Development of a Focused Library of Triazole-Linked Privileged-Structure-Based Conjugates Leading to the Discovery of Novel Phenotypic Hits against Protozoan Parasitic Infections

Article abstract of DOI:10.1002/cmdc.201700786

Protozoan infections caused by Plasmodium, Leishmania, and Trypanosoma spp. contribute significantly to the burden of infectious diseases worldwide, causing severe morbidity and mortality. The inadequacy of available treatments calls for cost- and time-ef

Full text of DOI:10.1002/cmdc.201700786

DOI: 10.1002/cmdc.201700786
Communications
Development of a Focused Library of Triazole-Linked
Privileged-Structure-Based Conjugates Leading to the
Discovery of Novel Phenotypic Hits against Protozoan
Parasitic Infections
Elisa Uliassi,[a] Lorna Piazzi,[a] Federica Belluti,[a] Andrea Mazzanti,[b] Marcel Kaiser,[c, d]
Reto Brun,[c, d] Carolina B. Moraes,[e, h] Lucio H. Freitas-Junior,[e, h] Sheraz Gul,[f] Maria Kuzikov,[f]
Bernhard Ellinger,[f] Chiara Borsari,[g] Maria Paola Costi,[g] and Maria Laura Bolognesi*[a]
Protozoan infections caused by Plasmodium, Leishmania, and
Trypanosoma spp. contribute significantly to the burden of in-
fectious diseases worldwide, causing severe morbidity and
mortality. The inadequacy of available treatments calls for cost-
and time-effective drug discovery endeavors. To this end, we
envisaged the triazole linkage of privileged structures as an ef-
fective drug design strategy to generate a focused library of
high-quality compounds. The versatility of this approach was
combined with the feasibility of a phenotypic assay, integrated
with early ADME-tox profiling. Thus, an 18-membered library
was efficiently assembled via Huisgen cycloaddition of pheno-
thiazine, biphenyl, and phenylpiperazine scaffolds. The result-
ing 18 compounds were then tested against seven parasite
strains, and counter-screened for selectivity against two mam-
malian cell lines. In parallel, hERG and cytochrome P450 (CYP)
inhibition, and mitochondrial toxicity were assessed. Remarka-
bly, 10-((1-(3-([1,1’-biphenyl]-3-yloxy)propyl)-1H-1,2,3-triazol-5-
yl)methyl)-10H-phenothiazine (7) and 10-(3-(1-(3-([1,1’-biphen-
yl]-3-yloxy)propyl)-1H-1,2,3-triazol-4-yl)propyl)-10H-phenothia-
zine (12) showed respective IC50 values of 1.8 and 1.9 mgmLÀ1
against T. cruzi, together with optimal selectivity. In particular,
compound 7 showed a promising ADME-tox profile. Thus, hit
7 might be progressed as an antichagasic lead.
economic stability of societies.[2] Malaria, leishmaniasis, and try-
panosomiasis are three of the most important human protozo-
an diseases in terms of morbidity and mortality. For them, disa-
bility-adjusted life years, that is, the number of healthy life
years lost to disability or premature death, is estimated to be
in the millions.[3] While over 85% of deaths, disease, and disa-
bility occur in tropical and subtropical regions of the world,
more temperate regions, including North America and Mediter-
ranean Europe, are also impacted by protozoan diseases.[3]
Nowadays, climate change and migration have added layers of
complexity to the control of human protozoan infection. The
enduring lack of an effective vaccine leaves chemotherapy as
the only means to combat these diseases. However, most of
the currently available medicines, which are decades old and
suffer from low efficacy, high toxicity, and increasing resistance,
by no means meet the clinical need. Therefore, the develop-
ment of new pharmaceutical tools to treat protozoan diseases
is now, more than ever, a global research priority.[4]
As for any therapeutic area, the identification of antiproto-
zoal new chemical entities can be undertaken using either
phenotypic- or target-based approaches. Clearly, both ap-
proaches have advantages and disadvantages, which have
been discussed in detail elsewhere.[5]
Due to improvements in cultivation and assay techniques of
protozoa, phenotypic screening has become feasible in high-
throughput mode, allowing screens of large compound collec-
tions. It has the intrinsic advantage of identifying compounds
that are active in the whole-parasite context, thus cell-permea-
ble, devoid of efflux issues and more likely to possess drug-like
Infectious diseases caused by protozoan parasites pose a sig-
nificant threat to human health, being responsible for more
than a million deaths annually.[1] They also blight the lives of
millions of people worldwide, affecting both the health and
[a] Dr. E. Uliassi, Dr. L. Piazzi, Dr. F. Belluti, Prof. M. L. Bolognesi
Department of Pharmacy and Biotechnology, Alma Mater Studiorum—Uni-
versity of Bologna, Via Belmeloro 6, 40126 Bologna (Italy)
E-mail: marialaura.bolognesi@unibo.it
[f] Dr. S. Gul, M. Kuzikov, Dr. B. Ellinger
Fraunhofer Institute for Molecular Biology and Applied Ecology Screening-
Port, 22525 Hamburg (Germany)
[g] Dr. C. Borsari, Prof. M. P. Costi
[b] Prof. A. Mazzanti
Department of Life Sciences, University of Modena and Reggio Emilia, Via
Campi 103, 41125 Modena (Italy)
Department of Industrial Chemistry “Toso Montonari”, Alma Mater Studio-
rum—University of Bologna, Viale del Risorgimento 4, 40136 Bologna (Italy)
[h] Dr. C. B. Moraes, Dr. L. H. Freitas-Junior
[c] Dr. M. Kaiser, Prof. R. Brun
Present address: Instituto Butantan & Department of Microbiology, Institute
of Biomedical Sciences, University of S¼o Paulo, 05508-900 S¼o Paulo
(Brazil)
Swiss Tropical and Public Health Institute, 4002 Basel (Switzerland)
[d] Dr. M. Kaiser, Prof. R. Brun
University of Basel, Petersplatz 1, 4003 Basel (Switzerland)
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under:
https://doi.org/10.1002/cmdc.201700786.
[e] Dr. C. B. Moraes, Dr. L. H. Freitas-Junior
Laboratꢀrio Nacional de BiociÞncias (LNBio), Centro Nacional de Pesquisa
em Energia e Materiais (CNPEM), 13083-100 Campinas (Brazil)
ChemMedChem 2018, 13, 1 – 7
1
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
features. Mammalian cell cytotoxicity assays are often run in
parallel with parasite-based activity screens, allowing rapid
elimination of compounds endowed with nonspecific cytotox-
icity. However, unless one uses an appropriate genetically
modified parasite that expresses readily detectable reporters,
the identification of specific target(s) of active compounds is a
necessary and complicated further step.[6]
100% atom efficiency, and the high stability of the resulting
1,4- and 1,5-triazole products make the Huisgen cycloaddition
ideally suited for this purpose. Furthermore, the triazole-con-
necting unit is a substructure present in several drug classes.
On this basis, a focused library of compounds 4–21 was gener-
ated (Figure 1). To best of our knowledge, this is the first time
that a rational design based on privileged structures combined
with such a synthetic protocol has been proposed.
For all these reasons, there has been a switch in emphasis to
phenotypic approaches in antiprotozoal drug discovery re-
search,[7] although target-based screening remains a major in-
fluence. Notably, several successful phenotypic campaigns
have been reported in the areas of malaria, human African try-
panosomiasis (HAT), and leishmaniasis, leading to clinical can-
didates.[5] Particularly, drug repurposing through phenotypic
screening is a faster and more cost-effective method than
de novo phenotypic approaches,[3] Clearly, this is an added
value for the development of medicines against diseases en-
demic to low-income countries.
Next, a rapid in silico ADME-tox analysis was performed
using QikProp (Table S1, Supporting Information).[16] Triazoles
4–21 were compliant for most of the computed descriptors
(ꢁ2 violations), having properties that fell within the ADME-
tox range of 95% of known drugs. These data were supportive
of the designed library drug-likeness.
Schemes 1 and 2 illustrate the simple and cost-effective syn-
thesis of 4–21 by Huisgen reaction between the suitable
alkyne (22–25) and azide (26–28) derivatives.
Along these lines, we sought a strategy that would meet the
decreased cost and sustainability requirements of the field and
tailored to our academic settings. Toward this goal, the use of
validated chemical scaffolds emerged as a valuable option for
assembling a focused chemical library.[8] It is widely recognized
that chemical libraries enriched for certain “privileged” sub-
structures are more likely to generate hits than randomly se-
lected structures.[9] This is because such privileged substruc-
tures are structurally predisposed to binding efficiently to a
wide range of targets.[10] In addition, their intrinsic drug-like
properties offer an opportunity for providing leads with en-
hanced potential.[8a] Building on these considerations, for our
library design (Figure 1), we selected three privileged substruc-
Scheme 1. Synthesis of 4–13: a) 29 or 30, KOtBu, DMSO, RT; b) selected
azide (26–28), 648C.
Scheme 2. Synthesis of 14–21: a) 29 or 31, KHCO3, toluene, reflux; b) select-
ed azide (26–28), 648C.
Azides 26–28 were prepared by two consecutive nucleophil-
ic substitution reactions, such that 1 was reacted with 1-
bromo-3-chloropropane (32) in the presence of Et3N to afford
33, which was then quantitatively converted into the azide 26
(Scheme 3). On the other hand, Williamson reaction between
m-hydroxybiphenyl (34) or p-hydroxybiphenyl (35) and 32
Figure 1. Design of a focused library of 4–21.
tures: 2-methoxyphenylpiperazine (2-MPP, 1), biphenyl (BP, 2),
and phenothiazine (PTZ, 3).[11] Intriguingly, they have been
largely explored,[12] but quite overlooked in the parasitic field.
Although there are no reports of G protein coupled receptors
(GPCRs) in kinetoplastids, the presence of 1, a classical GPCR-
recognizing motif,[13] could enable probing the host–pathogen
interactions as an indirect mechanism of action.[14]
Part of our strategy was to appropriately conjugate the se-
lected privileged fragments (Mr <200 Da) covalently, giving rise
to hit-like compounds (Mr ꢀ500 Da).[15] The ease of synthesis,
Scheme 3. Synthesis of 26: a) 32, Et3N, toluene, RT; b) NaN3, DMF, 608C.
&
ChemMedChem 2018, 13, 1 – 7
www.chemmedchem.org
2
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
for Trypanosomatidic Infections (NMTrypI)[17] and the Swiss
Tropical and Public Health Institute (SwissTPHI),[18] seven para-
site strains and two mammalian cell lines were exploited. In
the case of HAT, the strains used were Lister 427 Trypanosoma
brucei brucei (T. b. brucei), a cattle parasite, and STIB 900
T. b. rhodesiense, a human strain. Concerning Chagas disease
(CD), the widely used Tulahuen clone and the Y strain of
T. cruzi were selected. The panel also included Leishmania in-
fantum (L. infantum) and L. donovani, two species responsible
for serious visceral disease reported worldwide, including Med-
iterranean countries. Additionally, we chose the Plasmodium
falciparum chloroquine- and pyrimethamine-resistant K1 (P. fal-
ciparum K1) strain, which is responsible for the most common
multidrug-resistant malarial infections. Furthermore, to assess
potential cytotoxic effects, studies were carried out on rat skel-
etal myoblast L6 cells and the human lung adenocarcinoma
epithelial A549 cell line.
Scheme 4. Synthesis of 27 and 28: a) 32, K2CO3, acetone, reflux; b) NaN3,
DMF, 608C.
gave 36 and 37, which were subsequently transformed into
the corresponding azides 27 and 28 (Scheme 4).
Alkynes 22 and 23 were respectively obtained by reacting 3
with 3-bromopropyne (29) and 5-chloro-1-pentyne (30) in the
presence of KOtBu (Scheme 1), whereas 24 and 25 by holding
29 and 31 at reflux with KHCO3 (Scheme 2). Next, cycloaddition
under solvent- and catalyst-free conditions afforded a mixture
of 1,4- and 1,5-regioisomers of disubstituted 1,2,3-triazoles (4–
21, 6–56% yields). The obtained regioisomers were easily puri-
1
fied, and their geometry was established by H NMR NOE spec-
troscopy experiments. In the case of the 1,4-disubstituted re-
gioisomer, saturation of the CH2 protons at position 1 yielded
NOE on the H-5 signal of triazole, and saturation of the second
CH2 at position 4 yielded the same result, thus proving the
spatial interaction between the above-mentioned protons. In
the case of the 1,5-isomer, irradiation of the CH2 protons at po-
sition 1 of the triazole gave an enhancement of the CH2 pro-
tons at position 5 (Figures S1–S2, Supporting Information).
To investigate the therapeutic potential of 4–21, we set up a
Table 1 lists the antiprotozoal and cytotoxicity profiles of 4–
21, together with the corresponding selectivity indices (SI).
These results were analyzed according to the guidelines pro-
posed by the Special Programme for Tropical Disease Research
(TDR)[19] for early screening aimed to identify antiprotozoal hits.
With regard to T. brucei, only compounds 8 (IC50 T. b. rhode-
siense: 2.3 mgmLÀ1, inhibition at 50 mm T. b. brucei: 99.9%) and
19 (IC50 T. b. rhodesiense: 4.2 mgmLÀ1, inhibition at 50 mm
T. b. brucei: 95.2%) were found to inhibit both human and
cattle strains. Nevertheless, 8 exhibited no or very little selec-
tivity against mammalian cells (SI: 0.7 and 45% of A549 cell
screening pipeline against
a panel of laboratory strains
(Table 1). Thanks to a collaboration between New Medicines
Table 1. Antiprotozoal activities and cytotoxicities of compounds 4–21 in comparison with reference drugs.
Compound
T. b.
rhodesiense
T. cruzi
(Tulahuen)
IC50
L. donovani
P. falciparum
L6
Inhibition at 50 mm [%]
A549 cell
growth at
K1
IC50
SI[a]
SI[a]
IC50
SI[a]
IC50
SI[a]
IC50
L. infantum[b] T. b. brucei[c] T. cruzi (Y)[d] 10 mm [%]
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
39.2
2.3 42.7
2.1
19.5
19.3
48.6
0.9
1.0 14.7
10.4
6.0
51.4
1.9
4.4
3.0
6.0
3.3
9.2
20.2 2.0
44.1 89.5
41.9 90.0
33.8 90.0
37.3 90.0
4.9
46.4
11.3
63.5
91.4
90.8
3.8
10.3
56.0
10.5
3.9
22.3
5.7
14.1
20.5
35.6
6.8
48.0
94.1
39.6
65.5
99.9
99.8
23.2
7.6
41.8
24.4
15.2
20.4
23.1
70.8
2.1
12.4
26.8
32.9
25.0
79.8
20.2
20.9
21.8
40.2
10.2
35.2
18.2
33.4
49.6
À4.8
À31.9
À59.5
17.2
–
114.1
95.0
124.8
103.1
45.3
55.4
125.2
149.9
122.0
79.7
105.5
126.1
126.3
113.2
87.5
100.6
99.2
121.7
–
18.6
26.9
12.8
2.3
4.8
3.3
7.0
0.7
0.6
1.2
4.6
4.7
1.9
1.7
5.0
6.0
30.4 2.2
15.1 2.7
26.9 2.4
0.2 0.2
0.3 0.9
9.3 1.1
30.5 2.1
15.7 0.7
23.9 2.4
1.7 3.3
12.1 1.8
1.6 3.1
7.9 1.8
4.0 1.8
6.3 1.6
0.5 1.6
0.5 0.3
7.2
5.3
1.6
5.0
7.7
51.7
23.5
25.6
23.0
11.7
17.7
6.6
10.0
4.9
4.2
9.6
4.0
0.004
–
–
–
–
6.8
3.0
5.7
3.8
56.7 62.9
42.1 90.0
136.2 90.0
37.7 90.0
9.2 30.5
43.9 78.6
4.3 13.0
10.7 19.3
41.9 76.8
46.6 75.1
7.5 12.0
3.8 15.1
3.5 1.8
3.9 46.7
2.6 13.5
2.3 17.9
19.0
1.7
13.5
4.4
1.9
1.9
4.1
7.8
1.4
6.5
8.3
2.5
15.7 34.5
2.2 19.2
18.0
1.3
1.7
–
6.7
7.0
3.9
–
11.2
11.9
1.7 24.7
1.7 13.9
95.2
78.0
39.5
–
–
–
20
21
22.6
6.9
68.5
–
–
–
–
melarsoprol
benznidazole
miltefosine
chloroquine
podophyllotoxin
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
0.412
–
–
–
–
–
–
–
–
–
0.155
–
–
–
–
0.065
–
–
–
–
–
0.004
–
The experimental errors of IC50 values [mgmLÀ1] are within Æ50%; SD values of percent inhibition values agreed to Æ10%. [a] Selectivity index, calculated
as (IC50 for L6)/(IC50 for the respective parasite). [b] Amphotericin B (IC50 =1.4 mgmLÀ1) was used as positive control for L. infantum. [c] Pentamidine (IC50
3.1 mgmLÀ1) was used as positive control for T. b. brucei. [d] Benznidazole (IC50 =11.4 mgmLÀ1) was used as positive control for T. cruzi Y strain.
=
ChemMedChem 2018, 13, 1 – 7
www.chemmedchem.org
3
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
Table 2. In vitro early ADME-tox profiles of 4–21.[a]
growth at 10 mm). Conversely, 19 was toxic only against L6
cells (SI: 18) and nontoxic against A549 cells (100% of cell
growth). In addition, derivatives 18 and 21 yielded IC50 values
of 4.9 and 4.0 mgmLÀ1 against T. b. rhodesiense, despite modest
(SI: 15.7) and poor (SI: 1.7) selectivity profiles against L6. Both
compounds showed no cytotoxic effects on A549 cells at
10 mm. Notably, 8, 18, 19, and 21, the only active derivatives of
the library, feature a 2-MMP (1) substructure. However, none of
them fulfilled the hit criteria for T. b. rhodesiense (IC50 <
0.2 mgmLÀ1, and SI>100).[19]
An IC50 value of 1–2 mgmLÀ1 and SI>20 were identified as
requirements for L. donovani or L. infantum.[19] However, no hits
were associated with such a profile. In addition to the very
potent but highly cytotoxic compound 8 (see above), only 7,
bearing m-BP and PTZ substructures, resulted as a moderate
inhibitor of both Leishmania forms (IC50 L. donovani:
3.3 mgmLÀ1, 63.5% inhibition of L. infantum), with a good
safety profile (SI: 26.9 and 100% A549 cell growth at 10 mm).
Concerning T. cruzi, a broader range of activities against Tu-
lahuen clone (1.7–46.7 mgmLÀ1) was found, with clear struc-
ture–activity relationship patterns identified. As a general
trend, with the exception of 9, 11, and 13, we observed that
1,5-regioisomers (5, 7, 15, 17, 19, and 21) are more active than
the corresponding 1,4-isomers (4, 6, 14, 16, 18, and 20). This
reinforces the initial idea that the synthesis of both regioisom-
ers would have expanded the chemical space explored by
these compounds. Of note, several azoles have entered clinical
trials for the treatment of CD.[20] Compounds 7 and 12, which
share the m-BP and PTZ substructures while differing in the
connecting methylene units, fully fulfilled the anti-T. cruzi hit
requirements (IC50 <2.0 mgmLÀ1, SI>50).[21] In fact, they show
IC50 values for T. cruzi (Tulahuen) of 1.8 and 1.9 mgmLÀ1, along
with promising SI values of 49 and 51, respectively. However,
in the case of the T. cruzi Y clone, only the cytotoxic derivative
8 exhibited ~80% inhibition at 50 mm, whereas all the others
displayed percentages of inhibition below 49%. Correspond-
ingly, the reference compound benznidazole displayed signifi-
cantly different IC50 values against the two strains (Table 1).
These discrepancies may be due to the use of different T. cruzi
forms, but most probably because different assay setups were
used (amastigotes in L6 cells and trypomastigotes co-seeded
with human bone osteosarcoma epithelial cells, also covering
intracellular amastigotes, respectively).[22]
Regarding P. falciparum K1, none of compounds 4–21 fully
met the antimalarial hit criteria (IC50 <0.2 mgmLÀ1, SI>100).
However, 12, which carries again BP and PTZ substructures,
displayed potent antimalarial activity (IC50: 0.7 mgmLÀ1) and no
sign of toxicity against mammalian cells (SI: 136 and 100%
A549 cell growth at 10 mm).
Collectively, this early screening fostered the identification of
7 and 12 as promising trypanocidal hits against T. cruzi (Tula-
huen), according to the TDR criteria.
In parallel, prioritization of 4–21 was not based solely on
their antiprotozoal activities and selectivities, but focus was
placed on investigating their in vitro early toxicity profiles[19] in
terms of hERG and cytochrome P450 (CYP) inhibition, and mi-
tochondrial toxicity (786-O, human renal carcinoma cell line)
(Table 2). Compounds 4–21 were screened in each assay at
&
ChemMedChem 2018, 13, 1 – 7
www.chemmedchem.org
4
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
10 mm, and a traffic light system was used, as previously re-
ported.[23] Because azoles are well known for their CYP inhibito-
ry activity, and significant inhibition of CYPs is a major cause of
drug–drug interactions, especially in co-infected patients,[24]
this emerged as a critical issue.[25] Indeed, among the five
tested isoforms (CYP1A2, CYP2C9, CYP2C19, CYP2D6, and
CYP3A4), CYP3A4 and CYP2C19 were inhibited by 4–21 to vari-
ous extents. Notably, the anti-T. cruzi hit 7 was amongst the
compounds associated with the least overall inhibition of the
CYPs, while 12 was consistently more toxic. Perhaps the de-
creased flexibility of 7 with respect to 12, together with the
different triazole geometry is the reason for this behavior. With
regards to hERG inhibition, as this can induce cardiac toxicity,
which must be avoided in pre-disposed CD patients,[26] un-
fortunately all the 2-MPP derivatives, with the exception of 15
and 17, were associated with significant hERG liability, whereas
BP- and PTZ-based compounds (4–7 and 10–13) have modest
hERG liability. This is in agreement with QikProp prediction
(Table S1). Intriguingly, when substructure 1 is linked to the tri-
azole ring by one methylene unit, hERG inhibition is not ob-
served in the case of 1,5-regioisomers 15 and 17 with respect
to the corresponding 1,4-isomers 14 and 16.
Acknowledgements
This work was supported by MIUR (PRIN Funds 201274BNKN_
003) and the University of Bologna (RFO 2016). Some antiparasit-
ic activity and the early ADME-tox profiling were developed
within the international collaborative effort of the European
Union’s Seventh Framework Programme for research, technologi-
cal development and demonstration under grant agreement no.
603240 (NMTrypI—New Medicines for Trypanosomatidic Infec-
tions, http://fp7-nmtrypi.eu/).
Conflict of interest
The authors declare no conflict of interest.
Keywords: phenotypic screening
protozoan parasitic infections
· privileged scaffolds ·
[1] J. R. Herricks, P. J. Hotez, V. Wanga, L. E. Coffeng, J. A. Haagsma, M. G.
Basanez, G. Buckle, C. M. Budke, H. Carabin, E. M. Fevre, T. Furst, Y. A.
Halasa, C. H. King, M. E. Murdoch, K. D. Ramaiah, D. S. Shepard, W. A.
Stolk, E. A. Undurraga, J. D. Stanaway, M. Naghavi, C. J. L. Murray, PLoS
Neglected Trop. Dis. 2017, 11, e0005424.
[2] P. J. Hotez, B. Pecoul, S. Rijal, C. Boehme, S. Aksoy, M. Malecela, R. Tapia-
Conyer, J. C. Reeder, PLoS Neglected Trop. Dis. 2016, 10, e0003895.
[3] K. T. Andrews, G. Fisher, T. S. Skinner-Adams, Int. J. Parasitol. Drugs Drug
Resist. 2014, 4, 95–111.
Mitochondria are increasingly implicated in the etiology of
drug-induced toxicities.[27] Accordingly, we tested whether 4–
21 at 10 mm could affect mitochondrial viability. Remarkably, all
compounds displayed negligible mitochondrial toxicity.
Overall, the in vitro evaluation of early toxicity profile re-
vealed some liabilities of 4–21, which need to be addressed in
further hit-to-lead optimization stages.
[4] L. Monzote, A. Siddiq, Open Med. Chem. J. 2011, 5, 1–3.
[5] I. H. Gilbert, J. Med. Chem. 2013, 56, 7719–7726.
[6] M. L. Sykes, V. M. Avery, J. Med. Chem. 2013, 56, 7727–7740.
[7] A. Woodland, S. Thompson, L. A. Cleghorn, N. Norcross, M. De Rycker, R.
Grimaldi, I. Hallyburton, B. Rao, S. Norval, L. Stojanovski, R. Brun, M.
Kaiser, J. A. Frearson, D. W. Gray, P. G. Wyatt, K. D. Read, I. H. Gilbert,
ChemMedChem 2015, 10, 1809–1820.
[8] a) A. Cavalli, F. Lizzi, S. Bongarzone, F. Belluti, L. Piazzi, M. L. Bolognesi,
FEMS Immunol. Med. Microbiol. 2010, 58, 51–60; b) S. Bongarzone, M. L.
Bolognesi, Expert Opin. Drug Discovery 2011, 6, 251–268.
[9] K. L. Spear, S. P. Brown, Drug Discovery Today Technol. 2017, 23, 61–67.
[10] P. M. Watt, Future Med. Chem. 2009, 1, 257–265.
[11] E. Uliassi, L. Piazzi, F. Belluti, M. Kaiser, R. Brun, S. Gul, B. Ellinger, C. B.
Moraes, L. H. Freitas-Junior, C. Borsari, M. P. Costi, M. L. Bolognesi, Pro-
ceedings 2017, 1, 648.
[12] a) D. A. Horton, G. T. Bourne, M. L. Smythe, Chem. Rev. 2003, 103, 893–
930; b) A. Martinez, C. Gil in Privileged Scaffolds in Medicinal Chemistry:
Design, Synthesis Evaluation, The Royal Society of Chemistry, London,
2016, pp. 231–261.
[13] T. A. Esbenshade in G Protein-Coupled Receptors in Drug Discovery, Vol. 4
(Eds.: K. H. Lundstrom, M. L. Chiu), CRC Press, Boca Raton, 2005,
pp. 15–36.
[14] I. Pena, M. P. Manzano, J. Cantizani, A. Kessler, J. Alonso-Padilla, A. I. Bar-
dera, E. Alvarez, G. Colmenarejo, I. Cotillo, I. Roquero, F. de Dios-Anton,
V. Barroso, A. Rodriguez, D. W. Gray, M. Navarro, V. Kumar, A. Sherstnev,
D. H. Drewry, J. R. Brown, J. M. Fiandor, J. J. Martin, Sci. Rep. 2015, 5,
8771.
[15] C. A. Lipinski, Drug Discovery Today Technol. 2004, 1, 337–341.
[16] Schrçdinger Release 2017-4: QikProp, Schrçdinger, LLC, New York, NY
(USA), 2017.
[17] NMTrypI—New Medicines for Trypanosomatidic Infections: http://fp7-
nmtrypi.eu (accessed November 30, 2017).
In summary, we established a panel of parasite/mammalian
cell-based and early ADME-tox assays and screened a small fo-
cused library (compounds 4–21) that was rapidly and efficient-
ly assembled by combining three privileged substructures
through a catalyst- and solvent-free Huisgen cycloaddition.
When considering the TDR criteria,[19] we identified 7 (1,5-re-
gioisomer) and 12 (1,4-regioisomer) as novel hits against
T. cruzi that yield potent parasite growth inhibition, whilst
being nontoxic to mammalian cells. Intriguingly, both 7 and 12
feature the m-BP and the PTZ substructures conjugated to the
triazole through different methylene linkers. Although the in vi-
tro early ADME-tox assessment highlighted potential metabolic
liabilities for 12 (from moderate to significant inhibition against
hERG and all CYPs), positively, 7 inhibited only CYP3A4 and
CYP2C19, with no hERG and mitochondrial liability. Taken to-
gether, the new PTZ-triazole-m-BP chemotype represents an
attractive starting point for further medicinal chemistry efforts
aimed at developing new and sustainable antichagasic leads
with an improved ADME-tox profile.
Importantly, the approach described herein might be a
useful tool for antiprotozoal drug discovery: the multiple bio-
logical properties and the drug-likeness of privileged substruc-
tures together with the efficient triazole-based conjugation
strategy foreshadow a huge variety of interesting conjugates
yet to be explored.
[18] SwissTPHI—Swiss Tropical & Public Health Institute, Parasite Chemo-
therapy Unit: https://www.swisstph.ch/en/ (accessed November 30,
2017).
[19] S. Nwaka, B. Ramirez, R. Brun, L. Maes, F. Douglas, R. Ridley, PLoS Ne-
glected Trop. Dis. 2009, 3, e440.
ChemMedChem 2018, 13, 1 – 7
www.chemmedchem.org
5
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
[20] L. Gaspar, C. B. Moraes, L. H. Freitas, S. Ferrari, L. Costantino, M. P. Costi,
R. P. Coron, T. K. Smith, J. L. Siqueira-Neto, J. H. McKerrow, A. Cordeiro-
da-Silva, Curr. Med. Chem. 2015, 22, 4293–4312.
[24] a) E. Vaumourin, G. Vourc’h, P. Gasqui, M. Vayssier-Taussat, Parasites Vec-
tors 2015, 8, 545; b) J. A. Lindoso, M. A. Cunha, I. T. Queiroz, C. H. Mor-
eira, HIV/AIDS 2016, 8, 147–156.
[21] World Health Organization, Chagas Disease: in vitro Screening Model,
[25] K. Seden, S. Khoo, D. Back, N. Prevatt, M. Lamorde, P. Byakika-Kibwika, J.
Mayito, M. Ryan, C. Merry, AIDS 2013, 27, 675–686.
[26] F. S. Buckner, Adv. Parasitol. 2011, 75, 89–119.
http://www.who.int/tdr/grants/workplans/en/chagas_invitro.pdf
cessed November 30, 2017).
(ac-
[22] C. B. Moraes, M. A. Giardini, H. Kim, C. H. Franco, A. M. Araujo, S. Schenk-
man, E. Chatelain, L. H. Freitas, Sci. Rep. 2014, 4, 4703.
[27] J. A. Dykens, Y. Will, Drug Discovery Today 2007, 12, 777–785.
[23] E. Uliassi, G. Fiorani, R. L. Krauth-Siegel, C. Bergamini, R. Fato, G. Bianchi-
ni, J. C. Menendez, M. T. Molina, E. Lopez-Montero, F. Falchi, A. Cavalli, S.
Gul, M. Kuzikov, B. Ellinger, G. Witt, C. B. Moraes, L. H. Freitas-Junior, C.
Borsari, M. P. Costi, M. L. Bolognesi, Eur. J. Med. Chem. 2017, 141, 138–
148.
Manuscript received: December 15, 2017
Accepted manuscript online: && &&, 0000
Version of record online: && &&, 0000
&
ChemMedChem 2018, 13, 1 – 7
www.chemmedchem.org
6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

COMMUNICATIONS
E. Uliassi, L. Piazzi, F. Belluti, A. Mazzanti,
M. Kaiser, R. Brun, C. B. Moraes,
L. H. Freitas-Junior, S. Gul, M. Kuzikov,
B. Ellinger, C. Borsari, M. P. Costi,
M. L. Bolognesi*
&& – &&
An 18-membered library, rapidly and
efficiently assembled by combining
three privileged structures by catalyst-
and solvent-free Huisgen cycloaddition,
was screened in parasite/mammalian
cell-based and early ADME-tox assays.
Notably, compound 7 emerged as an
antichagasic phenotypic hit, with a
good ADME-tox profile. The simple and
versatile triazole-based conjugation
strategy of privileged scaffolds might
produce high-quality antiparasitic conju-
gates.
Development of a Focused Library of
Triazole-Linked Privileged-Structure-
Based Conjugates Leading to the
Discovery of Novel Phenotypic Hits
against Protozoan Parasitic Infections
ChemMedChem 2018, 13, 1 – 7
www.chemmedchem.org
7
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ