Enhancement of Benzothiazoles as Pteridine Reductase-1 Inhibitors for the Treatment of Trypanosomatidic Infections

Article abstract of DOI:10.1021/acs.jmedchem.8b02021

2-Amino-benzo[d]thiazole was identified as a new scaffold for the development of improved pteridine reductase-1 (PTR1) inhibitors and anti-trypanosomatidic agents. Molecular docking and crystallography guided the design and synthesis of 42 new benzothiazoles. The compounds were assessed for Trypanosoma brucei and Leishmania major PTR1 inhibition and in vitro activity against T. brucei and amastigote Leishmania infantum. We identified several 2-amino-benzo[d]thiazoles with improved enzymatic activity (TbPTR1 IC₅₀ = 0.35 μM; LmPTR1 IC₅₀ = 1.9 μM) and low μM antiparasitic activity against T. brucei. The ten most active compounds against TbPTR1 were able to potentiate the antiparasitic activity of methotrexate when evaluated in combination against T. brucei, with a potentiating index between 1.2 and 2.7. The compound library was profiled for early ADME toxicity, and 2-amino-N-benzylbenzo[d]thiazole-6-carboxamide (4c) was finally identified as a novel potent, safe, and selective anti-trypanocydal agent (EC₅₀ = 7.0 μM). Formulation of 4c with hydroxypropyl-β-cyclodextrin yielded good oral bioavailability, encouraging progression to in vivo studies.

Full text of DOI:10.1021/acs.jmedchem.8b02021

Subscriber access provided by READING UNIV
Article
Enhancement of benzothiazoles as Pteridine Reductase-1 (PTR1)
inhibitors for the treatment of Trypanosomatidic infections.
Pasquale Linciano, Cecilia Pozzi, Lucia Dello Iacono, Flavio di PIsa, Giacomo Landi,
Alessio Bonucci, Sheraz Gul, Maria Kuzikov, Bernhard Ellinger, Gesa Witt, Nuno Santarem,
Catarina Baptista, Caio Franco, Carolina Borsoi Moraes, Wolfgang Müller, Ulrike Wittig,
Rosaria Luciani, Antony Sesenna, Antonio Quotadamo, Stefania Ferrari, Ina Pöhner,
Anabela Cordeiro-da-Silva, Stefano Mangani, Luca Costantino, and Maria Paola Costi
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 25 Mar 2019
Downloaded from http://pubs.acs.org on March 25, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 81
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Costi, Maria Paola; Universita degli Studi di Modena e Reggio Emilia,
Dipartimento di Scienze della Vita
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Enhancement of benzothiazoles as Pteridine Reductase-1 (PTR1) inhibitors for
the treatment of Trypanosomatidic infections.
1*
2
2
2
2
Pasquale Linciano , Cecilia Pozzi , Lucia dello Iacono , Flavio di Pisa , Giacomo Landi , Alessio
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
3
3
3
4
Bonucci , Sheraz Gul , Maria Kuzikov , Bernhard Ellinger , Gesa Witt , Nuno Santarem ; Catarina
4
5
5
6
6
1
Baptista , Caio Franco , Carolina B. Moraes , Wolfgang Müller , Ulrike Wittig , Rosaria Luciani ,
1
1
1
7
4
Antony Sesenna , Antonio Quotadamo , Stefania Ferrari , Ina Pöhner , Anabela Cordeiro-da-Silva ,
2
*
1*
1*
Stefano Mangani ; Luca Costantino , Maria Paola Costi .
1University of Modena and Reggio Emilia, Via Campi 103, 41125 Modena, Italy
2University of Siena, Dipartimento di Biotecnologie, Chimica e Farmacia,Via Aldo Moro 2, 53100 Siena, Italy
3Fraunhofer Institute for Molecular Biology and Applied Ecology Screening Port, 22525 Hamburg, Germany
4Institute for Molecular and Cell Biology, 4150-180 Porto, Portugal and Instituto de Investigação e Inovação em
Saúde, Universidade do Porto and Institute for Molecular and Cell Biology, 4150-180 Porto, Portugal
5Laboratório Nacional de Biociências (LNBio), Centro Nacional de Pesquisaem Energia e Materiais (CNPEM), 13083-
100 Campinas – SP, Brazil; Present Address: Department of Microbiology, Institute of Biomedical Sciences, University
of Sao Paulo, 05508-900 Sao Paulo – SP, Brazil.
6Scientific Databases and Visualization Group, Heidelberg Institute for Theoretical Studies (HITS), D-69118
Heidelberg, Germany.
7Molecular and Cellular Modeling Group, Heidelberg Institute for Theoretical Studies (HITS), D-69118 Heidelberg,
Germany.
Abstract
2
-amino-benzo[d]thiazole was identified as new scaffold for the development of improved PTR1
inhibitors and anti-Trypanosomatidic agents. Molecular docking and crystallography guided the
design and synthesis of 42 new benzothiazoles. The compounds were assessed for T. brucei and L.
1
ACS Paragon Plus Environment

Page 3 of 81
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
major PTR1 inhibition and in-vitro activity against T. brucei and amastigote L. infantum. We
identified several 2-amino-benzo[d]thiazoles with improved enzymatic activity (TbPTR1 IC50
=
0.35 µM; LmPTR1 IC = 1.9 µM) and low µM anti-parasitic activity against T. brucei. The ten
5
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
most active compounds against TbPTR1, were able to potentiate the antiparasitic activity of
methotrexate when evaluated in combination against T. brucei, with a Potentiating Index between
1
.2 and 2.7. The compound library was profiled for early ADME-Toxicity and 2-amino-N-
benzylbenzo[d]thiazole-6-carboxamide (4c) was finally identified as a novel potent, safe and
selective anti-trypanocydal agent (EC50 = 7.0 µM). Formulation of 4c with hydroxypropyl-β-
cyclodextrin yielded good oral bioavailability, encouraging progression to in-vivo studies.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
INTRODUCTION
Kinetoplastidae are a group of flagellated protozoan parasites transmitted by insect vectors and
responsible for severe human pathologies. Human African Trypanosomiasis (HAT; also known as
African sleeping sickness), is caused by Trypanosoma brucei (T. brucei), the Chagas disease is
caused by Trypanosoma cruzi (T. cruzi) and three different manifestations of leishmaniasis
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(visceral, cutaneous and mucocutaneous) are caused by infection with different species of
1
Leishmania. In particular, HAT can arise in an acute or chronic form. Cardiac and kidney diseases,
among other serious complications, can lead to death if untreated. The second stage arises when the
parasite migrates across the blood-brain barrier, resulting in neurological symptoms that ultimately
can lead to coma and death.2
Although treatment options for these infections exist, their use is limited by several factors,
including toxicity, poor efficacy, difficult administration and cost. Furthermore, increasing levels of
drug resistance emphasize the need for new, improved and affordable drugs.3,4 Recently, after thirty
years of research in this field, fexinidazole, a 5-nitroimidazole, has been finally approved as the first
oral drug for the treatment of HAT.5,6
In order to accelerate the discovery of new treatments for Trypanosomatidic infections, one
approach already used for the treatment of bacterial infections and some parasitic diseases (e.g.
malaria) entails the use of drugs targeting folate enzymes (e.g. thymidylate synthase, TS or
dihydrofolate reductase, DHFR).7,8 As a result, the reduction of the cellular pool of deoxythymidine
8
monophosphate, which is essential for DNA replication, causes cell death. Trypanosomatids are
auxotrophic for folates and pterins; therefore, inhibiting the enzymes involved in the folate salvage
9
pathways should provide effective treatment. However, the use of DHFR inhibitors (e.g.
methotrexate (MTX) or pyrimethamine) for the treatment of these parasitic diseases is unsuccessful,
partly due to pteridine reductase
1
(PTR1) activity.10 PTR1 is
a
short-chain
dehydrogenase/reductase (SDR) that can reduce both conjugated and unconjugated pterins. Indeed,
3
ACS Paragon Plus Environment

Page 5 of 81
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
expression of PTR1 is promoted when parasitic DHFR is inhibited, ensuring survival of the
parasite.11–13 The entanglement between parasitic PTR1 and DHFR activity is corroborated by the
limited anti-parasitic efficacy of full PTR1 inhibitors developed to date,14–19 and by the necessity to
co-administer a DHFR inhibitor to observe a trypanocidal activity.15,17,20 Recently, T. brucei PTR1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
TbPTR1) inhibitors based on structural elaboration of 2-amino-1,3,4-thiadiazoles were proposed.17
The new compounds showed only micromolar activity against TbPTR1 and potentiated the activity
of MTX when used in combination24.
A double aromatic/heteroaromatic ring system was deemed necessary to implement the design of
the thiadiazole core. A two-ring system is expected to provide improved PTR1 inhibitors with
increased affinity for the protein, thus competing more strongly with the substrate pterin moiety.
Since 2010, a number of X-ray crystal structures have been made solved in which two
aromatic/heteroaromatic ring systems were bound to PTR1 as protein inhibitors mapping the active
site.17,21–24 From our preliminary studies on compound repositioning, riluzole and other 2-amino-
1
5
benzothiazole derivatives exhibited low micromolar affinity against PTR1. The aim of the present
work was to enlarge the benzothiazole library through a structure-based approach, to characterize
the biological activity against PTR1 and Trypanosomatidic parasites and evaluate the early ADME-
Toxicity profile to select at least one compound for progression to pharmacokinetic evaluation in
animal assays.
X-ray crystal structures with 2-amino-benzothiazole that could reveal the precise core active-site
interactions were not available. Therefore, we first obtained the X-ray crystal structure of (6-
(methylsulfonyl)benzo[d]thiazol-2-amine) with TbPTR1 and, based on this structure, we designed
and synthesized a library of 42 compounds. The selection and prioritization of compounds for in-
vivo evaluation was performed following the criteria considered for the Target Inhibitor Profile
(
TIP) 25,26 resulting in one compound selected for pharmacokinetic studies.
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS AND DISCUSSION
TbPTR1– NADP(H)–3a crystal structure
The chemical starting point of the present study relied on the benzothiazole 1b and 3a that
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
5
originated from previous virtual screening studies. As a result of their moderate inhibitor activity
against T. brucei (IC50 = 34.2 µM and 15.1 µM for 3a and 1b, respectively) and high ligand
efficiency (LE = 0.32 for both 3a and 1b), the two fragments, together with the 2-amino-
benzothiazole core scaffold, were considered to be important sub-structures at the outset of this
work.
Figure 1. Chemical structures and enzymatic inhibitor activities against T. brucei and Leishmania
major (L. major) PTR1 of the benzothiazoles fragments 1b and 3a.
Previous docking studies, performed on 1b and 3a against L. major PTR1 (LmPTR1) showed
binding at the active site with the benzothiazole core sandwiched between the cofactor nicotinamide
and phenylalanine, forming an array of hydrogen bonds with catalytic residues and the cofactor
phosphate and ribose.15 To provide crucial information that could assist the design of improved
inhibitors, crystallographic studies were attempted. 1b and 3a were used for crystallization trials
with TbPTR1 and the structure of the ternary complex TbPTR1–NADPH–3a was successfully
obtained.
5
ACS Paragon Plus Environment

Page 7 of 81
Journal of Medicinal Chemistry
1
2
3
4
5
The structure of TbPTR1 in complex with the cofactor NADP(H) and compound 3a (IC of 34.2
5
0

M, PDB ID 6GCL) was determined to 1.95 Å resolution (Figure 2). The crystal asymmetric unit
includes the whole functional TbPTR1 tetramer whose structure is highly conserved with those
previously described in the literature17,24,27 (subunits were completely traced apart from two surface-
exposed loops, including residues 104-112 and 143-15, that are usually poorly visible in TbPTR1
structures). The 2-amino-benzothiazole core of 3a occupies the catalytic cavity, as previously
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
15
predicted by molecular modelling, in which it exploits its extended aromatic system to generate
the peculiar π-sandwich interaction within Phe97 and the cofactor nicotinamide. A network of H-
bonds entailed by the amine moiety, donating two hydrogen bonds to the cofactor β-phosphate,
stabilizes the scaffold. In addition, hydroxyl of Ser95 and benzothiazole nitrogen both receives an
H-bond from the hydroxyl of the cofactor ribose. The sulfonic group of 3a is located in the active
site pocket lined by the side chains of Cys168, Val206, Leu209, Pro210, Met213, and Trp221.
Figure 2. Active site view of TbPTR1 (white cartoon, active site residues in sticks with white
carbons) in complex with the cofactor NADP(H) (in sticks, black carbons) and 3a (in sticks,
purple carbons). The inhibitor is surrounded by the omit map (green wire) contoured at the 2.5 
level. H-bonds are represented as red dashed lines. Atom color code: nitrogen (blue), oxygen
(red) and sulfur (yellow). (PDB ID: 6GCL)
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Benzothiazoles design
In accordance with the binding mode of the 2-amino-benzothiazole fragment observed in the X-ray
structure of TbPTR1-NADP(H)-3a and the architecture of the parasitic TbPTR1, new PTR1
inhibitors based on the benzothiazole scaffold were designed and synthesized. The catalytic pocket
of TbPTR1 is located in the deepest part of the active site and it is delimited by Phe97, Ser95,
Arg14, Asp161, Leu209, Pro210, Tyr174 and the nicotinamide moiety of NADPH. The entrance of
the pocket is a hydrophobic tunnel (hydrophobic pocket 1, Figure 3) formed by Val206, Trp221 and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Met213 where the p-amino benzoic acid (pABA) ring of a folic acid substrate is accommodated
28
(
see PDB: 3BMC ). Next to hydrophobic pocket 1, a second hydrophobic pocket is present (Figure
3
), delimited by Cys168, Val206, Phe97 and closed on the top by Phe171.
Figure 3. TbPTR1 active site. The catalytic, primary and secondary hydrophobic pockets are
highlighted. The surface of the protein is colored based on the hydrophobicity of the amino acid
residues (from red: hydrophobicity to white: hydrophilicity). The relevant residues delimiting the
pocket and NADP(H) are displayed as sticks. (PDB ID: 6GCL)
7
ACS Paragon Plus Environment

Page 9 of 81
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
An internal virtual library of around 500 synthetically accessible benzothiazole derivatives was
rationally designed in a structure-based approach. The crystal structure of 3a was taken as a
template for the design; the 2-amino-benzothiazole core scaffold anchors the molecule within the
catalytic pocket of the TbPTR1 active site and during the design it was retained unaltered and
decorated as reported in Figure 4. The compounds were designed based on the following
considerations. Aliphatic or aromatic side chains were appended in position 6 of the benzothiazole
ring to probe the hydrophobic pocket 1 or the secondary hydrophobic cleft (Figure 4A). The side-
chain on C-6 of the aromatic substituent was linked to the main core scaffold by means of four
different linkers, varying in flexibility, polarity, lipophilicity and length (e.g. ether, sulfide, sulfone
and amide) as shown in Figure 4B. Moreover, more sterically hindered amidic derivatives were also
explored to simultaneously fill both hydrophobic pockets (Figure 4C). Finally, hydroxyl groups
were introduced at carbon 6 and carbon 7 on the benzothiazole core scaffold to explore the H-bond
interactions with the residues of the active site (Figure 4D). The diverse substituents were properly
combined to generate a library that was screened by molecular docking with Autodock 4.2 on the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
JQ7 crystal structure (see Supporting Information).28,29
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Main structural components of the 500 compounds virtual library based on the 2-
amino-benzothiazoles core.
The docking model was first validated by re-docking the DX2 inhibitor21 into the parent crystal
structure with the resulting RMSD being within 0.83Å. Thereafter, the virtual benzothiazole based
library was docked and docking poses were filtered first according to the predicted scores (ΔGbinding
ranging from –6 to –9 kcal/mol). The ligand-protein complexes were then visually inspected and
compounds able to reproduce the binding pattern of the 2-amino-benzothiazole core scaffold
observed in the TbPTR1– NADP(H)–3a crystal structure were selected for synthesis (series 1-4,
Table 1-4). Figure 5 shows the most reliable predicted binding mode of the designed benzothiazoles
addressed by the synthesis.
9
ACS Paragon Plus Environment

Page 11 of 81
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Predicted binding mode by docking calculation against TbPTR1 of some representative
designed benzothiazoles. A) Predicted pose of 2e (green carbon) reproducing the binding mode of
2
-amino-benzothiazole in the catalytic pocket and showing the placement of the aromatic side chain
+
in the hydrophobic pocket 1. B) Superimposition of the ternary complex NADP –TbPTR1–Folate
(PDB ID: 3BMC, folic acid in green carbons) and the predicted binding mode of 4g (in orange
carbons). C) Superposition of the predicted binding mode of four homologous benzothiazoles
differing only for the linker (1c, 2c, 3c and 4c respectively in magenta, cerulean, green and yellow
carbons). D) Predicted binding mode of hydroxylated benzothiazole 5b (in magenta carbons).
Model atoms except for carbons are color coded with protein carbons (white), oxygen (red),
nitrogen (blue), sulfur (yellow) and phosphorous (orange). The protein backbone is shown as white
cartoon.
1
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
2 molecules with the best docking scores and poses were synthesized (Table 1-5). The four
homologous series, in accordance with the linker type, were divided in etheric (1a, c-g, Table 1),
sulfide (2a-h, Table 2), sulfone (3a, c-g, Table 3) and amide (4c-h, Table 4) benzothiazoles.
Derivatives 1h, 3b and 4a-b could not be synthesized. Assuming no more than one violation to the
rule of 5, (molecular weight, number of H-bond donor and acceptor, number of rotatable bonds, log
Po/w and polar surface area), all the designed compounds were in accordance with Lipinski’s rule
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Table SI-1). The compounds passed the check for pan-assay interference compounds (PAINS)
evaluated with the in silico tool FAFdrugs4.30,31
Chemistry
Etheric benzothiazoles 1a, c-g were prepared as reported in Scheme 1. 6-hydroxy-2-
aminobenzothiazole (6) was synthesized first through condensation between benzoquinone and
thiourea in refluxing ethanol in presence of concentrated hydrochloric acid as catalyst. 6 was further
reacted in standard S 2 condition with iodomethane (to prepare 1a) or with the appropriate benzyl-
N
bromide to obtain 1c-f. 1g was synthesized by hydrolyzing 1f with 1N NaOH in THF:MeOH 3:1 at
room temperature for 24 hours. The carboxylic acid thus obtained was further reacted with methyl
isonipecotate via standard coupling condition with EDC, HOBt in DMF at room temperature for 24
hours to give 1g.
O
N
S
N
S
S
a
b
NH2
NH2
R
H N
2
NH2
O
HO
O
1a R: CH3-
1c R: C H -CH -
6
6
5
2
1d R: (3,4-Cl-C H )-CH -
6
3
2
1e R: (4-CN-C H )-CH -
6
4
2
1f R: (4-(COOCH )-C H )-CH -
3
6
4
2
c,d
1g R: (4-(4-(COOCH )-C H -N)CO-C H )CH -
3
5
9
6
4
2
Scheme 1. Reagents and conditions: (a) HCl conc. (0.1 eq.), EtOH, refl., 6 h. (b) K CO (2.5 eq.),
2
3
DMF, r.t., 12 h. (c) 1N NaOH aq. (1.5 eq.), THF:MeOH 3:1, r.t., 24 h. (d) methyl isonipecotate (1
1
1
ACS Paragon Plus Environment

Page 13 of 81
Journal of Medicinal Chemistry
1
2
3
eq.), EDC (1 eq.), HOBt (1 eq.), DMF, 0°C to r.t, 24 h.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Synthesis of 6-(sulfone)-2-aminobenzothiazoles 3a, c-g was performed as reported in Scheme 2,
passing by the synthesis of the relative sulfides 2a-h. 4-nitrothiophenol was reacted first with the
appropriate benzyl bromide and K CO as base in DMF to afford 4-nitro-S-benzylbenzenes 7a-f
2
3
that were subsequently reduced to the respective anilines 8a-f with zinc and ammonium chloride in
refluxing methanol. Anilines 8a-f were then cyclized to benzothiazoles 2a-f with potassium
thiocyanate in glacial acetic acid using elemental bromine as oxidizing agent. For the synthesis of
sulfides 2g-h, 2f was hydrolyzed first to carboxylic acid and consequentially reacted in standard
coupling condition with methyl-isonipecotate (to give 2g) or benzylamine (to give 2h). Sulfides 2a-
g thus obtained were further oxidized to the respective sulfones 3a, c-g with m-CPBA in DCM at
room temperature for 24 hours. It was not possible to synthesize sulfones 3b and 3h due to
decomposition of the precursors 2b and 2h in presence of oxidizing agents.
NO2
NO2
NH2
a
b
R
R
HS
S
S
7a-f
8a-f
c
2
2
2
a: R: CH3-
b: R: CF3-
c: R: C H -CH -
N
6
5
2
2d: R: (3,4-Cl-C H -)CH -
2e: R: (4-CN-C H -)CH -
NH2
d,e
6
3
2
R
S
6 4 2
S
2
2
2
f: R: (4-(COOCH )-C H )CH -
g: R: (4-(4-(COOCH )-C H -N)CO-C H )CH -
h: R: (4-(C H CH NH)CO-C H )CH -
3 6 4 2
3
5
9
6
4
2
6
5
2
6
4
2
f
3
a: R: CH3-
N
3c: R: C H -CH -
NH2
6 5 2
R
3d
:
R: (3,4-Cl-C H -)CH -
6 3 2
S
S
3e: R: (4-CN-C H -)CH -
6
4
2
O O
3
f: R: (4-(COOCH )-C H )CH -
3 6 4 2
3
g: R: (4-(4-(COOCH )-C H -N)CO-C H )CH -
3
5
9
6
4
2
Scheme 2. Reagents and conditions: (a) K CO (2.5 eq.), DMF, r.t., 24 h. (b) Zn (2 eq.), NH Cl
2
3
4
1
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
2.5 eq.), MeOH, rifl. 2 h. (c) Br (2 eq.), KSCN (4 eq.), glacial acetic acid, T < 10 °C, 1-6 h. (d)
2
1
N NaOH aq. (1.5 eq.), THF:MeOH 3:1, r.t., 24 h. (e) methyl isonipecotate (1 eq.) or
benzylamine (1 eq.), EDC (1 eq.), HOBt (1 eq.), DMF, 0°C to r.t, 24 h. (f) m-CPBA (2 eq.) ,
DCM, 0°C → r.t., 24 h.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Synthesis of 6-amide-benzothiazoles 4c-r was performed as depicted in Scheme 3. Commercially
available 2-amino-6-carboxybenzothiazole was reacted with the appropriate amine 9c-f, i-j, l-r
through standard coupling conditions with EDC and HOBt as condensing agents, in DMF at room
temperature for 24 hours. Amines 9c-f, i-j, l-o were commercially available and directly used.
N
S
N
S
R
R1
a
NH2
R
NH2
HO
+
N
1N
R
H
O
O
9c-f, i-j, l-r
1
4c: R: C H -CH -; R : H-
6
5
2
1
4d: R: (3,4-Cl-C H -)CH -; R : H-
6
3
2
1
4e: R: (4-CN-C H -)CH -; R : H-
6
4
2
1
4f: R: (4-(4-(COOCH )-C H -)CH -; R : H-
3
6
4
2
b,c
1
4g: R: (4-(4-(COOCH )-C H -N-)CO-C H )CH -; R : H-
3
5
9
6
4
1
2
4h: R: (4-(C H CH NH)CO-C H )CH -; R : H-
6
5
2
6
4
2
1
4i: R: (3,4,5-(CH O-)C H -)CH -; R : H-
3
6
2
2
1
4
4
4
4
4
4
4
4
4
j: R: (2-NO -C H -)CH -; R : H-
2 6 4 2
d
1
k: R: (2-NH -C H -)CH -; R : H-
2
6
4
2
1
l: R: bis-(C H -)CH-; R : H-
6
5
1
m: R: 4-(4-(COOCH )-C H -N-; R : H-
3
5 9
1
n: R: 2-(2'-thiophenyl-)CH CH -; R : H-
2
2
1
o: R: 2-(3'-indolyl-)CH CH -; R : H-
2
2
1
p: R: (3,4-Cl-C H -)CH -; R : C H -CH -
6
3
2
6
5
2
q: R: C H -CH -; R : HO-CH CH -
6
5
2
1
2
2
1
r: R: (3,4-Cl-C H -)CH -; R : 3,4-(CH O-)C H -)CH -;
6
3
2
3
6
2
2
Scheme 3. Reagents and conditions: (a) EDC (1 eq.), HOBt (1 eq.), DMF, 0°C to r.t, 24 h; (b) 1N
NaOH aq. (1.5 eq.), THF:MeOH 3:1, r.t., 24 h. (c) methyl isonipecotate (1 eq.) or benzylamine (1
eq.), EDC (1 eq.), HOBt (1 eq.), DMF, 0°C to r.t, 24 h; (d) Zn (2 eq.), NH Cl (2.5 eq.), MeOH,
4
refl. 6 h.
1
3
ACS Paragon Plus Environment

Page 15 of 81
Journal of Medicinal Chemistry
1
2
3
Contrary, amines 9p-r were synthesized first through S 2 reaction between the appropriate
N
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
benzylamines and alkylbromides (Scheme 4).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R2
NH2
N
a
H
2
+
R Br
R
R
R1
R1
1
2
9
9
9
p: R: Cl-; R : Cl-; R : C H -CH -
6
5
2
1
2
q: R: H-; R ; H-; R ; HO-CH CH -
2
2
1
2
r: R: Cl-; R : Cl-; R : 4-(CH O-)C H -)CH -
3
6
2
2
2
Scheme 4. Reagents and conditions: (a) benzylamine (2.5 eq.), alkylbromide (1.0 eq.), K CO3
(2.5 eq.), DMF, r.t., 1-6 h.
6
-amido-4-hydroxy-benzothiazoles 4s-t were prepared as depicted in Scheme 5. 2-amino-6-
hydroxy-benzothiazol-6-carboxylic acid 10 was synthesized first by cyclization of the respective
aniline with potassium thiocyanate and bromine, in glacial acetic acid as previously described. 10
was further condensed with benzylamine (to give 4s) or 3,4-dichlorobenzylamine (to give 4t) in
standard coupling conditions (Scheme 5).
OH
OH
OH
NH2
a
N
S
b
R
R
N
S
NH2
H
NH2
HO
HO
N
O
O
O
10
4s R:H-
t R: Cl-
4
Scheme 5. Reagents and conditions: (a) Br (2 eq.), KSCN (4 eq.), glacial acetic acid, T < 10 °C,
2
1
-6 h. (b) benzyl amines (1 eq.), EDC (1 eq.), HOBt (1 eq.), DMF, 0°C to r.t, 24 h.
1
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Finally, for the synthesis of the catecholic 4-amido-benzothiazoles 5a-c, the benzothiazole core was
synthesized first starting from commercially available methyl 2-amino-4,5-dimethoxybenzoate by
reaction with potassium thiocyanate and bromine, in glacial acetic acid as previously describe. The
intermediate 11 was hydrolyzed to carboxylic acid 12 with aqueous NaOH and further condensed
with the appropriate benzyl amine by coupling reaction. The methoxy groups of intermediates 13a-c
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
were deprotected with BBr in DCM under nitrogen atmosphere to give the final products 5a-c
3
(Scheme 6).
O
O
O
O
O
OH
NH2
b
a
N
S
N
NH2
NH2
S
O
O
O
O
O
O
11
12
c
R1
N
R1
O
O
N
R
N
R
N
d
NH2
NH2
S
HO
S
O
OH
O
1
5
5
5
a R: (3,4-Cl-C H -)CH -; R : C H CH -
6
4
2
6
5
2
13a-c
1
b R: (3,4-Cl-C H -)CH -; R : H-
6
4
2
1
c R: C H CH -; R : H-
6
5
2
Scheme 6. Reagents and conditions: (a) Br (2 eq.), KSCN (4 eq.), glacial acetic acid, T < 10 °C,
2
1
-6 h. (b) 1N NaOH aq. (1.5 eq.), THF:MeOH 3:1, r.t., 24 h. (c) benzyl amines (1 eq.), EDC (1
eq.), HOBt (1 eq.), DMF, 0°C to r.t, 24 h. (d) BBr 1M in DCM (6 eq.), dry DCM, N , 0 °C to
3
2
r.t., overnight.
1
5
ACS Paragon Plus Environment

Page 17 of 81
Journal of Medicinal Chemistry
1
2
3
Relationships between the structural properties, the enzymatic activity and the binding mode
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to TbPTR1.
All 42 new benzothiazoles 1a-g, 2a-h, 3a,c-g, 4c-t and 5a-c were evaluated for their inhibitory
activity against T. brucei and L. major PTR1, initially at 50 µM and those compounds yielding > 60
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
%
inhibition were profiled in dose-response experiments to determine their potency. The results are
reported in Tables 1-5 and pyrimethamine was routinely utilised as a positive control compound
TbPTR1 IC 0.09 µM and LmPTR1 IC 13.6 µM). Almost all the synthesized compounds showed
(
50
50
a measurable IC (< 100 µM) in the range of 0.3 – 57.6 µM against TbPTR1 and 1.9 – 94 µM
5
0
against LmPTR1. Eight compounds against TbPTR1 (1a, 2a, 3d-f, 4i,s-t) and 17 compounds against
LmPTR1 (1a, c-d, 2a, 3c-f, 4g, i, m-r, t) yielded IC > 100 µM and therefore were considered
5
0
inactive. Overall, the compounds resulted from 2 to 100-fold more active against TbPTR1 than
LmPTR1. Only compounds 3a, 4j and 5a-c showed comparable activity against both PTR1
enzymes, whereas the only inversion in tendency was observed for 4s showing LmPTR1 IC = 2.48
5
0
µM but no inhibitory activity against TbPTR1. The entire structure activity relationship is resumed
in Figure 6. In addition, the LE against TbPTR1 was calculated for all the synthesized compounds.
LE, instead of IC , removes the bias due to the increased size of the ligands, highlighting the
5
0
structural modifications favourable or deleterious for the inhibitor. Focusing on TbPTR1, the
elongation of the 2-amino-benzothiazole main scaffold in a folic acid substrate-like shape resulted
in a worthwhile strategy. The introduction of an aromatic ring as a side chain, able to mimic the
pABA ring of folic acid, led to low/sub micromolar TbPTR1 inhibitors. Docking calculations
predicted the side aromatic ring to be mainly placed in hydrophobic pocket 1, forming face-to-face
and edge-to-face π-π interactions with Trp221 or Phe97, respectively. On the contrary, compounds
1
-3a or 4m, possessing a methyl group or a non-aromatic ring, respectively as the side chain, were
completely inactive. The only exceptions were compounds 1b and 2b possessing a CF group as
3
the side chain (TbPTR1 IC = 15.1 and 0.50 µM for 1b and 2b, respectively). In particular, the
5
0
1
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
replacement of the oxygen atom of the starting benzothiazole 1b with a sulphur atom in compound
b resulted in a 30-times improvement of the inhibitor activity against TbPR1. The efficacy of this
2
chemical modification is reflected in the meaningful improvement in LE for 2b (LE = 0.42)
compared to 1b (LE = 0.32). The role of the linker was subsequently analysed. Sulfide derivatives
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
TbPTR1 IC = 0.35 – 1.9 µM), followed by etheric derivatives (TbPTR1 IC = 0.67 – 15.1 µM)
5
0
5
0
resulted in the most active compounds of the library, compared to the homologous sulfone and
amide derivatives. The efficacy and the importance of the linker is reflected in the overall range of
LE of each subseries of compounds. Comparing benzothiazoles differing only for the linker
between the main core scaffold and the side chain, the higher LE was always observed for sulphide
(LE = 0.20 – 0.42) and etheric derivatives (LE = 0.18 to 0.32), followed by sulfones (LE = 0.14 –
0
.25) and amides (LE = 0.15 – 0.26). The introduction of small polar substituents on the aromatic
ring (e.g. nitrile, nitro, amino and methyl ester) was always tolerated (TbPTR1 IC 1e = 0.67 µM;
5
0
1
f = 1.7 µM; 2e = 0.35 µM; 2f = 0.90 µM; 3e = 3.00 µM) resulting in a LE ranging between 0.25
for 3e) and 0.31 (for 1e). On the contrary, bulky side chains (e.g. phenyl ring in 4l, indole in 4o or
trimethoxy in 4i) led to a drop both in activity and LE (< 0.19).
(
1
6
In our previous work we stated that the substitution of folate glutamate with the cyclic analogous
methyl isonipecotate was well tolerated for inhibitory activity against PTR1s. This structural
modification was therefore introduced on 2-amino-benzothiazole scaffolds to closely match the
structures of folic acid, resulting in compounds 1-4g and 2,4h. The improvement in activity for
these derivatives was only modest, as suggested in addition by a LE < 0.20, probably due to the
exposition of the folic acid side chain to the solvent (TbPTR1 IC 1g = 3.3 µM; 2g = 0.93 µM; 3g =
5
0
2
4.3 µM; 4g = 21.8 µM).
The introduction of a hydroxyl group in position 4 on the benzothiazole ring (4s-t) resulted in a
significant reduction in inhibitory activity against TbPTR1. From docking calculation, the hydroxyl
group appears to force the benzothiazole ring to assume an unusual pose within the active site. The
1
7
ACS Paragon Plus Environment

Page 19 of 81
Journal of Medicinal Chemistry
1
2
3
scaffold is flipped by 180° with the intra-ring sulphur pointing toward Tyr174, and the
benzothiazole displaced from the standard stacking geometry between Phe97 and NADP(H).
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
All the tested compounds were twice to 100-times less active against LmPTR1 than TbPTR1
reflecting the efficacy of the design which was mainly performed on the architecture of the TbPTR1
enzyme. Even though an overall comparable SAR among the two enzymes can be observed, these
differences are presumable due to the structural diversity of the distal region of the binding site of
the two enzymes. Contrary to TbPTR1, LmPTR1 possesses a more polar and solvent-exposed active
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
site entrance surface. Thus, the elongation of the 2-amino-benzothiazole main scaffold in a folic
acid substrate-like shape, pointed the side chain toward a polar area, preventing the capability of the
side benzyl ring to form π-π stacking or additional hydrophobic contacts (as in TbPTR1), probably
explaining the observed lower activity against LmPTR1 compared to TbPTR1.
Ramification, with a second aromatic ring,
is well tolerated
Introduction of hydroxyl in position-4
causes loss of the activity
R2
R1
N
S
NH2
X
Small substituents (-Cl, -CN, -COOCH ...) are
well tolerated.
3,
R
The elongation of the molecule in a pABA-like
shape enhances the activity.
Ether (-O-) and sulphide (-S-) linkers enhance the activity
Amide (-CONH-) linkers retain the activity
Sulfone (-SO2-) linkers cause drop of the activity
Aromatic rings are essential for activity allowing  stacking
with Phe97 and/or Trp221. Aliphatic moieties are unfavorable,
with the only noteworthy exception of -CF3.
Figure 6. Schematic highlights of the structure-activity relationship of the synthesized
compounds against TbPTR1 and LmPTR1. All the inhibitors resulted more active against
TbPTR1 than LmPTR1. The above reported structure-activity considerations are consistent for
both enzymes.
A notable exception is represented by three compounds of series 5 (5a-c), showing comparable
inhibitory activity against both enzymes. From docking calculations, these sub-classes of
1
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
benzothiazoles could assume a different binding mode (Figure 5D) compared to the poses of series
-4, allowing them to adapt to both TbPTR1 and LmPTR1 binding sites.
1
Comparison between crystallographic ternary complexes of the inhibitors with TbPTR1 and -
NADP(H) with the docked binding modes.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compounds showing IC50 values lower than 1 M and a low toxicity and seven compounds
belonging to series 4 bearing a high structural diversity were subjected to structural studies. The
structure of TbPTR1 was solved in complex with the cofactor NADP(H) and nine benzothiazole
derivatives (1e, 2b, 2d, 2g, 2h, 4c, 4d, 4g and 4l) showing IC50 values ranging from 0.5 M
(compounds 2b and 2h) to 21.8 M (compound 4g). In all structures, described in Table 6, the
shared benzothiazole core of the inhibitors adopts a highly conserved binding mode, retaining also
the same interaction patterns described for the forefather 3a.
1
9
ACS Paragon Plus Environment

Page 21 of 81
Journal of Medicinal Chemistry
1
2
3
Table 6. Crystal structures of TbPTR1 in complex with NADP(H) and the benzothiazole
compounds. A brief description of the main interactions entailed by the inhibitors in the catalytic
pocket is reported (the shared benzothiazole moiety forms conserved interactions in the cavity, as
already described for the forefather 3a). The picture column reports an active site view of TbPTR1
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(white cartoon and carbon atoms) in complex with the cofactor NADP(H) (in sticks, black carbons)
and 1e (in sticks, green carbons), 2b (in sticks, cyan carbons), 2d (in sticks, light blue carbons), 2g
(in sticks, brown carbons), 2h (in sticks, lilac carbons), and 4c (in sticks, orange carbons), 4d (in
sticks, dark green carbons), 4g (in sticks, dark red carbons), and 4l (in sticks, pink carbons).
Inhibitors are surrounded by their omit map (green wire) contoured at the 2.5  level. Water
molecules are represented as red spheres and H-bonds as red dashed lines. Atom color code:
nitrogen (blue), oxygen (red), sulfur (yellow), phosphorous (magenta), chlorine (dark grey) and
fluorine (grey).
Complex
Picture
Description


PDB: 6GCK
Resolution: 2.14 Å

4-cianophenyl moiety of 1e (IC of 0.67
5
0
M) is stabilized in pocket 1 by van der
Waals interactions with Phe97, Leu209,
TbPTR1 –
NADP(H) – 1e
Pro210,
Met213,
and
Trp221
(interactions shared within the tetramer).

1e populates four subunits of the
tetramer (100 % occupancy).
2
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6


PDB: 6GCQ
Resolution: 1.58 Å

Trifluoromethyl moiety of 2b (IC50 of
0
.5 M) is accommodated in the active
TbPTR1 –
NADP(H) – 2b
site pocket 1 (halogens directed towards
Phe97, Pro210, and Trp221 in all
enzyme subunits).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0


2b populates four subunits of the
tetramer (70-100% occupancy).
PDB: 6GCP
Resolution: 1.52 Å

The 3,4-dichlorophenyl moiety of 2d
IC of 0.9 M), stabilized in pocket 1
(
5
0
by van der Waals interactions with
Phe97, Leu209, Pro210, Met213, and
Trp221, is observed in two alternate
orientations (named 2d-A and 2d-B,
mutually rotated by 180°).
TbPTR1 –
NADP(H) – 2d

In chain B, only one conformation
(having the m-chlorine facing Pro210) is
visible.


2d populates four subunits of the
tetramer (90% occupancy).
PDB: 6GDO
Resolution: 1.78 Å

The molecule of 2g (IC of 0.93 M) is
5
0
completely rebuilt only in chain D,
whereas in chain C, only the cofactor is
visible (60 % occupancy).
TbPTR1 –
NADP(H) – 2g

Phenyl ring of 2g is stabilized in pocket
1
by van der Waals interactions with
Phe97, Leu209, Pro210, Met213, and
Trp221.

Piperidine moiety of 2g forms
hydrophobic interactions with Phe171
and Pro99.

2g populates three subunits of the
2
1
ACS Paragon Plus Environment

Page 23 of 81
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
tetramer (80% occupancy).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0


PDB: 6GEX
Resolution: 1.78 Å

The molecule of 2h (IC of 0.5 M) is
5
0
completely rebuilt only in chain D
highly flexible terminal phenyl ring)
(
whereas in chain C, only the cofactor is
detected (60 % occupancy).
TbPTR1 –
NADP(H) – 2h

Phenyl ring of 2g is stabilized in pocket
1 by van der Waals interactions with
Phe97, Leu209, Pro210, Met213, and
Trp221.
2h populates three subunits of the
tetramer (80% occupancy).


PDB: 6GD4
Resolution: 1.42 Å

The phenyl moiety on the variable tail of
4
c (IC50 of 10.0 M) is highly flexible
(excluded from our model, only
molecular core and amide linker of the
inhibitor are traced).
TbPTR1 –
NADP(H) – 4c


Cys168 is modified as S-oxy-cysteine
(CSX168) in the four subunits of the
tetramer.
4c populates four subunits of the
tetramer (50-80% occupancy).


6GEY
Resolution: 1.56 Å

The 3,4-dichlorophenyl of 4d (IC50 of
TbPTR1 –
NADP(H) – 4d
5
.4 M) is stabilized in the hydrophobic
pocket 1 by van der Waals interactions
with Phe97, Leu209, Pro210, Met213,
and Trp221.
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6


Different orientations of the 3,4-
dichlorophenyl moiety are observed
within the tetramer (Figure SI-1).
4d populates four subunits of the
tetramer (60-90% occupancy).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0


PDB: 6GD0
Resolution: 1.74 Å

The molecule of 4g (IC of 21.8 M) is
5
0
completely rebuilt only in chain A.

The phenyl moiety of 4g is stabilized in
pocket 1 by van der Waals interactions
with Phe97, Val206, Leu209, Pro210,
Met213, and Trp221.
TbPTR1 –
NADP(H) – 4g
The amide nitrogen of 4g forms a
network of water mediated interactions
within the active site.

4g populates three subunits of the
tetramer (90% occupancy).


PDB: 6GPD
Resolution: 1.52 Å

Two alternate states are observed in the
catalytic cavity: ligand-bound (aperture
of Trp221, A-conf) and ligand-free
(closure of Trp221, B-conf) states
(estimated to 50 %).
TbPTR1 –
NADP(H) – 4l
Phenyl moieties of 4l (IC50 of 18.5 M)
occupy the hydrophobic pockets 1 (van
der Waals interaction with Trp221 and
Met213) and 2 (Figure 3).

The amide nitrogen of 4l forms water
mediated interactions with Asp161 and
Gly205 (backbone oxygen).
4l populates three subunits of the
tetramer (50% occupancy).
2
3
ACS Paragon Plus Environment

Page 25 of 81
Journal of Medicinal Chemistry
1
2
3
The analysis of the ten ternary X-ray complexes (Figure 2 and Table 6) revealed a conserved
binding mode and some common interactions of the ten inhibitors within TbPTR1. In all ten
complexes, the benzothiazole core is always located within the catalytic pocket of TbPTR1,
interacting with the four key amino acids (Ser95, Phe97, Asp161 and Trp221) and NADP(H), as
previously described (Figure 7). The main contribution to binding originates from the π-stacking
between the nicotinamide moiety of NADP(H) and Phe97 and H-bonds with phosphate or ribose
hydroxyls of the cofactor. An additional conserved H-bond interaction is always observed between
the amino group in position 2 on the benzothiazole ring and the hydroxyl group of Ser95.
Additional weak electrostatic interaction with Asp161 was observed for sulfone (for 3a) or amide
(for 4d,g,l) benzothiazoles. Moreover, for all the inhibitors possessing an aromatic ring as the side
chain, mimicking the pABA moiety of the folate substrate (1e, 2d,g,h, 4g,l), the side-chain is
always located within the hydrophobic pocket 1 stabilized by van-der-Waals (vdW) interactions
with Leu209, Pro210, Met213, and Trp221, and mainly by face-to-edge π-π stacking with Phe97
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Figure 7).
2
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 7. Common binding interactions of benzothiazoles with the main amino acids delimiting
the TbPTR1 active site and NADPH based on the crystallographic ternary complex of compounds
1e, 2b,d,g-h, 3a, 4c-d,g,l.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The binding mode of the ten benzothiazoles crystalized was compared by superimposition with the
respective docking poses (Figure 8B-K). In both X-ray and docked poses, the main 2-amino-
benzothiazole core always shows the same binding mode and orientation within the TbPTR1
catalytic pocket. In addition, the orientation and placement of the side chain are well reproduced for
the inhibitors with small ‘tails’ attached to the main benzothiazole scaffold (such as for compounds
1
e, 2b,d, 3a, 4c,d,l). On the contrary, for compounds 4g (Figure 8D), 2g (Figure 8J) and 2h (Figure
8K) the distal part of the longer side chain is not well reproduced, since it points outside the
confined TbPTR1 pocket and can assume multiple conformations. An overall good correlation
between the TbPTR1 pIC50 of all inhibitors and the respective Autodock Docking Score was
observed (Figure 8A, Table SI-4), revealing the capability of the docking protocol and the design
and selection strategy adopted to foresee valid potential TbPTR1 inhibitors.
2
5
ACS Paragon Plus Environment

Page 27 of 81
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. (A) Correlation between the enzymatic inhibitor activity against TbPTR1 (expressed as
pIC ) and the Autodock Score. Colored dots correspond to the inhibitors for which the X-ray
50
complex is available. The color of the dots is in accordance with the chemical series: ether (blue),
sulfide (yellow), sulfone (red), amide (green). (B-K) Superimposition between the binding mode
observed by X-ray crystallography and the respective pose predicted by docking calculation for
inhibitors 4c (B, carbon in light orange for X-ray pose and in red for docked pose), 4d (C, carbon
in light green for X-ray pose and in dark green for docked pose), 4g (D, carbon in yellow for X-
ray pose and in orange for docked pose), 4l (E, carbon in purple for X-ray pose and in violet for
docked pose), 3a (F, carbon in purple for X-ray pose and in violet for docked pose), 1e (G, carbon
in light green for X-ray pose and in dark green for docked pose), 2b (H, carbon in cerulean for X-
ray pose and in blue for docked pose), 2d (I, carbon in cerulean for X-ray pose and in blue for
docked pose), 2g (J, carbon in purple for X-ray pose and in violet for docked pose), 2h (K, carbon
in yellow for X-ray pose and in orange for docked pose). The X-ray poses are colored in lighter
tone and the docked pose in a darker tone. Color code: nitrogen (blue), oxygen (red), sulfur
(yellow), phosphorous (magenta), chlorine (dark green), fluorine (green).
2
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 81
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Anti-parasitic activity as single agents and in combination
All synthesized benzothiazoles were evaluated first for in-vitro anti-parasitic activity against
cultured bloodstream forms of T. brucei. To evaluate a potential broad-spectrum anti-
trypanosomatidic activity all the compounds were in addition screened against the amastigote form
of Leishmania infantum (L. infantum) and trypomastigote form of T. cruzi, which reproduce a more
physiological and disease-relevant model. Human leukemic monocyte cell line THP-1 was infected
with L. infantum while osteosarcoma human U2OS cell-line was infected with trypomastigote
forms of the Y strain of T. cruzi. Compounds were screened at 10 µM against T. brucei and
intracellular amastigote L. infantum, and at 50 µM against intracellular amastigote T. cruzi during a
HTS phenotypic screening campaign that included thousands of compounds. The anti-parasitic
activity was expressed as % of parasitic cell growth inhibition at the defined compound
concentration. For compounds associated with > 60 % cell growth inhibition (against T. brucei and
L. infantum) and > 85 % cell growth inhibition (against T. cruzi), dose-response studies were
performed to determine the EC50 values (Table 1-5). The T. brucei assay relied on indirect
determination of parasite population viability by quantification of total DNA present in the well
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
32
using the Sybr Green I DNA fluorescent dye. Pentamidine was used as the reference compound
exhibiting an EC50 of 3.8 nM which is comparable with the value reported in literature.33
Amphotericin B and miltefosine were utilized as positive controls in the L. infantum assay yielding
EC 1.14 µM and 1.29 µM respectively, while benznidazole was used as reference compound in
5
0
the T. cruzi assay, exhibiting an EC of 2.4 µM, which is comparable with the value reported in
5
0
3
4
literature. Almost all tested benzothiazoles yielded weak anti-T. brucei activity ranging below 20
cell growth inhibition at 10 µM (Table 1-5). Only three compounds (4c, 4r and 4t) showed a %
cell growth inhibition > 60 % at 10 µM resulting in an EC of 7.0 µM (4c), 9.7 µM (4r) and 5.9
%
5
0
µM (4t) respectively. In contrast, no compounds showed an appreciable anti-parasitic activity
2
7
ACS Paragon Plus Environment

Page 29 of 81
Journal of Medicinal Chemistry
1
2
3
against the two intracellular forms of L. infantum and T. cruzi (% cell growth inhibition < 21 % at
0 µM, Table 1-5).
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
As stated before, the effective inhibition of parasite growth can be observed if both PTR1 and
DHFR enzymes are inhibited.17 On this basis, we studied the effect of the PTR1 inhibitors in
combination with MTX, a known DHFR inhibitor, against T. brucei to observe the potentiation
effect that the PTR1 inhibitors have on MTX parasite growth inhibition. With this aim, we selected
some inhibitors showing the highest inhibition effect against TbPTR1, namely, compounds 1d-e,
2b,d-h, 3e and 4c. The selected compounds were combined at 10 M with MTX at 4 M
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
(
corresponding to the EC against T. brucei). Only compound 4c was tested at 1 M, because it
3
0
showed an anti-parasitic activity >80% at 10 M which is an inhibition potency too high to allow
any potentiation effects in combination to be observed. The anti-parasitic activity was expressed as
%
cell growth inhibition and is reported in Figure 9 and Table 7. The potency in combination
studies was estimated through determination of a Potentiating Index (PI) given by eq. 1.
eq. 1
A PI higher than 1 reflects the capacity of the compounds to increase the anti-parasitic activity of
MTX beyond what would be expected by the simple addition of the individual effects (Figure 9 and
Table 7). A PI effect was observed for all compounds except for 2d and 2b. Compounds 1e and 2f
were most potent compounds showing PI >2.
2
8
ACS Paragon Plus Environment