Synthesis and evaluation of 1,3,4-oxadiazole derivatives for development as broad-spectrum antibiotics

Article abstract of DOI:10.1016/j.bmc.2019.115097

The reality and intensity of antibiotic resistance in pathogenic bacteria calls for the rapid development of new antimicrobial drugs. In bacteria, trans-translation is the primary quality control mechanism for rescuing ribosomes arrested during translation. Because trans-translation is absent in eukaryotes but necessary to avoid ribosomal stalling and therefore essential for bacterial survival, it is a promising target either for novel antibiotics or for improving the activities of the protein synthesis inhibitors already in use. Oxadiazole derivatives display strong bactericidal activity against a large number of bacteria, but their effects on trans-translation were recently questioned. In this work, a series of new 1,3,4-oxadiazole derivatives and analogs were synthesized and assessed for their efficiency as antimicrobial agents against a wide range of gram-positive and gram-negative pathogenic strains. Despite the strong antimicrobial activity observed in these molecules, it turns out that they do not target trans-translation in vivo, but they definitely act on other cellular pathways.

Full text of DOI:10.1016/j.bmc.2019.115097

Bioorganic&MedicinalChemistryxxx(xxxx)xxxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and evaluation of 1,3,4-oxadiazole derivatives for development as
broad-spectrum antibiotics
Cédric Tressea, Richard Radigueb, Rafael Gomes Von Borowskic, Marion Thepautc,
Hong Hanh Lea, Fanny Demayc, Sylvie Georgeaultc, Anne Dhalluinb, Annie Trautwetterc,
⁎
Gwennola Ermelc, Carlos Blancoc, Pierre van de Weghea, Mickaël Jeana, Jean-Christophe Giardb,
,
⁎
Reynald Gilletc,
a Univ. Rennes, INSERM, Chemistry Oncogenesis Stress Signaling (COSS) Group U1242, 35000 Rennes, France
b Université de Caen Normandie, EA4655 U2RM, Antibio-résistance Group, Caen, France
c Univ. Rennes, CNRS, Institut de Génétique et Développement de Rennes (IGDR) UMR6290, 35000 Rennes, France
A R T I C L E I N F O
A B S T R A C T
Keywords:
The reality and intensity of antibiotic resistance in pathogenic bacteria calls for the rapid development of new
antimicrobial drugs. In bacteria, trans-translation is the primary quality control mechanism for rescuing ribo-
somes arrested during translation. Because trans-translation is absent in eukaryotes but necessary to avoid ri-
bosomal stalling and therefore essential for bacterial survival, it is a promising target either for novel antibiotics
or for improving the activities of the protein synthesis inhibitors already in use. Oxadiazole derivatives display
strong bactericidal activity against a large number of bacteria, but their effects on trans-translation were recently
questioned. In this work, a series of new 1,3,4-oxadiazole derivatives and analogs were synthesized and assessed
for their efficiency as antimicrobial agents against a wide range of gram-positive and gram-negative pathogenic
strains. Despite the strong antimicrobial activity observed in these molecules, it turns out that they do not target
trans-translation in vivo, but they definitely act on other cellular pathways.
Antibiotics
Oxadiazoles
Ribosome
tmRNA
Trans-translation
1. Introduction
even when pathogenic bacteria do manage to survive in the absence of
trans-translation, they generally lose their virulence12–15. Therefore,
1,3,4-oxadiazoles are five-membered ring heterocyclic compounds
that have a very wide range of biological activities, making them im-
portant construction motifs for the development of new drugs1. Among
the various pharmacological activities attributed to 1,3,4-oxadiazole
heterocycles, their strong antimicrobial activity is of particular in-
terest2,3,4. Recently, after high-throughput screening, several 1,3,4-ox-
adiazole compounds active against a wide array of bacteria were
identified as inhibitors of trans-translation, the process that delivers
stalled ribosomes during bacterial protein synthesis5,6. In bacteria,
trans-translation is performed by transfer-messenger RNA (tmRNA) and
its partner SmpB, small protein B7. Two alternative ribosome rescue
factors A and B (ArfA and ArfB) can take over when trans-translation is
missing or overwhelmed. Ribosomal stalling is a serious issue for cells
and without tmRNA, bacteria that are devoid of arfA (including Shigella
flexneri, Helicobacter pylori, Francisella tularensis, Mycobacterium tu-
berculosis, and Legionella pneumophila) cannot survive8–11. In addition,
trans-translation inhibitors may act as antibiotics, and have minimal
side effects on ribosomes of eukaryotic hosts16,17. KKL-35 is one of the
most active oxadiazole compounds, showing strong bactericidal activity
against various non-related bacteria, and it is thus a promising molecule
for future development5. However, so far neither its molecular targets
nor its exact mechanism of action have been solved. In fact, we and
others have recently demonstrated that trans-translation is not the only
target of KKL-35 in vivo11,18. In order to improve the potential anti-
trans-translation activity of oxadiazole compounds, we designed and
synthesized a new series of derivatives, and studied their antimicrobial
activities. Despite their strong antimicrobial effects, we were able to
show that 1,3,4-oxadiazoles compounds target other pathways in vivo.
⁎ Corresponding authors at: Unité de Recherche Risques Microbiens (U2RM), Laboratoire de Microbiologie, CHU Côte de Nacre, 14033 Caen Cedex, France (J.-C.
Giard). Univ. Rennes 1, CNRS, Institut de Génétique et Développement de Rennes (IGDR), 35000 Rennes, France (R. Gillet).
E-mail addresses: jean-christophe.giard@unicaen.fr (J.-C. Giard), reynald.gillet@univ-rennes1.fr (R. Gillet).
https://doi.org/10.1016/j.bmc.2019.115097
Received 12 July 2019; Received in revised form 6 September 2019; Accepted 9 September 2019
0968-0896/©2019ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:CédricTresse,etal.,Bioorganic&MedicinalChemistry,https://doi.org/10.1016/j.bmc.2019.115097

C. Tresse, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxxx
these compounds was then evaluated against a panel of 24 bacterial
strains (Table S1). Only CT1-69 acts as a broad-spectrum antibiotic
(Table 1), emphasizing the importance of the compound’s structure.
This is seen for instance with LHH-55, where a simple inversion of the
KKL-35 amide functional group results in an absence of antibiotic ac-
tivity. The results are the same when the heterocycle is changed, such as
with the pyridine substitution in LHH-84, or when it is slightly mod-
ified, as in LHH-19, where the oxygen atom was replaced by sulfur.
At the same time, we replaced the fluorine atom in KKL-35 with a
trifluoromethyl group. Thus, LHH-32 and CT1-98 were prepared
through the synthesis of LHH-23 and CT1-96, respectively (Scheme 5).
Only LHH-32 turned out to have remarkable antibacterial activity
against a small panel of pathogenic bacteria (Table 1). Finally we did
some variations on the eastern part of KKL-35 by modifying the
chlorophenyl eastern part while keeping the trifluoromethyl substituent
on the western benzene ring. The displacement of the chlorine atom
from para to the meta position led to CT1-115 whereas the replacement
of the benzene ring by a pyridine provided CT1-83 (Scheme 6). These 2
new derivatives exhibited interesting broad spectrum antibiotic activ-
ities (Table 1).
Scheme 1. Structural modifications of KKL-35.
2. Results and discussion
2.1. Design of KKL-35 analogs
Since the oxadiazole KKL-35 shows very strong antibacterial ac-
tivity 5, we started with this molecule, designing different derivatives of
its structure by varying either its heterocyclic core or the lateral aro-
matic groups (Scheme 1). First, we focused on the role played by the
amide functional group. For this purpose, we prepared the compound
LHH-55, which corresponds to the KKL-35 retroamide (Scheme 2). The
heterocyclic core was generated via one-pot, two-step synthesis starting
with commercially available 4-fluorobenzyl hydrazine, first adding
ethyl oxalyl chloride, then tosyl chloride. The isolated ethyl ester LHH-
39 was then converted into the target compound by adding 4-chlor-
oaniline with tBuOK in THF. After this, three other compounds were
synthesized by varying just the heterocyclic core. This was done by
replacing the oxygen atom with a sulphur atom (LHH-19), by removing
one of the nitrogen atoms from the heterocyclic ring (CT1-69), or by
substituting the oxadiazole moiety with a pyridine (LHH-84). The
1,3,4-thiadiazole LHH-19 was obtained after three steps starting from
4-fluoro-benzaldehyde, with the key step being the formation of the
heterocycle from the intermediate LHH-12 compound. Several reaction
conditions were tested in order to obtain the compound. The best re-
sults were observed when the 1,3,4-thiadiazol-2-amine derivative LHH-
30 was placed in the presence of acyl chloride, using pyridine as sol-
vent. The low yield was mainly due to the difficulty of its isolation
(Scheme 3). The two other new compounds were obtained in a standard
way. For the first, the formation of 2-oxazolamine CT1-56 produced the
derivative CT1-69 after an amide-coupling reaction with TBTU. For the
2.2. Antimicrobial activity of the synthetic oxadiazole compounds
We tested the antimicrobial activity of the newly synthesized 1,3,4-
oxadiazole compounds against a selection of 24 bacterial species that
included international reference strains and clinical isolates (Table S1).
Antimicrobial susceptibility testing was first performed by agar disk
diffusion, loading the cellulose disks with either 50 µg of a compound or
10 μL DMSO as a control. KKL-35 was used as a reference for com-
parison with the newly synthesized molecules (Table 1). Whatever the
strain, no growth inhibition is observed around the control disks, de-
monstrating that the solvent used to dissolve the compounds does not
affect the analysis (data not shown). Of the 12 molecules tested, 8 of
them (LHH-55, LHH-39, LHH-19, LHH-77, LHH-84, CT1-56, CT1-96,
and CT1-98) did not appear to exhibit any antibacterial activity, as
growth of all strains was observed up to the edges of the loaded disks
(6 mm inhibition diameter, corresponding to the disk diameters, data
not shown). The other four molecules (CT1-69, LHH-32, CT1-83, and
CT1-115) did display antimicrobial activity. Interestingly, three of
them (CT1-69, CT1-83, and CT1-115) exhibited a wide spectrum of
action and strong antimicrobial activity (diameters over 12 mm). In
fact, they seemed to be active against almost all of the gram-positive
strains tested, with the exception of C. perfringens and S. pyogenes.
However, the only gram-negative strain affected was B. fragilis ATCC
25285 (by CT1-83). Interestingly, CT1-69, CT1-83 and CT1-115 also
affected the growth of the mycobacteria M. fortuitum, and CT1-83 had
an effect against M. abscessus (Table 1).
second,
a Suzuki-Miyaura cross-coupling reaction and amidation
starting from 2-amino-6-chloropyridine led to LHH-84. In both cases,
despite various attempts to optimize the reaction conditions, the last
steps produced low yields, again due to the difficulties encountered in
isolating the final products (Scheme 4). The antimicrobial activity of
Scheme 2. Preparation of LHH-55.
2

C. Tresse, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxxx
Scheme 3. Preparation of LHH-19.
3. Effects of the changes to the parent KKL-35 molecule on
antimicrobial activity
3.1. Determination of minimum inhibitory concentrations
We determined the MICs of CT1-69, CT1-83, and CT1-115 mole-
The retroinversion of the KKL-35 amide functional group led to a
total loss of antimicrobial activity in the resulting LHH-55 compound.
The amide functional group is therefore indispensable to the anti-
microbial activity of the oxadiazole derivatives, and it was conserved in
the following modified molecules. Variations to the heterocyclic core
led to a loss of activity when replacing the oxygen atom with a sulfur
atom (LHH-19), or when substituting the oxadiazole moiety by a pyr-
idine (LHH-84). On the other hand, removing one of the nitrogen atoms
from the heterocyclic ring (CT1-69) improved activity against all of the
bacteria already sensitive to KKL-35. Moreover, CT1-69 also exhibited
antimicrobial activity against Enterococci, which are resistant to KKL-
35 (Table 1). The third type of modification done on the aryl group of
the KKL-35 parent was certainly the most interesting one. Indeed, an-
tibacterial activities greatly improve when the fluorine atom is replaced
by a trifluoromethyl group. Finally, we did further biological studies on
three of the most promising compounds (CT1-69, CT1-83, and CT1-
115), chosen because their MIC and IC50 values are significantly lower
than those of KKL-35.
cules for 10 strains, including 1 g-negative, 7 g-positive, and 2
Mycobacterium strains. Table 2 shows the MIC values (or ranges of va-
lues) obtained after three independent experiments. When the bacterial
growth inhibition diameters obtained by the agar diffusion method
were less than 9 mm, we did not determine the corresponding MICs.
KKL-35 was again used as control, since the main goal was to identify
molecule(s) having a MIC at least twice as low as its, a sign of better
antimicrobial activity. In this context, CT1-69 did not seem more effi-
cient than the reference molecules. On the other hand, CT1-83 dis-
played lower MICs for both Enterococci and M. fortuitum. In addition,
CT1-115 seemed equal to KKL-35 against B. fragilis ATCC 25285, C.
perfringens, L. monocytogenes, M. abscessus, M. fortuitum, S. aureus ATCC
29213, and S. pyogenes, and is much stronger against E. faecalis ATCC
29212, E. faecium AUS0004, and S. epidermidis, while having very low
(0.0625–1 mg/L) MICs (Table 2).
Scheme 4. Synthesis of CT1-69 and LHH-84.
3

C. Tresse, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxxx
Table 1
the growth inhibition of L. monocytogenes (Table 3). Note that for this
last strain, the diameter of inhibition (15 mm) observed around the disk
loaded with ofloxacin and CT1-69 was lower than that measured with
just ofloxacin (21 mm), suggesting that CT1-69 may inhibit ofloxacin in
L. monocytogenes.
Measurement of the antimicrobial activity of various synthetic oxadiazole
molecules against a panel of 24 bacterial strains (Diameter of zone of inhibition
in mm).
Strains
KKL-35 CT1-69 CT1-83 CT1-115 LHH-32
3.3. Cytotoxicity
Staphylococcus aureus
Staphylococcus aureus ATCC
29213
17
12
20
16
30
22
28
24
ND
9
The cytotoxicity of the lead compounds was assessed using eight
mammalian cell lines: Huh7, Caco-2, MDA-MB-231, HCT116, PC-3,
MCF-7, NCI-H727, and a fibroblast cell line (Table 4). Cell viability was
determined by MTT assay. Overall, the IC50 values are in the same
range (14–25 µM) as for KKL-35, except in the case of CT1-83, which is
not cytotoxic below 25 µM.
Staphylococcus epidermidis
Streptococcus pyogenes
Bacillus subtilis
15
6
18
6
28
10
24
13
23
27
9
ND
6
12
6
15
6
24
11
20
13
6
Clostridium perfringens
Enterococcus faecalis ATCC
29212
6
11
10
Eentrococcus faecium AUS0004
E. faecium HM1070
Kocuria rhizophila
6
12
13
18
17
6
23
24
28
31
6
21
23
30
28
6
9
8
11
14
16
6
3.4. E. coli resistance to 1,3,4-oxadiazole compounds is linked to efflux
11
13
6
Listeria monocytogenes
Acinetobacter baumannii
Bacteroides fragilis ATCC 25285
Enterobacter cloacae
Escherichia coli
Considering the lack of antimicrobial effects on Enterobacteriaceae,
our intention was to study how E. coli resists 1,3,4-oxadiazole deriva-
tives. We began by doing disk diffusion assays, and by calculating the
MICs in E. coli BW25113 for the three compounds previously identified
as being the most promising, CT1-69, CT1-83, and CT1-115. However,
no inhibition zone was seen in the wild-type strain, as already noticed
with KKL-355,18, and the MIC was higher than 128 µM for all of the
tested compounds (Table 5). Experiments were then conducted using a
strain deleted for TolC. Indeed, in E. coli the antimicrobial drug efflux
pathway is composed of the outer membrane porin protein TolC com-
bined with cytoplasmic membrane pumps belonging to five major su-
perfamilies. We and others have in fact documented growth inhibition
in the presence of KKL-35 when the AcrAB pump was inactivated, in-
dicating that KKL-35 is ejected by the AcrAB-TolC multidrug efflux
pump in E. coli. TolC inactivation in the presence of 1,3,4-oxadiazoles
results in restricted growth, showing an inhibition halo of 19 mm in
disk inhibition assays and a MIC of 0.25 µM for CT1-115, the most
potent derivative, as compared to KKL-35′s inhibition halo of 22 mm
and a MIC of 2 µM. These data confirm that the 1,3,4-oxadiazole
compounds described here are ejected by the same pathway as KKL-35.
Since the TolC system exists only in gram-negative bacteria, these re-
sults also explain why the compounds are mainly active in gram-posi-
tive bacterial strains or in strains lacking efficient efflux (Table 1).
6
8
16
6
11
6
9
6
6
6
6
6
6
6
6
Escherichia coli ATCC 25922
Klebsiella pneumoniae
Morganella morganii EB6
Morganella morganii EB8
Proteus mirabilis
6
8
6
6
6
6
6
6
6
6
6
6
6
6
ND
ND
ND
ND
6
6
6
6
6
6
6
6
6
Proteus vulgaris
6
10
6
8
6
Pseudomonas aeruginosa
Mycobacterium abscessus
Mycobacterium fortuitum
6
6
6
6
7
14
19
9
ND
ND
9
12
23
ND, Not Determined.
3.2. Synergistic effects with other antibiotics
We observed a synergistic effect between an antibiotic and a test
compound as soon as the diameter of growth inhibition measured for
the association increased by at least 5 mm over the highest value ob-
tained in individual tests of either of the two molecules. The results are
presented in Table 3 for the 6 strains E. faecalis ATCC 29212, E. faecium
AUS0004, B. fragilis ATCC 25285, C. perfringens, L. monocytogenes and S.
pyogenes. Depending on the strain, we combined the CT1-69, CT1-83,
and CT1-115 molecules with the antibiotics to which the strain had
previously been shown to be either “resistant” or “intermediate” ac-
cording to EUCAST guidelines. No effects were detected using the
strains B. fragilis ATCC 25285, C. perfringens, or S. pyogenes (data not
shown). On the other hand, for E. faecalis ATCC 29212, significant sy-
nergistic effects were observed with the associations streptomycin/CT1-
115, and gentamicin/CT1-115 (Table 3). Moreover, the combination of
gentamicin/CT1-83 showed greater growth inhibitions of E. faecium
AUS0004 than the compound alone (Table 3). Lastly, CT1-115 asso-
ciated with moxalactam also displayed a significant synergistic effect on
3.5. Correlation between the antimicrobial activity of oxadiazole
compounds and the in vivo inhibition of trans-translation
Since the links between the antimicrobial activity of KKL-35 and its
in vivo activity on trans-translation were recently called into ques-
tion11,18, we performed the same inhibitory assays on various strains
mutated for rescuing systems which are involved in protein synthesis
quality control. Indeed, since ribosome rescue is essential, the double
deletion of ssrA and arfA is lethal in E. coli8. Assuming that a compound
Scheme 5. Synthesis of LHH-32 and CT1-98.
4

C. Tresse, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxxx
Scheme 6. Synthesis of CT1-83 and CT1-115.
1,3,4-oxadiazole compound activity in cells. Towards this aim, we re-
cently developed a reliable in vivo double-fluorescence reporter system
for the simultaneous quantification of both trans-translation and the
associated proteolysis activities in bacteria. In this system, two different
fluorescent proteins are simultaneously used in one strain: red mCherry
and green fluorescent protein (GFP). If trans-translation is inhibited,
bacteria will glow green, and if inhibitors target proteases, the bacteria
glow both green and red 18. We chose the ΔtolC strain for the experi-
ments, as this has a higher sensitivity to 1,3,4-oxadiazoles than the
wild-type strain. The concentrations used were less than or equal to the
minimum inhibitor concentrations calculated for each compound in E.
coli. However, as previously shown for KKL-35 18, the fluorescence
levels did not change for these molecules as compared to the positive
controls (data not shown). All together, these results suggested that the
antibacterial effects of oxadiazoles in E. coli might not be due to the
inhibition of tmRNA or ClpP, challenging the theory that trans-trans-
lation is the main target for oxadiazole derivatives in vivo.
Table 2
Minimum inhibitory concentrations (mg/L) of various oxadiazole molecules
against a group of bacterial strains.
Strains
Molecules
KKL-35
CT1-69
[0.5–4]
0.25
CT1-83
[2–4]
CT1-115
2
Staphylococcus aureus ATCC [0.5–4]
29213
Staphylococcus epidermidis
Streptococcus pyogenes
Clostridium perfringens
Enterococcus faecalis ATCC
29212
[0.5–4]
[1–2]
[0.0625–0.25]a
[4–32]
[32–64] [32– > 64] 32
4
NDb
4
[2–4]
16
[2–16]
[2–4]
[0.125–1]
Enterococcus faecium
AUS0004
16
[1–4]
2
[0.0625–0.25]
Listeria monocytogenes
Bacteroides fragilis ATCC
25285
4
[32–64]
ND
8
2
8
[2–4]
[1–2]
Mycobacterium abscessus
Mycobacterium fortuitum
64
32
ND
[16–32] [32–64]
[16–32]
[32– > 64]
8
a
4. Conclusion
Bold values are significantly different than the MICs observed with the KKl-
35 reference molecule.
b
ND, no detectable antimicrobial activity (see Table 1).
The three new oxadiazole compounds CT1-69, CT1-83, and CT1-
115 have antimicrobial activities that are higher than the KKL-35 re-
ference molecule against a large panel of gram-positive pathogenic
strains. Furthermore, these new compounds show synergistic activity
with conventional antibiotics and have low toxicities. Taking into ac-
count their negative trans-translation in vivo assay results in E. coli, the
overall global activity of the three compounds on a large panel of pa-
thogenic bacteria suggests that oxadiazole derivatives may have one or
several other major cellular targets in addition to trans-translation. So
far, we have failed to obtain spontaneous, UV, or nitrosoguanidine-in-
duced resistant mutants with wild-type as well as delta tolC strains,
which suggests that the 1,3,4-oxadiazole target is either important for
survival, or there are multiple targets. However, our cellular assays
were essentially performed using E. coli cells, and we cannot rule out
the idea that the anti-trans-translational effects are improved in vitro or
in certain bacteria, as recently suggested for Francisella tularensis or
Mycobacterium tuberculosis2,6. Our goal now is to further improve the
inhibited trans-translation, we expected that it would display a stronger
inhibition of bacterial growth in the absence of ArfA. Since ArfB plays
also a partial role in cell viability in the absence of SsrA and ArfA19, we
constructed mutant cells deficient in (arfA, arfB) in a ΔtolC background.
However, we observed similar inhibition diameters around the disks for
strains deficient in tolC or (tolC, arfA, arfB) (Table 6). We thus per-
formed the same experiments in the absence of the putative tmRNA, but
also ClpP targets. Indeed, when Clp is absent or inhibited, trans-trans-
lated proteins will not degrade and the process of trans-translation is
incomplete. If a molecule targets trans-translation, in the absence of
that process, the molecule should no longer be effective. Yet the mo-
lecules studied inhibit growth in the same way as in strains deficient for
(tolC, ssrA) and (tolC, clpP), suggesting that they affect other targets
present in those mutants (Table 6).
The next step was to use a fluorescence reporter system to check
Table 3
Effects of the combination of oxadiazole molecules and antibiotics on the growth inhibition of three gram-positive bacterial strains.
Molecules Bacterial strains
Enterococcus faecalis ATCC 29212
Enterococcus faecium AUS0004
Listeria monocytogenes
Antibiotica
Antibiotic
Antibiotic
No ATB RIF 5 ERY 15 HLS 300 GME 30 No ATB RIF 5 IPM 10 API 2 ERY 15 HLS 300 GME 30 No ATB MOX 30 FOS 200 OFX 5
None
21
20
20
23
23
20
20
20
23
24
16
18
19
22
24
14
13
16
23
23
9
6
6
6
22
22
21
26
26
17
17
18
25
25
15
13
17
30
33
6
21
19
15c
30
29
DMSO
6b
9
6
10
15
22
26
6
6
6
6
6
CT1-69
CT1-83
CT1-115
11
20
22
13
24
24
13
24
24
12
22
22
16
29
28
15
29
30
20
17
a
Antibiotics (ATBs) used: RIF (rifampicin); ERY (erythromycin); HLS (streptomycin); GME (gentamicin); IPM (imipenem); API (ampicillin); MOX (moxalactam);
FOS (fosfomycin); and OFX (ofloxacin). The numbers underneath their abbreviations indicate the disk loads (µg).
b
c
The growth inhibition diameters (mm) observed when the molecule or antibiotic is used alone are italicized.
The growth inhibition diameters (mm) that indicate significant synergistic effects are bold.
5

C. Tresse, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxxx
Table 4
IC50 values (µM) of cytotoxicity tests of oxadiazole derivatives on various mammalian cell lines.
Molecules
Huh7 liver
Caco-2 colon
MDA-MB-231 Breast
HCT116 colon
PC-3 prostate
MCF-7 breast
NCI lung
Fibro skin
KKL-35
CT1-69
CT1-83
CT1-115
20
14
25
17
21
22
18
> 25
> 25
> 25
> 25
3
3
5
2
2
4
> 25
20
> 25
11
> 25
23
> 25
10
> 25
17
> 25
15
> 25
10
5.2. Bacterial strains and growth conditions
Table 5
Minimum inhibitory concentrations of oxadiazole compounds on wild type and
ΔtolC strains of Escherichia coli BW25113.
Mol 10 mM Ø inhibition (mm)
Escherichia coli BW25113 deletion stains and derivatives used in this
study were previously described18. All the other bacterial strains are
listed in Supplementary Table 1. Identification of the different species
MIC BW25113 MIC
MBC
BW25113
ΔtolC
ΔtolC
was carried out from culture of 20
4 h and from a 4 days culture for
KKL-35
CT1-69
CT1-83
CT1-115
6
6
6
6
> 128
> 128
> 128
> 128
2
> 32
ND
Mycobacterium, by MALDI-TOF (Bruker MALDI Biotyper CA System,
Germany). Strains belonging to the Enterobacteriaceae (8 strains), En-
terococcaceae (3 strains), Staphylococcaceae (3 strains) families and
Kocuria rhizophila, Acinetobacter baumannii, Pseudomonas aeruginosa,
ND
1
> 32
16
0.25
MIC. Minimum inhibitory concentration.
MBC. Minimum bactericidal concentration.
ND. Not determined.
Listeria monocytogenes and Bacillus subtilis species were grown 20
4 h
2 °C. Strepto-
4 h in an aerobic atmosphere + 5%
CO2 on trypticase soy agar + 5% horse blood (Biomerieux, France), at
35 2 °C. The species Clostridium perfringens and Bacteroides fragilis
were cultured 20 4 h anaerobically (AnoxomatTM, Mart® Micro-
biology B.V., Netherlands), on TSH agar at 35 2 °C. The two Myco-
in aerobic conditions, on Trypticase-Soja agar, at 35
coccus pyogenes was grown 20
Table 6
Antimicrobial activity of the lead compounds against E. coli mutants.
Bacterial strain
KKL-35
CT1 69
CT1 83
CT1 115
bacterium strains were cultured for 4 days in a humid aerobic atmo-
BW25113 ΔtolC
BW25113 ΔarfA arfB
tolC
13
1
11.33
0.6 23.6
0.6
21.3
3.2
3.2
sphere TSH on agar at 35
stored at −80 °C.
2 °C. After identification, the strains were
12.3
3.5 14
1.8
1.1
22.0.5
0.7 22.6
BW25113 ΔclpP tolC
BW25113 ΔssrA tolC
14.3
13
2.1 13.3
1.4 13
27
27
2.6
2.6
22.3
22.3
2.5
2.3
2
5.3. Screening of the antimicrobial activity
Disc diffusion assay results are in mm. Results are the mean of three in-
dependent experiments and the standard deviations are reported.
The antimicrobial activity of the compounds was measured ac-
cording to the diffusion method in agar medium. Mueller-Hinton (MH)
medium was used for bacteria other than slow-growing bacteria and
MHF (5% horse blood added to MH) medium was used for strains of S.
pyogenes, L. monocytogenes, C. perfringens, B. fragilis ATCC 25285 as well
as for Mycobacterium strains (M. abscessus and M. fortuitum). The surface
of the agar was seeded from a swab of a standardized 0.5 McFarland
inoculum (around 108 Colony Forming Units (CFU)/ml). Sterile cellu-
lose discs were disposed on its surface and loaded with 50 μg of each
compound (10 μL of a 5 g/L solution into DMSO). For each strain, a
control was carried out by depositing 10 μL of DMSO on a cellulose disc.
activity, selectivity, and cellular behavior of 1,3,4-oxadiazoles com-
pounds and find their other cellular targets.
5. Experimental procedures
5.1. General information
All reagents and solvents were purchased from commercial sup-
pliers or when necessary were purified/dried according to W. L. F.
The media were incubated at 35
2 °C aerobically, except for S.
Armarego and C. L. L. Chai20
.
1H and 13C NMR spectra were recorded
pyogenes (incubated aerobically +5% CO2) and C. perfringens and B.
fragilis ATCC 25285 (incubated anaerobically). The two Mycobacterium
strains were incubated in humid aerobic conditions for 4 days at
on a Bruker DMX 500 or Avance 300 instrument using TMS and CDCl3
respectively as the internal standard. δ values are given in parts per
million (ppm) and coupling constants (J values) are given in Hertz (Hz).
Signal multiplicity is reported as follows: s, singlet; bs, broad singlet; d,
doublet; t, triplet; q, quadruplet; qt, quintet; sext, sextuplet; hept,
heptuplet; dd, doublet of doublets; dq, doublet of quadruplets; td, tri-
plet of doublets; tq, triplet of quadruplets; dsept, doublet of septuplets;
ddd, doublet of doublets of doublets; and m, multiplet. High-resolution
mass spectrometry (HRMS) analysis was done using a Waters Q-Tof II, a
Micromass ZabSpec TOF, a Bruker micrO-TOF Q II, or an LTQ Orbitrap
XL instrument for ESI. X-ray crystallographic data were collected on an
APEXII crystal diffractometer. Optical rotations were recorded on a
PerkinElmer 341 polarimeter. Melting points were obtained on a hot
bench. TLC analysis was performed using precoated Merck TLC silica
gel 60 F254 plates. Column chromatography purifications on silica gel
were done using Merck silica Gel 60 (70–230 mesh), those on neutral
alumina used Merck aluminium oxide 90 active, and the preparative
thin-layer chromatography purifications were performed using Merck
Silica Gel 60 PF254. The petroleum ether (PE) used for purifications was
at a low boiling point (40–60 °C).
35
2 °C. All other strains are incubated for 20
4 h. Lastly, the
diameters of the growth inhibitions around the discs were measured.
5.4. Antimicrobial susceptibility testing
MICs of tested molecules as well as antibiotics were determined
using the broth microdilution method according to EUCAST (http://
www.eucast.org/) guidelines. The strains used for these determinations
were C. perfringens, E. faecalis ATCC 29212, E. faecium AUS0004, L.
monocytogenes, S. aureus ATCC 29213, S. epidermidis, S. pyogenes, B.
fragilis ATCC 25285, M. abscessus and M. fortuitum and the growth
conditions were those as described above.
5.5. Effects of the combination molecule-antibiotic
In order to determine whether the activity of antibiotics combined
with our molecules was able to be modified, synergistic tests were
carried out. To do this, the selected antibiotics were those for which the
6

C. Tresse, et al.
Bioorganic&MedicinalChemistryxxx(xxxx)xxxx
categorization as “intermediate” or “resistant” were obtained by stan-
dard antibiogram (according to the EUCAST guidelines). For this pur-
pose, the agar diffusion technique was used. A swab of bacterial stan-
dard inoculum (0.5 MacFarland suspension), was used to seed the
surface of medium. MH medium was used for E. faecalis ATCC 29212
and E. faecium AUS0004 and MHF for B. fragilis ATCC 25285, C. per-
fringens, L. monocytogenes and S. pyogenes. Standard discs containing the
selected antibiotics were put on the surface and then loaded with 50 μg
of each anti-microbial compound. The control tests corresponded to the
antibiotic discs not supplemented with compounds and to discs only
loaded with 50 μg of molecules to be tested. The medium were in-
cubated as mentioned above. A synergistic effect was observed when
the diameter of the growth inhibition was larger (at least 5 mm) than
the aureoles obtained with the antibiotic or the molecule alone.
RGVB, FD performed in vivo trans-translation experiments; RR, SG, AD,
AT, GE, CB and JCG performed microbiology experiments; CB, JCG,
PVW, and RG wrote the paper. RG supervised the project.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmc.2019.115097.
References
1. de Oliveira CS, Lira BF, Barbosa-Filho JM, Lorenzo JG, de Athayde-Filho PF.
Synthetic approaches and pharmacological activity of 1,3,4-oxadiazoles: a review of
the literature from 2000–2012. Molecules. 2012;17:10192–101231.
2. Goralski TD, Dewan KK, Alumasa JN, et al. Inhibitors of ribosome rescue arrest
growth of francisella tularensis at all stages of intracellular replication. Antimicrob
Agents Chemother. 2016;60:3276–3282.
5.6. Toxicity test
3. Chandrakantha B, Shetty P, Nambiyar V, Isloor N, Isloor AM. Synthesis, character-
ization and biological activity of some new 1,3,4-oxadiazole bearing 2-flouro-4-
methoxy phenyl moiety. Eur J Med Chem. 2010;45:1206–1210.
4. Patel NB, Patel JC. Synthesis and antimicrobial activity of 3-(1,3,4-oxadiazol-2-il)
quinazolin-4(3H)-ones. Sci Pharm. 2010;78:171–193.
Cytotoxicity tests were performed on the ImPACcell robotic plat-
form (BIOSIT, Université de Rennes 1). This featured high-throughput
multiparameter image analysis, with both high-content screening and
high-content analysis. The platform is equipped with an Olympus mi-
croscope and Compix SimplePCI software; a Zeiss Axio Imager M1
microscope with a Zeiss camera and AxioVision software; and imaging
systems including an ArrayScan VTI Cellomics reader (Thermo Fisher),
Hamilton STARlet and NIMBUS workstations, and a Scienion spotter.
For the tests, seven different cell lines were used: human hepatocellular
carcinoma (HuH7); colorectal adenocarcinoma (Caco-2); breast ade-
nocarcinoma (MDA-MB231); colorectal carcinoma (HCT116); prostatic
adenocarcinoma (PC3); lung carcinoma (NCL-H727); and breast cancer
(MCF7). The residual cell percentages reported correspond to viable
cells compared to the average viable cells in the DMSO control.
Viability of 100% represents no cytotoxicity or inhibition of cell
growth, while under 25–30% is considered cytotoxic and 0% represents
acute cytotoxicity.
5. Ramadoss NS, Alumasa JN, Cheng L, et al. Small molecule inhibitors of trans-trans-
lation have broad-spectrum antibiotic activity. Proc Natl Acad Sci USA.
2013;110:10282–10287.
6. Alumasa JN, Keiler KC. Clicking on trans-translation drug targets. Front Microbiol.
2015;6:498.
7. Giudice E, Macé K, Gillet R. Trans-translation exposed: understanding the structures
and functions of tmRNA-SmpB. Front Microbiol. 2014;5:113.
8. Chadani Y, Ono K, Ozawa S, et al. Ribosome rescue by Escherichia coli ArfA (YhdL) in
the absence of trans-translation system. Mol Microbiol. 2010;78:796–808.
9. Starosta AL, Lassak J, Jung K, Wilson DN. The bacterial translation stress response.
FEMS Microbiol Rev. 2014;38:1172–1201.
10. Keiler KC, Feaga HA. Resolving nonstop translation complexes is a matter of life or
death. J Bacteriol. 2014;196:2123–2130.
11. Brunel R, Charpentier X. Trans-translation is essential in the human pathogen
Legionella pneumophila. Sci Rep. 2016;6:37935.
12. Julio SM, Heithoff DM, Mahan MJ. ssrA (tmRNA) plays a role in Salmonella enterica
serovar Typhimurium pathogenesis. J Bacteriol. 2000;182:1558–1563.
13. Huang C, Wolfgang MC, Withey J, Koomey M, Friedman DI. Charged tmRNA but not
tmRNA-mediated proteolysis is essential for Neisseria gonorrhoeae viability. EMBO J.
2000;19:1098–1107.
14. Okan NA, Mena P, Benach JL, Bliska JB, Karzai AW. The smpB-ssrA mutant of
Yersinia pestis functions as a live attenuated vaccine to protect mice against pul-
monary plague infection. Infect Immun. 2010;78:1284–1293.
Declaration of Competing Interest
15. Svetlanov A, Puri N, Mena P, Koller A, Karzai AW. Francisella tularensis tmRNA
system mutants are vulnerable to stress, avirulent in mice, and provide effective
immune protection. Mol Microbiol. 2012;85:122–141.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper
16. Vioque A, de la Cruz J. Trans-translation and protein synthesis inhibitors. FEMS
Microbiol Lett. 2003;218:9–14.
17. Li J, Ji L, Shi W, Xie J, Zhang Y. Trans-translation mediates tolerance to multiple
antibiotics and stresses in Escherichia coli. J Antimicrob Chemother.
2013;68:2477–2481.
Acknowledgments
This work was supported by the Direction Générale de l’Armement
(#ANR-14-ASTR-0001, France) and the Agence Nationale pour la
Recherche under the frame of the Joint JPI-EC-AMR Project named
“Ribotarget - Development of novel ribosome-targeting antibiotics”
(SNF No. 40AR40_185777, Europe).
18. Macé K, Demay F, Guyomar C, et al. A genetic tool to quantify trans-translation
activity in vivo. J Mol Biol. 2017;429:3617–3625.
19. Chadani Y, Ono K, Kutsukake K, Abo T. Escherichia coli YaeJ protein mediates a
novel ribosome-rescue pathway distinct from SsrA- and ArfA-mediated pathways.
Mol Microbiol. 2011;80:772–785.
20. Armarego WLF, Chai CLL. Purification of laboratory chemicals. Amsterdam:
Butterworth-Heinemann; 2003.
Author contributions
CT, HHL, PVW and MJ performed chemical experiments; MT,
7