Docking Assays
To explore the potential binding mode of the dithiocarbamate to MβLs, we docked DC1 into the
active sites of NDM-1, CphA (in lieu of ImiS, because the target has no crystal structure yet reported,
and shares 96% sequence identity with CphA) and L1. The resulting lowest–energy conformations
of those clusters are shown in Figure 5, in which the binding energies are
−
5.22,
−
5.99, and 6.64
−
kcal/mol for the NDM-1/DC1, CphA/DC1 and L1/DC1 complexes, respectively. Figure 5a–c clearly
shows that DC1 has a similar binding pattern to NDM-1, CphA, and L1. The hydroxyl oxygen atoms
and carbonyl oxygen atoms form coordination bonds with Zn (II) ions of NDM-1, CphA and L1 active
sites, and the bond distances are 1.8 and 2.7 Å, 1.7 and 2.7 Å, 1.8 and 2.1 Å, respectively. Further, the
hydroxyl oxygen atom in DC1 also forms coordinate bonds and hydrogen bonds with His120NDM-1
His118CphA and His116L1, in the active centers of NDM-1, CphA and L1, respectively. Moreover, in the
complexes NDM-1/DC1, CphA/DC1 and L1/DC1, the hydroxyl hydrogen atom at the benzene ring
forms a hydrogen bond with the Asp124NDM-1, Asp120CphA and Asp120L1, in the active sites of the
,
three enzymes. Also, the sulfur atom in DC1 forms hydrogen bonds with Lys211NDM-1, Asn220 NDM-1
Lys224CphA, Asn233CphA, and Ser221L1, Leu222L1, Tyr293L1, in the active centers of NDM-1, CphA and
L1, respectively. DC8 was docked into the active sites of NDM-1, CphA and L1, and the results show
that it had a similar binding pattern to MβLs as DC1 (Figure S3).
,
Figure 5. Cont.

Biomolecules 2019, 9, 699
10 of 13
Figure 5. Lowest–energy conformations of DC1, docked into the active site of NDM-1 (PDB code 4EYL)
), CphA (PDB code 2QDS) ( ) and L1 (PDB code 2AIO) ( ). Three enzymes were depicted as follows:
(
a
b
c
backbone as cartoon in green; selected residues are shown as sticks, colored by element (H, white;
C, skyblue; N, blue; O, red); Zn (II) ions were shown as magenta spheres; compound DC1 was also
shown as sticks, with the same color code as amino acid residues, except C in is depicted as white and
S as yellow. Characteristic short distances between inhibitors and the protein are indicated by dashed
lines. These figures were generated with PyMOL.
3.6. Cytotoxicity Assays
The potential toxicity of dthiocarbamates DC1, DC3, DC7 and DC9 as representatives was assayed
using the mouse fibroblast cells (L929) treated with different concentrations of inhibitors (15.625, 31.25,
62.5, 125, 250 and 500 µM). As shown in Figure 6, when the concentration was as high as 250 µM for
the tested compounds, the survival rate of L929 mouse cells was higher than 80%, indicating that the
dthiocarbamates are basically non-toxic or low-toxic to L929 mouse cells.
Figure 6. Percent cell viability (relative to viability without compound) of L-929 mouse fibroblastic
cells, in the presence of DC1, DC3, DC7 or DC9, at concentrations ranging from 15.6 to 500 µM.
4. Conclusions
In summary, in response to the enormous biomedical importance of M
derivatives were synthesized and characterized by 1H and 13C NMR and HRMS. Bioactivity evaluation
showed that the dithiocarbamates DC1, DC8 and DC10 had broad-spectrum inhibition effect on M Ls
NDM-1, VIM-2, IMP-1, ImiS, and L1 belonging to three different groups, exhibiting IC50 values <
26 M. DC1 was found to be the best inhibitor of ImiS (IC50 < 0.22 M). MIC assays revealed that
DC1-2, DC4, DC8 and DC10 increased antimicrobial effects of cefazolin and imipenem against E.
βLs, ten dithiocarbamate
β
µ
µ

Biomolecules 2019, 9, 699
11 of 13
coli-BL21–producing NDM-1, ImiS or L1, and DC1 showed the best inhibitory effects on the three
bacteria tested, resulting in a 16-fold reduction in the MICs of both antibiotics. Enzyme kinetics and
ITC assays indicated that DC1 exhibited a reversible and partially mixed inhibition of NDM-1, ImiS and
L1, with Ki values of 0.29, 0.14, and 5.06 µM, respectively. Docking studies suggest that the hydroxyl
and carbonyl groups of DC1 form coordinate bonds with the Zn (II) ions in the active center of NDM-1,
ImiS and L1, thereby inhibiting the activity of the enzymes.
Supplementary Materials: The supplementary materials are available online. http://www.mdpi.com/2218-273X/
9/11/699/s1.
Author Contributions: The manuscript was written through contributions from all authors. K.Y. designed the
experiments and revised the draft; L.-W.X. and Y.G. conducted the synthetic experiments, data analysis and
preparation of original draft; Y.L., L.-Y.S., H.G. and J.-Q.L. performed enzymatic purification, ITC assays and
toxicity evaluation.
Funding: This work was supported by grants (21572179, 2019KW-068) to K.Y. from the National Natural Science
Foundation of China and the Shaanxi Province Project for International Science and Technology Cooperation.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
Drawz, S.M.; Bonomo, R.A. Three decades of beta-lactamase inhibitors. Clin. Microbiol. Rev. 2010, 23,
160–201. [CrossRef] [PubMed]
Fisher, J.F.; Meroueh, S.O.; Mobashery, S. Bacterial resistance to beta-lactam antibiotics: Compelling
opportunism, compelling opportunity. Chem. Rev. 2005, 105, 395–424. [CrossRef] [PubMed]
Bush, K. Past and Present Perspectives on
[CrossRef] [PubMed]
β-Lactamases. Antimicrob. Agents Chemother. 2018, 62, e01076-18.
Ambler, R.P. The Structure of
[CrossRef]
β-Lactamases. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 1980, 289, 321–331.
Crowder, M.W.; Spencer, J.; Vila, A.J. Metallo-beta-lactamases: Novel weaponry for antibiotic resistance in
bacteria. Acc. Chem. Res. 2006, 39, 721–728. [CrossRef]
Bebrone, C. Metallo-beta-lactamases (classification, activity, genetic organization, structure, zinc coordination)
and their superfamily. Biochem. Pharmacol. 2007, 74, 1686–1701. [CrossRef]
Meini, M.R.; Llarrull, L.I.; Vila, A.J. Overcoming differences: The catalytic mechanism of metallo-
β-lactamases.
FEBS Lett. 2015, 589, 3419–3432. [CrossRef]
Linciano, P.; Cendron, L.; Gianquinto, E.; Spyrakis, F.; Tondi, D. Ten Years with New Delhi
Metallo- -lactamase-1 (NDM-1): From Structural Insights to Inhibitor Design. ACS Infect. Dis. 2019
5, 9–34. [CrossRef]
β
,
9.
Mohamed, M.S.; Hussein, W.M.; McGeary, R.P.; Vella, P.; Schenk, G.; Abd El-Hameed, R.H. Synthesis and
kinetic testing of new inhibitors for a metallo-beta-lactamase from Klebsiella pneumonia and Pseudomonas
aeruginosa. Eur. J. Med. Chem. 2011, 46, 6075–6082. [CrossRef]
10. Toney, J.H.; Hammond, G.G.; Fitzgerald, P.M.D.; Sharma, N.; Balkovec, J.M.; Rouen, G.P.; Olson, S.H.;
Hammond, M.L.; Greenlee, M.L.; Gao, Y.D. Succinic Acids as Potent Inhibitors of Plasmid-borne IMP-1
Metallo-β-lactamase. J. Biol. Chem. 2001, 276, 31913–31918. [CrossRef]
11. Hiraiwa, Y.; Saito, J.; Watanabe, T.; Yamada, M.; Morinaka, A.; Fukushima, T.; Kudo, T. X-ray crystallographic
analysis of IMP-1 metallo-β-lactamase complexed with a 3-aminophthalic acid derivative, structure-based
drug design, and synthesis of 3,6-disubstituted phthalic acid derivative inhibitors. Bioorg. Med. Chem. Lett.
2014, 24, 4891–4894. [CrossRef] [PubMed]
12. Simm, A.M.; Loveridge, E.J.; Crosby, J.; Avison, M.B.; Walsh, T.R.; Bennett, P.M. Bulgecin A: A novel inhibitor
of binuclear metallo-beta-lactamases. Biochem. J. 2005, 387, 585–590. [CrossRef] [PubMed]
13. Toney, J.H.; Fitzgerald, P.M.; Grover-Sharma, N.; Olson, S.H.; May, W.J.; Sundelof, J.G.; Vanderwall, D.E.;
Cleary, K.A.; Grant, S.K.; Wu, J.K. Antibiotic sensitization using biphenyl tetrazoles as potent inhibitors of
Bacteroides fragilis metallo-beta-lactamase. Chem. Biol. 1998, 5, 185–196. [CrossRef]

Biomolecules 2019, 9, 699
12 of 13
14. King, A.M.; Reid-Yu, S.A.; Wang, W.; King, D.T.; Pascale, G.D.; Strynadka, N.C.; Walsh, T.R.; Coombes, B.K.;
Wright, G.D. AMA overcomes antibiotic resistance by NDM and VIM metallo-Î2-lactamases. Nature 2014
510, 503–506. [CrossRef] [PubMed]
,
15. Walter, M.W.; Valladares, M.H.; Adlington, R.M.; Amicosante, G.; Baldwin, J.E.; Frère, J.M.; Galleni, M.;
Rossolini, G.M.; Schofield, C.J. Hydroxamate Inhibitors of Aeromonas hydrophila AE036 Metallo-B-lactamase.
Bioorg. Chem. 1999, 27, 35–40. [CrossRef]
16. Chiou, J.; Wan, S.; Chan, K.F.; So, P.K.; He, D.; Chan, E.W.; Chan, T.H.; Wong, K.Y.; Tao, J.; Chen, S. Ebselen as
a potent covalent inhibitor of New Delhi metallo-β-lactamase (NDM-1). Chem. Commun. 2015, 51, 9543–9546.
[CrossRef]
17. Carta, F.; Aggarwal, M.; Maresca, A.; Scozzafava, A.; McKenna, R.; Masini, E.; Supuran, C.T. Dithiocarbamates
Strongly Inhibit Carbonic Anhydrases and Show Antiglaucoma Action in Vivo. J. Med. Chem. 2012, 55,
1721–1730. [CrossRef]
18. Supuran, C.T. Carbonic anhydrases: Novel therapeutic applications for inhibitors and activators. Nat. Rev.
Drug Discov. 2008, 7, 168–181. [CrossRef]
19. Lee, S.J.; Caboni, P.; Tomizawa, M.; Casida, J.E. Cartap hydrolysis relative to its action at the insect nicotinic
channel. J. Agric. Food Chem. 2004, 52, 95–98. [CrossRef]
20. Hou, X.; Ge, Z.; Wang, T.; Guo, W.; Cui, J.; Cheng, T.; Lai, C.; Li, R.; Hou, X.; Ge, Z. Dithiocarbamic acid
esters as anticancer agent. Part 1: 4-Substituted-piperazine-1-carbodithioic acid 3-cyano-3,3-diphenyl-propyl
esters. Bioorg. Med. Chem. Lett. 2006, 16, 4214–4219. [CrossRef]
21. Scozzafava, A.; Supuran, C.T. Carbonic anhydrase inhibitors: Synthesis of N-morpholylthiocar
bonylsulfenylamino aromatic/heterocyclic sulfonamides and their interaction with isozymes I, II and IV.
Bioorg. Med. Chem. Lett. 2000, 10, 1117–1120. [CrossRef]
22. Horita, Y.; Takii, T.; Chiba, T.; Kuroishi, R.; Maeda, Y.; Kurono, Y.; Inagaki, E.; Nishimura, K.; Yamamoto, Y.;
Abe, C. Synthesis of new sugar derivatives and evaluation of their antibacterial activities against
Mycobacterium tuberculosis. Bioorg. Med. Chem. Lett. 2009, 19, 6313–6316. [CrossRef] [PubMed]
23. Ohtake, N.; Imamura, H.; Jona, H.; Kiyonaga, H.; Shimizu, A.; Moriya, M.; Sato, H.; Nakano, M.; Ushijima, R.;
Nakagawa, S. Novel dithiocarbamate carbapenems with anti-MRSA activity. Bioorg. Med. Chem. 1998, 9,
1571–1578. [CrossRef]
24. Len, C.; Boulognemerlot, A.; Postel, D.; Gino Ronco, A.; Villa, P.; Goubert, C.; Jeufrault, E.; Bernard Mathon, A.;
Simon, H. Synthesis and Antifungal Activity of Novel Bis(dithiocarbamate) Derivatives of Glycerol. J. Agric.
Food Chem. 1996, 44, 2856–2858. [CrossRef]
25. Rafin, C.; Veignie, E.; Sancholle, M.; Postel, D.; Len, C.; Villa, P.; Ronco, G. Synthesis and antifungal activity
of novel bisdithiocarbamate derivatives of carbohydrates against Fusarium oxysporum f. sp. lini. J. Agric.
Food Chem. 2000, 48, 5283–5287. [CrossRef]
26. Wang, M.M.; Chu, W.C.; Yang, Y.; Yang, Q.Q.; Qin, S.S.; Zhang, E. Dithiocarbamates: Efficient
metallo-beta-lactamase inhibitors with good antibacterial activity when combined with meropenem. Bioorg.
Med. Chem. Lett. 2018, 28, 3436–3440. [CrossRef]
27. Zhai, L.; Zhang, Y.L.; Kang, J.S.; Oelschlaeger, P.; Xiao, L.; Nie, S.S.; Yang, K.W. Triazolylthioacetamide: A
Valid Scaffold for the Development of New Delhi Metallo-beta-Lactmase-1 (NDM-1) Inhibitors. ACS Med.
Chem. Lett. 2016, 7, 413–417. [CrossRef]
28. King, D.T.; Worrall, L.J.; Gruninger, R.; Strynadka, N.C.J. New Delhi Metallo-
into β-Lactam Recognition and Inhibition. J. Am. Chem. Soc. 2012, 134, 11362–11365. [CrossRef]
29. Li nard, B.M.; Garau, G.; Horsfall, L.; Karsisiotis, A.I.; Damblon, C.; Lassaux, P.; Papamicael, C.; Roberts, G.C.;
β-Lactamase: Structural Insights
é
Galleni, M.; Dideberg, O. Structural basis for the broad-spectrum inhibition of metallo-beta-lactamases by
thiols. Org. Biomol. Chem. 2008, 6, 2282–2294. [CrossRef]
30. Spencer, J.; Read, J.; Sessions, R.B.; Howell, S.; Blackburn, G.M.; Gamblin, S.J. Antibiotic Recognition
by Binuclear Metallo-β-Lactamases Revealed by X-ray Crystallography. J. Am. Chem. Soc. 2005, 127,
14439–14444. [CrossRef]
31. Morris, G.M.; Huey, R.; Lindstrom, W.; Sanner, M.F.; Belew, R.K.; Goodsell, D.S.; Olson, A.J. AutoDock4
and AutoDockTools4: Automated Docking with Selective Receptor Flexibility. J. Comput. Chem. 2010, 30,
2785–2791. [CrossRef] [PubMed]
32. Stote, R.H.; Karplus, M. Zinc binding in proteins and solution: A simple but accurate nonbonded
representation. Proteins-Struct. Funct. Bioinform. 2010, 23, 12–31. [CrossRef] [PubMed]

Biomolecules 2019, 9, 699
13 of 13
33. Xiang, Y.; Chen, C.; Wang, W.M.; Xu, L.W.; Yang, K.W.; Oelschlaeger, P.; He, Y. Rhodanine as a Potent Scaffold
for the Development of Broad-Spectrum Metallo-beta-lactamase Inhibitors. ACS Med. Chem. Lett. 2018, 9,
359–364. [CrossRef]
34. Yang, H.; Aitha, M.; Hetrick, A.M.; Richmond, T.K.; Tierney, D.L.; Crowder, M.W. Mechanistic and
spectroscopic studies of metallo-
β-lactamase NDM-1. Biochemistry 2012, 51, 3839–3847. [CrossRef] [PubMed]
35. Aitha, M.; Marts, A.R.; Bergstrom, A.; Møller, A.J.; Moritz, L.; Turner, L.; Nix, J.C.; Bonomo, R.A.; Page, R.C.;
Tierney, D.L. Biochemical, mechanistic, and spectroscopic characterization of metallo-β-lactamase VIM-2.
Biochemistry 2014, 53, 7321–7331. [CrossRef] [PubMed]
36. Crawford, P.A.; Sharma, N.; Chandrasekar, S.; Sigdel, T.; Walsh, T.R.; Spencer, J.; Crowder, M.W.
Over-expression, purification, and characterization of metallo-beta-lactamase ImiS from Aeromonas veronii
bv. sobria. Protein Expr. Purif. 2004, 36, 272–279. [CrossRef]
37. Crowder, M.; Walsh, T.L.; Pettit, M.; Spencer, J. Overexpression, purification, and characterization of the
cloned metallo-beta-lactamase L1 from Stenotrophomonas maltophilia. Antimicrob. Agents Chemother. 1998
42, 921–926. [CrossRef]
,
38. Cadag, E.; Vitalis, E.; Lennox, K.P.; Zhou, C.L.E.; Zemla, A.T. Computational analysis of pathogen-borne
metallo -lactamases reveals discriminating structural features between B1 types. BMC Res. Notes 2012, 5, 96.
β
[CrossRef]
39. Yuan, Q.; He, L.; Ke, H. A Potential Substrate Binding Conformation of β-Lactams and Insight into the Broad
Spectrum of NDM-1 Activity. Antimicrob. Agents Chemother. 2012, 56, 5157–5163. [CrossRef]
40. Zhang, Y.J.; Wang, W.M.; Oelschlaeger, P.; Chen, C.; Lei, J.E.; Lv, M.; Yang, K.W. Real-Time Monitoring of
NDM-1 Activity in Live Bacterial Cells by Isothermal Titration Calorimetry: A New Approach To Measure
Inhibition of Antibiotic-Resistant Bacteria. ACS Infect. Dis. 2018, 4, 1671–1678. [CrossRef]
41. Celenza, G.; Vicario, M.; Bellio, P.; Linciano, P.; Perilli, M.; Oliver, A.; Blazquez, J.; Cendron, L.; Tondi, D.
Phenylboronic Acid Derivatives as Validated Leads Active in Clinical Strains Overexpressing KPC-2: A Step
against Bacterial Resistance. ChemMedChem 2018, 13, 713–724. [CrossRef] [PubMed]
42. Wayne, P.A. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically: Approved
Standard, 9th ed.; Clinical and Laboratory Standards Institute: Wayne, PA, USA, 2012; M07-A9.
©
2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).