Structure-guided approach to identify a novel class of anti-leishmaniasis diaryl sulfide compounds targeting the trypanothione metabolism

Article abstract of DOI:10.1007/s00726-019-02731-4

Leishmania protozoans are the causative agent of leishmaniasis, a neglected tropical disease consisting of three major clinical forms: visceral leishmaniasis (VL), cutaneous leishmaniasis, and mucocutaneous leishmaniasis. VL is caused by Leishmania donova

Full text of DOI:10.1007/s00726-019-02731-4

Amino Acids
https://doi.org/10.1007/s00726-019-02731-4
ORIGINAL ARTICLE
Structure‑guided approach to identify a novel class
of anti‑leishmaniasis diaryl sulfde compounds targeting
the trypanothione metabolism
Gianni Colotti1 · Francesco Saccoliti3 · Marina Gramiccia4 · Trentina Di Muccio4 · Jay Prakash5 · Sunita Yadav6 ·
Vikash Kumar Dubey6 · Giulio Vistoli7 · Theo Battista2 · Stefano Mocci2 · Annarita Fiorillo2 · Aasia Bibi2 ·
Valentina Noemi Madia3 · Antonella Messore3 · Roberta Costi3 · Roberto Di Santo3 · Andrea Ilari1
Received: 18 January 2019 / Accepted: 3 April 2019
© Springer-Verlag GmbH Austria, part of Springer Nature 2019
Abstract
Leishmania protozoans are the causative agent of leishmaniasis, a neglected tropical disease consisting of three major clinical
forms: visceral leishmaniasis (VL), cutaneous leishmaniasis, and mucocutaneous leishmaniasis. VL is caused by Leishma-
nia donovani in East Africa and the Indian subcontinent and by Leishmania infantum in Europe, North Africa, and Latin
America, and causes an estimated 60,000 deaths per year. Trypanothione reductase (TR) is considered to be one of the best
targets to fnd new drugs against leishmaniasis. This enzyme is fundamental for parasite survival in the human host since
it reduces trypanothione, a molecule used by the tryparedoxin/tryparedoxin peroxidase system of Leishmania to neutralize
the hydrogen peroxide produced by host macrophages during infection. Recently, we solved the X-ray structure of TR in
complex with the diaryl sulfde compound RDS 777 (6-(sec-butoxy)-2-((3-chlorophenyl)thio)pyrimidin-4-amine), which
impairs the parasite defense against the reactive oxygen species by inhibiting TR with high efciency. The compound binds
to the catalytic site and engages in hydrogen bonds the residues more involved in the catalysis, namely Glu466′, Cys57 and
Cys52, thereby inhibiting the trypanothione binding. On the basis of the RDS 777–TR complex, we synthesized structur-
ally related diaryl sulfde analogs as TR inhibitors able to compete for trypanothione binding to the enzyme and to kill the
promastigote in the micromolar range. One of the most active among these compounds (RDS 562) was able to reduce the
trypanothione concentration in cell of about 33% via TR inhibition. RDS 562 inhibits selectively Leishmania TR, while it
does not inhibit the human homolog glutathione reductase.
Keywords Trypanothione metabolism · Trypanothione reductase · Structure-based drug design · Diaryl sulfde inhibitors
Abbreviations
FAD
GSSG
hGR
Flavin dinucleotide
Oxidized glutathione
Human glutathione reductase
Handling Editor: E. Agostinelli.
3
* Andrea Ilari
Dip. Chimica e Tecnologie del Farmaco, Istituto
Pasteur-Fondazione Cenci Bolognetti, Università Sapienza,
P.le Aldo Moro 5, 00185 Rome, Italy
andrea.ilari@uniroma1.it
Gianni Colotti
4
5
6
gianni.colotti@uniroma1.it
Dipartimento di Malattie Infettive, Istituto Superiore di
Sanità Viale Regina Elena, 299, 00161 Rome, Italy
Roberto Di Santo
roberto.disanto@uniroma1.it
Department of Biosciences and Bioengineering, Indian
Institute of Technology Guwahati, Assam 981039, India
1
Istituto di Biologia e Patologia Molecolari, Consiglio
School of Biochemical Engineering, Indian Institute
of Technology (Banaras Hindu University) Varanasi,
Varanasi 221005, India
Nazionale delle Ricerche, IBPM-CNR, c/o Dip. Scienze
Biochimiche Università Sapienza, P.le Aldo Moro 5,
00185 Rome, Italy
7
Dip. di Scienze Farmaceutiche, Università degli Studi di
Milano, via Mangiagalli 25, 20133 Milan, Italy
2
Dip. Scienze Biochimiche, Università Sapienza, P.le Aldo
Moro 5, 00185 Rome, Italy
Vol.:(0123456789)
1 3

G. Colotti et al.
IR
Infrared
NADPH Reduced nicotinamide adenine dinucleotide
TR
Trypanothione reductase
Trypanothione synthetase
Trypanothione [N1,N8-bis(glutathionyl)
spermidine]
TryS
TS2
T(SH)2
TXN
Reduced trypanothione
Tryparedoxin
TXNPx Tryparedoxin peroxidase I
VL
Visceral leishmaniasis
Introduction
Leishmania parasites are the causative agent of leishma-
niasis, a neglected tropical disease infecting numerous
mammals including humans throughout the world. Human
leishmaniasis consists of three major clinical forms: visceral
leishmaniasis (VL), which is fatal if left untreated, cutane-
ous leishmaniasis, which can heal spontaneously but leaves
disfguring scars and mucocutaneous leishmaniasis, which
is not self-healing and can potentially be fatal. VL, which
is caused by Leishmania donovani in East Africa and the
Indian subcontinent and by Leishmania infantum in Europe,
North Africa, and Latin America, causes an estimated
60,000 deaths annually (Desjeux 2004; Hotez et al. 2012).
Leishmania spp., together with other protozoan pathogens
responsible for diseases such as sleeping sickness (Trypa-
nosoma brucei) and Chagas disease (Trypanosoma cruzi),
belong to the Trypanosomatidae (trypanosomatids) family.
The therapy against the diseases caused by Trypanosomati-
dae, and in particular against VL, mainly relies on treatments
implying the use of drugs such as antimony, amphotericin,
paromomycin, miltefosine, aminoquinoline or sitamaquine,
unsatisfactory in terms of their safety and efcacy (Burza
et al. 2018). For these reasons there is an urgent need to fnd
new drugs against these diseases, which afect mainly people
living in tropical and subtropical countries that have inad-
equate research capacities and public health infrastructures.
As recently suggested by Zulfqar and colleagues, advances
in this feld can be achieved by targeting proteins essential
for parasite survival but that are absent in the human host
(Zulfqar et al. 2017).
Fig. 1 Trypanothione metabolism. In the fgure, the following abbre-
viations are used: ARG arginase, ODC ornithine decarboxylase,
AdoMetDC S-adenosylmethionine decarboxylase, SpdS spermidine
synthase, TR trypanothione reductase, TryS trypanothione synthetase,
TXN tryparedoxin, TXNPx tryparedoxin peroxidase I, TS2 trypan-
othione [N1,N8-bis(glutathionyl)spermidine], T(SH)2 reduced trypan-
othione. The structure formulas of spermidine, glutathione and tryp-
anothione are displayed
been shown that mutants displaying a partial trisomy of TR
locus, where two TR alleles are disrupted by gene target-
ing, show attenuated infectivity and decreased ability to
survive within the macrophages (Dumas et al. 1997). These
data, together with the fact that all the attempts to date to
obtain TR null mutants have failed, demonstrate that TR is
an essential enzyme for the parasite survival. For this rea-
son, this protein is an attractive target for the development
of potential new drugs against trypanosomatids since it is
also absent in the human host (Colotti et al. 2013a; Colotti
and Ilari 2011). Indeed the human homolog, glutathione
reductase (hGR), that maintains glutathione in a reduced
state, displays a signifcantly diferent substrate binding site
in terms of both volume and charge distribution, suggesting
the possibility of identifying molecules able to selectively
target TR (Colotti et al. 2013a).
In contrast to the mammalian redox defense machinery
that is based on glutathione, trypanosomatid parasites pos-
sess trypanothione [N1,N8-bis(glutathionyl)spermidine]
(TS2) as the main detoxifying agent against oxidative dam-
age. This dithiol is synthesized by trypanothione synthetase
(TryS) and is reduced to T(SH)2 by the trypanothione reduc-
tase (TR). T(SH)2 is then used by the couple tryparedoxin/
tryparedoxin peroxidase I (TXN/TXNPx) to neutralize
hydrogen peroxide produced by macrophages during infec-
tion (Fig. 1) (Colotti et al. 2013a; Ilari et al. 2017). It has
The crystal structure of L. infantum TR (Baiocco et al.
2009a) that has been solved by Baiocco et al. is very simi-
lar to the structure of other trypanosomatid TRs solved
to date, such as those from Crithidia fasciculata, T. cruzi
and T. brucei (Bailey et al. 1993; Patterson et al. 2011;
1 3

Structure‑guided approach to identify a novel class of anti‑leishmaniasis diaryl sulfde…
Saravanamuthu et al. 2004). It is a twofold symmetrical
homodimer in which each subunit is formed by three
domains, i.e., the interface domain (residues 361–488), the
NADPH-binding domain (residues 161–288) and the FAD
binding domain (residues 1–160 and 289–360). The tryp-
anothione binding site resides in a large cavity at the inter-
face between the two monomers, formed by the residues
of the FAD binding domain of one subunit and those of
the interface domain of the other. The mechanism of tryp-
anothione reduction catalyzed by TR involves NADPH,
which transfers two electrons via FAD to the Cys52–Cys57
disulfde bridge. TS2 then binds to the protein and Cys52,
deprotonated by the couple His461′–Glu466′, nucleophili-
cally attacks the trypanothione disulfde bridge with the
formation of a mixed disulfde. Finally, the attack of Cys57
on Cys52 promotes the release of the reduced substrate
(Baiocco et al. 2009a).
NADPH-binding site, not present in the human homolog
glutathione reductase (hGR) (Turcano et al. 2018).
The screening of our in house library of diaryl sulfdes
allowed us to identify a compound (RDS 777) able to
inhibit TR in sub-micromolar range (Ki =0.25 0.18 µM).
The X-ray structure of LiTR-777 complex showed that the
inhibitor is bound to both the trypanothione binding site
and the NADPH-binding site. In particular, the monomer
A binds three RDS 777 molecules, whereas the monomer B
binds four RDS 777 molecules. RDS 777 b and c are bound
in both monomers to the NADPH-binding site, whereas the
RDS 777 a in the monomer A and a and d in monomer B
are bound to the trypanothione-binding site. RDS 777 in
position a interacts with the hydrophobic Val53 and Val58
and, furthermore, it is hydrogen bound to the residues more
involved in the catalysis, namely His 461′, Glu466′, Cys57,
Cys52 and Glu467′ (Saccoliti et al. 2017).
Antimony-containing drugs, still among the most used
against VL, and other metals and metal-containing com-
pounds can inhibit TR by binding to the two catalytic
cysteines of the active site (Baiocco et al. 2011; Colotti et al.
2013b, 2018; Ilari et al. 2012, 2015). In addition to metal
compounds, several organic compounds have been shown to
inhibit TR activity by binding to the trypanothione binding
site: 2-iminobenzimidazoles (Holloway et al. 2007), polyam-
ine analogs (Dixon et al. 2005), quinoline-based compounds
(Spinks et al. 2009), pyrimidopyridazine-based scafolds
(Spinks et al. 2009), azole-based compounds (Baiocco et al.
2013), lunarine analogs (Hamilton et al. 2006) and diaryl
sulfdes (Saccoliti et al. 2017; Stump et al. 2008) have all
been reported. Some of these compounds were identifed
using high-throughput screenings, employing the colorimet-
ric method for detection of TR activity (Martyn et al. 2007;
Spinks et al. 2009) developed by Hamilton et al. (2003).
Recently our group identifed among the GlaxoSmith-
Kline LeishBOX, containing 192 molecules selected by
high-throughput whole-cell phenotypic screening from a
collection of 1.8 millions of compounds, carboxamide mol-
ecules able to selectively inhibit TR with high potency, not
active on the human homolog glutathione reductase, and
able to kill Leishmania parasites. Among them, the com-
pound A1/7 (N-{4-methoxy-3-[(4-methoxyphenyl)sul-
famoyl]phenyl}-5-nitrothiophene-2-carboxamide) is the
only compound present also in the ChagasBOX and in the
HATBOX, and for this reason, represents a lead compound
to fnd new drugs against all the trypanosomatid diseases
(Ilari et al. 2018). We were also able to develop a new
luminescence-based assay to screen a library of 120,000
compounds, which allowed us to fnd new scafolds able to
inhibit TR. Among these molecules, compound 3 (2-(dieth-
ylamino)ethyl 4-((3-(4-nitrophenyl)-3-oxopropyl)amino)
benzoate) was shown to inhibit TR in the micromolar range
by binding to a unique cavity placed at the entrance of the
This binding mode of RDS 777 differs from that of
the other diaryl sulfde derivatives reported in literature
(Stump et al. 2008). Indeed, the diaryl sulfde derivative
(3-({5-chloro-2-[(40-methylbiphenyl-4-yl)thio]phenyl}
amino)-N-(3,4-dichlorobenzyl)-N,N-dimethylpropan-
1-ammonium chloride)) modeled by Stump et al. accom-
modates in the trypanothione-binding cavity in the so-called
“mepacrine”-binding site distant from the two catalytic
cysteines. In the monomer B of LiTR-777 X-ray structure, an
additional molecule of RDS 777 binds to the trypanothione-
binding site (molecule d) close to the entrance of the tryp-
anothione-binding site, establishes a stacking interaction
with the pyrimidineamine moiety of the molecule a and is
hydrogen bound to Glu467′.
In this paper, we report a structure-based anti-leishmania-
sis drug design study: we have synthesized a series of diaryl
compounds on the basis of the structure of RDS 777-TR
complex and in particular of the sites a and d. Our studies
allow the identifcation of a compound (RDS 562) able to
inhibit TR, to kill the promastigote in the micromolar range
and to reduce the intracellular concentration of reduced
trypanothione. The compound specifcally inhibits Leish-
mania TR, while it does not inhibit hGR. The inhibition
assays demonstrate that RDS 562 inhibits TR by competing
with trypanothione binding, and the docking experiments
show the molecular basis of TR inhibition by the organic
compound.
Materials and methods
Materials
Trypanothione disulfde (Bachem), NADPH (Panreac-Appli-
chem), human glutathione reductase (Sigma-Aldrich) and
oxidized glutathione (GSSG) (Sigma-Aldrich) were used for
1 3

G. Colotti et al.
the experiments. The L. donovani (BHU-1081) was obtained
from Prof. Shyam Sundar, Banaras Hindu University, India,
and cultivated M199 liquid media supplemented with 15%
heat inactivated fetal bovine serum (FBS), 100 U/mL peni-
cillin and 100 µg/ml streptomycin.
in isopropanol (3.97 mL). To this solution, the proper ben-
zenethiol (1.984 mmol) and DIPEA (1.984 mmol) were
added. The vial was sealed and heated by microwave at
160 °C for 5 min. After cooling, the mixture was treated with
water (3 mL) and extracted with ethyl acetate (3 × 3 mL).
The organic extracts were collected, washed with brine
(3 × 2 mL), dried over Na2SO4 and evaporated at reduced
pressure. Raw materials were chromatographed on silica gel
(n-hexane:acetone 5:1 as eluent) to furnish pure RDS 802
and RDS 866.
Chemistry: general
Melting points were determined with a Büchi 530 capillary
apparatus and are uncorrected. Infrared (IR) spectra were
recorded on a Perkin-Elmer Spectrum-One spectrophotom-
For each compound, new achieved yields (%) are
reported.
1
eter. H NMR spectra were recorded on a Bruker AC 400
spectrometer. Merck silica gel 60 F254 plates were used for
analytical TLC. Developed plates were visualized by UV
light. Column chromatography was performed on silica gel
(Merck; 70–230 mesh) or aluminum oxide (Merck; 70–230
mesh). Compounds purities were always >95% determined
by high-pressure liquid chromatography (HPLC). HPLC
analysis was carried out using Shimadzu LC-10AD VP and
CTO-10AC VP. Column used was generally Suplex pKb-
100 (250 mm×4.6 mm, 5 μm). Microwave reactions were
conducted using a CEM Discover system unit (CEM Corp.,
Matthews, NC, USA). The machine consists of a continu-
ous focused microwave-power delivery system with operator
selectable power output from 0 to 300 W. The temperature of
the contents of the vessel was monitored using a calibrated
infrared temperature control mounted under the reaction ves-
sel. All experiments were performed using a stirring option
whereby the contents of the vessel are stirred by means of a
rotating magnetic plate located below the foor of the micro-
wave cavity and a Tefon-coated magnetic stir bar in the ves-
sel. Solvents were reagent grade and, when necessary, puri-
fed and dried by standard methods. Organic solutions were
dried over anhydrous sodium sulfate (Merck). Concentration
of solution after reactions and extractions involved the use of
a rotary evaporator operating at reduced pressure of approxi-
mately 20 Torr. Analytical results agreed to within 0.40%
of the theoretical values. Dimethylsulfoxide-d6 99.9% (code
44139-2) and deuterochloroform 98.8% (code 41675-4) of
isotopic purity (Aldrich) were used.
General procedure (GP‑B) for the synthesis of diaryl
sulfde derivatives RDS 562, RDS 615 and RDS 771
The proper halopyrimidine (4.96 mmol) was added to
a well-stirred solution of Pd2(dba)3 (49.6 µmol), DPPF
(99.18 µmol), DIPEA (5.45 mmol) and the proper ben-
zenethiol (4.96 mmol) in anhydrous DMF (7.5 mL) at
room temperature. The reaction was stirred under refux
for the proper time and then cooled to room temperature,
treated with water (50 mL) and extracted with chloroform
(3 × 50 mL). The combined organic phases were washed
with brine (5×20 mL), dried over Na2SO4 and evaporated
at reduced pressure. The crude products were purifed by
column chromatography yielding pure derivatives RDS 562,
RDS 615 and RDS 771. For each compound, yield (%), reac-
tion time and chromatographic system are reported.
6‑(sec‑Butoxy)‑2‑((4‑fuorophenyl)thio)
pyrimidin‑4‑amine (RDS 802)
Compound RDS 802 was prepared from intermediate 6-(sec-
butoxy)-2-chloropyrimidin-4-amine and 4-fuorothiophenol
by means of GP-A. Yield 50%.
6‑(sec‑Butoxy)‑2‑((phenyl)thio)pyrimidin‑4‑amine
(RDS 615)
Intermediate 6-(sec-butoxy)-2-chloropyrimidin-4-amine
and fnal products RDS 832, RDS 939 and RDS 1256 have
been synthesized according to literature (Costi et al. 2000;
Massa et al. 1994; Saccoliti et al. 2017).
Compound RDS 615 was prepared from intermediate 6-(sec-
butoxy)-2-chloropyrimidin-4-amine and benzenethiol by
means of GP-B. 55%; 6.5 h; SiO2/n-hexane:acetone 5:1.
Chemical, physical, and analytical data of the described
compounds are previously reported (Costi et al. 2000; Massa
et al. 1994).
6‑(sec‑Butoxy)‑2‑((4‑chlorophenyl)thio)
pyrimidin‑4‑amine (RDS 771)
General procedure (GP‑A) for the synthesis of diaryl
sulfde derivatives RDS 802, RDS 866
Compound RDS 771 was prepared from intermediate 6-(sec-
butoxy)-2-chloropyrimidin-4-amine and 4-chlorothiophenol
by means of GP-B. 69%; 6.5 h; SiO2/n-hexane:acetone 5:1.
In a microwave vial, intermediate 6-(sec-butoxy)-2-chlo-
ropyrimidin-4-amine (0.25 M; 0.992 mmol) was dissolved
1 3

Structure‑guided approach to identify a novel class of anti‑leishmaniasis diaryl sulfde…
2009; Mukhopadhyay et al. 1996; Saudagar et al. 2013;
6‑(sec‑Butoxy)‑2‑((3‑methoxyphenyl)thio)
pyrimidin‑4‑amine (RDS 866)
Wintner et al. 2010). In brief, L. donovani mid-log phase
promastigotes (1×107 cells) cultures were incubated with
IC50 dose of all synthesized drugs (Table 1) and RDS 777.
Following incubation for 24 h, 1×107 cells were harvested
by centrifugation (10,000 rpm, 10 min, 4 °C) and washed
twice with PBS. Pellet was suspended in 100 µl buffer
(50 mM HEPES at pH 8.0) containing 5 mM EDTA) taken
in a dark tube, 100 µL of 2 mM monobromobimane (dis-
solve in ethanol) was added, and suspension was incubated
at 70 °C for 3 min. Ice-cold 200 µL of 25% trichloroacetic
acid were added to the suspension, which was incubated on
ice for 20 min. The denatured protein and cell debris were
removed by centrifugation. Acid-soluble thiols were sepa-
rated by ion-paired, reverse phase HPLC on an ion-paired
Ultrasphere C18 column with linear gradient of 0–90%
methanol in 0.25% acetic acid (pH 3.5). Bimane-T(SH)2
complex formation absorbance was taken at 254 nm, using
a Dionex Ultimate 3000 instrument ftted with a Dionex
RF-2000 absorbance detector. Absorbance of standard
T(SH)2–bimane complex formation was also measured at
254 nm. Experiments were performed in duplicate and the
results are expressed as mean SD. Data were analyzed
using Student’s t test (one tailed) with the help of statistical
software PRISM 5 (Graphpad software). The criterion for
statistical signifcance between the groups was as follows: p
value 0.05 was considered signifcant and marked as *, 0.01
as highly signifcant and marked as **, 0.001 as very highly
signifcant and marked as ***.
Compound RDS 866 was prepared from intermediate 6-(sec-
butoxy)-2-chloropyrimidin-4-amine and 3-methoxythiophenol
by means of GP-A. Yield 79%.
2‑Chloro‑6‑(phenylthio)pyrimidin‑4‑amine (RDS
562)
Compound RDS 562 was prepared from 4-amino-2,6-dichlo-
ropyrimidine and benzenethiol by means of GP-B. 7%; 5 h;
Al2O3/CHCl3:n-hexane 9:1.
Inhibition of promastigote growth
Promastigote (the extracellular stage of the Leishmania life
cycle) growth inhibition was evaluated using L. infantum
strain (MHOM/TN/80/IPT1). To estimate the 50% inhibitory
concentration (IC50), the MTT (3-[4.5-dimethylthiazol-2-yl]-
2.5-diphenyltetrazolium bromide) micromethod (Sereno and
Lemesre 1997) was used throughout the experiments with
modifcation (Saccoliti et al. 2017).
Promastigotes were grown in Schneider’s Drosophila
medium (SIGMA) containing 10% fetal calf serum (FCS)
(GIBCO-BRL) and 2% gentamicin (50 mg/L) (Sigma) at
22 °C.
Parasites were adjusted to 1×106 parasites/mL, and 200 µL
of suspension was seeded in triplicate in 96-well fat bot-
tom microplates and incubated with varying concentrations
of compound. In a frst screening 0.01, 0.1, 1, 10, 100 μM
concentrations were used to evaluate the activity range of the
drugs. In the second screening, the following drug concentra-
tions were used to calculate the IC50: 0.16, 0.32, 0.65, 1.31,
2.62, 5.25, 10.50, 21.00, 42.00, 84.00, 168.00, 320.00 μM.
Amphotericin B (IC50 0.5 µM) (Euroclone) was used as
control. Each experiment was done in triplicate for each drug
concentration and two independent experiments were per-
formed. After 72 h of incubation, 30 µL of MTT was added
to each well and plates were further incubated for 2 h. The
absorbance at 550 nm was measured with a 96-well scanner
(DINEX Technologies, VA, USA). Anti-leishmanial activ-
ity per experimental point was expressed as mean percent-
age of inhibition with respect to vehicle treated parasites
(0%). In addition, the promastigote vitality was followed by
microscopic observation after 72 h. Data were analyzed and
plotted by Prism (GraphPad, San Diego, USA).
Enzymatic assay
LiTR was expressed and purifed as previously reported
(Baiocco et al. 2009b). Enzymatic inhibition assays were
carried out at 20 °C using a diode array Hewlett-Packard
HP8452A spectrophotometer.
All the diaryl sulfde compounds were frst tested at fnal
concentration of 10 μM. The solution containing bufer
(HEPES 50 mM, NaCl 40 mM at pH 7.4), TR 5 nM, TS2
150 μM and inhibitor 10 μM was allowed to equilibrate for
2 min in a quartz cuvette. Assays were initiated by addition
of NADPH 100 μM.
Table 1 IC50 values for each
Compound
IC50 (µM)
tested compound vs. L. infantum
promastigotes. The data are
expressed as mean standard
error of two independent
experiments performed in
triplicate
RDS 562
RDS 615
RDS 802
RDS 832
RDS 939
RDS 1256
11
31
18
15
16
2
2
1
1
1
Intracellular measurement of trypanothione
concentration
The intracellular trypanothione (T(SH)2) level was estimated
in the procedure as reported previously (Bhattacharya et al.
6.7 0.2
1 3

G. Colotti et al.
Absorbance decrease was followed at 340 nm, indicating
NADPH oxidation, and the percentage of inhibition was cal-
culated by measuring the velocity of NADPH oxidation with
respect to the experiment performed in absence of inhibi-
tor. NADPH concentration was calculated using the molar
extinction coefcient ε=6222/M/cm.
with and without a penalty (equal to −3.0 kcal/mol) to con-
straint the ligands within the site a to better evaluate the
mutual preference between the two binding sites. For each
simulated inhibitor, ten poses were generated and ranked by
the ChemPLP scoring function with a speed equal to 1. The
poses were then minimized and rescored by using ReScore+
(Vistoli et al. 2017).
As 10 µM RDS 562 displayed a percentage of inhibi-
tion higher than 30%, various conditions of inhibitor (0 µM,
6 µM, 8 µM, 10 µM) and substrate (50 µM, 100 µM, 150 µM)
were tested to determine the inhibition constant, Ki of the
compound through Dixon plot analysis.
Results
Due to solubility/aggregation problems, determination of
RDS 562 inhibition was not possible at higher concentra-
tions. Every experiment was carried out in triplicate.
In order to verify selectivity, RDS 562 was tested on glu-
tathione reductase, the homologous human protein of TR, in
the same conditions of the screening assay: GR 5 nM, GSSG
150 µM, RDS 562 (0, 8, 10, 12, 20 µM) in bufer 20 mM
HEPES, 40 mM NaCl pH 7.4.
Synthesis of diaryl sulfde compounds
Compounds RDS 562, RDS 615, RDS 802, RDS 832, RDS
939 and RDS 1256 have been synthesized according to
Scheme 1.
The synthesis of such derivatives has been already
reported in literature by our research group (Costi et al.
2000); however, in the current work some steps of the pre-
viously described synthetic pathway have been changed for
both safety and saving time reasons. In particular, all the
synthetic steps involving the use of benzenethiol reagents in
nucleophilic aromatic substitution have been properly modi-
fed. The originally reported synthetic route was based on
the use of in situ prepared potassium thiophenolates, which
are well-known dangerous and difcult to handle reagents.
Moreover, in general, reactions involving such reactants usu-
ally require long time to achieve the product. For instance,
despite its notable yield, nucleophilic aromatic substitu-
tion reaction leading to reference compound RDS 777 was
reported to take place in 1 week (Costi et al. 2000).
Thus, according to the synthetic approach already applied
for the re-synthesis of RDS 777 (Saccoliti et al. 2017), the
introduction of thiophenol groups on the central pyrimidine
core was achieved through alternative synthetic routes. In
particular, microwave-assisted or Pd-catalyzed batch reac-
tions have been employed, in order to overcome the afore-
mentioned limits of the previously reported pathway.
The commercially available product 4-amino-2,6-dichlo-
ropyrimidine underwent a nucleophilic aromatic substitution
in the presence of sodium sec-butoxide, which was in situ
prepared from anhydrous sec-butanol and metallic sodium,
as previously reported (Costi et al. 2000; Massa et al. 1994;
Saccoliti et al. 2017).
The linear regression analyses were performed with
Microsoft Excel (version 2013).
Docking experiments
The TR structure in complex with RDS 777 was retrieved
from PDB (Id: 5ebk) and prepared by deleting ions, water
molecules and crystallization additives and by adding
hydrogen atoms. To be compatible with physiological pH,
Asp, Glu, Lys and Arg residues were considered ionized
while His and Cys residues neutral by default. The protein
structure was then minimized by fxing the backbone atoms
to preserve the resolved folding. After deleting the bound
inhibitors, the so refned TR structure underwent docking
simulations. The conformational profle of the considered
ligands was explored by Monte Carlo procedures as imple-
mented in the VEGA software (Pedretti et al. 2002), which
generates 1000 minimized conformers for each ligand.
The so-derived lowest energy structure underwent docking
simulations after a fnal optimization by PM7-based semi-
empirical calculations (Stewart 2013). Docking simulations
were performed by PLANTS (Korb et al. 2009) and attention
was focused on three possible binding sites as derived by the
resolved structure: the frst (named site a) corresponds to the
resolved pose in the catalytic site, the second (site d) is adja-
cent to site a and slightly more external towards the entrance
of the catalytic pocket while the third binding site (site c) is
rather superfcial at the entrance of the NADPH-binding site.
Hence, docking searches were focused within a 15 Å radius
sphere around the resolved RDS 777 pose corresponding
to site a as well as within a 10 Å radius sphere around the
resolved RDS 777 pose corresponding to site c. Notice that
the sphere size for the frst search was selected to encompass
both binding sites a and d and the simulations were repeated
The reaction was performed in the same solvent for 4 h,
leading to intermediate 6-(sec-butoxy)-2-chloropyrimidin-
4-amine as main product.
Applying the recently reported procedure employed for
the re-synthesis of RDS 777 (Saccoliti et al. 2017), diaryl
sulfde derivatives RDS 562, RDS 615 and RDS 771 have
been synthesized through a modifed Buchwald–Hartwig
C-S cross coupling reaction by using the proper thiophenol
in the presence of Pd(dba)3, DPPF and DIPEA in anhydrous
1 3

Structure‑guided approach to identify a novel class of anti‑leishmaniasis diaryl sulfde…
NH2
NH2
ii
N
N
iv
Cl
N
S
Cl
N
S
O
O
RDS 832
RDS 562
NH2
N
N
Cl
Cl
NH2
R2
NH2
R2
NH2
i
R1
R1
ii or iii
N
N
iv
N
S
N
O
Cl
N
O
S
N
O
O
O
3a-d
4c,d
RDS 802: R1 = F; R2 = H
RDS 615: R1 = H; R2 = H
RDS 771: R1 = Cl; R2 = H
RDS 939: R1 = Cl; R2 = H
RDS 1256: R1 = H; R2 = OCH3
NH2
N
RDS 866: R1 = H; R2 = OCH3
N
O
Cl
Reagents and conditions: (i) sec-BuOHa, Na, reflux, 4h, Ar, 60%; (ii) proper thiophenol, Pd2(dba)3, DPPF, DIPEA, DMFa, reflux, 5-6,5h, Ar, 7-69%;
(iii) proper thiophenol, DIPEA, iPrOH 0.25M, MW, 160°C, 5 min, 50-79%; (iv) H2O2 30%, AcOH gl., 60°C, 4-7.5 h, 23-87%.
Scheme 1 Synthesis of compounds RDS 562, RDS 615, RDS 802, RDS 832, RDS 939 and RDS 1256
DMF. Alternatively, a previously reported microwave-
assisted procedure (Luo et al. 2002) was applied for the syn-
thesis of thioethers RDS 802 and RDS 866. Notably, such
reaction, which was performed in the presence of the proper
thiophenol and DIPEA in isopropanol 0.25 M, took place in
only 5 min at 160 °C, yielding the products in good yields.
Sulfone derivatives RDS 832, RDS 939 and RDS 1256
have been synthesized from the corresponding thioether
intermediates RDS 562, RDS 771 and RDS 866 in the pres-
ence of hydrogen peroxide 30% in glacial acetic acid at
60 °C, according to literature (Costi et al. 2000).
0.3 μM), no changes in promastigote morphology, motil-
ity and growth were observed with respect to control cells.
Table 1 reports the IC50 values of the tested compounds,
showing that all synthesized compounds are able to kill pro-
mastigote in micromolar range. RDS 562, the compound
more active against trypanothione metabolism (see below),
at dose 168 μM, induced 100% promastigote mortality,
whereas at ‘sublethal’ concentrations, ranging from 84.00 to
0.65 μM, showed a dose-dependent leishmanistatic activity.
Lower concentrations, ≤0.3 μM, did not afect promastigote
growth (Fig. 2).
Activity against L. infantum promastigote
Intracellular measurement of trypanothione
concentration
Diaryl sulfdes induced a dose-dependent anti-proliferative
efect on L. infantum promastigotes that was observed by
both MTT enzymatic assay and microscopic observations.
The results of the frst screening showed that all com-
pounds induced 90% promastigote mortality at a concen-
tration around 100 μM. At this concentration, RDS com-
pounds induced mortality observed by microscopy, i.e., the
parasites appeared as rounded, afagellated and vacuolated
cells, arranged in clusters along the edges of the micro-
plate well. At ‘sublethal’ concentrations (between 1 and
100 μM), a dose-dependent leishmanistatic activity was
observed. Indeed, the parasites showed motility inhibition
(with a reduced fagellum) and body swelling, not present
in control cells. At submicromolar concentrations (below
The promastigote cells treated with IC50 dose of all drugs
for 24 h showed signifcant changes in the levels of intracel-
lular T(SH)2. Interestingly all the RDS compounds with the
exception of RDS 777 and RDS 562 display an increase in
T(SH)2 concentration in the cell (data not shown) indicating
that TR is not the main target for these compounds.
On the contrary a decrease of 33.15% of T(SH)2 was
observed in RDS 562- and of 38.5% in RDS 777-treated
cells, as compared to untreated control cells (Fig. 3) indicat-
ing that these compounds afect the trypanothione metabo-
lism and act preferentially by inhibiting TR. As reported in
Fig. 3, the inhibitor concentrations used for the experiments
correspond to the IC50 values of RDS 777 and RDS 562,
1 3

G. Colotti et al.
experiments were carried out at various concentrations of
TS2 and RDS 562, while fxed TR and NADPH concentra-
tions (5 nM and 100 μM, respectively) were maintained.
After starting the reaction by the addition of NADPH, the
absorbance decrease at 340 nm, indicative of NADPH oxi-
dation, was measured (Baiocco et al. 2009a). As shown in
Fig. 4a, RDS 562 inhibited the binding of TS2 to TR. Each
line in the Dixon plot represents linear regression analysis
of reciprocal average ftted rates of TS2 reduction for dif-
ferent substrate concentrations, as a function of inhibitor
concentration. The KM and kcat of TR used for the Ki calcu-
lation were 23.0 1.0 μM and 11.4 0.3 s−1, respectively.
The value of Ki calculated from the Dixon plot analysis was
12.0 1.0 μM, higher than that of Sb(III) (1.5 μM) (Baiocco
et al. 2009a). The Lineweaver–Burk plot, representing linear
regression analysis of reciprocal average ftted rates of TS2
reduction for diferent inhibitor concentrations as a function
of substrate concentrations, clearly demonstrates that RDS
562 is a competitive inhibitor. Indeed, as shown in Fig. 4b,
Vmax does not change whereas the apparent KM increases as
a function of the inhibitor concentration increase (Fig. 4b).
RDS 562 was tested also on hGR and, as shown in Fig. 4c,
it is not efective on the human homolog of TR (Fig. 4c).
Fig. 2 Growth inhibition induced by RDS 562. % of growth inhibi-
tion calculated on L. infantum promastigote stage after treatment
with various concentrations of RDS 562. The data are expressed as
mean standard error of two independent experiments performed in
triplicate
Docking experiments
The docking experiments have been performed using all the
synthesized compounds, but we focalized our discussion on
RDS 562 which is the only one able to reduce the trypan-
othione concentration in cells and therefore acting preferen-
tially on the trypanothione reductase enzyme.
Fig. 3 Measurement of intracellular thiol level in both untreated
Leishmania donovani (control) and drug (RDS 562 and RDS 777)
treated Leishmania donovani promastigotes. Results are expressed
as average value of area of the peak of two independent experiments
(n=2). Percentage changes in the case of two samples (RDS777
and RDS562) were calculated considering average peak value of
control as 100. Results are expressed as mean SD. Data were ana-
lyzed using Student’s t test (one tailed) with the help of statistical
software PRISM 5 (Graphpad software). The criterion for statistical
signifcance between the groups was as follows: p value, 0.05 was
considered signifcant and marked as *, 0.01 as highly signifcant and
marked as **, 0.001 as very highly signifcant and marked as ***
As explained under methods, the sites a and d are adjacent
and partly overlapped and therefore a single docking simu-
lation can comprise both sites, thus evaluating the relative
preference of each ligand for them. A bird’s eye analysis of
all docking results and of relative scoring functions reveals
that all simulating ligands are properly accommodated
within the site a only when applying the constraint pen-
alty. When docking search is left unconstrained, the ligands
markedly prefer to engage the site d. The clear preference is
refected into all computed docking scores which shows bet-
ter values when considering the unconstrained docking sim-
ulation thus suggesting that the binding to site d allows the
ligand to stabilize stronger interactions compared to site a.
For example, the primary score ChemPLP reveals a decrease
average of 4.3 kcal/mol shifting from −61.9 to −57.6 kcal/
mol. Notably, the comparison of docking scores between
site a and d (Table 2) reveals that the tighter contacts involve
both polar and hydrophobic interactions thus suggesting
that the ligand polarity should not signifcantly infuence
the preference between binding sites. Figure 5 compares the
best pose of RDS 562 within site a and site d. In the site
a (see Fig. 5a), RDS 562 is seen to assume a fipped pose
respectively. Thus, since the concentration of RDS 777 used
in the experiments (29.43 μM) is three times higher than the
concentration of RDS 562 used in the experiment (11 μM),
it follows that RDS 562 results to be more active than RDS
777 on the trypanothione metabolism.
Enzymatic assays
Kinetic studies were performed on RDS 562 endowed with
the highest activity against the promastigote forms of L.
donovani and vs. trypanothione levels. Steady state kinetic
1 3

Structure‑guided approach to identify a novel class of anti‑leishmaniasis diaryl sulfde…
Fig. 4 Inhibition assays. a
Dixon plot of LiTR inhibition
with RDS 562. Linear regres-
sion was performed using four
inhibitor concentrations (0 µM
RDS 562; 6 µM RDS 562;
8 µM RDS 562; 10 µM RDS
562) and three TS2 concentra-
tions (50 µM TS2, flled circles,
R2 =0.9386; 100 µM, flled
squares TS2, R2 =0.9941;
150 µM TS2, flled diamonds,
R2 =0.9875). Each point repre-
sent the mean of three experi-
ments; the assay was performed
in bufer Hepes 20 mM, 40 mM
NaCl pH 7,4, in presence
of 5 nM LiTR, and 100 µM
NADPH. b Lineweaver–Burk
plot. RDS 562 inhibits LiTR
in a competitive fashion; linear
regression was performed with
four inhibitor concentrations
(0 µM RDS 562, flled squares,
R2 =0.9502; 6 µM RDS 562,
flled diamonds, R2 =0.9746;
8 µM RDS 562, flled triangles,
R2 =0.9103; 10 µM RDS 562,
flled circles, R2 =0.8356).
Each line is the mean of three
experiments; the assay was
performed in bufer Hepes
20 mM, 40 mM NaCl pH 7,4, in
presence of 5 nM LiTR, 50 µM,
100 µM and 150 µM of TS2 and
100 µM NADPH. c Inhibition
assays on TR and GR. RDS
562 is not active against hGR
(IC50 >100 µM) (flled circles).
On the contrary, LiTR activity
decreases as a function of RDS
562 concentrations (flled dia-
monds). Both the experiments
were performed in bufer Hepes
20 mM, 40 mM NaCl pH 7.4,
in presence of 5 nM enzyme,
150 µM of TS2 and 100 µM of
NADPH. Each line is expres-
sion of three independent
experiments (RDS 562 vs hGR
R2 =0.8017; RDS 562 vs LiTR
R2 =0.8930)
compared to the corresponding resolved RDS 777 pose by
which the 4-aminopyrimidine ring elicits a set of H-bonds
involving Ser14, Tyr110 and Thr335 of chain A, while the
phenyl ring approaches the chain B by contacting His461′,
Pro462′ and Glu466′. Thus and similarly to what was seen
for RDS 777, the binding site a comprises both subunits
even though with a diferent involvement. Indeed, while for
RDS 777 the key polar contacts were stabilized by residues
1 3

G. Colotti et al.
Table 2 Representative docking scores for the three considered bind-
RDS 777, RDS 562 is seen to elicit H-bonds which involve
the 4-aminopyrimidine ring with Glu466′ and Glu467′. The
pyrimidine ring also stabilizes a clear π–π stacking interac-
tion with Phe396′, while the phenyl ring is mostly engaged
in hydrophobic contacts which involve Pro398′, Leu399′ and
Pro462′. The only involvement of the chain A is represented
by Lys61, which approaches the ligand phenyl ring stabiliz-
ing charge transfer interactions. The analysis of the docking
results for the other two compounds containing a thioether
linker reveals binding modes rather similar to that above
described for both binding sites a and d. A slightly more
heterogeneous behavior is shown by the three derivatives
including an oxidized sulphonyl linker. In detail RDS 832 is
able to retain similar poses both in the site a where the sul-
phonyl group elicits a strong H-bond with His461′ and in the
site d where the sulphonyl approaches Lys61. The increase
of the steric hindrance in RDS 939 has a rather destabiliz-
ing efect especially for the binding site a and indeed few
computed RDS 939 poses appear suitably arranged there.
The site d appears to be more able to accommodate these
sulphonyl derivatives with an increased relevance of the
polar contacts elicited by Lys61 as mentioned above. On
these grounds, one may understand why the further increase
of size and polarity render the most active RDS 1256 a sort
of exception since it is unable to engage the site a even
ing sites for RDS 562
RDS 562
LJ_CHARMM
(kcal/mol)
Elect_DD (kcal/ CHEMPLP
mol)
(kcal/mol)
Site d
Site a
Site c
−25.51
−21.63
−22.83
−6.95
−4.03
−9.84
−9.84
−53.63
−54.77
The primary ChemPLP score is computed by PLANTS and encodes
for both ionic and non-polar contacts while LJ_CHARMM and Elec_
DD are computed by VEGA and encode for the van der Waals con-
tacts as calculated by the Lennard-Jones term of the CHARMM force
feld and the ionic interactions by using a distance dependent dielec-
tric function, respectively (all scores are expressed in kcal/mol)
of the chain B, and the chain A takes part only by hydropho-
bic contacts, the key H-bonds of RDS 562 involve the chain
A, while the chain B elicits almost only hydrophobic con-
tacts. Despite the described diferences, both ligands (RDS
777 and RDS 562) emphasize the pivotal role of His461′,
which stabilizes π–π stacking with both aromatic rings plus
a weak H-bond with the thioether linker. Figure 5b shows
the putative complex of RDS 562 within the binding site d,
which is almost completely lined by residues of the chain B.
In detail and similarly to the corresponding resolved pose of
Fig. 5 Docking results. Main
interactions stabilizing the
putative complexes of RDS 562
within the binding site a (a),
site d (b) and site c (d). Main
interactions stabilizing RDS
1256 in the binding site d (c)
1 3

Structure‑guided approach to identify a novel class of anti‑leishmaniasis diaryl sulfde…
when performing constrained docking simulations and in
both docking runs it occupies the binding site d assuming a
slightly extended pose by which its sulphonyl group is able
to contacts Arg472′ while the pyrimidine ring maintains its
interactions with Glu466′ and Glu467′ (see Fig. 5c).
Finally, Fig. 5d shows the best putative complex as
computed for RDS 562 within the binding site c where it
approaches Arg222 and Arg228 with which it can stabilize
both H-bonds and charge transfer interactions. The com-
puted complex appears to be further stabilized by π–π stack-
ing involving Tyr221 which is also engaged by a halogen
bond with chlorine atom plus hydrophobic contacts with
Ile285 and Val194. All simulated compounds show similar
interaction patterns with the sulphonyl derivatives which
elicits strong polar interactions with the surrounding argi-
nine residues.
homologous GR. The value of Ki calculated from the Dixon
plot analysis for RDS 562 is an order of magnitude higher
than that calculated for RDS 777 (0.25 µM) and more than
six times higher than that of Sb(III) (1.5 µM), the active
form of antimonials, the most used drug against Leishmania-
sis, calculated by Baiocco et al. (2009a) and about two times
higher than that of compound 1 (4.6 2.5 µM) (Baiocco
et al. 2013).
The docking studies performed on TR with RDS 562
starting from the binding sites of RDS 777 on the com-
plex structure TR-777 shows that whereas this compounds
preferentially binds to the so-called site a, RDS 562 binds
preferentially to site d. Indeed, the structural analysis per-
formed on the complex LiTR-777 show that RDS 777 binds
to both the TR subunit A and B in the site a, whereas only
in the B subunit binds to the site d with an occupancy of 0.8
indicating a higher afnity of RDS 777 for the a site; on the
contrary when RDS 562 docking search is performed on TR
unconstrained, the ligand markedly prefers to engage the site
d. The clear preference is refected into all computed dock-
ing scores which show better values when considering the
unconstrained docking simulation thus suggesting that the
binding site d allows the ligand to stabilize stronger interac-
tions compared to site a (Table 2, Fig. 5).
The analysis of the computed docking scores for the site
c reveals, on average, worst values compared to both bind-
ing site a and d. For example, the primary score ChemPLP
shows a decrease average of 8 kcal/mol when comparing
site d and c and of 3.7 kcal/mol between site a and site c.
The general worsening of the docking scores for the site c is
clearly explainable when considering that it is on the protein
surface and thus the ligands are not completely surrounded
by interacting residues. Considering the described interac-
tions, it comes as no surprise that the only docking scores
which shows better values for the site c is that describing
the ionic contacts based on a distance dependent dielectric
constant.
The data presented in this paper show that we succeeded
to fnd a new diaryl sulfde compound able to kill Leishma-
nia parasites with a higher efciency with respect to RDS
777. Indeed the IC50 of RDS 777 is 29 1 μM, whereas the
IC50 of RDS 562 is 11 2 μM. Our data show that RDS 562
target specifcally the trypanothione metabolism by decreas-
ing its concentration in the cell through TR inhibition.
Even if the ability to inhibit TR of RDS 562 is an order of
magnitude lower than that of RDS 777, the docking studies
show that whereas RDS 777 binds to four diferent TR sites
(a–d), RDS 562 is able to bind preferentially to the site d
in the trypanothione binding cavity whereas its afnity for
the other binding sites is negligible. More importantly, RDS
562 is able to reduce the intracellular reduced trypanothione
concentration of about 33% when added to the medium at a
concentration of 11 μM (=IC50), whereas RDS 777 is able
to reduce the reduced trypanothione concentration of 38%
when added to the medium at a concentration three times
higher (29 μM).
Discussion
Leishmania parasites rely on a unique and essential thiol-
based redox metabolism based on trypanothione (Comini
and Flohé 2013; Krauth-Siegel and Comini 2008; Oza
et al. 2005), while a system of redox homeostasis based on
catalase and glutathione reductase is absent in the parasite.
Therefore, inhibition of the enzymes of the trypanothione
pathway is one of the most attractive options for anti-kine-
toplastid drug discovery.
On the basis of the crystal structure of the diaryl sulfde
RDS 777 in complex with LiTR, we synthesized a series
of diaryl sulfdes able to inhibit the Leishmania growth in
the promastigote stage displaying a IC50 ranging from 6 to
30 µM (Table 1). Among them RDS 1256 and RDS 562 are
inhibitors of Leishmania parasite growth more potent than
RDS 777 but only RDS 562 besides RDS 777 is able to
signifcantly decrease the amount of reduced trypanothione
in the Leishmania donovani promastigotes. As shown by
the inhibition assays, RDS 562 is able to selectively inhibit
the trypanothione reductase, displaying an inhibition con-
stant of 12.0 1.0 µM, while it does not inhibit the human
In conclusion, among diaryl sulfdes, RDS 562 which is
able to kill Leishmania parasites and to decrease trypan-
othione concentration in the promastigote stage, by compet-
ing with trypanothione binding, without of-target efects,
can represent a new promising compound to find new
drugs against visceral leishmaniasis; in addition, since the
residues lining the trypanothione binding cavity of TR are
strictly conserved, it could also represent a promising mol-
ecule for the treatment of other related neglected diseases
caused by Trypanosomatids such as Chagas’ disease, caused
1 3

G. Colotti et al.
Colotti G, Fiorillo A, Ilari A (2018) Metal- and metalloid-containing
drugs for the treatment of trypanosomatid diseases. Front Biosci
(Landmark Ed) 23:954–966
by Trypanosoma cruzi, and the African sleeping sickness,
caused by T. brucei, diseases against which no efective and
safe therapies are available.
Comini MA, Flohé L (2013) Trypanothione-based redox metabo-
lism of trypanosomatids. In: Jäger T, Koch O, Flohé L (eds)
Trypanosomatid diseases. Wiley-VCH Verlag GmbH & Co.
KGaA, Hoboken, pp 167–199. https://doi.org/10.1002/97835
27670383.ch9
Acknowledgements We gratefully acknowledge CNCCS CNR
(National Collection of Chemical Compounds and Screening Center
2018) to AI and MIUR PRIN 20154JRJPP to GC which supported the
experiments reported in these studies. This paper is dedicated to the
memory of our friend and colleague Prof. Emilia Chiancone, passed
away sadly in December 2018.
Costi R et al (2000) Structure-activity relationship studies on poten-
tial non-nucleoside DABO-like inhibitors of HIV-1 reverse
transcriptase. Antivir Chem Chemother 11:117–133. https://
doi.org/10.1177/095632020001100204
Desjeux P (2004) Leishmaniasis: current situation and new perspec-
tives. Comp Immunol Microbiol Infect Dis 27:305–318. https://
doi.org/10.1016/j.cimid.2004.03.004
Compliance with ethical standards
Conflict of interest The authors declare that they have no confict of
Dixon MJ, Maurer RI, Biggi C, Oyarzabal J, Essex JW, Bradley
M (2005) Mechanism and structure–activity relationships of
norspermidine-based peptidic inhibitors of trypanothione
reductase. Bioorg Med Chem 13:4513–4526. https://doi.
org/10.1016/j.bmc.2005.04.039
interest.
Research involving human participants and/or animals This article
does not contain any studies with human participants or animals per-
formed by any of the authors.
Dumas C et al (1997) Disruption of the trypanothione reductase gene
of Leishmania decreases its ability to survive oxidative stress in
macrophages. EMBO J 16:2590–2598. https://doi.org/10.1093/
emboj/16.10.2590
Informed consent Informed consent was obtained from all individual
participants included in the study.
Hamilton CJ, Saravanamuthu A, Eggleston IM, Fairlamb AH (2003)
Ellman’s-reagent-mediated regeneration of trypanothione
in situ: substrate-economical microplate and time-dependent
inhibition assays for trypanothione reductase. Biochem J
369:529–537. https://doi.org/10.1042/BJ20021298
References
Hamilton CJ, Saravanamuthu A, Poupat C, Fairlamb AH, Eggleston
IM (2006) Time-dependent inhibitors of trypanothione reduc-
tase: analogues of the spermidine alkaloid lunarine and related
natural products. Bioorg Med Chem 14:2266–2278. https://doi.
org/10.1016/j.bmc.2005.11.004
Bailey S, Smith K, Fairlamb AH, Hunter WN (1993) Substrate interac-
tions between trypanothione reductase and N1-glutathionylsper-
midine disulphide at 0.28-nm resolution. Eur J Biochem FEBS
213:67–75
Baiocco P, Colotti G, Franceschini S, Ilari A (2009a) Molecular basis
of antimony treatment in leishmaniasis. J Med Chem 52:2603–
2612. https://doi.org/10.1021/jm900185q
Holloway GA et al (2007) Discovery of 2-iminobenzimidazoles as a
new class of trypanothione reductase inhibitor by high-through-
put screening. Bioorg Med Chem Lett 17:1422–1427. https://
doi.org/10.1016/j.bmcl.2006.11.090
Baiocco P, Franceschini S, Ilari A, Colotti G (2009b) Trypanothione
reductase from Leishmania infantum: cloning, expression, purif-
cation, crystallization and preliminary X-ray data analysis. Protein
Pept Lett 16:196–200
Hotez PJ, Savioli L, Fenwick A (2012) Neglected tropical diseases
of the Middle East and North Africa: review of their prevalence,
distribution, and opportunities for control. PLoS Negl Trop Dis
6:e1475. https://doi.org/10.1371/journal.pntd.0001475
Ilari A et al (2012) A gold-containing drug against parasitic polyam-
ine metabolism: the X-ray structure of trypanothione reductase
from Leishmania infantum in complex with auranofn reveals a
dual mechanism of enzyme inhibition. Amino Acids 42:803–
811. https://doi.org/10.1007/s00726-011-0997-9
Baiocco P, Ilari A, Ceci P, Orsini S, Gramiccia M, Di Muccio T, Colotti
G (2011) Inhibitory efect of silver nanoparticles on trypanothione
reductase activity and Leishmania infantum proliferation. ACS
Med Chem Lett 2:230–233. https://doi.org/10.1021/ml1002629
Baiocco P et al (2013) Inhibition of Leishmania infantum trypanothione
reductase by azole-based compounds: a comparative analysis with
its physiological substrate by X-ray crystallography. ChemMed-
Chem 8:1175–1183. https://doi.org/10.1002/cmdc.201300176
Bhattacharya A, Biswas A, Das PK (2009) Role of a diferentially
expressed cAMP phosphodiesterase in regulating the induction
of resistance against oxidative damage in Leishmania donovani.
Free Radic Biol Med 47:1494–1506. https://doi.org/10.1016/j.
freeradbiomed.2009.08.025
Ilari A, Fiorillo A, Baiocco P, Poser E, Angiulli G, Colotti G (2015)
Targeting polyamine metabolism for fnding new drugs against
leishmaniasis: a review. Mini Rev Med Chem 15:243–252
Ilari A, Fiorillo A, Genovese I, Colotti G (2017) Polyamine–trypan-
othione pathway: an update. Future Med Chem 9:61–77. https
://doi.org/10.4155/fmc-2016-0180
Ilari A, Genovese I, Fiorillo F, Battista T, De Ionna I, Fiorillo A,
Colotti G (2018) Toward a drug against all kinetoplastids: from
leishbox to specifc and potent trypanothione reductase inhibi-
tors. Mol Pharm 15:3069–3078. https://doi.org/10.1021/acs.
molpharmaceut.8b00185
Burza S, Croft SL, Boelaert M (2018) Leishmaniasis. Lancet 392:951–
970. https://doi.org/10.1016/S0140-6736(18)31204-2
Colotti G, Ilari A (2011) Polyamine metabolism in Leishmania: from
arginine to trypanothione. Amino Acids 40:269–285. https://doi.
org/10.1007/s00726-010-0630-3
Korb O, Stutzle T, Exner TE (2009) Empirical scoring functions for
advanced protein-ligand docking with Plants. J Chem Inf Model
49:84–96. https://doi.org/10.1021/ci800298z
Colotti G, Baiocco P, Fiorillo A, Bof A, Poser E, Chiaro FD, Ilari A
(2013a) Structural insights into the enzymes of the trypanothione
pathway: targets for antileishmaniasis drugs. Future Med Chem
5:1861–1875. https://doi.org/10.4155/fmc.13.146
Krauth-Siegel RL, Comini MA (2008) Redox control in trypano-
somatids, parasitic protozoa with trypanothione-based thiol
metabolism. Biochim Biophys Acta 1780:1236–1248. https://
doi.org/10.1016/j.bbagen.2008.03.006
Colotti G et al (2013b) Metal-based compounds as prospective
antileishmanial agents: inhibition of trypanothione reductase by
selected gold complexes. ChemMedChem 8:1634–1637. https://
doi.org/10.1002/cmdc.201300276
1 3

Structure‑guided approach to identify a novel class of anti‑leishmaniasis diaryl sulfde…
Sereno D, Lemesre JL (1997) Use of an enzymatic micromethod to
Luo GL, Chen L, Poindexter GS (2002) Microwave-assisted synthesis
of aminopyrimidines. Tetrahedron Lett 43:5739–5742. https://doi.
org/10.1016/S0040-4039(02)01190-5
quantify amastigote stage of Leishmania amazonensis in vitro.
Parasitol Res 83:401–403
Spinks D et al (2009) Investigation of trypanothione reductase as a drug
target in Trypanosoma brucei. ChemMedChem 4:2060–2069.
https://doi.org/10.1002/cmdc.200900262
Martyn DC, Jones DC, Fairlamb AH, Clardy J (2007) High-throughput
screening afords novel and selective trypanothione reductase
inhibitors with anti-trypanosomal activity. Bioorg Med Chem
Lett 17:1280–1283. https://doi.org/10.1016/j.bmcl.2006.12.016
Massa SDS, Costi R, Artico R, Loi M, Doa AG, Scano M, La Colla
P (1994) Synthesis of novel HEPT analogues with anti-HIV-1
activity. Med Chem Res 4:554–562
Stewart JJ (2013) Optimization of parameters for semiempirical meth-
ods VI: more modifcations to the NDDO approximations and
re-optimization of parameters. J Mol Model 19:1–32. https://doi.
org/10.1007/s00894-012-1667-x
Stump B, Eberle C, Kaiser M, Brun R, Krauth-Siegel RL, Diederich F
(2008) Diaryl sulfde-based inhibitors of trypanothione reductase:
inhibition potency, revised binding mode and antiprotozoal activi-
ties. Org Biomol Chem 6:3935–3947. https://doi.org/10.1039/
b806371k
Mukhopadhyay R, Kapoor P, Madhubala R (1996) Antileishmanial
efect of a potent S-adenosylmethionine decarboxylase inhibitor:
CGP 40215A. Pharmacol Res 33:67–70. https://doi.org/10.1006/
phrs.1996.0011
Oza SL, Shaw MP, Wyllie S, Fairlamb AH (2005) Trypanothione bio-
synthesis in Leishmania major. Mol biochem Parasitol 139:107–
116. https://doi.org/10.1016/j.molbiopara.2004.10.004
Patterson S et al (2011) Dihydroquinazolines as a novel class of Trypa-
nosoma brucei trypanothione reductase inhibitors: discovery, syn-
thesis, and characterization of their binding mode by protein crys-
tallography. J Med Chem 54:6514–6530. https://doi.org/10.1021/
jm200312v
Turcano L et al (2018) Identifcation and binding mode of a novel
Leishmania Trypanothione reductase inhibitor from high through-
put screening. PLoS Negl Trop Dis 12:e0006969. https://doi.
org/10.1371/journal.pntd.0006969
Vistoli G, Mazzolari A, Testa B, Pedretti A (2017) Binding space
concept: a new approach to enhance the reliability of docking
scores and its application to predicting butyrylcholinesterase
hydrolytic activity. J Chem Inf Model 57:1691–1702. https://doi.
org/10.1021/acs.jcim.7b00121
Pedretti A, Villa L, Vistoli G (2002) VEGA: a versatile program to
convert, handle and visualize molecular structure on Windows-
based PCs. J Mol Graph Model 21:47–49
Wintner EA et al (2010) A monobromobimane-based assay to meas-
ure the pharmacokinetic profle of reactive sulphide species in
blood. Br J Pharmacol 160:941–957. https://doi.org/10.111
1/j.1476-5381.2010.00704.x
Saccoliti F et al (2017) Inhibition of Leishmania infantum tryp-
anothione reductase by diaryl sulfide derivatives. J Enzyme
Inhib Med Chem 32:304–310. https://doi.org/10.1080/14756
366.2016.1250755
Zulfqar B, Shelper TB, Avery VM (2017) Leishmaniasis drug discov-
ery: recent progress and challenges in assay development. Drug
Discov Today 22:1516–1531. https://doi.org/10.1016/j.drudi
s.2017.06.004
Saravanamuthu A, Vickers TJ, Bond CS, Peterson MR, Hunter WN,
Fairlamb AH (2004) Two interacting binding sites for quinacrine
derivatives in the active site of trypanothione reductase: a tem-
plate for drug design. J Biol Chem 279:29493–29500. https://doi.
org/10.1074/jbc.M403187200
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional afliations.
Saudagar P, Saha P, Saikia AK, Dubey VK (2013) Molecular mecha-
nism underlying antileishmanial effect of oxabicyclo[3.3.1]
nonanones: inhibition of key redox enzymes of the pathogen.
Eur J Pharm Biopharm 85:569–577. https://doi.org/10.1016/j.
ejpb.2013.08.014
1 3