Journal Pre-proofs

1,2,4-Triazole-3-thione analogues with an arylakyl group at position 4 as met-

allo-β-lactamase inhibitors

Laurent Gavara, Federica Verdirosa, Laurent Sevaille, Alice Legru,

Giuseppina Corsica, Lionel Nauton, Paola Sandra Mercuri, Filomena Sannio,
Filomena De Luca, Margot Hadjadj, Giulia Cerboni, Yen Vo Hoang, Patricia
Licznar-Fajardo, Moreno Galleni, Jean-Denis Docquier, Jean-François
Hernandez

PII: S0968-0896(22)00357-1
DOI: https://doi.org/10.1016/j.bmc.2022.116964
Reference: BMC 116964

To appear in: Bioorganic & Medicinal Chemistry

Received Date: 29 March 2022

Revised Date: 22 July 2022
Accepted Date: 6 August 2022

Please cite this article as: L. Gavara, F. Verdirosa, L. Sevaille, A. Legru, G. Corsica, L. Nauton, P. Sandra
Mercuri, F. Sannio, F. De Luca, M. Hadjadj, G. Cerboni, Y. Vo Hoang, P. Licznar-Fajardo, M. Galleni, J-D.
Docquier, J-F. Hernandez, 1,2,4-Triazole-3-thione analogues with an arylakyl group at position 4 as metallo-β-
lactamase inhibitors, Bioorganic & Medicinal Chemistry (2022), doi: https://doi.org/10.1016/j.bmc.2022.116964

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. 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.

© 2022 Published by Elsevier Ltd.

 1,2,4-Triazole-3-thione analogues with an arylakyl group at
position 4 as metallo--lactamase inhibitors
Laurent Gavara,a* Federica Verdirosa,b Laurent Sevaille,a Alice Legru,a Giuseppina Corsica,b
Lionel Nauton,c Paola Sandra Mercuri,d Filomena Sannio,b Filomena De Luca,b Margot
Hadjadj,a Giulia Cerboni,b Yen Vo Hoang,a Patricia Licznar-Fajardo,e Moreno Galleni,d Jean-
Denis Docquierb,f* and Jean-François Hernandeza*
aInstitut des Biomolécules Max Mousseron, Univ Montpellier, CNRS, ENSCM, Montpellier,
France. bDipartimento
di Biotecnologie Mediche, Università di Siena, I-53100 Siena, Italy.
cInstitut de Chimie de Clermont-Ferrand, Université Clermont-Auvergne, CNRS, Clermont-

Ferrand, France. dLaboratoire des Macromolécules Biologiques, Centre d’Ingénierie des

Protéines-InBioS, Université de Liège, Institute of Chemistry B6a, Sart-Tilman, 4000 Liège,
Belgium. eHSM, Univ Montpellier, CNRS, IRD, CHU Montpellier, France. fLaboratoire de
Bactériologie Moléculaire, Centre d’Ingénierie des Protéines-InBioS, Université de Liège, B-
4000 Liège, Belgium

*Corresponding authors:
E-mail addresses: [email protected] (J.-F. Hernandez);
[email protected] (L. Gavara); [email protected] (J.-D. Docquier).

Note: This work is dedicated to the memory of a friend and former colleague, Dr Otto Dideberg.

Keywords: Metallo--Lactamase; 1,2,4-triazole-3-thione; bacterial resistance; -lactam

antibiotic.

Abbreviations list
ACE, angiotensin-converting enzyme; CLSI, Clinical and Laboratory Standards Institute;
DCM, dichloromethane; DFT, density functional theory; DMF, dimethylformamide; DMSO,
dimethylsulfoxide; DPT, di(2-pyridyl) thionocarbonate; FIC, fractional inhibitory
concentration; HEPES, 4-(2-Hydroxyethyl)-1-piperazine-ethanesulfonic acid; IMP,
imipenemase; KPC, Klebsiella pneumoniae Carbapenemase; MBL, metallo--lactamase;
MEM, meropenem; MIC, minimum inhibitory concentration; NDM, New Delhi Metallo--
lactamase; OXA, oxacillinase; PDB, protein data bank; SBL, serine--lactamase; VIM, Verona
Integron-borne Metallo--lactamase.

1
Abstract
Metallo-β-lactamases (MBLs) represent an increasingly serious threat to public health because
of their increased prevalence worldwide in relevant opportunistic Gram-negative pathogens.
MBLs efficiently inactivate widely used and most valuable β-lactam antibiotics, such as
oxyiminocephalosporins (ceftriaxone, ceftazidime) and the last-resort carbapenems. To date,
no MBL inhibitor has been approved for therapeutic applications. We are developing inhibitors
characterized by a 1,2,4-triazole-3-thione scaffold as an original zinc ligand and few promising
series were already reported. Here, we present the synthesis and evaluation of a new series of
compounds characterized by the presence of an arylalkyl substituent at position 4 of the triazole
ring. The alkyl link was mainly an ethylene, but a few compounds without alkyl or with an
alkyl group of various lengths up to a butyl chain were also synthesized. Some compounds in
both sub-series were micromolar to submicromolar inhibitors of tested VIM-type MBLs. A few
of them were broad-spectrum inhibitors, as they showed significant inhibitory activity on
NDM-1 and, to a lesser extent, IMP-1. Among these, several inhibitors were able to
significantly reduce the meropenem MIC on VIM-1- and VIM-4- producing clinical isolates by
up to 16-fold. In addition, ACE inhibition was absent or moderate and one promising compound
did not show toxicity toward HeLa cells at concentrations up to 250 M. This series represents
a promising basis for further exploration. Finally, molecular modelling of representative
compounds in complex with VIM-2 was performed to study their binding mode.

2
1. Introduction

Carbapenemases, or carbapenem-hydrolysing β-lactamases, are enzymes belonging to several

molecular classes (e. g. the class A KPC-type, the class B VIM- and NDM-types and the class
D OXA-type enzymes such as OXA-48 or OXA-40), which are able to efficiently inactivate
carbapenems. Carbapenems, a sub-class of β-lactam antibiotics, typically show highly potent
and broad-spectrum antibacterial activity against relevant opportunistic pathogens, and are thus
considered last resort therapeutics for the treatment of serious infections caused by antibiotic-
resistant bacteria. In addition, some carbapenemases exhibit a broad-substrate profile, thus
conferring resistance to other β-lactams antibiotics, such as penicillins and the widely used
oxyiminocephalosporins.1-3 Their rapid spread among multidrug-resistant opportunistic Gram-
negative pathogens rapidly evolving towards ultra- and pan-drug resistance phenotypes is of
major concern and exacerbates the current scarcity of therapeutic solutions useful to treat
nosocomial infections caused by such ultra-resistant isolates. This alarming perspective has
now been globally acknowledged for a few years, and the World Health Organization (WHO)
has designated the fight against carbapenem-resistant Gram-negative bacteria (Acinetobacter
baumannii, Pseudomonas aeruginosa and Enterobacterales) as a critical priority for antibiotic
discovery and development.4

Carbapenemases are divided into two large families of hydrolases depending on their catalytic
mechanism.5 Those using a catalytic serine belong to either molecular class A (e. g. KPC-type)
or class D (e. g. OXA-48-type) serine-β-lactamases (SBL).6 Metallo-β-lactamases (MBLs) are
zinc-dependent hydrolases and belong to molecular class B.7 Three distinct MBL subclasses
are known (B1, B2, and B3), which show different zinc requirements, active site architecture
and substrate profiles.8 Acquired MBLs are increasingly found in Gram-negative bacteria,
subclass B1 VIM-, NDM-, and IMP-types being the most clinically relevant enzymes. These
are highly worrying because of their broad substrate profile, their global spreading not only in
the hospital setting but also in the community,9 and the absence of approved inhibitors. Indeed,
this situation contrasts with the recent approval of several β-lactamase inhibitors (e. g.
avibactam and vaborbactam) targeting SBL-type carbapenemases, most notable KPC-type
enzymes and OXA-48.10,11

Identifying a broad-spectrum inhibitor targeting both all subclass B1 enzymes, ideally the VIM-
, NDM- and IMP-type MBLs, is a challenge because of the significant structural differences
within their active sites.10,12,13 The majority of reported MBL inhibitors contain a zinc-
coordinating group and many different metal-binding pharmacophores have been utilized.14

3
Among the most frequently reported are the thiol group15-18 and the carboxylate group, single
or multiple.19,20 While the majority of inhibitors forms ternary complexes with their targets,
some act by zinc stripping,21-23 with potential risk of insufficient selectivity toward human
metallo-enzymes.24

Currently, the most advanced inhibitors25 are the thiazole-4-carboxylate ANT2681, which
preclinical evaluation was recently achieved26 and the bicyclic boronates taniborbactam27-30 and
QPX7728.31 In particular, the latter are ultrabroad-spectrum inhibitors of both serine- and
metallo-carbapenemases and taniborbactam associated to cefepime is currently under phase III
clinical trials in patients with complicated urinary tract infections.

Several other series of compounds possessing a heterocycle as a zinc ligand have also been
reported.32-36 In particular, since the discovery by an in silico study that a 1,2,4-triazole-3-thione
compound could inhibit a MBL (i.e. the B3 sub-class L1),32 this scaffold has retained much
attention. A crystallographic study showed that this heterocyclic motif simultaneously
coordinated the two zinc ions present in the L1 active site by its N2 and S3 atoms.37 The same
original binding mode was later observed within the active sites of the B1 sub-class VIM-238-41
and NDM-140 enzymes. These studies as well as other random virtual and experimental
screenings, which also included IMP-1 support that the 1,2,4-triazole-3-thione scaffold is well
adapted for binding the dinuclear active site of these enzymes.40,42,43

We already reported several series of 1,2,4-triazole-3-thione compounds diversely substituted

at positions 4 and 5 and identified in each of these series potent inhibitors with a more or less
large spectrum of activity against MBLs of high clinical impact (i.e. VIM-, NDM- and IMP-
type).39,41,44-46 Among these, a few compounds were found to reduce the minimum inhibitory
concentration (MIC) of meropenem toward multi-resistant VIM-type MBL-producing clinical
isolates up to 16-fold. However, none showed synergistic activity on NDM-1-producing
clinical isolates. The main reasons were probably their insufficient capacity to penetrate and/or
a too low inhibition potency for the enzyme (Ki > 1 M). For instance, we reported Schiff base
analogues (Figure 1A, C, D) as potent and broad-spectrum inhibitors.39 Unfortunately, none
could potentiate the antibacterial activity of meropenem against any MBL-producing clinical
isolate. To solve this strong limitation, we are developing series of compounds where the
hydrazone-like bond (i.e. N=CH) was replaced by a stable CH2-CH2 link. We recently reported
a series of inhibitors derived from a Schiff base sub-family represented by the compound shown
in Figure 1A (i.e. with a 2-(o-benzoic)ethyl group at position 4 as in Figure 1B).46 Whereas
some analogues showed synergistic activity on VIM-producing clinical isolates, the inhibition

4
spectrum was restricted to VIM-type enzymes and all compounds were only moderately or not
active against NDM-1, in contrast to their Schiff base counterparts.

We now report on a series of new stable 4-phenethyl analogues of 1,2,4-triazole-3-thione-based

Schiff base inhibitors belonging to sub-families represented by compounds shown in Figure 1C
(i.e. 2,4-dihydroxyphenyl at position 4) and 1D (i.e. p-benzyloxyphenyl at position 4). We also
widened the variety of aryl substituents at position 4 and changed the length of the alkyl
segment between the aryl group and the triazole (Figure 1E). Overall, the synthetic compounds
showed lower to similar inhibitory potencies against both tested enzymes compared to the
Schiff base analogues. However, compared to the 4-[2-(o-benzoic)ethyl] series a few of them
exhibited a broader spectrum of inhibition, including on both VIM-type MBLs, NDM-1 and
IMP-1. In addition, some compounds showed synergistic activity against VIM-producing K.
pneumoniae clinical isolates.

Figure 1. Structure of previously reported 1,2,4-triazole-3-thione Schiff base analogues with diverse 4-
substituents39 (A, C, D) and of a 2-(o-benzoic)ethyl-containing analogue46 (B) and their inhibitory potency on
selected MBLs, and general structure of synthesized analogues with an arylalkyl substituent at position 4 (E). NI,
No or poor inhibition (< 30% at 100 M). ND, Not determined.

2. Results and Discussion

2.1. Chemistry
The formation of the 1,2,4-triazole-3-thione ring was performed following the general pathway
described in Scheme 1.47 Amines R2-NH2 were first reacted with dipyridylthionocarbonate

5
(DPT) to yield the intermediate isothiocyanates, which were subsequently treated in the same
pot with hydrazides R1-CONHNH2 to form the thiosemicarbazide derivatives. Their
cyclodehydration under basic conditions led to the expected 1,2,4-triazole-3-thione compounds.

Scheme 1. Synthesis of 4-phenylalkyl-1,2,4-triazole-3-thione derivatives. Reagents and conditions: (a) DPT,

DMF, sealed tube, 55 °C, 3h; (b) R1-CONHNH2, DMF, 55 °C, 3h; (c) aqueous KOH or NaHCO3, 100 °C, 3h.

The hydrazide R1-CONHNH2 precursors of substituent at position 5 were obtained in two steps
from the corresponding carboxylic derivatives R1-CO2H via the ethyl ester followed by
hydrazine treatment as previously described.45,48 Most amine R2-NH2 precursors of the
substituent at position 4 were commercially available. A few of them were prepared in two
steps from the corresponding benzaldehydes as described in Scheme 2A. In this pathway, a
benzaldehyde derivative was first treated with nitromethane and ammonium acetate in acetic
acid to give the nitro-alkenes 1-6, which were reduced to the amines 7-12 using LiAlH4.

The amine precursor of compounds 25, 33, 37, 39, 44, 46, 48 and 51 (Table 1) was the 3-
(aminomethyl)-1(3H)-isobenzofuranone 14. Compound 14 was obtained from 2-
carboxybenzaldehyde (Scheme 2B). Nitromethanation of this compound did not lead to the
nitro-alkene product as seen in Scheme 2A but to the lactone 13 by reaction between the
intermediate hydroxyl group and the carboxylic function. 13 was then reduced to give 14 as
described in Scheme 2B.

Scheme 2. Synthesis of the phenethylamines R2-NH2 7-12 (A) and 14 (B). Reagents and conditions: (a)
Nitromethane, NH4AcO, AcOH, 120 °C, 15h; (b) LiAlH4, THF, 80 °C, 15h; (c) AcOH, HCl, Zn powder, 16h, r.t.

6
In the case of compounds possessing a 2-(dihydroxyphenyl)ethyl at position 4 (2,4- for 21, 38,
41, 45, 2,3- for 42, 3,4- for 43 (Table 1)), it was not possible to prepare the corresponding amine
precursors 2-(dihydroxyphenyl)ethan-1-amine by following the general pathway. Indeed, the
nitromethanation of the corresponding aldehyde was not efficient. Therefore, these amine
precursors were obtained from the protected dihydroxybenzaldehydes. Two kinds of protected
derivatives were used. The 2-(2,4-bis(benzyloxy)phenyl)ethan-1-amine 9 was used for the
synthesis of compounds 21, 38 and 41, first yielding the protected intermediates. A final
deprotection step was therefore carried out. However, while hydrogenolysis in the presence of
a palladium catalyst was not possible because of the sulphur, other usual acidic conditions,
including the use of the mixture HBr/AcOH, led to poor yields. Indeed, we observed the
formation of an important secondary product resulting from the irreversible migration of one
protecting benzyl group to an adjacent aromatic carbon. This unwanted reaction was reduced
and yields were improved by using AlCl3 in the presence of N,N-dimethylaniline (Scheme 3).49
In this reaction, AlCl3 is supposed to coordinate the benzyl ether oxygen atom followed by the
nucleophilic addition of N,N-dimethylaniline on the benzyl CH2 group, leading to the cleavage
of the ether bond. Compounds 42, 43 and 45 were prepared from the corresponding 2-
(dimethoxyphenyl)ethan-1-amines 10 (2,3-dimethoxy), 11 (3,4-dimethoxy) and 12 (2,4-
dimethoxy), respectively. In this case, the final methoxy cleavage was performed using boron
tribromide (Scheme 4).

Scheme 3. Cleavage of O-benzyl protecting groups for the preparation of compounds 21, 38, 41. Reagents and
conditions: (a) N,N-Dimethylaniline, AlCl3, DCM, r.t., 10-20 min.

7
Scheme 4. Cleavage of dimethoxy groups for the preparation of compounds 42, 43, 45. Reagents and conditions:
(a) 1 M BBr3 in dry DCM, 0 °C then r.t. overnight.

2.2. Evaluation of inhibitory potency toward purified MBLs.

Compounds were tested against up to five representative subclass B1 MBLs, including VIM-1,
VIM-2, VIM-4, NDM-1 and IMP-1 (Tables 1 and 2). The compounds presented in Table 2
were also tested against the subclass B3 MBL L1 but were all poorly active (≤ 45% inhibition
at 100 M) or inactive against this enzyme. Testing was first performed at one concentration
(100 or 200 M) and Ki values were measured for compounds exhibiting > 75% inhibition at
these concentrations.
The results obtained for a first series of 37 compounds (15-51) are presented in Table 1. The
substituent at position 5 of the triazole ring was chosen among aromatic groups, which were
previously found to be favourable for MBL inhibition. In addition to phenyl, it also included o-
toluyl, 2-hydroxy-5-methoxy-phenyl, naphth-2-yl, o-, m- and p-biphenyl, N-methyl-pyrrol-2-
yl, benzyl and naphth-2-ylmethyl. At position 4, compounds 15-51 possessed a phenethyl-
derived substituent. As mentioned above, the choice of these substituents was initially based on
the previously published Schiff base series39 and included a 2,4-dihydroxyphenyl and a p-
benzyloxyphenyl group. However, a larger panel of substituents was explored as a function of
their commercial availability. In addition, because nitromethanation of 2-carboxybenzaldehyde
led to an isobenzofuranone group instead of the desired nitro-alkene derivative, several
compounds incorporating this substituent were prepared (i.e. 25, 33, 37, 39, 44, 46, 48, 51).
Overall, although some compounds were not active or were modest inhibitors of all tested
enzymes (e.g. 17, 24, 39, 47, 49-51), the others inhibited at least one tested enzyme with Ki
values in the micromolar to submicromolar range (0.41 to 42 M). Among these, several
strongly inhibited one to both tested VIM-type enzymes only (i.e. 15, 25, 27, 31-33, 45, 48).
Furthermore and most interestingly, 15 compounds (i.e. 18, 20-22, 26, 29, 30, 36, 37, 38, 41-
43) exhibited a broader spectrum of inhibition, with a detectable inhibitory activity on either

8
NDM-1 or IMP-1, or both, in addition to inhibiting one to both tested VIM-type enzymes. This
is in contrast with the series of 4-[2-(o-benzoic)ethyl] compounds, which displayed a restricted
inhibition spectrum (i.e. potent inhibition of VIM-type enzymes but not of NDM-1 or IMP-1),46
suggesting that the presence of a carboxylic group was not favorable for NDM-1 and IMP-1
inhibition in the 4-phenethyl series.

Table 1. Inhibitory activity of 1,2,4-triazole-3-thiones 15-51 with combined R1 and phenethyl-

based R2 against various MBLs.

Cpd Structure Ki (M)a or (% inhibition at b100 or c200 M)

R1 R2 VIM-1 VIM-2 VIM-4 NDM-1 IMP-1
15 3.2 ± 0.2 NId 2.1 ± 0.2 NI NI

16 NDe NI ND NI NI

17 ND NI NI NI (55%)a

18 3.1 ± 0.2 (30%)b NI 15.4 ± ND

3.1

19 ND NI ND NI NI

20 2.1 ± 0.4 6.1 ± 0.4 2.3 ± 0.1 12.0 ± 42.0 ±

2.2 3.5
21 5.1 ± 0.4 NI 4.1 ± 0.3 42f 27.0 ±
4.2

22 (34%)b 18.9 ± NI NI 5.3 ± 1.8

0.5
23 (50%)b (50%)c (50%)c NI NI

9
24 ND NI NI (45%)b (45%)b

25 0.75 ± NI 7.8 ± 0.7 (51%)b NI

0.10

26 3.3 ± 0.2 9.4 ± 0.9 2.4 ± 0.2 NI 14f

27 ND ND 1.6 ± 0.1 NI (56%)b

28 (47%)b NI NI NI NI

29 1.9 ± 0.2 3.9 ± 0.2 1.5 ± 0.1 6.8 ± 0.9 20.2 ±

1.0

30 2.2 ± 0.1 (41%)b 2.5 ± 0.1 9.5 ± 2.4 4.6 ± 0.3

31 1.6 ± 0.1 3.9 ± 0.3 1.9 ± 0.1 NI NI

32 11.1 ± NI 0.54 ± NI (35%)b

0.6 0.03

33 1.2 ± 0.1 3.3 ± 0.2 0.9 ± 0.1 (68%)b (47%)b

34 3.7 ± 0.4 NI NI NI NI

35 ND NI (35%)b NI NI

36 3.3 ± 0.2 NI 3.2 ± 0.2 NI 24.0 ±

2.8
37 6.3 ± 2.0 NI 8.2 ± 0.8 28f 9.8 ± 2.1

10
38 27.0 ± (35%)b 60.0 ± 15.2 ± NI
1.9 2.0 1.1

39 (30%)b NI (34%)a NI (54%)b

40 ND NI NI (69%)b NI

41 0.41 ± 1.3 ± 0.4 0.82 ± 14.3 ± NI

0.07 0.07 3.2

42 0.70 ± 1.8 ± 0.1 0.80 ± 22.4 ± 20.0 ±

0.07 0.05 2.5 3.4

43 1.3 ± 0.1 1.7 ± 0.1 0.68 ± 17.4 ± 9.2 ± 0.9

0.05 2.2
44 ND NI ND NI NI

45 0.76 ± 1.9 ± 0.2 0.85 ± (38%)b (44%)b

0.05 0.05

46 ND NI ND NI (48%)b

47 NI NI NI NI NI

48 2.2 ± 0.1 (38%)b 2.5 ± 0.2 (31%)b (30%)b

49 NI NI NI NI NI

11
 50 NI NI NI NI NI

51 NI NI NI NI (30%)b

aKinetics were monitored at 30 °C by following the absorbance variation observed upon reporter substrate
hydrolysis. Ki’s were determined when inhibition > 75% and values are mean ± SD. Assays performed in triplicate.
bPercentage of inhibition in the presence of 100 M of inhibitor. cPercentage of inhibition in the presence of 200

M of inhibitor. dNo or poor inhibition (< 30% or < 50% inhibition at 100 or 200 µM, respectively). eNot
determined. fApparent Ki value, not determined due to non linear v0/vi vs [I] plot.

Although no general rule could be drawn from the study of structure-activity relationships,
potent inhibitors in this series often showed a m-biphenyl (41-43, 45) or 2-hydroxy-4-methoxy-
phenyl (29-33) substituent as R1, and phenethyl (15, 26, 29) or a hydroxylated phenethyl (20,
21, 31, 36, 41-43, 45) substituent as R2. Some of these substituents were already shown
favourable in the parent Schiff base series.39 In addition, the constrained phenethyl analogues
phenylcyclopropyl (18, 30) and phthalidylmethyl (25, 33, 37, 48) were often well
accommodated. Interestingly, one of the most potent IMP-1 inhibitor (i.e. 22, Ki value of 5.3
M) possessed a p-benzyloxyphenyl at position 4, previously found the most favourable for
IMP-1 inhibition in the same Schiff base series.39
Among the active compounds, it is noteworthy that they often exhibited a better inhibition of
VIM-1 and VIM-4 enzymes than of VIM-2 (e. g. 15, 21, 30, 32, 36, 37, 48). This different
behaviour may come from the existence of residue variability in the VIM-type MBL active sites
as for instance the presence of Val223 and His224 residues for VIM-1 and VIM-4, where Ile
and Tyr are found in VIM-2, respectively. In fact, all potent VIM-2 inhibitors (Ki values in the
low micromolar range) were also potent VIM-1/4 inhibitors and they often displayed a
hydroxylated phenethyl moiety at position 4 (i.e. 20, 31, 41-43, 45).

Finally, the most interesting inhibitors were compounds 20 (phenyl in 5, p-hydroxyphenethyl

in 4), 29 (2-hydroxy-4-methoxy-phenyl in 5, phenethyl in 4), 42 and 43 (m-biphenyl in 5, 2,3-
or 3,4-dihydroxyphenethyl in 4, respectively.) as they displayed submicromolar to micromolar
Ki values against all five enzymes. In addition, although inactive against IMP-1, compound 41
(m-biphenyl in 5 and 2,4-dihydroxyphenethyl in 4) potently inhibited all VIM-type enzymes
and to a lesser extent NDM-1 (Ki value of 14.3 M). Compound 30 (2-hydroxy-4-methoxy-
phenyl in 5, cyclopropylphenyl in 4) also showed, despite being poorly active on VIM-2, among
the lowest measured Ki values (<10 μM) on NDM-1 and IMP-1, while retaining potent

12
inhibitory activity on VIM-1 and VIM-4. These results further highlight the potential of
triazole-thione compounds to be successfully optimized and yield broad-spectrum MBL
inhibitors.

A second series of 11 related analogues (52-62) was prepared to explore the importance of the
alkyl link between N4 of the triazole ring and a phenyl group, substituted or not (Table 2). This
link was either absent (52, 53) or being one (54-59), three (60, 61), or four carbon atoms-long
(62). None of these compounds potently inhibited NDM-1, IMP-1 (with the exception of
compound 54) or the B3 subclass MBL L1. More interesting results were obtained for VIM-
type enzymes, whatever the link length. In particular, several compounds (55, 56, 58, 60, 52)
strongly inhibited both VIM-type enzymes with Ki values in the micromolar to submicromolar
range. The absence of link (i.e. 52, 53) was the least favourable, while a benzyl moiety allowed
potent VIM inhibition (i.e. 54-56, 58).

Again, VIM-2 behaved differently than VIM-1 and VIM-4 as it was not inhibited by compound
54. In this sub-series, VIM-2 inhibition was promoted by either a carboxylic group attached to
the 4-benzyl group (i.e. 55, 56) or by methoxy and/or hydroxy group(s) present on the 5-phenyl
substituent (i.e. 58). In fact, a methoxy group at this position in the Schiff base analogue
JMV4690 was shown to establish an important hydrogen bond with the indole NH group of
Trp87 in the VIM-2 active site (pdb code 6YRP39). Compared to the corresponding 4-phenethyl
compound 1 (i.e. phenyl ring in both 4- and 5-substituents), the further increase of the alkyl
chain length with one (compound 60) or two (compound 62) carbon atom(s) led to high
inhibitory potency against all VIM-type MBLs. Compounds 60 and 62 are among the best VIM-
type MBL inhibitors in this series. It is interesting to note that one previously reported analogue
only differing from 62 by the presence of a sulfur instead of one CH2 behaved similarly with
regard to VIM-type enzymes.41 Therefore, this substituent configuration might be the basis for
a future series. Finally, as already observed for compounds 49-51 (Table 1) a benzyl group at
position 5 (i.e. 59, 61) was not favourable to MBL inhibition.

13
Table 2. Inhibitory activity of 1,2,4-triazole-3-thiones 42-62 with combined R1 and R2 against
various MBLs.

Structure Ki (M)a or (% inhibition at 100 M)

Cpd R1 R2 VIM-1 VIM-2 VIM-4 NDM-1 IMP-1 L1
52 3.2 ± 16.0 ± NIb NI (30%) NI
0.1 1.0

53 NDc ND NI NI NI NI

54 0.72 ± NI 1.5 ± NI 8.0 ± (35%)

0.04 0.1 0.8

55 3.6 ± 2.7 ± 1.1 ± NI NI (45%)

0.3 0.2 0.1
56 1.4 ± 1.8 ± 0.40 ± NI NI ND
0.1 0.1 0.02
57 NI (50%) NI NI NI NI

58 0.59 ± 4.0 ± 1.8 ± NI NI (40%)

0.20 0.6 0.8

59 23.0 ± NI NI NI (57%) (40%)

3.0

60 0.56 ± 0.74 ± 0.90 ± NI 14.0 ± NI

0.09 0.04 0.10 1.7

61 (30%) NI NI NI (63%) (32%)

62 1.2 ± 0.23 ± 0.43 ± (55%) NI (50%)

0.1 0.02 0.03

14
aKinetics were monitored at 30 °C by following the absorbance variation observed upon reporter substrate
hydrolysis. Ki’s were determined when inhibition > 75% and values are mean ± SD. Assays performed in triplicate.
bNo or poor inhibition (< 30% at 100 µM). cNot determined.

2.3. In vitro antibacterial synergistic activity

A total of 20 compounds (Table 3) were selected on the basis of their inhibitory activity to
evaluate their antibacterial synergistic activity with meropenem (a carbapenem antibiotic) on
MBL-producing multidrug-resistant clinical isolates (VIM-1- and VIM-4-producing K.
pneumoniae and NDM-1-producing E. coli). We first checked that none of the tested
compounds showed intrinsic antibacterial activity when tested alone (MIC > 128 g/mL). MIC
values of meropenem in the absence and presence of inhibitors (tested at a fixed concentration
of 32 μg/mL to allow direct comparison with previously reported data41,45,46) were determined
using the CLSI broth microdilution method and are reported in Table 3. Some compounds (e.
g. 15, 25, 31, 33, 45, 48, 56, 58, 60) were able to reduce the MIC of meropenem by 8 to 16-fold
on both K. pneumoniae isolates (Table 3). Four to eight-fold MIC reduction on both VIM-1-
and VIM-4-producing bacteria was also observed for compounds 21, 29, 30, 41, 55 and 62.
Curiously, the 41 isomers 42 and 43 showed similar potentiation activity as 41 on the VIM-4-
producing clinical isolate as expected (similar Ki values on VIM-4) but are inactive on the VIM-
1-producing one, despite close Ki values on this enzyme. This might potentially rely on strain-
specific factors, such as the composition of the outer membrane (OMPs, lipopolysaccharides)
considering that these two strains were epidemiologically unrelated and may show significantly
different genotypes. Furthermore, these two compounds contain a catechol moiety, which can
potentially act as a siderophore group known to improve uptake through bacterial membranes
via the iron acquisition system.50 These results would not support a drastic improvement of the
synergistic activity (only 43 showed a slightly better activity than 41) when a catechol group is
present in this series of triazole-thione analogues. Overall, no clear correlation between
structure, inhibitory potencies and microbiological activity could be drawn. Finally, no
significant activity was observed on the NDM-1-producing E. coli isolate, very probably
because most of these compounds are less potent NDM-1 inhibitors.

15
Table 3. Antibacterial synergistic activity of compounds on VIM-1- and VIM-4-producing K.
pneumoniae and NDM-1-producing E. coli clinical isolates with meropenem determined by the
broth microdilution method.
Cpd MEM MIC (g/mL)a and Ki (μM) values of selected inhibitors
(32 K. pneumoniae 7023 K. pneumoniae E. coli SI-004M
g/mL) (blaVIM-1+) VA416/02 (blaNDM-1+)
(blaVIM-4+)
MEM MIC Ki (μM)b MEM Ki (μM) MEM Ki (μM)
None 16 - 16 - 64 -
15 2 3.2 ± 0.2 2 2.1 ± 0.1 64 NIc
20 4 2.1 ± 0.4 4 2.3 ± 0.1 64 12.0 ± 2.2
21 4 5.1 ± 0.4 2 4.1 ± 0.2 NDd 42
25 2 0.75 ± 0.10 2 7.8 ± 0.7 64 (51%)e
26 4 3.3 ± 0.2 4 2.4 ± 0.2 ND NI
29 4 1.9 ± 0.2 2 1.5 ± 0.1 64 6.8 ± 0.9

30 4 2.2 ± 0.1 2 2.5 ± 0.1 32 9.5 ± 2.4

31 2 1.6 ± 0.1 2 1.9 ± 0.1 ND NI

33 2 1.2 ± 0.1 2 0.93 ± 0.04 64 (68%)e

41 2 0.41 ± 0.07 4 0.82 ± 0.07 32 14.3 ± 3.2

42 16 0.70 ± 0.07 4 0.80 ± 0.05 32 22.4 ± 2.5
43 16 1.3 ± 0.1 2 0.68 ± 0.05 32 17.4 ± 2.2

45 1 0.76 ± 0.05 2 0.85 ± 0.05 64 (38%)e

48 2 2.2 ± 0.1 2 2.5 ± 0.2 32 (31%)e

54 4 0.72 ± 0.04 4 1.5 ± 0.1 ND NI

55 4 3.6 ± 0.3 2 1.1 ± 0.1 64 NI
56 2 1.4 ± 0.1 2 0.40 ± 0.02 32 NI
58 2 0.59 ± 0.20 2 1.8 ± 0.8 32 NI
60 2 0.56 ± 0.09 2 0.90 ± 0.10 64 NI
62 2 1.2 ± 0.1 4 0.43 ± 0.03 64 (55%)e
aMEM, meropenem. bFrom Tables 1 and 2. cNo or poor inhibition (< 30% at 100 µM). dNot determined.
ePercentage of inhibition in the presence of 100 M of inhibitor.

16
In contrast with the Schiff base analogues,39 none of which showing potentiation activity in
similar assays, many compounds in this series did prove to significantly decrease the
meropenem MIC. These results confirmed the beneficial effect of replacing the hydrazone-like
bond by a non-hydrolysable one. Furthermore, a chequerboard analysis was carried out with
meropenem and compounds 43 and 60, selected based on their potentiation activity and
availability, using the VIM-4 producing K. pneumoniae clinical isolate (VA-416/02). With both
compounds, an average FIC index ≤ 0.5 was determined (0.48 and 0.25 for 43 and 60,
respectively), supporting a synergistic drug-drug interaction between the MBL inhibitor and
the antibiotic.51

2.4. Inhibitor selectivity and ACE inhibition

To probe the selectivity of some representative compounds, selected based on their activity in
microbiological assays and chemical diversity, an Angiotensin-Converting Enzyme (ACE, a
Zn-dependent peptidase) inhibition test was carried out with a fluorogenic substrate (see
Experimental section 4.2.3 for details). Captopril, a potent inhibitor of ACE was used as the
inhibition control and showed 96 % inhibition of ACE when tested at a final concentration of
100 nM. Compounds 33, 41, 54, 56 and 60, all showing among the best potentiation of
meropenem in microbiological assays, were tested at 100 μM. Compounds 33, 54 and 56 did
not show any inhibitory activity on ACE, while 41 and 60 showed a moderate inhibition of the
enzyme, with a percentage of inhibition equal to 35 and 50 %, respectively (SD < 10 %).
Although it is difficult to provide a structural basis for these results, it could be noted that the
compounds showing moderate ACE inhibition are characterized by a higher number of carbon
atoms (2 or 3) separating the triazole-thione moiety from the aryl group in the R2 side chain
than that in inactive molecules (1 carbon atom). Interestingly, the extent of ACE inhibition
seems to increase with the length of the alkyl linker, likely allowing a better flexibility of the
R2 substituent and thus potentially a better accommodation in the ACE active site. Nonetheless,
the selectivity of the tested compounds for metallo-β-lactamases remains favorable. Indeed, and
assuming a competitive mechanism of inhibition, the resulting Ki values (≈30 and 55 µM for
60 and 41, respectively; computed using a Km value of ACE for the fluorogenic tripeptide
substrate equal to 110 μM52) would be at least ≈30-fold higher than the inhibition constants
observed on VIM-type MBLs.

2.5. Cytotoxicity assays on human cells

17
The potential cytotoxicity of compound 41 was assessed using a membrane integrity assay
(HeLa cells). 41 was found to not induce cell lysis at concentrations up to 250 M. This result
was confirmed using a cell viability assay (HeLa cells, 1,500 cells/well), in which no cytotoxic
effects could be observed after up to 72h of incubation in the presence of 250 M of this
compound. It is in agreement with the absence of cytotoxicity generally observed for previous
series.39,41,45

2.6. Molecular modelling

We investigated the putative binding mode of compounds 41, 56 and 62 within VIM-2 active
site via docking experiments.

In experimental 3D structures of VIM-2 complexes with triazole-thione inhibitors (e. g. PDB

codes 7PP0, 7OVF), the hydroxide anion is absent as the two zincs are coordinated by the S
and N atoms of the triazole-thione moiety. This specific interaction leads to an increase of the
distance between the two zincs (from 3.7 Å to 4.2 Å) (as seen in structures of complexes where
the hydroxide anion is present, e.g. PDB codes 5ACW, 5LSC, 5NI0, 5NHZ, 6DD0, 6DD1,
6KW1, 6O5T, 6TGI, 6Y6J, 6YRP, 7A5Z, 7A60, 7OVE, 7OVF, 7OVH, 7PP0). Therefore, the
docking was optimised to reflect this mode of binding allowing to grasp potential interactions
of the compounds with the residues Arg228, Phe61 and Trp87, as a function of the nature of
the substituents present at positions 4 and 5.

Therefore, the docking experiments were performed with a VIM-2 model generated from
7PP046 available in the Protein data bank using AutoDock VINA 1.2.0.53 7PP0 is the
crystallographic structure of the complex formed between VIM-2 and the phenethyl analogue
JMV7038 (Ki = 0.34 M, Figure S1), which differs from 41 by possessing a phenyl ring meta-
substituted with a flexible morpholinyl-ethoxy moiety and a 2-(o-benzoic)ethyl group at
positions 5 and 4 of the triazole ring, respectively (Figure S1). The structure showed that the
1,2,4-triazole-3-thione core of JMV7038 simultaneously coordinated both active site zinc ions,
displacing the catalytic hydroxide anion and increasing the distance Zn1-Zn2, as observed in
other structures of 1,2,4-triazole-3-thione/VIM-2 complexes (PDB codes 6YRP, 5ACW, 6TGI,
7OVE, 7OVF, 7OVH). Among significant interactions, the carboxylate of the benzoic group is
stabilized through H-bond with the Asn233 backbone nitrogen and by additional water
mediated interactions. The benzoic phenyl ring also establishes a distorted π-π interaction with
the His263 imidazole. The m-alkoxy phenyl substituent in position 5 adopts two orientations

18
mutually rotated by ~ 180°. The phenyl moiety of both conformations is within van der Waals
contact to the Trp87 indole. But the morpholinyl-ethyl group is located in the solvent exposed
area and is too flexible to be resolved.

As already observed for the structure of VIM-2 in complex with the 1,2,4-triazole-3-thione
Schiff base JMV4690 (PDB code 6YRP,39 Figure S1), which also possesses a benzoic group at
the same position as JMV7038, the close Arg228 does not make electrostatic interaction with
the compound carboxylate groups. In fact, although expected, this interaction was prevented by
the bulkiness of the benzoyl cycle. Interestingly, the same behaviour was observed for unrelated
carboxylic-containing VIM-2 inhibitors (e. g. see PDB codes 4UA4, 5LCA, 5LM6, 5O7N).
Indeed, their acidic function also interacted with the Asn233 backbone nitrogen and water
molecules, while Arg228 kept the same position in all structures. Anyway, compound 56 also
possesses a carboxylic group but not the two others 41 and 62, the question arises as to whether
the Arg228 side-chain should be let freely moving or fixed. Indeed, docking is strongly
influenced by electrostatic interactions, which might inappropriately prevail over other ones.
Therefore, docking was performed with free and fixed Arg228.

2.6.1. Method validation with JMV7038

We first docked the original ligand in VIM-2 to check that our protocol allowed the
crystallographic positioning. It was not possible when the arginine side-chain was let moving,
as it favoured electrostatic interaction between the compound carboxylate and the arginine
guanidinium. In contrast, blocking the arginine orientation seen in 7PP0.pdb allowed to very
well reproduce the crystal pose, including the two opposite positionings of the 5-substituent
(Figure S2). Although the morpholinyl-ethyl moiety was not resolved in the crystal structure,
the solvent-free docking proposed possible interactions between the morpholino oxygen atom
with either the Tyr67 hydroxyl group (positioning 1) or the Asn150 NH2 group (positioning 2).

2.6.2 Compound 41.

This compound is characterized by a m-biphenyl and a 2,4-dihydroxy-phenethyl substituents at
positions 5 and 4, respectively. When the arginine side-chain was fixed, the correct positioning
of the 1,2,4-triazole-3-thione core was prevented by conflicts occurring between the compound
p-hydroxyl and biphenyl groups and Arg228 and Trp87, respectively. When the arginine was
free, the 1,2,4-triazole-3-thione core could be more adequately placed. The 2,4-
dihydroxyphenyl group could take two orientations, both of which interacting with Tyr67,
while the biphenyl could establish -stacking interaction with Phe61. However, some biphenyl
orientations could clash with Trp87. When comparing all sixty-six VIM-2 structures available

19
in the Protein Data Bank, the Trp87 position is highly conserved and rarely disturbed by a
ligand. But it is the case in the presence of a triazolylthioacetamide inhibitor (PDB code
5LSC54). So, we used the 5LSC structure to dock compound 41 letting Arg228 free. In this
case, the different orientation of Trp87 largely improved the positioning of the 1,2,4-triazole-
3-thione core. In addition, the biphenyl group made interesting - stacking interactions with
Phe61 and the p-hydroxyl group interacted with the Asp63 side-chain, which was further
stabilized by the Tyr67 phenol group (Figure 2A).

2.6.3. Compound 56
This compound significantly differs from JMV7038 and compound 41 as the link between the
triazole and the 4-substituent phenyl ring was one atom shorter, making this substituent less
mobile. When the arginine was let flexible, the compound carboxylate group interacted with
the Arg228 guanidinium and the Tyr67 hydroxyl group. However, this dominant interaction
did not allow the correct positioning of the 1,2,4-triazole-3-thione core between the two zinc
atoms. As observed with JMV7038 docking, fixing the arginine side-chain resolved the
problem. In that case, no interaction between the compound carboxylate and Arg228 was
established. The benzoic moiety displayed a different orientation compared to the JMV7038
one and established interactions with the same water molecules as well as Tyr67 (Figure 2B).

2.6.4. Compound 62
This compound is characterized by a four-atom long alkyl link and the absence of any functional
groups on the two substituents. Whether arginine was free or not, no significant change was
observed. The highly hydrophobic butylphenyl moiety was interacting within a hydrophobic
zone delimited by Tyr67 and Phe61 aromatic rings (Figure 2C).

Overall, it was possible to obtain for both compound the conserved crystallographic positioning
of the 1,2,4-triazole-3-thione that we assumed to be a prerequisite to MBL inhibition. Apart the
zinc-binding moiety, the favoured poses of both substituents was significantly different than
those observed for JMV7038 because of their structural distance (i.e. a bulky biphenyl group
in 41 or a variable link size in 56 and 62). In particular, while the benzoic group of JMV7038
(and also JMV4690) established hydrogen bonds in the Asn233 pocket, both 4-substituents of
docked compounds were proposed to interact with or close to Tyr67, leading to important side-
chain movement(s) in the hydrophobic patch including Phe61 and Tyr67 (Figure 2). This study
afforded valuable information to further optimize these compounds.

20
Figure 2. Binding mode of compounds 41 (A), 56 (B) and 62 (C) in VIM-2 (blue ribbon and yellow side chains)
studied by molecular modeling. Compounds (orange) were docked using the structure 7PP0.pdb, excepted
compound 41, which used 5LSC.pdb. In the case of compound 62 (C), only the pose obtained with fixed Arg228
was shown. Protein side chains, which have moved from their position in 7PP0 (yellow) are in green. All three

21
compounds interact with the two zinc ions as determined for JMV7038. The images were produced using UCSF
Chimera.55

The scores (i.e. binding affinity in kcal/mol) and ligand efficiencies (LE) were calculated for
best poses and are presented in Table S1. The low ligand efficiency calculated for JMV7038
(0.20) compared to other compounds is probably due to the fact that a significant portion of its
structure (i.e. the morpholinylethyl moiety) does not establish stable interaction with the VIM-2
active site.

3. Conclusion

When a direct comparison could be made with their corresponding Schiff base analogues (i.e.
15, 21, 26-28, 38, 41, 50),39 the compounds presented in this study were globally better MBL
inhibitors, with a few exceptions (i.e. 28, 38). In particular, compound 41 was much more potent
against VIM-type enzymes and also inhibited NDM-1. When possible to compare with the 4-
[2-(o-benzoic)ethyl] series,46 which inhibition spectrum was restricted to VIM-type MBLs, the
absence of the carboxylic group (i.e. unsubstituted phenethyl group at position 4 for compounds
15, 26, 29, 34) generally led to a decrease in inhibitory potency toward VIM-2. But importantly,
in one case, a broader inhibition spectrum was observed as compound 29 also inhibited NDM-1
and IMP-1. In fact, several other compounds of the present series, all possessing at least one
phenol group on the 4-substituent, showed broad-spectrum of inhibition (20, 41-43).
Compared to previous series, while several aryl groups of substituents at position 4 were already
known to be favourable for MBL inhibition (i.e. 2,4-dihydroxyphenyl and p-benzyloxyphenyl),
other new moieties unsubstituted on their phenyl group were here shown useful. It is the case
of phenylcyclopropyl (i.e. 18, 30) or isobenzofuranone (i.e. 25, 33, 37, 48). Furthermore,
increasing the alkyl length to a propyl or a butyl yielded very potent VIM-type MBL inhibitors
(60 and 62, respectively).
In contrast to the Schiff base analogues, which were devoid of synergistic activity when tested
in clinical isolates in combination with a -lactam antibiotic,39 but as observed for the 4-[2-(o-
benzoic)ethyl] series,46 some compounds were able to potentiate the activity of meropenem
against two VIM-producing clinical isolates of K. pneumoniae.
Therefore, because of its activity in bacteria and its potential to inhibit both VIM-type, NDM-
1 and IMP-1 MBLs, this series deserves further development.

22
4. Experimental section

4.1. Chemistry

Hydrazides R1-CO-NHNH2 and amines R2-NH2 were prepared as described in Supplementary

material part.

4.1.1. General procedure for the preparation of 4-alkyl-1,2,4-triazole-3-thiones diversely

substituted at position 5
The procedure followed the synthetic pathway reported by Deprez-Poulain et al.47

4.1.1.1. Thiosemicarbazide intermediates

To a solution of the amine R2-NH2 (1 mmol, 1 equiv.) in anhydrous DMF (4 mL) (in the case
of hydrochloride salt, 1.5 equiv. of Na2CO3 (159 mg) was added) was added DPT (244 mg,
1.05 mmol, 1.05 equiv.). The reaction mixture was stirred at 55 °C in a sealed tube for 1h30.
The hydrazide R1-CONHNH2 (1.1 mmol, 1.1 equiv.) was then added and the mixture was again
heated at 55 °C for 1h30, and allowed to cool to room temperature. The solution was diluted in
EtOAc and the organic phase was extracted five times with water, dried over MgSO4 and
concentrated in vacuum. If necessary, the product was purified by gel column chromatography
(EtOAc/Hexane).

4.1.1.2. Cyclization
The thiosemicarbazide intermediates were solubilized in a mixture of water and ethanol (2:3)
and KOH (3 equiv., 3 mmol) was added. The reaction mixture was refluxed for 2h. The mixture
was then neutralized with saturated aqueous KHSO4 and extracted twice with DCM. The
organic phases were mixed, dried over MgSO4, filtered and evaporated under vacuum. The
residues were purified by gel column chromatography (EtOAc/Hexane), reverse phase HPLC
or recrystallization.

The purity of all compounds was determined to be ≥95% by 1H NMR and LC-MS analysis.
Their characteristics are presented in the Supplementary material part.

4.2. Biology

4.2.1. Metallo--Lactamase inhibition assays

4.2.1.1. VIM-type enzymes, NDM-1 and IMP-1

The inhibition potency of the compounds has been assessed as previously reported.39,41 The rate
of hydrolysis of a reporter substrate (150 μM imipenem or meropenem), by a purified MBL

23
enzyme (VIM-1, VIM-2, VIM-4, NDM-1, and IMP-1; enzyme concentration in the assay
ranged 1-70 nM) was measured at 30 °C in 50 mM HEPES buffer (pH 7.5) in the absence and
presence of several concentrations of the inhibitor (0.5 µM –1 mM), by following the
absorbance variation at  = 300 nm.
The inhibition constants (Ki) were determined on the basis of a model of competitive inhibition
by analysing the dependence of the ratio v0/vi (v0, hydrolysis velocity in the absence of inhibitor;
vi, hydrolysis velocity in the presence of inhibitor) as a function of [I] as already described.56
The slope of the plot of v0/vi vs [I], which corresponds to KmS/( KmS + [S])Ki (where KmS is the
Km value of the reporter substrate and [S] its concentration in the reaction mixture) and allowed
the calculation of the Ki value. Alternatively, a Dixon plot analysis was carried out by measuring
the initial hydrolysis rates in the presence of variable concentrations of inhibitor and substrate.
This allowed Ki values to be determined and supported the hypothesis that the various
compounds behaved as competitive inhibitors of the various tested enzymes. The assays were
performed in triplicate.

4.2.1.2. L1
L1 inhibition was measured as previously described.39

4.2.2. Microbiological assays

The minimum inhibitory concentrations (MICs) of meropenem (MEM) were determined using
Mueller-Hinton broth and a bacterial inoculum of 5 × 104 CFU/well, as recommended by the
CLSI,57 in both the absence and presence of a fixed concentration (32 μg/ml) of an inhibitor, as
previously described.41,46 VIM-1- and VIM-4-producing K. pneumoniae (strains 7023 and VA-
416/02, respectively) and NDM-1-producing Escherichia coli (SI-004M) clinical isolates
present in our collection were used. Chequerboard analysis, used to confirm the nature of the
interaction between meropenem and selected active inhibitors (synergistic vs additive activity),
was carried out and interpreted as previously described.41 All experiments were performed in
triplicate.

4.2.3. Angiotensin-converting enzyme (ACE) inhibition assays

The potential inhibitory activity of compounds on the Zn-dependent ACE was determined
essentially as described in Sentandreu and Toldrá.58 ACE enzyme from rabbit lung was
purchased from Sigma (St Louis, Miss., USA; cat. no. A6778). The fluorogenic tripeptide
substrate o-aminobenzoylglycyl-p-nitro-L-phenylalanyl-L-proline (Abz-Gly-Phe(NO2)-Pro)59
was purchased from FluoProbes (Interchim, Montluçon, France). The enzyme was resuspended

24
at 0.3 U/mL (≈150 μg/mL) in 150 mM Tris (pH, 8.3), 1 μM ZnCl2, 50 % glycerol buffer and
kept at -20 °C until use. Fluorescence of the reaction product Abz-Gly-OH (i. e. generated after
cleavage of the peptide with ACE) was measured in the reaction buffer (150 mM Tris, 750 mM
NaCl, pH 8.3) using a monochromators-equipped Envision microplate reader (Perkin-Elmer,
Waltham, Mass., USA) and excitation and emission wavelengths of 316 and 413 nm,
respectively. Purified Abz-Gly-OH was used to establish the linearity between measured
fluorescence and product concentration (range, 0-50 μM) in the reaction buffer. The rate of
hydrolysis of the fluorogenic substrate was measured in both the absence (v0) and presence (vi)
of 100 μM compound by monitoring the increase of blank-corrected fluorescence for up to 90
minutes, in the reaction buffer (final reaction volume, 300 μL) and 1.5 mUnits of ACE per well.
The percentage of inhibition was computed as 100 - [(vi / vo) × 100]. Controls included substrate
alone and reactions in the presence of 100 nM captopril, a well-known potent inhibitor of ACE.

4.2.4. Cell toxicity assay

The potential cytotoxic activity of compound 41 was evaluated on HeLa cell cultures using the
commercially available membrane integrity assay (CytoTox 96® non-radioactive cytotoxicity
assay, Promega, Madison, WI, U.S.A.) as previously described.39 The cytotoxicity of
compounds was also assessed using the RealTime-GloTM MT Cell Viability Assay (Promega).39
The assays were performed in triplicate.

4.3. Molecular modelling

The geometric optimization of the compound structures was done using Gaussian 16 at DFT
level of theory with B3LYP hybrid functional and 6-31g basis set. The resulting structures were
registered as mol2 files. AM1 charges were calculated using Chimera software and
Antechamber. Then, in AutoDockTools (ADT),60 the final files (pdbqt) were generated where
AM1 charges were kept and all possible rotations were free.
The docking studies were performed with AutoDock Vina 1.2.0.53 This new version allows
docking on zinc metallo-enzymes. The method consists in placing fictional points (TZ) to
complete the Zn coordination spheres. They will serve to anchor the 1,2,4-triazole-3-thione core
of the ligands. Files for the docking were prepared from: (i) the structure of complex VIM-
2/JMV7038 (PDB code 7PP0),46 which was treated as follows: water molecules (with the
exception of structural waters 5 and 177), small molecules (acetate, DMSO and ethylene glycol)
and the third zinc ion were removed; The protein was protonated with Dockprep module of
Chimera and the program was let to itself protonate His residues, while checking the absence

25
of conflict in the H-bond network; AM1 charges were calculated using Chimera55 and
Antechamber and the file was registered as mol2 format. The center and dimensions of the
docking grid box were obtained using ADT software and the size of the search zone was 30 Å
x 30 Å x 30 Å. The structure 5LSC.pdb,54 which was also used for compound 41 was similarly
treated; (ii) compounds and protein pdbqt files prepared with ADT. For the protein, the side-
chain of Arg228 was let free or fixed.
Molecular graphics and analyses were performed with UCSF Chimera, developed by the
Resource of Biocomputing, Visualization and Informatics at the University of California, San
Francisco, with support from NIH P41-GM103311.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

Part of this work was supported by Agence Nationale de la Recherche (ANR-14-CE16-0028-

01, including fellowship to L.S.). We thank Mr Pierre Sanchez for mass spectrometry analyses.

Appendix A. Supplementary material

Supplementary data to this article can be found online at…

References

[1] Nordmann P, Naas T, Poirel L. Global spread of carbapenemase-producing

Enterobacteriaceae. Emerg Infect Dis. 2011;17:1791-1798. DOI: 10.3201/eid1710.110655

[2] Walsh TR, Toleman MA. The emergence of pan-resistant Gram-negative pathogens merits
a rapid global political response. J Antimicrob Chemother. 2012;67:1-3.
DOI: 10.1093/jac/dkr378

[3] Reygaert WC. An overview of the antimicrobial resistance mechanisms of bacteria. AIMS
Microbiol. 2018;4:482-501. DOI: 10.3934/microbiol.2018.3.482

[4] World Health Organization, Global priority list of antibiotic-resistant bacteria to guide
research, discovery and development of new antibiotics, 27 february 2017.

26
[5] Tooke CL, Hinchliffe P, Bragginton EC, Colenso CK, Hirvonen VHA, Takebayashi Y,
Spencer J. -Lactamases and -lactamase inhibitors in the 21st century. J Mol Biol.
2019;431:3472-3500. DOI: 10.1016/j.jmb.2019.04.002

[6] Bush K. Past and present perspectives on -lactamases. Antimicrob Agents Chemother.
2018;62: e01076-18. DOI: 10.1128/AAC.01076-18

[7] Bahr G, Gonzalez LJ, Vila AJ. Metallo--lactamase inhibitors in the age of multidrug
resistance: from structure and mechanism to evolution, dissemination, and inhibitor design.
Chem Rev. 2021;121:7957-8094. DOI: 10.1021/acs.chemrev.1c00138

[8] Palzkill T. Metallo--lactamase structure and function. Ann. N. Y. Acad. Sci. 2013;1277:91-
104. DOI: 10.1111/j.1749-6632.2012.06796.x

[9] Gajamer VR, Bhattacharjee A, Paul D, Deshamukhya C, Singh AK, Pradhan N, Tiwari HK.
Escherichia coli encoding blaNDM-5 associated with community-acquired urinary tract
infections with unusual MIC creep-like phenomenon against imipenem. J Glob Antimicrob
Resist. 2018;14:228-232. DOI: 10.1016/j.jgar.2018.05.004

[10] Docquier J-D, Mangani S. An update on -lactamase inhibitor discovery and development,
Drug Resist Updat. 2018;36:13-29. DOI: 10.1016/j.drup.2017.11.002

[11] Gonzalez-Bello C, Rodriguez D, Pernas M, Rodriguez A, Colchon E. -Lactamase

inhibitors to restore the efficacy of antibiotics against superbugs. J Med Chem. 2020;63:1859-
1881. DOI: 10.1021/acs.jmedchem.9b01279

[12] McGeary RP, Tan DT, Schenk G. Progress toward inhibitors of metallo--lactamases.
Future Med Chem. 2017;9:673-691. DOI: 10.4155/fmc-2017-0007

[13] Palacios AR, Rossi M-A, Mahler GS, Vila AJ. Metallo--lactamase inhibitors inspired on
snapshots from the catalytic mechanism. Biomolecules 2020;10:854.
DOI: 10.3390/biom10060854

[14] Yan Y-H, Li G, Li G-B. Principles and current strategies targeting metallo--lactamases
mediated antibacterial resistance. Med Res Rev. 2020;40:1558-1592. DOI: 10.1002/med.21665

[15] Liénard BM, Garau G, Horsfall L, Karsisiotis AI, Damblon C, Lassaux P, Papamicael C,
Roberts GCK, Galleni M, Dideberg O, Frère J-M, Schofield CJ. Structural basis for the broad-
spectrum inhibition of metallo--lactamases by thiols. Org Biomol Chem. 2008;6:2282-2294.
DOI: 10.1039/b802311e

27
[16] Lassaux P, Hamel M, Gulea M, Delbrück H, Mercuri PS, Horsfall L, Dehareng D, Kupper
M, Frère J-M, Hoffmann K, Galleni M, Bebrone C. Mercaptophosphonate compounds as broad-
spectrum inhibitors of the metallo-β-lactamases. J Med Chem. 2010;53:4862-4876. DOI:
10.1021/jm100213c

[17] Gonzalez MM, Kosmopoulou M, Mojica MF, Castillo V, Hinchliffe P, Pettinati I, Brem J,
Schofield CJ, Mahler G, Bonomo RA, Llarrull LI, Spencer J, Vila AJ. Bisthiazolidines: a
substrate-mimicking scaffold as an inhibitor of the NDM-1 carbapenemase. ACS Inf Dis.
2015;1:544-554. DOI: 10.1021/acsinfecdis.5b00046

[18] Kaya C, Konstantinović J, Kany AM, Andreas A, Kramer JS, Brunst S, Weizel L, Rotter
MJ, Frank D, Yahiaoui S, Müller R, Hartmann RW, Haupenthal J, Proschak E, Wichelhaus TA,
Hirsch AKH. N-Aryl mercaptopropionamides as broad-spectrum inhibitors of metallo-β-
lactamases. J Med Chem. 2022;65:3913-3922. DOI: 10.1021/acs.jmedchem.1c01755

[19] Toney JH, Hammond GG, Fitzgerald PM, Sharma N, Balkovec JM, Rouen GP, Olson SH,
Hammond ML, Greenlee ML, Gao YD. Succinic acids as potent inhibitors of plasmid-borne
IMP-1 metallo--lactamase. J Biol Chem. 2001;276:31913-31918. DOI:
10.1074/jbc.M104742200

[20] Chen AY, Thomas PW, Stewart AC, Bergstrom A, Cheng Z, Miller C, Bethel CR,
Marshall SH, Credille CV, Riley CL, Page RC, Bonomo RA, Crowder MW, Tierney DL, Fast
W, Cohen SM. Dipicolinic acid derivatives as inhibitors of New Delhi Metallo-β-lactamase-1.
J Med Chem. 2017;60:7267-7283. DOI: 10.1021/acs.jmedchem.7b00407

[21] King AM, Reid-Yu SA, Wang W, King DT, De Pascale G, Strynadka NC, Walsh TR,
Coombes BK, Wright GD. Aspergillomarasmine A overcomes metallo-β-lactamase antibiotic
resistance. Nature 2014;510:503-506. DOI: 10.1038/nature13445

[22] Bergstrom A, Katko A, Adkins Z, Hill J, Cheng Z, Burnett M, Yang H, Aitha M, Mehaffey
MR, Brodbelt JS, Tehrani KH, Martin NI, Bonomo RA, Page RC, Tierney DL, Fast W, Wright
GD, Crowder MW. Probing the interaction of aspergillomarasmine A with metallo--lactamase
NDM-1, VIM-2, and IMP-7. ACS Infect Dis. 2018;4:135-145. DOI:
10.1021/acsinfecdis.7b00106

[23] Samuelsen O, Astrand OAH, Fröhlich C, Heikal A, Skagseth S, Carlsen TJO, Leiros HS,
Bayer A, Schnaars C, Kildahl-Andersen G, Lauksund S, Finke S, Huber S, Gjoen T, Andresen
AMS, Okstad OA, Rongved P. ZN148 is a modular synthetic metallo--lactamase inhibitor that

28
reverses carbapenem-resistance in Gram-negative pathogens in vivo. Antimicrob Agents
Chemother. 2020;64:e02415-19. DOI: 10.1128/AAC.02415-19

[24] Chen AY, Adamek RN, Dick BL, Credille CV, Morrison CN, Cohen SM. Targeting
metalloenzymes for therapeutic intervention. Chem Rev. 2019;119:1323-1455.
DOI: 10.1021/acs.chemrev.8b00201

[25] Reddy N, Shungube M, Arvidsson PI, Baijnath S, Kruger HG, Govender T, Naicker T. A
2018-2019 patent review of metallo--lactamase inhibitors. Expert Opin Ther Pat.
2020;30:541-555. DOI: 10.1080/13543776.2020.1767070

[26] Davies DT, Leiris S, Sprynski N, Castandet J, Lozano C, Bousquet J, Zalacain M, Vasa S,
Dasari PK, Pattipati R, Vempala N, Gujjewar S, Godi S, Jallala R, Sathyap RR, Darshanoju
NA, Ravu VR, Juventhala RR, Pottabathini N, Sharma S, Pothukanuri S, Holden K, Warn P,
Marcoccia F, Benvenuti M, Pozzi C, Mangani S, Docquier J-D, Lemonnier M, Everett M.
ANT2681: SAR studies leading to the identification of a metallo-lactamase inhibitor with
potential for clinical use in combination with meropenem for the treatment of infections caused
by NDM-producing Enterobacteriaceae. ACS Infect Dis. 2020;6:2419-2430.
DOI: 10.1021/acsinfecdis.0c00207

[27] Burns CJ, Daigle D, Liu B, McGarry D, Pevear DC, Trout RE. -Lactamase inhibitors.
WO Patent WO 2014/ 089365 A1.

[28] Krajnc A, Brem J, Hinchliffe P, Calvopiña K, Panduwawala TD, Lang PA, Kamps JJAG,
Tyrrell JM, Widlake E, Saward BG, Walsh TR, Spencer J, Schofield CJ. Bicyclic boronate
VNRX-5133 inhibits metallo- and serine β-lactamases. J Med Chem. 2019;62:8544-8556. DOI:
10.1021/acs.jmedchem.9b00911

[29] Liu B, Trout REL, Chu GH, McGarry D, Jackson RW, Hamrick JC, Daigle DM, Cusick
SM, Pozzi C, De Luca F, Benvenuti M, Mangani S, Docquier J-D, Weis WJ, Pevear DC, Xerri
L, Burns CJ. Discovery of Taniborbactam (VNRX-5133): a broad spectrum serine- and
metallo--lactamase inhibitor for carbapenem-resistant bacterial infections. J Med Chem.
2020;63:2789-2801. DOI: 10.1021/acs.jmedchem.9b01518

[30] Hamrick JC, Docquier JD, Uehara T, Myers CL, Six DA, Chatwin CL, John KJ,
Vernacchio SF, Cusick SM, Trout REL, Pozzi C, De Luca F, Benvenuti M, Mangani S, Liu B,
Jackson RW, Moeck G, Xerri L, Burns CJ, Pevear DC, Daigle DM. VNRX-5133
(Taniborbactam), a broad-spectrum inhibitor of serine- and metallo-β-lactamase, restores

29
activity of cefepime in Enterobacterales and Pseudomonas aeruginosa. Antimicrob Agents
Chemother. 2020;64:e01963-19. DOI: 10.1128/AAC.01963-19

[31] Hecker SJ, Reddy KR, Lomovskaya O, Griffith DC, Rubio-Aparicio D, Nelson K,
Tsivkovski R, Sun D, Sabet M, Tarazi Z, Parkinson J, Totrov M, Boyer SH, Glinka TW,
Pemberton OA, Chen Y, Dudley MN. Discovery of cyclic boronic acid QPX7728, an ultra-
broad-spectrum inhibitor of serine and metallo--lactamases. J Med Chem. 2020;63:7491-
7507. DOI: 10.1021/acs.jmedchem.9b01976

[32] Olsen L, Jost S, Adolph HW, Pettersson I, Hemmingsen L, Jørgensen FS. New leads of
metallo--lactamase inhibitors from structure-based pharmacophore design, Bioorg Med Chem.
2006;14:2627-2635. DOI: 10.1016/j.bmc.2005.11.046

[33] Toney JH, Fitzgerald PM, Grover-Sharma N, Olson SH, May WJ, Sundelof JG,
Vanderwall DE, Cleary KA, Grant SK, Wu JK, Kozarich JW, Pompliano DL, Hammond GG.
Antibiotic sensitization using biphenyl tetrazoles as potent inhibitors of Bacteroides fragilis
metallo-beta-lactamase. Chem Biol. 1998;5:185-196. DOI: 10.1016/s1074-5521(98)90632-9

[34] Hussein WM, Fatahala SS, Mohamed ZM, McGeary RP, Schenk G, Ollis DL, Mohamed
MS. Synthesis and kinetic testing of tetrahydropyrimidine-2-thione and pyrrole derivatives as
inhibitors of the metallo-β-lactamase from Klebsiella pneumonia and Pseudomonas
aeruginosa. Chem Biol Drug Des. 2012;80:500-515. DOI: 10.1111/j.1747-0285.2012.01440.x

[35] Zhai L, Zhang YL, Kang JS, Oelschlaeger P, Xiao L, Nie SS, Yang KW.
Triazolylthioacetamide: a valid scaffold for the development of New Delhi Metallo-β-
Lactamase-1 (NDM-1) inhibitors. ACS Med Chem Lett. 2016;7:413-417. DOI:
10.1021/acsmedchemlett.5b00495

[36] Muhammad Z, Skagseth S, Boomgaren M, Akhter S, Fröhlich C, Ismael A, Christopeit T,

Bayer A, Leiros HS. Structural studies of triazole inhibitors with promising inhibitor effects
against antibiotic resistance metallo-β-lactamases. Bioorg Med Chem. 2020;28:115598. DOI:
10.1016/j.bmc.2020.115598

[37] Nauton L, Kahn R, Garau G, Hernandez J-F, Dideberg O. Structural insights into the
design of inhibitiors of the L1 metallo--lactamase from Stenotrophomonas maltophilia. J Mol
Biol. 2008;375:257-269. DOI: 10.1016/j.jmb.2007.10.036

30
[38] Christopeit T, Carlsen TJ, Helland R, Leiros HK. Discovery of novel inhibitor scaffolds
against the metallo--lactamase VIM-2 by surface plasmon resonance (SPR) based fragment
screening. J Med Chem. 2015;58:8671-8682. DOI: 10.1021/acs.jmedchem.5b01289

[39] Gavara L, Sevaille L, De Luca F, Mercuri P, Bebrone C, Feller G, Legru A, Cerboni G,

Tanfoni S, Baud D, Cutolo G, Bestgen B, Chelini G, Verdirosa F, Sannio F, Pozzi C, Benvenuti
M, Kwapien K, Fischer M, Becker K, Frère J-M, Mangani S, Gresh N, Berthomieu D, Galleni
M, Docquier J-D, Hernandez J-F. 4-Amino-1,2,4-triazole-3-thione-derived Schiff bases as
metallo--lactamase inhibitors. Eur J Med Chem. 2020;208:112720.
DOI: 10.1016/j.ejmech.2020.112720

[40] Spyrakis F, Santucci M, Maso L, Cross S, Gianquinto E, Sannio F, Verdirosa F, De Luca

F, Docquier J-D, Cendron L, Tondi D, Venturelli A, Cruciani G, Costi MP. Virtual screening
identifies broad-spectrum -lactamase inhibitors with activity on clinically relevant serine- and
metallo-carbapenemases. Sci Rep. 2020;10:12763. DOI: 10.1038/s41598-020-69431-y

[41] Legru A, Verdirosa F, Hernandez J-F, Tassone G, Sannio F, Benvenuti M, Conde P-A,
Bossis G, Thomas CA, Crowder MW, Dillenberger M, Becker K, Pozzi C, Mangani S,
Docquier J-D, Gavara L. 1,2,4-Triazole-3-thione compounds with a 4-ethyl alkyl/aryl sulfide
substituent are broad-spectrum metallo--lactamase inhibitors with re-sensitization activity.
Eur J Med Chem. 2021;226:113873. DOI: 10.1016/j.ejmech.2021.113873

[42] Vella P, Hussein WM, Leung EW, Clayton D, Ollis DL, Mitić N, Schenk G, McGeary RP.
The identification of new metallo--lactamase inhibitor leads from fragment-based screening.
Bioorg Med Chem Lett. 2011;21:3282-3285. DOI: 10.1016/j.bmcl.2011.04.027

[43] Spyrakis F, Celenza G, Marcoccia F, Santucci M, Cross S, Bellio P, Cendron L, Perilli M,

Tondi D. Structure-based virtual screening for the discovery of novel inhibitors of New Delhi
Metallo--lactamase-1. ACS Med Chem Lett. 2018;9:45-50. DOI:
10.1021/acsmedchemlett.7b00428

[44] Gavara L, Verdirosa F, Legru A, Mercuri PS, Nauton L, Sevaille L, Feller G, Berthomieu
D, Sannio F, Marcoccia F, Tanfoni S, De Luca F, Gresh N, Galleni M, Docquier J-D, Hernandez
J-F. 4-(N-Alkyl- and -acyl-amino)-1,2,4-triazole-3-thione analogs as metallo--lactamase
inhibitors: impact of 4-linker on potency and spectrum of inhibition. Biomolecules
2020;10:1094. DOI: 10.3390/biom10081094

31
[45] Gavara L, Legru A, Verdirosa F, Sevaille L, Nauton L, Corsica G, Mercuri PS, Sannio F,
Feller G, Coulon R, De Luca F, Cerboni G, Tanfoni S, Chelini G, Galleni M, Docquier J-D,
Hernandez J-F. 4-Alkyl-1,2,4-triazole-3-thione analogues as metallo--lactamase inhibitors.
Bioorg Chem. 2021;113:105024. 10.1016/j.bioorg.2021.105024

[46] Verdirosa F, Gavara L, Sevaille L, Tassone G, Corsica G, Legru A, Feller G, Chelini G,

Mercuri PS, Tanfoni S, Sannio F, Benvenuti M, Cerboni G, De Luca F, Bouajila E, Vo Hoang
Y, Licznar-Fajardo P, Galleni M, Pozzi C, Mangani S, Docquier J-D, Hernandez J-F. 1,2,4-
Triazole-3-thione analogues with a 2-ethylbenzoic acid at position 4 as VIM-type metallo--
lactamase inhibitors. ChemMedChem 2022 ; In press. DOI: 10.1002/cmdc.202100699

[47] Deprez-Poulain RF, Charton J, Leroux V, Deprez BP. Convenient synthesis of 4H-1,2,4-
triazole-3-thiols using di-2-pyridylthionocarbamate. Tetrahedron Lett. 2007;48:8157-8162.
https://doi.org/10.1016/j.tetlet.2007.09.094

[48] Sevaille L, Gavara L, Bebrone C, De Luca F, Nauton L, Achard M, Mercuri P, Tanfoni S,

Borgianni L, Guyon C, Lonjon P, Turan-Zitouni G, Dzieciolowski J, Becker K, Bénard L,
Condon C, Maillard L, Martinez J, Frère J-M, Dideberg O, Galleni M, Docquier J-D, Hernandez
J-F. 1,2,4-Triazole-3-thione compounds as inhibitors of dizinc metallo--lactamase.
ChemMedChem 2017;12:972-985. DOI: 10.1002/cmdc.201700186

[49] Akiyama T, Hirofuji H, Ozaki S. AlCl3-N,N-dimethylaniline: a new benzyl and allyl ether
cleavage reagent. Tetrahedron Lett. 1991;32:1321-1324. DOI: https://doi.org/10.1016/S0040-
4039(00)79656-0

[50] Negash KH, Norris JKS, Hodgkinson JT. Siderophore-antibiotic conjugate design: new
drugs for bad bugs? Molecules 2019;24:3314. DOI: 10.3390/molecules24183314

[51] European Committee for Antimicrobial Susceptibility Testing (EUCAST) of the European
Society of Clinical Microbiology and Infectious Diseases (ESCMID). EUCAST Definitive
Document E.Def 1.2. Terminology relating to methods for the determination of susceptibility
of bacteria to antimicrobial agents. Clin Microb Infect. 2000; 6: 503-508. DOI: 10.1046/j.1469-
0691.2000.00149.x

[52] Sentandreu MA, Toldrá F. A rapid, simple and sensitive fluorescence method for the assay
of angiotensin-I converting enzyme, Food Chem. 2006;97: 546-554. DOI:
https://doi.org/10.1016/j.foodchem.2005.06.006.

32
[53] Eberhardt J, Santos-Martins D, Tillack AF, Forli S. AutoDock Vina 1.2.0: New docking
methods, expanded force field, and python bindings. J Chem Inf Model. 2021;61:3891-3898.
DOI: 10.1021/acs.jcim.1c00203

[54] Christopeit T, Yang KW, Yang SK, Leiros HKS. The structure of the metallo--lactamase
VIM-2 in complex with a triazolylthioacetamide inhibitor. Acta Crystallogr. F 2016;72:813-
819. DOI: 10.1107/S2053230X16016113

[55] Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE.
UCSF Chimera, a visualization system for exploratory research and analysis. J Comput Chem.
2004;25:1605-1612. DOI: 10.1002/jcc.20084

[56] Docquier J-D, Lamotte-Brasseur J, Galleni M, Amicosante G, Frère J-M, Rossolini GM.
On functional and structural heterogeneity of VIM-type metallo--lactamases. J Antimicrob
Chemother. 2003;51:257-266. DOI: 10.1093/jac/dkg067

[57] Clinical Laboratory Standard Institute, Methods for dilution antimicrobial susceptibility
tests for bacteria that grow aerobically, Document M07-A10, 2015, Twelfth Edition, Wayne,
PA, USA.
[58] Sentandreu MA, Toldrá F. A fluorescence-based protocol for quantifying angiotensin-
converting enzyme activity. Nat Protoc. 2006;1:2423-2427. DOI: 10.1038/nprot.2006.349
[59] Carmel A, Yaron A. An intramolecularly quenched fluorescent tripeptide as a fluorogenic
substrate of angiotensin-I-converting enzyme and of bacterial dipeptidyl carboxypeptidase. Eur
J Biochem. 1978;87:265-273. DOI: 10.1111/j.1432-1033.1978.tb12375.x
[60] Sanner MF. Python: a programming language for software integration and development. J
Mol Graph Model. 1999;17:57–61. PMID: 10660911.

33
Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.

☐The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests:

34
- Fourty-eight 1,2,4-triazole-3-thione compounds with an arylalkyl moiety at position 4 are
reported.

- Several analogues were broad-spectrum inhibitors of clinically important MBLs.

- Several analogues restored meropenem activity on VIM-1/4-producing clinical isolates.

- The binding mode of three compounds to VIM-2 was studied by molecular modelling.

35
36