Anti-mycobacterial activity of 1,3-diaryltriazenes

Article abstract of DOI:10.1016/j.ejmech.2014.02.065

The rapid generation and spread of the drug resistant tuberculosis has led to an ongoing demand for novel compounds for therapeutic use. Identification and study of compounds with the ability to inhibit Mycobacterium tuberculosis is of paramount importance. For this reason, a library of substituted 1,3-diaryltriazenes based on the acting component of the anti-trypanosomal drug, diminazene aceturate was created and evaluated for its potential as anti-tubercular agent. Several compounds were identified with sub-micro molar inhibitory concentrations against M. tuberculosis and other clinically relevant mycobacterial species such as Mycobacterium bovis, Mycobacterium avium and Mycobacterium ulcerans. Although the library of the compounds showed a considerable acute cytotoxicity, a genotoxicity could not be observed. Finally, the triazene 14 was selected with the best biological properties (IC ₅₀ = 3.26 μM, NI₅₀ = 24.22 μM, SI = 7.44). The compound 14 showed the ability to inhibit the growth of intracellular replicating and multi-drug resistant M. tuberculosis. The results suggest the molecule to be an interesting scaffold for further study and optimization.

Full text of DOI:10.1016/j.ejmech.2014.02.065

European Journal of Medicinal Chemistry 77 (2014) 193e203
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Anti-mycobacterial activity of 1,3-diaryltriazenes
Davie Cappoen a, Jure Vajs b, Cynthia Uythethofken a, Andrej Virag b, Vanessa Mathys c,
b
d,
Marijan Kocevar , Luc Verschaeve e, Martin Gazvoda b, Slovenko Polanc b,
ꢀ
Kris Huygen a , Janez Kosmrlj
a Scientific Service Immunology, O.D. Communicable & Infectious Diseases, Scientific Institute of Public Health (Site Ukkel), Engelandstraat 642,
,
b,
*
*
ꢀ
B-1180 Ukkel, Belgium
b
ꢀ
ꢀ
Faculty of Chemistry and Chemical Technology, University of Ljubljana, Askerceva 5, SI-1000 Ljubljana, Slovenia
c Program Tuberculosis and Mycobacteria, O.D. Communicable & Infectious Diseases, Scientific Institute of Public Health (Site Ukkel), Engelandstraat 642,
B-1180 Ukkel, Belgium
d Program Toxicology, O.D. Public Health and Surveillance, Scientific Institute of Public Health (Site Elsene), J. Wytsmanstraat 14, B-1050 Brussels, Belgium
e Department of Biomedical Sciences, University of Antwerp, Universiteitsplein 1, B-2610 Wilrijk, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
The rapid generation and spread of the drug resistant tuberculosis has led to an ongoing demand for
novel compounds for therapeutic use. Identification and study of compounds with the ability to inhibit
Mycobacterium tuberculosis is of paramount importance. For this reason, a library of substituted 1,3-
diaryltriazenes based on the acting component of the anti-trypanosomal drug, diminazene aceturate
was created and evaluated for its potential as anti-tubercular agent. Several compounds were identified
with sub-micro molar inhibitory concentrations against M. tuberculosis and other clinically relevant
mycobacterial species such as Mycobacterium bovis, Mycobacterium avium and Mycobacterium ulcerans.
Although the library of the compounds showed a considerable acute cytotoxicity, a genotoxicity could
Received 9 December 2013
Received in revised form
19 February 2014
Accepted 28 February 2014
Available online 1 March 2014
Keywords:
1,3-Diaryltriazene
Tuberculosis
Antibiotic lead
Acute toxicity
SAR
not be observed. Finally, the triazene 14 was selected with the best biological properties (IC50 ¼ 3.26
mM,
NI50 ¼ 24.22 M, SI ¼ 7.44). The compound 14 showed the ability to inhibit the growth of intracellular
m
replicating and multi-drug resistant M. tuberculosis. The results suggest the molecule to be an interesting
scaffold for further study and optimization.
Genotoxicity
Ó 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
carboxamide) (Chart 1, structure a), a synthetic analogue of the
naturally occurring purine precursor 5-amino-1H-imidazole-4-
Identification and development of the compounds to target
Mycobacterium tuberculosis, the etiological agent of tuberculosis, is
still of utmost importance. According to the figures presented by
the World Health Organization (WHO) in 2013 this report states
that about 8.6 million people fell ill with TB in 2012 from which
450,000 with multidrug-resistant pulmonary TB. Although more
rapid diagnosis and treatment have decreased the mortality by 45%
since 1990, 1.3 million patients died, including 320,000 people that
were TB/HIV confected. TB is still one of the greatest killers
worldwide due to a single infectious agent, second only to HIV/AIDS
[1]. Apart from the reports on antibacterial and antifungal activity
[2], triazenes are best represented in cancer chemotherapy [3].
carboxamide, and temozolomide (4-methyl-5-oxo-2,3,4,6,8-
pentazabicyclo-[4.3.0]nona-2,7,9-triene-9-carboxamide) (Chart 1,
structure b) are both antineoplastic compounds [4e6]. As alkylat-
ing agents, both compounds share a similar mechanism of action by
methylating DNA at the N-7 or O-6 positions of guanine residues.
This methylation causes somatic point mutations represented by
C:G / T:A transitions in the DNA helix and finally triggers the
death of tumour cells [6,7]. The library of 1,3-diaryltriazenes pre-
sented in this study are based on the structure of diminazene (4-[2-
(4-carbamimidoylphenyl)iminohydrazinyl]-benzenecarbox-
imidamide; Chart 1, structure c), the acting component of the anti-
trypanosomal drug diminazene aceturate. The activity of dimina-
zene against protozoa is based on its capacity of minor groove
binding to RNA and DNA duplexes [8]. It has been shown that
diminazene can act as an antiviral compound, but shows weak
potential as antitumour agents [9]. Potential DNA binding proper-
ties rise caution in TB chemotherapy, in which patients are treated
Dacarbazine
(5-(3,3-dimethyl-1-triazenyl)imidazole-4-
* Corresponding authors.
E-mail addresses: kris.huygen@wiv-isp.be (K. Huygen), janez.kosmrlj@fkkt.uni-
ꢀ
lj.si (J. Kosmrlj).
http://dx.doi.org/10.1016/j.ejmech.2014.02.065
0223-5234/Ó 2014 Elsevier Masson SAS. All rights reserved.

194
D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
Chart 1. The structures of: a) dacarbazine, b) temozolomide, c) diminazene and d) 3-acetyl-1,3-bis(2-chloro-4-nitrophenyl)triazene [10].
ꢀ
for several months. However, observations by Cimbora-Zovko et al.
report on a halogen-substituted analogue which did not bind into
the minor grove of DNA but instead induces reactive oxidative
species [10] (Chart 1, structure d). In this work, we have evaluated a
library of twenty-six 1,3-diaryltriazenes, substituted with different
electron-withdrawing groups (bromo, chloro, fluoro, nitro, cyano
and trifluoromethyl) at the benzene ring.
genes to express the ATP independent bacterial luciferase of Vibrio
harveyi driven by a constitutive hsp60 promoter. The addition of the
luciferase substrate n-decanal, elicits a luminescent signal of green-
blue light (490 nm) which can be measured by luminometry. The
influence of the compounds or other xenobiotics on the bacillary
growth can be measured as a reduction in the intensity of the
luminescent signal. Serial dilutions of the investigated compounds
were used to calculate the concentrations at which 50% (IC50), 90%
(IC90) and 99% (IC99) of the M. tuberculosis growth was inhibited
after 6 days. The IC50 of the compounds against M. tuberculosis was
used to derive a structure activity relation as shown in Table 1. It
was observed that the position and nature of the electron-
withdrawing substitution group at the 1,3-diaryltriazene largely
influenced the activity. Comparing the active compounds 2, 10 and
2. Results and discussion
2.1. Chemistry
The structures of 1,3-diaryltriazenes and their N-acyl
substituted derivatives tested in this study are shown in Fig. 1.
Several triazenes were synthesized as described before (2 [10],
10 [10], 11 [10], 14 [11], 15 [12], 16 [13], 17 [14], 18 [11], 19 [10], 20
[15] and 24 [10]). Most of them were recently examined for their
antitumour activity [10]. New 1,3-diaryltriazenes were prepared
from the corresponding substituted anilines on treatment with
either sodium nitrite in hydrochloric acid or with isopentyl nitrite
in dichloromethane (Scheme 1). An introduction of the acyl func-
tionality to the triazene nitrogen of the 1,3-diaryltriazene was
achieved using the appropriate acid chloride in the presence of
triethylamine (Scheme 1).
An attempt to obtain 1,3-diaryltriazene 9a from 4-fluoro-3-
nitroaniline in dichloromethane as a solvent resulted in a com-
plex mixture of products, which was not analysed further. When 4-
fluoro-3-nitroaniline reacted with isopentyl nitrite in a mixture of
toluene/petroleum ether (1:1), the triazene 9 was formed (Scheme
2). It is the product of SNAr displacements of one of the fluorine
atoms at the initially formed triazene 9a with the starting aniline.
An alternative pathway that first involves the substitution of the
fluorine atom at the starting aniline with another molecule of an-
iline and subsequent reaction with diazonium salt is ruled out.
Namely, 4-fluoro-3-nitroaniline remains unchanged in toluene/
petroleum ether solution (1:1) at rt for 2 days.
13 (IC50 ¼ 1.58, 0.03 and 0.57
compounds 19 and 20 (IC50 ¼ 2.82 and 0.06
diaryltriazenes and the compounds 25 and 26 (IC50 ¼ 1.69 and
1.45 M) of the 3-[(2E)-3-arylprop-2-enoyl]-1,3-diaryltriazenes we
mM) of the 1,3-diaryltriazenes with the
m
M) of the 3-acetyl-1,3-
m
could observe that all compounds with a nitro substitution at the
para position R4 showed an increased potency to reduce the in vitro
M. tuberculosis growth at sub-micro molar concentrations. Like-
wise, comparing the compound 10 (IC50 ¼ 0.03
mM) with the an-
M) in which the NO2
alogues 5 and 11 (IC50 ¼ 85.51 and 92.96
m
groups were replaced with a halogen F and Cl respectively, the
potency was reduced for the latter two molecules.
The compounds 1, 8, 12, 13 and 14 (IC50 ¼ 6.88, 6.26, 0.77, 1.97
and 3.26 m
M) with a CF3 group at the meta position R3 also showed
increased activity over their analogues. When the CF3 group was
moved from the meta R3 position to the ortho position R2 (com-
pounds 3 and 5 (IC50 ¼ 58.94 and 85.51
mM)) the activity against
M. tuberculosis was abrogated. An exception to this rule was the
compound 10 (IC50 ¼ 0.03
mM), which was substituted with a CF3
group at the ortho R2 position but as mentioned, the compound 10
carried an electron-withdrawing group NO2 at the para R4 position.
Analogues synthesized with both NO2 and CF3 substitutions, the
compounds 10 and 13, showed IC50 values of respectively 0.03 and
0.57
M. tuberculosis in this experiment than the first line M. tuberculosis
drug INH (IC50 ¼ 0.04 M), used as a positive control in this assay.
mM. The compound 10 was even more potent in vitro against
2.2. Biology
m
In previous publications, we and others have shown that
luminometry is a rapid and reproducible tool to test the anti-
mycobacterial activity of novel anti-tuberculous compounds [16e
19]. Screening the library of compounds for in vitro activity
against M. tuberculosis was done with the luminescent
M. tuberculosis H37Rv laboratory strain (H37Rvlux). The H37Rvlux
strain has multiple copies of pSMT1 plasmid carrying the necessary
An acute toxicity of the compounds was examined in a Neutral
Red dye Uptake assay (NRU), shown in Table 1 [20]. This test
method makes use of the ability of living cells to bind and incor-
porate neutral red in their lysosomal structures. The J774 A.1 mu-
rine macrophage cell line was selected as a cellular model,
macrophages being the primary host cell for M. tuberculosis. A
toxicity was measured as reduced uptake of the neutral red dye by

D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
195
Fig. 1. Structures of 1,3-diaryltriazenes and their N-acyl derivatives.
Scheme 1. The synthesis of 1,3-diaryltriaezenes and their N-acyl derivatives. Reagents and conditions: (a) ArNH2 (1 mmol), NaNO2 (0.5 mmol)/H2O (3 mL), 5% HCl (2.5 mL), 0 ꢀC
then rt for 18e33 h; (b) ArNH2 (1 mmol)/CH2Cl2 (4 mL), i-PeONO (1 mmol)/CH2Cl2 (4 mL), 0 ꢀC then rt for 0.5e22 h; (c) 1,3-diaryltriazene (1 mmol), acid chloride (1.1 mmol), Et3N
(2 mmol) in acetone (5 mL), rt, 5e40 min.
Scheme 2. Formation of the triazene 9.

196
D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
Table 1
toxicity by half. For practical reasons, all compounds with a calcu-
lated selectivity index lower than the arbitrarily chosen value 7
were excluded from further studies.
Growth inhibition of M. tuberculosis by the triazene analogues and acute toxicity
against macrophages.
SIe
d
Cmpd.
ICa50
ICb90
ICc99
M)
23.90
NI50
Next, the susceptibility of other clinically relevant mycobacterial
species was tested as shown in Table 2 for the compounds with an
acceptable selectivity index. Mycobacterium bovis, together with
Mycobacterium africanum and Mycobacterium microti, constitute
the M. tuberculosis complex. Mycobacterium avium subsp. avium is
pathogenic for birds but can cause opportunistic infections in AIDS
patients, whereas M. avium subsp. paratuberculosis is the etiological
agent of Johne’s disease or paratuberculosis (in ruminants) which
may be linked to Crohn’s disease in humans. Mycobacterium
ulcerans causes Buruli ulcer, a necrotizing skin disease affecting
mostly children in certain rural areas of West-Africa, and may lead
to irreversible disabilities when left untreated [21e24]. The study
showed that the other mycobacterial species were also susceptible
to the 1,3-diaryltriazene compounds at concentrations comparable
to that of M. tuberculosis. However, the low IC50 values observed for
M. tuberculosis and (to a slightly lesser extent) for M. bovis for the
(
m
NI50/IC50
1
2
3
4
5
6
7
8
6.88
1.58
13.41
52.68
11.63
N.D.f
N.D.
N.D.
28.18
N.D.
26.03
21.10
0.61
N.D.
2.09
6.87
24.22
9.69
N.D.
7.7
7.5
4.12
>100.00
>100.00
>100.00
20.36
>100.00
13.41
>100.0
0.54
9.53
58.94
62.34
85.51
15.51
>100.00
6.25
95.66
0.03
92.96
1.97
>100.00
>100.00
>100.00
26.02
>100.00
16.77
>100.0
1.54
>100.00
10.02
10.25
7.68
15.49
>100.00
>100.00
72.06
21.96
1.58
N.D.
N.D.
N.D.
1.8
N.D.
4.2
9
0.2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
INHg
26.2
N.D.
6.22
10.72
7.44
2.12
N.D.
N.D.
8.12
4.55
17.47
N.D.
N.D.
N.D.
N.D.
3.04
2.7
>100.00
4.22
0.57
3.26
4.57
6.80
6.47
7.77
97.48
>100.00
6.81
>100.00
>100.00
26.86
N.D.
compounds 10 and 20 (IC50 ¼ 0.03 and 0.06
mM), could not be
55.22
12.90
1.07
N.D.
N.D.
N.D.
observed for the other mycobacterial species.
2.82
0.06
8.27
0.64
Because the DNA intercalating properties of diminazene, the
danger of genotoxicity was addressed. Early signs of genotoxicity
can be observed by studying the activation of DNA repair mecha-
nisms. The regulatory SOS operon is a key to the repair of early
cellular DNA damage and its activation is mediated by a multiple
component system. As a result of DNA interruption, binding of
ssDNA fragments to the recA protein increases the enzymatic af-
finity for the lexA repressor protein. The decrease in lexA frees the
SOS box consensus sequence and thereby facilitates the binding of
the general transcription machinery to the recN promotor sequence
and transcription of the SOS operon. Based on this molecular
mechanism, the VITOTOXÔ model from Gentaur can detect acti-
vation of the SOS operon by an integrated lux operon [25]. The
model employs two recombinant Salmonella typhymurium reporter
strains. The integrated lux operon of the Genox strain T104 (recN2-
4) is expressed under transcriptional control of the recN promoter
and in the Cytox pr1 strain it is driven by the strong constitutive pr1
promotor, the latter strain mainly functioning as an internal con-
trol. Both recombinant reporter strains lack the necessary oxidative
machinery to metabolize xenobiotics. By the addition of the S9 liver
fraction the genotoxicity of the investigated substance can be
distinguished from that of its metabolites. Closely correlated with
the AMES test (91%), with a high specificity of 94%, this model al-
lows an early elimination of novel compound classes due to
genotoxicity.
83.13
52.02
71.28
72.33
1.69
>100.0
80.77
>100.0
>100.0
4.37
>100.0
84.41
>100.0
>100.0
9.21
N.D.
5.12
3.94
N.D.
1.45
0.04
3.83
0.11
8.41
0.13
N.D.
a,b,cThe concentration of the analogues at which respectively a growth inhibition of
50%, 90% and 99% of the M. tuberculosis H37Rv lab strain was observed as compared
to the negative control. Presented results are the mean of triplicate values with a
SD < 10%.
d
The concentration of the analogues at which viability of the macrophages was
reduced by 50%. Presented results are the mean of eight replications with a
SD < 10%.
e
The selectivity index (SI), calculated by dividing the NId50 by the IC5e0
N.D. Not Done.
INH, one of the first line M. tuberculosis antibiotics, isoniazid used as a positive
.
f
g
control.
the macrophage culture and the NI50 was determined as the con-
centration at which this reduction reaches 50%. The selectivity in-
dex (SI) was calculated as a function of the NI50 divided with the
IC50. An acute toxicity against the J774 macrophages caused by the
compounds was studied for compounds with IC90 values below
10
macrophages was reported at concentrations lower than 30
except for the 1,3-diaryltriazenes 1 and 18 (NI50 ¼ 52.68 and
mM. For most compounds a 50% reduction in viability of the
mM
Emission of luminescence by the strains was recorded in real
time after the addition of the compounds every 5 min during a 4 h
period. Compounds were tested at 25.00 mM, 2.50 mM and 0.25 mM
in the presence and absence of the metabolizing S9 fraction. Gen-
otoxicity was deduced from the signal to noise ratio (S/N), being the
55.22
mM), which carry different substitutions. The compounds
with the NO2 group at the R4 position that were observed to have a
strong potency, showed also a high acute toxicity with NI50 values
below 10
m
M. The compound 9 (NI50 ¼ 21.10
mM) showed a reduced
Table 2
The 50% in vitro growth inhibitory concentration of the derivatives against other Mycobacteria.
Compound
M. tuberculosis H37Rvlux
M. bovis AN5lux
M. avium ssp. paratuberculosis ATCC19698lux
M. avium ssp. avium ATCC15769lux
M. ulcerans1615lux
50% Growth inhibition (mM)
1
2
6.88
1.55
0.03
0.57
3.26
6.81
0.06
8.15
1.30
0.28
0.76
5.50
6.24
0.31
7.80
2.44
0.64
0.99
4.13
10.24
0.45
8.75
2.10
0.38
0.88
7.99
11.90
0.37
21.38
3.37
0.59
10
13
14
18
20
1.06
10.74
12.72
1.62
In vitro growth inhibition was measured by luminometry after 6 days exposure to the compounds. Inhibition calculated from triplicates cultures (SD values < 10%).

D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
197
light produced by the bacterial suspension exposed to the com-
detected by the COMET assay for 0.13, 0.25, 1.25, 2.50 and 5.00
of the compound 13, but because of the exceeded NI50 values ob-
tained with the NRU assay, the observations made at 12.50 M were
regarded inconclusive and omitted from Fig. 4. All concentrations
0.13, 0.25, 1.25, 2.50, 5.00, 12.50 and 25.00 M of the compound 14
were studied in the COMET assay as well and no increase of DNA in
the TAIL fraction could be observed. For the reason of genotoxicity
and toxicity, the compounds 10 and 13 were omitted from the
further study.
mM
pounds divided by the light produced by a non-exposed bacterial
suspension. S/N values for the Genox (recN2-4) higher than 1.5
were interpreted as positive and implied genotoxic properties by
the compounds. For the Cytox (pr1) strain, S/N values above 1.5 are
considered as direct influence of the compound on the bacterial
luciferase or the luminescent signal itself and regarded a false
positive. S/N values lower than 0.8 for the Cytox strain are indica-
tive of direct cytotoxicity and exclude false negatives. As shown in
Fig. 2, the compounds 1, 2, 10, 13, 14, 18 and 20 resulted in values of
the Cytox control strain between 0.8 and 1.5 with and without the
S9 liver extract. The luminescence emitted from the Genox strain
did not exceed S/N values significant higher as 1.5 after addition of
the compounds in absence and presence of the S9 extract. However,
when the Genox strain was incubated together with S9 and the
compounds 2, 18 or 20 the luminescent signal increased, exceeding
1.5, indicating activation of the SOS operon as a sign of early gen-
otoxicity caused by the metabolites of the compounds 2, 18 and 20.
No genotoxicity was found for the compounds 1, 10, 13 and 14 both
in the absence and in the presence of the S9 liver fraction.
Regarding the structural differences between the group of 2, 18 and
20 and the group of 1, 10, 13 and 14 it was noticed that all members
of the latter group carried CF3 moiety while none had a CF3 side
chain in the former group.
To investigate the importance of the CF3 side chain for the
protection against degradation, as well as to assess the metabolic
stability of the candidate, the potency of the molecules was eval-
uated when incubated together with S9. The concentration at
which a 90% growth reduction of M. tuberculosis could be observed
by the compounds was measured both in the absence and in the
presence of the S9 fraction as shown in Fig. 3. It was confirmed that
the potency of the compounds 2, 18 and 20, the derivatives without
the CF3 moiety, was reduced significantly by the addition of the S9
liver fraction p < 0.0001. This resulted in increased IC90 values for
m
m
Macrophages are the major host cell for M. tuberculosis infection
[27]. For most of its infectious cycle, M. tuberculosis resides within
the early endosome of the macrophages where it replicates.
Therefore it was important to investigate if the selected 1,3-
diaryltriazene 14 is able to reach its target through the mem-
brane structure of the macrophages. In the macrophage infection
assay set up, cells were first infected with M. tuberculosis H37Rvlux
.
After 24 h of infection, the infected J774 murine macrophages were
treated with the compound 14 for 3 days. The treatment was fol-
lowed by washing and the subsequent lysis of the J774 monolayer
after a cell viability control by tryptophan blue staining. The
luminescence emitted by the cell lysate was compared with
infected macrophages which received no treatment. If the bacterial
load inside the cells was diminished it was concluded that the
compound was able to reach its target and kill it. As shown in Fig. 5,
the compound 14 could indeed reduce the bacterial load inside the
macrophages and after three days of exposure at 15.55
M. tuberculosis load inside the macrophage monolayer was reduced
by 71.67%. At 7.79 M the reduction still remained 62.02%. At 3.88
and 1.94 M the reduction dropped to respectively 43.81% and
mM, the
m
m
6.32% as compared to the bacterial load measured in the lysate from
untreated M. tuberculosis infected macrophages.
To extend the study, antimicrobial activity against a multidrug
resistant M. tuberculosis isolate was determined with the fluoro-
metric BACTEC MGIT 960Ô Mycobacteria Growth Indicator (Fig. 6)
[28]. This standardized automated fluorometric assay relies an
these compounds: for compound
3.84 ꢁ 0.35 M to 7.95 ꢁ 0.48 M in the presence of S9. For the
derivative 18 the IC90 rose from 26.75 ꢁ 0.24 M to 67.26 ꢁ 7.42
and finally the IC90 without S9 of the compound 20 of
M. The potency
2
the IC90 rose from
m
m
oxygen-quenched
fluorochrome,
tris(4,7-diphenyl-1,10-
m
mM
phenanthroline)ruthenium(II) chloride pentahydrate, embedded
in silicone at the bottom of the tube. During a mycobacterial growth
within the tube, free oxygen is utilized and is replaced with CO2.
With depletion of free oxygen, the fluorochrome is no longer
inhibited, resulting in fluorescence within the MGIT tube when
visualized under UV light. The intensity of a fluorescence is directly
proportional to the extent of the oxygen depletion and the growth
of the mycobacterial strain. Drug susceptibility can be measured in
4e6 days.
The growth inhibition induced by a compound was calculated
by comparing the growth index of the M. tuberculosis culture
exposed to different compound concentrations with the growth
index of an untreated control culture 100 times diluted upon
inoculation (C/100). As a resistant M. tuberculosis strain, a LAM-1
clinical isolate was selected. This MDR M. tuberculosis strain
shows resistance against isoniazid, rifampicin, rifabutine and pro-
thionamide. The susceptibility to the compound 14 was tested at a
0.63 ꢁ 0.04
m
M was increased to 9.70 ꢁ 0.87
m
observed for the compounds 1, 10, 13 and 14 that had a CF3 moiety
was not altered by addition of the S9 fraction.
Further study of the genotoxic properties was performed for the
compounds 10, 13 and 14 using a Comet assay [26]. With this assay,
a fragmentation of the genomic DNA can be detected. Briefly, J774
A.1 cells were exposed to serial dilutions of the compounds and
immobilized onto agarose. Immobilized cells were lysed and the
remaining nuclei were submitted to electrophoresis. Through the
pore exclusion of the agarose, fragmented DNA migrated further
with the applied current. After staining with a gel red it can be
observed that the genomic DNA of an undamaged nucleus main-
tains its spherical organization (HEAD) whereas fragmented DNA
migrates further across the gel (TAIL). The ratio DNA in the TAIL/
DNA in the HEAD is a measure of DNA fragmentation due to the
genotoxicity of the compounds for the J774 A.1 murine macrophage
like monocytes.
concentration range of 25.00
to the untreated control diluted 100 times upon inoculation, was
found for a concentration between 12.5 M and 6.25 M, con-
firming the susceptibility of this MDR strain for the compound. INH
did not inhibit the growth at the critical concentration of 0.73 M,
mMe1.25 mM. A growth index similar
At low concentrations of 0.13
mM of the compound 10 there was
m
m
a significant increase of an amount of DNA in the TAIL (p < 0.0001
ManneWhitney test). The amount of DNA fragmentation was lower
m
at 0.25
mM but still a significant increase (p < 0.05 ManneWhitney
confirming the resistance of the strain towards this first line
test). The lower degree of DNA fragmentation at this concentration
could be explained by cytotoxic effects caused by the compound. At
antibiotic.
1.25
mM the concentration of the compound 10 exceeded its NI50
3. Conclusions
concentration and the results could not be interpreted. The results
indicated fragmentation of the genomic DNA of the macrophages
caused by the compound 10. DNA fragmentation could not be
A library of substituted 1,3-diaryltriazenes was synthesized in a
convenient way followed by a biological evaluation for their

198
D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
Fig. 2. The VITOTOXÔ assay for the detection of early signs of genotoxicity caused by the compounds. Maximum recorded S/N in a time span of 4 h by the Genox (recN2-4) and
Cytox (pr1) reporter strains. 4NQO; 4-nitroquinolone-1-oxide genotoxic positive control in samples without S9 liver fraction. Bap; benzo[a]pyrene, the positive control, only turns
genotoxic after S9 metabolization.
potential as anti-tubercular agents. Several compounds were
detected as ‘hits’ with an IC50 below 1 M against M. tuberculosis.
further testing showed a certain level of toxicity for J774 A.1 mac-
rophages at low concentrations (30 M) but due to the high po-
m
m
From the structure activity relation it could be deduced that com-
pounds carrying a NO2 group at the para R4 position or a CF3
functionality at the meta R3 position were more active than ana-
logues with alternative electron-withdrawing groups at those po-
sitions. Changing the position of these groups resulted in
abrogation of the potency as well. The compounds selected for
tency of some, an acceptable selectivity index above 7 could be
calculated for the compounds 1, 2, 10, 13, 14, 18 and 20. For these
1,3-diaryltriazenes the activity against M. bovis, M. ulcerans and two
M. avium subspecies was confirmed, broadening their applicability.
Genotoxicity was studied and detected for the compounds 2, 18 and
20 after the addition of the S9 liver extract, but not in the absence of

D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
199
Fig. 5. Inhibition of an intracellular replication of M. tuberculosis inside macrophages
by the compound 14. Growth inhibition of M. tuberculosis H37Rv inside J774 A.1
macrophages. Results are presented as mean ꢁ SD RLU of triplicate cultures. A treat-
ment of the infected cells started 24 h post infection and lasted for 3 days.
the growth of a multi-drug resistant M. tuberculosis LAM-1 strain as
well as the intracellular replication of M. tuberculosis, resulting in a
growth reduction of 71.67% at 15.55
mM. As the anti-tubercular
Fig. 3. The 90% in vitro growth inhibitory concentration of the derivatives against
M. tuberculosis in the presence and in the absence of the S9. IC90, the concentration at
which 90% of the M. tuberculosis growth is inhibited, was measured in the presence
and absence of the S9 liver extract. Results are shown as an average of the triplicate
with SD bars. Significance was studied with ManneWhitney test p < 0.0001.
potency was accompanied with an acute cytotoxicity, combined
with the DNA binding properties of the anti-trypanosomal drugs
diminazene, the development of 1,3-dirayltriazenes should be done
with caution. Given the fact that TB chemotherapy requires inten-
sive treatment for several months, it is unlikely that these com-
pounds could find their application as a first line drugs for the
treatment of the drug susceptible strains. However with the in-
crease in incidence of multi-, extensively- and total-drug resistant
TB, when other treatment options are exhausted, this compound
class might prove of use.
S9. It indicated that a liver metabolism could be responsible for
converting the compounds into genotoxic metabolites. This could
not be observed for 1, 10, 13 and 14. After analysis of the structural
differences, it was found that the latter four compounds all carried
a CF3 functional group, preventing the metabolic destruction by the
S9 extract. Next the MIC concentration against M. tuberculosis of the
triazenes 1, 2, 10, 13, 14, 18 and 20 was studied in the presence or in
the absence of S9. It was observed that the potency of the de-
rivatives 2, 18 and 20 was significantly reduced and the molecules
most probably metabolized in a non-potent metabolite whereas
this was not the case for 1, 10, 13 and 14 where no reduction in
efficacy was observed. This reinforced the importance of a CF3
moiety as a functional group on these 1,3-diaryltriazene molecules.
Further investigation of the possible genotoxic effects for the three
most promising compounds showed that the compound 10 did
indeed cause the DNA breakage and that for the triazene 13 the
results were inconclusive due to cytotoxicity. In the case of 14, the
DNA breakage could not be observed at the concentrations tested.
4. Experimental protocols
4.1. Chemistry
Starting materials for the synthesis of the examined compounds
were used as obtained from the commercial sources (Aldrich, Fluka,
The compound 14 (IC50 ¼ 3.26
m
M, NI50 ¼ 24.22 M, SI ¼ 7.44) was
m
ultimately selected as the best representative of the library. In
further studies it was found that this compound was able to inhibit
Fig. 6. Susceptibility of an MDR M. tuberculosis LAM-1 strain to the triazene 14. Growth
of MDR M. tuberculosis was monitored in BACTEC MGIT 960Ô. a,bGrowth was
compared to an untreated bacterial suspension and an untreated bacterial suspension
diluted 100 times (C/100) to study the minimal inhibitory concentration MIC.
Resistance was confirmed with isoniazid (INH).
Fig. 4. Detection of DNA fragmentation by Comet assay. Genomic DNA fragmentation
after 24 h exposure to the different concentrations of the compounds 10, 13 and 14. The
ManneWhitney U test was used to confirm the significance.
c

200
D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
Alfa Aesar, Maybridge Chemical Company Ltd). Melting points were
determined on a Kofler micro hot stage and are uncorrected. NMR
spectra were recorded at 29 ꢀC with a Bruker Avance DPX 300
spectrometer (1H NMR spectra at 300 MHz; 13C NMR spectra at
75.5 MHz) and with a Bruker Avance III 500 spectrometer (1H NMR
spectra at 500 MHz; 13C NMR spectra at 125.8 MHz). Proton spectra
are referenced to TMS as an internal standard; the carbon shifts are
given against the central line of the solvent signal (DMSO-d6 at
4.1.1.3. 2,20-(Triazene-1,3-diyl)-bis(6-fluorobenzonitrile)
(4).
Method A; reaction time: 20 h; yield: 92%; mp 182.5e184.5 ꢀC
(MeOH/H2O); IR: 3181, 2230, 1616, 1585, 1520, 1475, 1458,
1418 cmꢃ1 1H NMR (CDCl3):
; d 6.99e7.15 (2H, m), 7.54e7.68 (4H,
m),10.22 (s,1H); HRMS for C14H8F2N5þ ([M þ H]þ): calcd 284.0742,
found 284.0739. Anal. for C14H7F2N5 ꢄ 1/8H2O: calcd C, 58.90; H,
2.56; N, 24.53; found C, 58.84; H, 2.42; N, 24.23.
d
¼ 39.5 ppm; CDCl3 at
d
¼ 77.1 ppm). In the 13C NMR spectra of
4.1.1.4. 1,3-Bis(4-fluoro-2-(trifluoromethyl)phenyl)triazene
Method A; reaction time: 19 h; yield: 84%; mp 155.3e157.5 ꢀC
(5).
compounds 1, 3e9, 12 and 13 the resonances of were broadened to
the baseline and are thus not reported. IR spectra were obtained
with a PerkineElmer Spectrum 100, equipped with a Specac
Golden Gate Diamond ATR as a solid sample support. MS spectra
were recorded with an Agilent 6224 Accurate Mass TOF LC/MS
spectrometer. Elemental analyses (C, H, N) were performed with
Perkin Elmer 2400 Series II CHNS/O Analyzer. TLC was carried out
on Fluka silica-gel TLC-cards.
(MeOH/H2O); IR: 3358, 1612, 1592, 1522, 1483, 1449, 1433, 1420,
1320 cmꢃ1
;
1H NMR (CDCl3):
d
7.22e7.32 (2H, m), 7.33e7.48 (2H,
þ
m), 7.62e7.89 (2H, m), 9.94 (1H, s); HRMS for C14H8F8N3
([M þ H]þ): calcd 370.0585, found 370.0581. Anal. for C14H7F8N3:
calcd C, 45.54; H, 1.91; N, 11.38; found C, 45.54; H, 1.88; N, 11.40.
4.1.1.5. 2,20-(Triazene-1,3-diyl)-bis(5-fluorobenzonitrile)
Method B; reaction time: 0.5 h; yield: 40%; mp 158e160 ꢀC (MeOH/
(6).
All tested compounds possessed a purity of ꢂ95% as verified by
NMR spectroscopy and elemental microanalysis. The figures, ob-
tained from C, H, N analyses, were in all cases within ꢁ0.35% of the
calculated values.
H2O); IR: 3221, 2229, 1603, 1592, 1584, 1523, 1495, 1479, 1439,
1409 cmꢃ1; 1H NMR (DMSO-d6):
d 7.37e8.06 (6H, m), 13.44 (1H, s);
HRMS for C14H8F2N5þ ([M þ H]þ): calcd 284.0742, found 284.0737.
Anal. for C14H7F2N5 ꢄ 1/6H2O: calcd C, 58.74; H, 2.58; N, 24.47;
found C, 58.76; H, 2.49; N, 24.23.
4.1.1. General procedures for the synthesis of 1,3-diaryltriazenes
Method A: A solution of sodium nitrite (34.5 mg, 0.5 mmol) in
water (3 mL) was added dropwise to the suspension of the
appropriate aniline (1 mmol) in 5% water solution of HCl (2.5 mL) at
0 ꢀC. The reaction mixture was stirred at room temperature for 18e
23 h. Then, the solid material was filtered off and washed with
water at 0 ꢀC yielding the corresponding triazene.
4.1.1.6. 3,30-(Triazene-1,3-diyl)-bis(6-fluorobenzonitrile)
Method B; reaction time: 1.5 h; yield: 58%; mp 205e207 ꢀC (pe-
(7).
troleum ether/EtOAc); IR: 3285, 2235, 1613, 1600, 1582, 1522, 1487,
1454, 1405 cmꢃ1
;
1H NMR (DMSO-d6):
d
7.46e8.30 (6H, m), 12.93
(1H, s); HRMS for C14H6F2N5 ([MꢃH]e): calcd 282.0597, found
282.0591. Anal. for C14H7F2N5 ꢄ 1/8EtOAc: calcd C, 59.19; H, 2.74; N,
23.80; found C, 59.26%; H, 2.73%; N, 23.83%.
ꢃ
Method B: A solution of isoamyl nitrite (117 mg, 1 mmol) in
dichloromethane (4 mL; toluene, 2.5 mL in the case of 9) was added
dropwise to the solution of the appropriate aniline (1 mmol) in
dichloromethane (4 mL; petroleum ether, 2.5 mL in the case of 9) at
4.1.1.7. 1,3-Bis(4-fluoro-3-(trifluoromethyl)phenyl)triazene
Method B; reaction time: 0.5 h; yield: 56%; mp 105.4e107 ꢀC (pe-
(8).
0
ꢀC. The reaction mixture was stirred for 0.5e22 h at room tem-
perature and evaporated to dryness under reduced pressure. Tri-
azenes 8 and 9 were isolated by column chromatography. The crude
products were always crystallized from the appropriate solvent.
Method C: A selected 3-arylpropenoic acid chloride (1.1 mmol)
was slowly added to the mixture of the appropriate 1,3-
diaryltriazene (1 mmol) and ethylamine (0.282 mL, 2 mmol) in
acetone (5 mL). The reaction mixture was stirred at room temper-
ature for 5e40 min. The product was isolated as follows: (a) Trie-
thylamine hydrochloride was filtered off, the solution was
evaporated to dryness and treated with a solvent (1 mL of methanol
was used in the case of 21 and 25; 2 mL of methanol and 0.5 mL of
acetone for the triazene 26) to obtain the desired product. (b) A
solid material was separated by filtration, then suspended in water
(2 mL) to remove triethylamine hydrochloride and the final product
was filtered off (an isolation of 22 and 23).
troleum ether); IR: 3339, 2987, 1618, 1513, 1493, 1464, 1421, 1321,
1304 cmꢃ1; 1H NMR (DMSO-d6):
d 7.43e8.09 (6H, m), 12.87 (1H, s);
HRMS for C14H8F8N3þ ([M þ H]þ): calcd 370.0585, found 370.0582.
Anal. for C14H7F8N3: calcd C, 45.54; H, 1.91; N, 11.38; found C, 45.70;
H, 1.95; N, 11.31.
4.1.1.8. N-(4-fluoro-3-nitrophenyl)-4-(3-(4-fluoro-3-nitrophenyl)tri-
azene-1-yl)-2-nitroaniline (9). Method B; reaction time: 22 h; yield:
37%; mp 193e195 ꢀC (petroleum ether/EtOAc); IR: 3255, 2987,1615,
1590, 1521, 1496, 1443, 1401, 1343, 1316 cmꢃ1; 1H NMR (DMSO-d6):
d
7.37e7.45 (1H, m), 7.57e7.67 (2H, m), 7.70e7.90 (3H, m), 8.00e
þ
8.21 (3H, m), 9.45 (1H, s), 12.86 (1H, s); HRMS for C18H12F2N7O6
([M
C
þ
H]þ): calcd 460.0812, found 460.0812. Anal. for
18H11F2N7O6 ꢄ 0.25C2H60: calcd C, 47.19; H, 2.68; N, 20.82; found
C, 47.31; H, 2.41; N, 20.59.
4.1.1.1. 1,3-Bis[(2-fluoro-3-trifluoromethyl)phenyl]triazene
Method A; reaction time: 20 h; yield: 75%; mp 111e113 ꢀC (MeOH/
H2O); IR: 2988, 1633, 1596, 1535, 1489, 1470, 1418, 1325, 1313 cmꢃ1
(1).
4.1.1.9. 4,40-(Triazene-1,3-diyl)-bis[2-(trifluoromethyl)benzonitrile]
(12). Method A; reaction time: 18 h; yield: 91%; mp 215e217 ꢀC
(MeOH/H2O); IR: 2227, 1610, 1592, 1518, 1490, 1455, 1407, 1362,
;
1H NMR (CDCl3):
d
7.22e7.33 (2H, m), 7.40e7.49 (2H, m), 7.83e7.93
1330, 1314 cmꢃ1
;
1H NMRþ(CDCl3):
d 7.67e7.94 (6H, m), 10.14 (1H,
(2H, m), 10.02 (1H, s); HRMS for C14H8F8N3 ([M þ H]þ): calcd
370.0585, found 370.0586. Anal. for C14H7F8N3: calcd C, 45.54; H,
1.91; N, 11.38; found C, 45.33; H, 1.95; N, 11.32.
s); HRMS for C16H8F6N5 ([M þ H]þ): calcd 384.0678, found
384.0672. Anal. for C16H7F6N5 ꢄ 1/4H2O: calcd C, 49.56; H, 1.95; N,
18.06; found C, 49.76; H, 1.80; N, 17.67.
þ
4.1.1.2. 1,3-Bis[(2-fluoro-5-trifluoromethyl)phenyl]triazene
(3).
4.1.1.10. 1,3-Bis[(4-nitro-3-(trifluoromethyl)phenyl)]triazene
(13).
Method A; reaction time: 20 h; yield: 98%; mp 125.1e126.3 ꢀC
Method A; reaction time: 23 h; yield: 87%; mp 207e208 ꢀC (MeOH/
(MeOH/H2O); IR: 3313, 1622, 1611, 1530, 1513, 1489, 1478, 1448,
H2O); IR: 3218, 1610, 1595, 1532, 1514, 1482, 1455, 1411, 1393, 1334,
1413, 1335 cmꢃ1
;
1H NMR (CDCl3):
d
7.23e7.34 (2H, m), 7.39e7.5ꢃ4
1320 cmꢃ1
;
1H NMR (DMSO-d6):
d
7.94e8.08 þ(4H, m), 8.23e8.31
(2H, m), 7.91e8.00 (2H, m), 10.04 (1H, s); HRMS for C14H6F8N3
([MꢃH]e): calcd 368.0439; found 368.0438. Anal. for C14H7F8N3:
calcd C, 45.54; H, 1.91; N, 11.38; found C, 45.29; H, 1.93; N, 11.31.
(2H, m), 13.68 (1H, s); HRMS for C14H8F6N5O4 ([M þ H]þ): calcd
424.0475, found 424.0467. Anal. for C14H7F6N5O4 ꢄ 1/3H2O: calcd C,
39.18; H, 1.80; N, 16.32; found C, 39.36; H, 1.59; N, 16.08.

D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
201
4.1.1.11. 3-[(2E)-3-(2,4-dimethoxyphenyl)prop-2-enoyl]-1,3-bis[(4-
chloro-2-trifluoromethyl)-phenyl]-triazene (21). Method C; reaction
time: 5 min; yield: 80%; mp 154.5e156.5 ꢀC (ethanol); IR:1682,
NMR (DMSO-d6): d 20.4, 56.1, 113.5, 116.3, 120.8, 121.0, 121.4, 123.3,
123.4, 123.8, 124.1, 127.7, 128.5, 132.4, 133.0, 141.4, 141.9, 145.9, 147.7,
148.5,149.8,151.2,165.9,168.3; MS (FAB) m/z 662 ([Mþ1]þ, 0.2),154
(100), 137 (70), 71 (46); Anal. for C24H17Br2N5O8: calcd C, 43.46; H,
2.58; N, 10.56; found C, 43.65; H, 2.75; N, 10.30.
1603, 1566, 1359, 1310, 1260, 1214, 1162, 1134, 1049, 995 cmꢃ1 1H
;
NMR (CDCl3):
d
3.87 (3H, s), 3.91 (3H, s), 6.49 (1H, d, J ¼ 2.3 Hz), 6.56
(1H, dd, J ¼ 2.3, 8.6 Hz), 7.14 (1H, d, J ¼ 8.5 Hz), 7.53e7.61 (2H, m),
7.62e7.72 (4H, m), 7.79 (1H, d, J ¼ 2.3 Hz), 8.21 (1H, d, J ¼ 15.8 Hz);
4.2. Biology
13C NMR (CDCl3):
d 55.50, 55.51, 98.5, 105.5, 113.5, 116.8, 119.3,
122.0,122.4,127.0,127.6,128.1,130.4,131.5,132.6,132.8,132.9,133.1,
134.9, 135.9, 142.4, 144.2, 160.5, 163.3, 167.8; MS (EI) m/z 591 (Mþ,
11),191 (100),179 (22); Anal. for C25H17Cl2F6N3O3: calcd C, 50.69; H,
2.89; N, 7.09; found C, 50.86; H, 2.92; N, 6.88.
4.2.1. Materials
7H9 Growth medium was purchased as a powder from BD Sci-
ence (Franklin Lakes, NJ, USA). DMEM, glutamax, non-essential
amino acids, sodium pyruvate, gentamycin, 2-mercaptoethanol,
PBS were from GIBCO Invitrogen (Carlsbad, CA, USA). OADC,
mycobactin J, penicillin, fungizone and hygromycin were from
Roche (Basel, CH) and Triton X-100, glycerol, Tween 80 and n-
decanal were purchased from Sigma Aldrich (St. Louis, MO, USA).
Strains and growth conditions e M. tuberculosis H37Rv (Amer-
ican Type Culture Collection 27294) is known to be sensitive to the
five first line anti-tuberculosis drugs (streptomycin, isoniazid,
rifampin, ethambutol and pyrazinamide). The LAM-1 strain is a
clinical isolate from a patient diagnosed with TB. This strain has
been spoligotyped for identification and characterized for antibiotic
resistance by the national TB reference lab of the Institute of Public
Health of Belgium. With the resistance to isoniazid, rifampicin,
rifabutin and prothionamid this strain was classified as a multi-
drug resistant (MDR) strain. Four other strains used for screening
were M. bovis strain AN5, M. avium ssp. avium ATCC 15769, M. avium
ssp. paratuberculosis ATCC19698 and M. ulcerans 1615. All strains
were cultivated in 7H9 medium supplemented with 10% oleic acid-
albumin-dextrose-catalase (OADC) and 0.2% glycerol. For M. bovis,
the glycerol was replaced with 0.05% Tween 80. The growth me-
dium of M. avium ssp. paratuberculosis was further supplemented
4.1.1.12. 3-[(2E)-3-(2,4-dimethoxyphenyl)prop-2-enoyl]-1,3-bis[(4-
chloro-3-trifluoromethyl)-phenyl]-triazene (22). Method C; reaction
time: 5 min; yield: 84%; mp 160e161 ꢀC (ethanol); IR (KBr):1681,
1602, 1563, 1478, 1354, 1314, 1217, 1180, 1168, 1126, 1008 cmꢃ1 1H
;
NMR (CDCl3):
d
3.88 (3H, s), 3.92 (3H, s), 6.51 (1H, d, J ¼ 2.4 Hz), 6.57
(1H, dd, J ¼ 2.4, 8.5 Hz), 7.30 (1H, dd, J ¼ 2.4, 8.5 Hz), 7.48e7.54 (2H,
m), 7.58 (1H, d, J ¼ 8.6 Hz), 7.66 (1H, d, J ¼ 8.5 Hz), 7.67 (1H, dd,
J ¼ 2.4, 8.6 Hz), 7.74 (1H, d, J ¼ 15.8 Hz), 7.92 (1H, d, J ¼ 2.4 Hz), 8.14
(1H, d, J ¼ 15.8 Hz); 13C NMR (CDCl3):
d 55.46, 55.52, 98.6, 105.5,
113.8, 116.8, 121.1, 122.4, 122.5, 126.5, 128.5, 129.4, 129.5, 132.36,
132.38, 132.39, 132.8, 133.1, 133.6, 134.5, 142.9, 146.9, 160.7, 163.3,
168.1; MS (EI) m/z 591 (Mþ, 0.1), 191 (100), 148 (13); Anal. for
C
25H17Cl2F6N3O3: calcd C, 50.69; H, 2.89; N, 7.09; found C, 50.74; H,
2.96; N, 6.92.
4.1.1.13. 3-[(2E)-3-(3,5-dimethoxyphenyl)prop-2-enoyl]-1,3-bis[(4-
chloro-3-trifluoromethyl)-phenyl]-triazene (23). Method C; reaction
time: 5 min; yield: 73%; mp 158e160 ꢀC (ethanol); IR: 1689, 1619,
1595, 1480, 1344, 1313, 1208, 1159, 1137, 1123 cmꢃ1 1H NMR
;
(CDCl3):
d
3.85 (6H, s), 6.56 (1H, dd, J ¼ 2.3 Hz), 6.78 (2H, d,
with mycobactin J (2 mg/mL) required for optimal in vitro growth.
J ¼ 2.3 Hz), 7.31 (1H, dd, J ¼ 2.4, 8.4 Hz), 7.52 (1H, d, J ¼ 2.4 Hz),
7.49e7.70 (4H, m), 7.91 (1H, d, J ¼ 2.4 Hz), 7.91 (1H, d, J ¼ 15.7 Hz);
Both M. avium strains and the M. bovis strain were grown at 39 ꢀC,
M. ulcerans was grown at 32 ꢀC and M. tuberculosis was grown at
37 ꢀC.
13C NMR (CDCl3):
d 55.4,103.1,106.4,116.2,121.3,122.3,122.4,126.4,
128.5, 129.4, 129.7, 132.5, 133.2, 133.4, 133.5, 134.0, 136.3, 146.5,
146.8, 161.2, 167.0; MS (FAB) m/z 592 ([Mþ1]þ, 1.0) 207 (100), 191
(72); Anal. for C25H17Cl2F6N3O3: calcd C, 50.69; H, 2.89; N, 7.09;
found C, 50.77; H, 2.94; N, 7.01.
4.2.2. Monitoring mycobacterial growth by luminometry
The minimal inhibitory concentration (MIC) against mycobac-
teria of all synthesized compounds was evaluated by testing serial
dilutions. The in vitro assay was based on a method in which
luminescent mycobacteria transformed with pSMT1 luciferase re-
porter plasmid is used. The tested compounds were solubilized in
DMSO (SigmaeAldrich) at stock concentrations of 10 mM. Serial
dilutions of each compound were made in liquid 7H9 medium
[Middlebrook 7H9 broth based (Difco)] þ 10% FCS (Gibco). Volumes
4.1.1.14. 3-[(2E)-3-(4-acetoxy-3-methoxyphenyl)prop-2-enoyl]-1,3-
bis[2-chloro-4-nitrophenyl]-triazene (25). Method C; reaction time:
5 min; yield: 88%; mp 170e173 ꢀC (methanol); IR: 1767, 1622, 1527,
1350, 1257, 1194, 1132, 1045, 1000 cmꢃ1 1H NMR (CDCl3):
; d 2.34
(3H, s), 3.89 (3H, s), 7.12 (1H, d, J ¼ 8.3 Hz), 7.19 (1H, d, J ¼ 1.9 Hz),
7.30 (1H, dd, J ¼ 1.9, 8.1 Hz), 7.50 (1H, d, J ¼ 8.7 Hz), 7.59 (1H, d,
J ¼ 15.7 Hz), 7.64 (1H, d, J ¼ 8.9 Hz), 8.01 (1H, d, J ¼ 15.7 Hz), 8.22
(1H, dd, J ¼ 2.4, 8.9 Hz), 8.30 (1H, dd, J ¼ 2.4, 8.7 Hz), 8.34 (1H, d,
of 20 mL of the serial dilutions were added in triplicate to 96-well,
flat-bottomed micro well plates. The bacterial suspension was
made by thawing and dissolving a frozen Mycobacteria pellet in
7H9-10% FCS. The dissolved pellet was passed through a 5.0 mM
filter (Millipore) to eliminate clumps and left for 1 h to recover at
37 ꢀC, 5% CO2. Next, the bacterial suspension was diluted in 7H9-
10% FCS to obtain 50,000 Relative Light Units (RLU)/mL and a vol-
J ¼ 2.3 Hz), 8.45 (1H, d, J ¼ 2.4 Hz); 13C NMR (CDCl3):
d
20.6, 56.0,
112.6, 115.1, 120.2, 121.1, 122.72, 122.77, 123.5, 125.4, 126.1, 131.6,
132.3,133.1,134.4,139.8,142.2,147.2,147.8,148.8,149.4,151.6,166.0,
168.6; MS (EI) m/z 574 (Mþ, 2.2), 154 (100), 136 (75), 71 (70), 55
(77); Anal. for C24H17Cl2N5O8: calcd C, 50.19; H, 2.98; N,12.19; found
C, 50.44; H, 3.07; N, 11.86.
ume of 180
mL of bacteria was added to each well. A bacterial
replication was analysed by luminometry after 6 days of incubation.
The bacterial suspension from each well was collected, and trans-
ferred to a black 96-well plate to evade cross luminescence be-
tween wells. The luminescent signal was evoked by addition of the
substrate for the bacterial luciferase, 1% n-decanal in ethanol to
each well by the multi plus reader from Promega and the light
emission in each well was measured.
4.1.1.15. 3-[(2E)-3-(4-acetoxy-3-methoxyphenyl)prop-2-enoyl]-1,3-
bis[2-bromo-4-nitrophenyl]-triazene (26). Method C; reaction time:
40 min; yield: 94%; mp 180e181 ꢀC (diisopropyl ether/acetone); IR:
1769, 1699, 1623, 1527, 1467, 1347, 1256, 1193, 1150, 1121, 1038,
999 cmꢃ1; 1H NMR (DMSO-d6):
d 2.34 (3H, s), 3.90 (3H, s), 7.12 (1H,
d, J ¼ 8.2 Hz), 7.19 (1H, d, J ¼ 1.8 Hz), 7.31 (1H, dd, J ¼ 1.8, 8.2 Hz),
7.48 (1H, d, J ¼ 8.7 Hz), 7.58 (1H, J ¼ 15.8 Hz), 7.62 (1H, d, J ¼ 8.7 Hz),
8.02 (1H, d, J ¼ 15.8 Hz), 8.27 (1H, dd, J ¼ 2.4, 8.7 Hz), 8.35 (1H, dd,
4.2.3. Activity against multi-drug resistant M. tuberculosis
Antimicrobial activity of the compounds was tested using the
LAM-1 strain using the BACTECÔ MGITÔ 960 TB detection system.
J ¼ 2.4, 8.7 Hz), 8.52 (1H, d, J ¼ 2.4 Hz), 8.63 (1H, d, J ¼ 2.4 Hz); 13
C

202
D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
The compounds were dissolved in DMSO at stock concentrations of
1 mg/mL. Serial dilutions of each compound were made in 7H9
containing 10% OADC, at 83-fold the final concentrations.
M. tuberculosis LAM-1 was pre-cultured in a 4 mL BACTEC vial to a
4.2.6. Vitotox assay
Observations on genotoxicity were done with Vitotox (Gentaur,
Kampenhout, BE) and the included protocol was followed. In brief,
TA104 RecN2-4 (genox) and TA104 (cytox) Salmonella typhimurium
bacteria were cultivated at 36 ꢀC for 16 h in poor 869 medium. The
bacterial culture was diluted 250 times, incubated for 1 h at 36 ꢀC
and kept on ice. The bacterial culture was divided by 10 once more.
S9 was added to the designated þS9 cultures to test the genotoxic
effects of the metabolites of the compounds. The bacterial sus-
pensions were then incubated, shaking at 36 ꢀC, and the lumines-
cent signal was measured for 4 h with a 5 min interval.
growth index (GI) of 300. Then 100
inoculated into a new 4 mL BACTEC vial together with 100
serial dilutions of the compounds. As a positive control for the
resistance, the MDR LAM-1 culture was inoculated with 0.1
mL of this pre-culture was
mL of the
mM
isoniazid. The GI was measured each day. To determine the IC99, the
cultures, treated with the compounds were compared with an
untreated culture diluted 100 times upon inoculation.
4.2.7. Comet assay
4.2.4. Inhibition of intracellular M. tuberculosis growth
Possible DNA breakage effects of the derivatives on J774 A.1 cells
were investigated by the alkaline comet assay. The J774 A.1 cells
were grown in DMEM þ 10% FCS until a semi confluent layer of cells
was obtained. The cells were trypsinized, washed and seeded at
100,000 cells per well of a 24 well plate and left for recovery at
37 ꢀC, 5% CO2. The following day, the testing compounds were
dissolved in DMSO as a stock concentration of 1 mM. Serial di-
lutions of each compound were made in DMEM þ 10% FCS to obtain
the final concentrations. The J774 A.1 cells were washed and
exposed to the derivatives by adding 1 mL of the serial dilutions to
each well. The plates were incubated at 37 ꢀC, 5% CO2 for 24 h. After
The compounds were tested on the murine J774 A.1 macrophage
cell line infected with M. tuberculosis H37Rvlux. The J774 macro-
phages were grown at 37 ꢀC, 5% CO2 in complete DMEM medium
until a semi confluent layer was formed. The macrophages were
washed in fresh complete DMEM medium and seeded in a flat-
bottomed 96-well micro well plate at
a cell density of
40,000 cells per well. The cells were left to recover overnight and
were washed three times in complete DMEM medium.
M. tuberculosis H37Rvlux was grown at 37 ꢀC in 7H9 containing 10%
FCS and 0.2% hygromycin to an OD580 of 0.6e1.0. The fully-grown
bacterial suspension was measured and brought into complete
DMEM-Pen/Fung [DMEM medium containing 0.1% penicillin and
0.8% fungizone but without gentamicin]. The synthesized com-
pounds were dissolved in DMSO at stock concentrations of 10 mM.
Serial dilutions of the peptides were made in DMEM-Pen/Fung at
two times the concentration of each compound to be tested. A
incubation, the cells were trypsinized, washed with PBS, and 10
mL
of cell suspension was dissolved in 300 L low melting point
m
agarose. The dissolved cell suspension was then placed onto a
frosted microscope slide and left on ice for 5 min. The slide was
subsequently placed in a jar containing lysing solution for 1 h. After
lysis, the agarose cell suspension was subjected to electrophoresis
for 20 min at 300 mA. The slides were washed with neutralization
buffer for 5 min and dried in ice cold ethanol for 10 min. Staining of
the DNA was done with gel red (SigmaeAldrich). For the quantifi-
cation of the DNA migration, a fluorescence microscope was used
and the percentage DNA in the comet tail of the cells nuclei core
was calculated in proportion to the total DNA present in the nuclei
(comet head þ tail) by appropriate imager software from Meta-
systems (Altlussheim, Germany).
volume of 100
containing 4000 RLU of bacteria (multiplicity of infection of 0.1)
and 100 L of the serial compound dilutions were added to the
mL of the bacterial suspension in DMEM-Pen/Fung
m
macrophage cultures. To measure the effects of the compounds on
intracellular growth of M. tuberculosis, the infected macrophages
were washed three times on day 5 to remove all extracellular
bacteria, incubated 1 h with 1% gentamicin to kill the residual
extracellular bacteria, lysed with 200
and the wells washed four times with 200
transferred in a 2.5 mL tube together with the 4 PBS washings. One
hundred L of 1% n-decanal in ethanol was added to the tube and
mL 1% Triton X-100 (Sigma)
mL PBS. The lysate was
Acknowledgements
m
the luminescence was measured. RLU values shown were obtained
from six replicate cultures. Cell viability of the macrophage culture
was observed by Trypan blue with a microscope.
This work was partially supported by the Research Foundation
Flanders (FWO-Vlaanderen) (Grant G.0020.10N) and by funding
from the European Community’s Seventh Framework Program
(FP7/2007-2013) under grant agreement FP7-223681. Slovenian
Research Agency is gratefully acknowledged for the financial sup-
4.2.5. Assessment of cytotoxicity
Inhibitory effects on the J774 A.1 murine macrophage cell line
were determined for the derivatives by a neutral red uptake assay
as described before. The J774 A.1 cells were grown in DMEM þ 10%
FCS until a semi confluent layer of cells was obtained. The cells were
trypsinized, washed and 40,000 cells were seeded per well of a 96-
well plate and left for recovery at 37 ꢀC, 5% CO2. The following days,
the compounds were dissolved in DMSO to stock concentrations of
ꢀ
port (Project P1-0230). We thank Dr. Bogdan Kralj and Dr. Dusan
ꢀ
ꢀ
Zigon (Mass Spectrometry Centre, Jozef Stefan Institute, Ljubljana,
Slovenia) for recording some of the mass spectra. This work was
also partly supported with the infrastructure of the EN-FIST Centre
of Excellence, Dunajska 156, SI-1000 Ljubljana, Slovenia.
Appendix A. Supplementary data
10
mM. A serial dilution of each compound was made in
DMEM þ 10% FCS. The J774 A.1 cells were washed and exposed to
the derivatives by adding the serial dilutions of the compounds to
the wells. The plates were left for incubation at 37 ꢀC, 5% CO2 for
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.02.065.
24 h. After exposure, the cells were washed with 200
200 L neutral red working solution (Sigma) was added per well.
Subsequently the plates were incubated for 3 h at 37 ꢀC, 5% CO2. The
wells were washed with 200 L PBS and 200 L of an ethanol/acetic
mL PBS and
References
m
[1] World Health Organization, Global Tuberculosis Report, 2012. http://www.
who.int/tb/publications/global_report/en/.
m
m
[2] A.O. Ombaka, A.T. Muguna, J.M. Gichumbi, Antibacterial and antifungal ac-
tivities of novel hydroxytriazenes, Journal of Environmental Chemistry and
Ecotoxicology 4 (2012) 133e136.
[3] D.B. Kimball, M.M. Haley, Triazenes: a versatile tool in organic synthesis,
Angewandte Chemie International Edition 41 (2002) 3338e3351.
acid (50%) mixture. The plates were left on the shaker until the
colour became homogeneous purple and the optical density was
measured at 530 nm (NR max) and 620 nM (reference wavelength)
with the Paradigm detection platform.

D. Cappoen et al. / European Journal of Medicinal Chemistry 77 (2014) 193e203
203
[4] T.A. Connors, P.M. Goddard, K. Merai, W.C.J. Ross, D.E.V. Wilman, Tumor
inhibitory triazenes structural requirements for an active metabolite,
Biochemical Pharmacology 25 (1976) 241e246.
[5] J.A. Hickman, Investigation of the mechanism of action of antitumour dime-
thyltriazenes, Biochimie 60 (1978) 997e1002.
[6] T. Fukushima, H. Takeshima, Anti-glioma therapy with temozolomide and
status of the DNA-repair gene MGMT, Anticancer Research 29 (2009) 4845e
4854.
[17] D. Forge, D. Cappoen, D. Stanicki, T. Huang, L. Verschaeve, K. Huygen,
J.J. Vanden Eynde, 1,4-Diarylpiperazines and analogs as anti-tubercular
agents: synthesis and biological evaluation, European Journal of Medicinal
Chemistry 49 (2012) 95e101.
[18] D. Cappoen, D. Forge, F. Vercammen, V. Mathys, M. Kiass, V. Roupie,
R. Anthonissen, L. Verschaeve, J.J. Vanden Eynde, K. Huygen, Biological eval-
uation of bisbenzaldehydes against four Mycobacterium species, European
Journal of Medicinal Chemistry 63 (2013) 731e738.
-
[7] F. Marchesi, M. Turriziani, G. Tortorelli, G. Avvisati, F. Torino, L. De Vecchis,
Triazene compounds: mechanism of action and related DNA repair systems,
Pharmacological Research 56 (2007) 275e287.
[19] P. Claes, D. Cappoen, B. Mbala, J. Jacobs, B. Mertens, V. Mathys, L. Verschaeve,
K. Huygen, N. De Kimpe, Synthesis and antimycobacterial activity of ana-
logues of the bioactive natural products sampangine and cleistopholine, Eu-
ropean Journal of Medicinal Chemistry 67 (2013) 98e110.
[20] G. Repetto, A. del Peso, J.L. Zurita, Neutral red uptake assay for the estimation
of cell viability/cytotoxicity, Nature Protocols 3 (2008) 1125e1131.
[21] J.J. Cortez-Escalante, A.M. Santos, G.C. Garnica, A.L. Sarmento, C.N. Castro,
G.A.S. Romero, Mediastinitis and pericardial effusion in a patient with AIDS
and disseminated Mycobacterium avium infection: a case report, Revista da
Sociedade Brasileira de Medicina Tropical 45 (2012) 407e409.
[22] R.J. Chiodini, W.M. Chamberlin, J. Sarosiek, R.W. McCallum, Crohn’s disease
and the mycobacterioses: a quarter century later. Causation or simple asso-
ciation? Critical Reviews in Microbiology 38 (2012) 52e93.
[23] T. Einarsdottir, K. Huygen, Buruli ulcer, Human Vaccines 7 (2012) 1198e1203.
[24] G.S. Hotter, D.M. Collins, Mycobacterium bovis lipids: virulence and vaccines,
Veterinary Microbiology 151 (2011) 91e98.
[25] L. Verschaeve, J. Van Gompel, L. Thilemans, L. Regniers, P. Vanparys, D. Van der
Lelie, VITOTOXÒ bacterial genotoxicity and toxicity test for the rapid
screening of chemicals, Environmental and Molecular Mutagenesis 33 (1999)
240e248.
[26] R.R. Tice, E. Agurell, D. Anderson, B. Burlinson, A. Hartmann, H. Kobayashi,
Y. Miyamae, E. Rojas, J.C. Ryu, Y.F. Sasaki, Single cell gel/comet assay: guide-
lines for in vitro and in vivo genetic toxicology testing, Environmental and
Molecular Mutagenesis 35 (2000) 206e211.
[8] D.S. Pilch, M.A. Kirolos, X. Liu, G.E. Plum, K.J. Breslauer, Berenil [1,3-Bis(40-
amidinophenyl)triazene] binding to DNA duplexes and to a RNA duplex: ev-
idence for both intercalative and minor groove binding properties,
Biochemistry 34 (1995) 9962e9976.
[9] E.D. Clercq, O. Dann, Diaryl amidine derivatives as oncornaviral DNA poly-
merase inhibitors, Journal of Medicinal Chemistry 23 (1980) 787e795.
ꢀ
ꢀ
[10] T. Cimbora-Zovko, A. Brozovic, I. Piantanida, G. Fritz, A. Virag, B. Alic, V. Majce,
ꢀ
M. Kocevar, S. Polanc, M. Osmak, Synthesis and biological evaluation of 4-
nitro-substituted 1,3-diaryltriazenes as a novel class of potent antitumor
agents, European Journal of Medicinal Chemistry 46 (2011) 2971e2983.
ꢀ
ꢀ
[11] B. Stefane, M. Kocevar, S. Polanc, Nitrosation with sodium hexani-
trocobaltate(III), The Journal of Organic Chemistry 62 (1997) 7165e7169.
[12] J.A. Sweatlock, T.A. Gasiewicz, The interaction of 1,3-diaryltriazenes with the
Ah receptor, Chemosphere 15 (1986) 1687e1690.
[13] D.T. Hill, K.G. Stanley, J.E. Karoglan Williams, B. Loev, P.J. Flower,
J.P. McCafferty, E. Macko, C.E. Berkoff, C.B. Ladd, 1,3-Diaryltriazenes: a new
class of anrectic agents, Journal of Medicinal Chemistry 26 (1983) 865e869.
[14] G. Vernin, C. Siv, J. Metzger, C. Párkányi, Synthesis of 1,3-diaryltriazenes and
their derivatives by aprotic decomposition of arylamines, Synthesis (1977)
691e693.
ꢀ
[15] A. Virag, A. Meden, M. Kocevar, S. Polanc, Synthesis and characterization of
new triazenide salts, The Journal of Organic Chemistry 71 (2006) 4014e4017.
[16] D. Cappoen, J. Jacobs, V.T. Nguyen, S. Claesens, G. Diels, R. Anthonissen,
M. Fauville, L. Verschaeve, K. Huygen, N. De Kimpe, Straightforward
palladium-mediated synthesis and biological evaluation of benzo[j]phenan-
thridine-7,12-diones as anti-tuberculosis agents, European Journal of Medic-
inal Chemistry 48 (2012) 57e68.
[27] M.J. Fenton, M.W. Vermeulen, Immunopathology of tuberculosis: roles of
macrophages and monocytes, Infection and Immunity 64 (1996) 683e690.
[28] B.A. Hanna, A. Ebrahimzadeh, L.B. Elliot, M.A. Morgan, S.M. Novak, S. Rusch-
Gerdes, M. Acio, D.F. Dunbar, T.M. Holmes, C.H. Rexen, C. Savthyakumar,
A.M. Vannier, Multicenter evaluation of the BACTEC MGIT 960 system for
recovery of mycobacteria, Journal of Clinical Microbiology 37 (1999) 748e752.