Pyrrolopyrimidine vs Imidazole-Phenyl-Thiazole Scaffolds in Nonpeptidic Dimerization Inhibitors of Leishmania infantum Trypanothione Reductase

Article abstract of DOI:10.1021/acsinfecdis.8b00355

Disruption of protein-protein interactions of essential oligomeric enzymes by small molecules represents a significant challenge. We recently reported some linear and cyclic peptides derived from an α-helical region present in the homodimeric interface of Leishmania infantum trypanothione reductase (Li-TryR) that showed potent effects on both dimerization and redox activity of this essential enzyme. Here, we describe our first steps toward the design of nonpeptidic small-molecule Li-TryR dimerization disruptors using a proteomimetic approach. The pyrrolopyrimidine and the 5-6-5 imidazole-phenyl-thiazole α-helix-mimetic scaffolds were suitably decorated with substituents that could mimic three key residues (K, Q, and I) of the linear peptide prototype (PKIIQSVGIS-Nle-K-Nle). Extensive optimization of previously described synthetic methodologies was required. A library of 15 compounds bearing different hydrophobic alkyl and aromatic substituents was synthesized. The imidazole-phenyl-thiazole-based analogues outperformed the pyrrolopyrimidine-based derivatives in both inhibiting the enzyme and killing extracellular and intracellular parasites in cell culture. The most active imidazole-phenyl-thiazole compounds 3e and 3f inhibit Li-TryR and prevent growth of the parasites at low micromolar concentrations similar to those required by the peptide prototype. The intrinsic fluorescence of these compounds inside the parasites visually demonstrates their good permeability in comparison with previous peptide-based Li-TryR dimerization disruptors.

Full text of DOI:10.1021/acsinfecdis.8b00355

Subscriber access provided by ALBRIGHT COLLEGE
Article
Pyrrolopyrimidine vs imidazole-phenyl-thiazole scaffolds in non-peptidic
dimerization inhibitors of Leishmania infantum trypanothione reductase
Alejandro Revuelto, Marta Ruiz-Sanquiteria, Hector De Lucio, Ana María Gamo,
Alejandra A Carriles, Kilian Gutierrez, Pedro A. Sánchez-Murcia, Juan A. Hermoso,
Federico Gago, María-José Camarasa, Antonio Jimenez-Ruiz, and Sonsoles Velazquez
ACS Infect. Dis., Just Accepted Manuscript • DOI: 10.1021/acsinfecdis.8b00355 • Publication Date (Web): 15 Apr 2019
Downloaded from http://pubs.acs.org on April 15, 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 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
Pyrrolopyrimidine vs imidazole-phenyl-
thiazole scaffolds in non-peptidic
dimerization inhibitors of Leishmania
infantum trypanothione reductase
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
a
a
b
a
Alejandro Revuelto, Marta Ruiz-Santaquiteria, Héctor de Lucio, Ana Gamo, Alejandra
c
b
d
c
A. Carriles, Kilian Jesús Gutiérrez, Pedro A. Sánchez-Murcia, Juan A. Hermoso,
d
a
b
a
Federico Gago, María-José Camarasa, Antonio Jiménez-Ruiz, and Sonsoles Velázquez *
a
Instituto de Química Médica (IQM-CSIC), E-28006 Madrid, Spain
b
Departamento de Biología de Sistemas, Universidad de Alcalá, E-28805 Alcalá de Henares,
Madrid, Spain
c
Department of Crystallography and Structural Biology, Institute of Physical Chemistry
“
Rocasolano,”(IQFR-CSIC), E-28006 Madrid, Spain
d
Área de Farmacología, Departamento de Ciencias Biomédicas, Unidad Asociada al IQM-CSIC,
Universidad de Alcalá, E-28805 Alcalá de Henares, Madrid, Spain
* Corresponding author: Dr. Sonsoles Velázquez
Instituto de Química Médica (IQM-CSIC), C/ Juan de la Cierva 3
E-28006 Madrid (Spain) Phone number: (+34) 912587689
e-mail: iqmsv29@iqm.csic.es
1
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 2 of 66
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
Disruption of protein-protein interactions of essential oligomeric enzymes by small
molecules represents a significant challenge. We recently reported some linear and cyclic
peptides derived from an -helical region present in the homodimeric interface of
Leishmania infantum trypanothione reductase (Li-TryR) that showed potent effects on both
dimerization and redox activity of this essential enzyme. Here we describe our first steps
towards the design of non-peptidic small-molecule Li-TryR dimerization disruptors using
a proteomimetic approach. The pyrrolopyrimidine and the 5-6-5 imidazole-phenyl-
thiazole -helix-mimetic scaffolds were suitably decorated with substituents that could
mimic three key residues (K, Q and I) of the linear peptide prototype (PKIIQSVGIS-Nle-K-
Nle). Extensive optimization of previously described synthetic methodologies was
required. A library of 15 compounds bearing different hydrophobic alkyl and aromatic
substituents was synthesized. The imidazole-phenyl-thiazole-based analogues
outperformed the pyrrolopyrimidine-based derivatives in both inhibiting the enzyme and
killing extracellular and intracellular parasites in cell culture. The most active imidazole-
phenyl-thiazole compounds 3e and 3f inhibit Li-TryR and prevent growth of the parasites
at low micromolar concentrations similar to those required by the peptide prototype. The
intrinsic fluorescence of these compounds inside the parasites visually demonstrates their
good permeability in comparison with previous peptide-based Li-TryR dimerization
disruptors.
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
Keywords- -helix mimetics, protein-protein interactions, small-molecule dimerization
disruptor, trypanothione reductase, Leishmania infantum
2
ACS Paragon Plus Environment

Page 3 of 66
ACS Infectious Diseases
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
3
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 4 of 66
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
Leishmaniasis is a parasitic disease caused by protozoa of the genus Leishmania, whose 20
recognized species can infect humans leading to three main clinical syndromes: cutaneous
leishmaniasis (CL), mucocutaneous leishmaniasis (MCL) and visceral leishmaniasis (VL).
VL is the most severe form of the disease, which is caused by L. donovani and L. infantum
species. This disease is endemic in 98 countries where 300,000 out of more than one million
of new reported cases every year belong to VL, which is responsible for around 40,000
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
deaths per year. Unfortunately, despite being one of the major neglected tropical diseases
in terms of number of cases and death rate, drug development has not kept pace with its
severity and prevalence.2-5 There is no effective vaccine for leishmaniasis and currently
available treatments possess several drawbacks related to high toxicity, side-effects and
the emergence of resistance. All of the above caveats justify the research for novel and
more effective antileishmanial drugs.
6
Trypanothione reductase (TryR), one of the few genetically and chemically validated
drug targets in trypanosomatids (including the genus Leishmania), is an enzyme critical for
antioxidant defense.7-9 The trypanothione/trypanothione reductase (Try/TryR) couple is
analogous to, but distinct from, its mammalian counterpart based on the
glutathione/glutathione reductase (GSH/GR) pair. The absence of TryR in mammalian
cells, the significant differences between the host and parasite enzymes (including
opposite net charges in the active sites of the two enzymes) and its vital role for the
parasite make TryR an attractive target for rational drug design.10-12 Although a large
variety of potent and selective competitive TryR inhibitors have been described that are
good trypanocides in vitro, there are scarce reports of compounds that are also effective in
4
ACS Paragon Plus Environment

Page 5 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
vivo. Besides, all known inhibitors target the active site through covalent inactivation of
the catalytic cysteine residues and/or electrostatic interactions with negatively charged
1
3
protein residues. One of the disadvantages of TryR as a target is that the survival of the
parasite is only affected when its activity is reduced by more than 90%. This implies that
competitive inhibitors must have a very high binding affinity. For this reason, non-
competitive or irreversible inhibitors are known to be more attractive alternatives for
targeting this enzyme. Since the biologically functional form of TryR is a homodimer, our
research group is pursuing an alternative inhibition strategy that attempts to disrupt
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
1
4
protein-protein interactions (PPIs) at the homodimer interface. Potential druggable sites
for PPIs disruptors were explored by analysis of the interface through a combination of
molecular modelling and site-directed mutagenesis. As a result, E436 was identified as a
key residue for the structural integrity and catalytic activity of the enzyme. As a “proof of
concept” of this novel approach we designed and tested a small library of linear peptides
representing rational variations of the -helix spanning residues P435 to M447 of
Leishmania infantum TryR (Li-TryR) which contains the validated hot spot E436. From these
studies, the 13-mer linear peptides PKIIQSVGICMKM (1a, TRL14) and the chemically
stable PKIIQSVGIS-Nle-K-Nle derivative (1b, TRL35) emerged as potent inhibitors of both
the activity and the dimerization of the enzyme at low micromolar concentrations.14
Different peptidomimetics were later explored to stabilize the helical structure of these short
peptides and increase their metabolic stability. Constrained cyclic peptides15-17 and /3-
peptide foldamers18 not only retained activity but also provided higher proteolytic
stability (up to 40-fold) in comparison to their linear peptide counterparts. However, we
5
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 6 of 66
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
found it necessary to prepare structurally complex conjugates with cell-penetrating
peptides to facilitate cellular uptake of all the peptide-based inhibitors and, thus, kill the
parasites in vitro.
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
In addition to peptide-based approaches, the development of non-peptidic -helix
mimetics that reproduce key features of short peptides for inhibiting PPIs is an active area
of research.19-22 Hamilton and coworkers first demonstrated23 that a terphenyl scaffold
substituted with appropriate functional groups can reproduce the spatial orientation of the
key side-chains found predominantly at the i, i+3 or i+4 and i+7 positions at a common face
in -helical peptides. Since this pioneering work was published, numerous terphenyl-
inspired scaffolds, also termed proteomimetics, have been developed to improve water
solubility, synthetic accessibility, and/or flexibility.20,24,25 To increase the water solubility,
the strategy of benzene-to-heterocycle replacement leading to more hydrophilic and/or
amphiphilic scaffolds (e.g. terpyridine, terephthalamide or 5-6-5 imidazole-phenyl-
thiazole scaffolds)26-29 has been widely used. The synthetic difficulty can be reduced by
using solid-phase methods (e.g. N-alkylated oligobenzamide or pyrrolopyrimidine
scaffolds)30-33 or multicomponent reactions (e.g. phenyl-imidazole-phenyl).34 The
conformational flexibility can be restricted by intra-molecular hydrogen bonds (e.g.
terephthalamide, trispyridylamide or enaminone scaffolds). Interestingly, several
proteomimetics were shown to disrupt -helix-mediated PPIs in vitro and in vivo thus
highlighting a great therapeutic potential.25,35,36
Here we describe our first steps in the rational design of small-molecule non-peptidic
TryR dimerization inhibitors with enhanced drug-like properties that mimic the key
6
ACS Paragon Plus Environment

Page 7 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
interactions between the prototype peptide and the enzyme and hence produce the same
biological effects. Alanine-scanning studies and molecular dynamics (MD) simulations of a
Li-TryR monomer in complex with the peptide prototype 1b (TRL35) suggested that the side-
chains of residues Lys-2, Gln-5 and Ile-9 located at i, i+3 and i+7 positions, respectively, of 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
14
helical peptide may have large contributions to the inhibitory potency. Since these residues
are arranged on the same face of the α-helix we thought that they should be suitable for a
proteomimetic replacement approach (Figure 1a,b). From the diverse array of scaffolds
32
reported in the literature to mimic an -helix, we focused on pyrrolopyrimidines and 5-
6-5 imidazole-phenyl-thiazoles28,29 because of their synthetic accessibility, conformational
rigidity and water solubility. Target proteomimetics 2a-i and 3a-e bearing a variety of Lys,
Gln and Ile side-chains mimetic substituents were synthesized (Figure 1c). The biological
evaluation of these two novel families of small-molecule TryR inhibitors in
activity/dimerization enzymatic assays and their in vitro leishmanicidal effects against
promastigotes and extracellular and intracellular amastigotes of Leishmania infantum was
also undertaken. The water solubility of the most potent inhibitors was also
experimentally measured. In addition, the structure of the complex between Li-TryR and
pyrrolopyrimidine 2f solved by X-ray crystallography revealed a binding site for this
novel TryR inhibitor.
7
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 8 of 66
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
a)
b)
c)
NH2
N
H
Lys (R1)
Gln (R2)
N
NH2
O
Ri = Lys
N
N
N
H
N
O
R HN
N
1
N
N
Lys
O
N
R3 Ile
O
Ri+3 = Gln
Ri+7 = Ile
S
N
O
NH2 Gln (R2)
R3 Ile
3a R3 = isobutyl
3b R3 = neopentyl
3c R3 = CH2Ph
3d R3 = (CH2)2Ph
3e R3 = (CH ) naphthyl
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
a R1 = H R3 = NH-sec-butyl
2b R1 = H R3 = NH-iso-butyl
c R1 = H R3 = NHCH Ph
d R1 = H R3 = NH(CH ) Ph
2
2
2
2 2
2e R1 = H R3 = NH(CH ) Ph
2 3
2 2
2f R1 = H R3 = NH(CH ) naphthyl
g R1 = H R3 = NHCH biphenyl
2 2
3f R3 = (CH ) biphenyl
2 2
2
2
2
h R1 = H R =
N
3
NH
2
i
R1
=
R =
3 NH(CH ) naphthyl
2 2
NH2
Figure 1. a) Selected lysine (Lys), glutamine (Gln) and isoleucine (Ile) residues corresponding to
positions i, i+3 and i+7 in the prototype peptide 1b; b) These residues are located on the same face
of the helix; c) Target proteomimetics 2a-i and 3a-f based on a pyrrolopyrimidine or a 5-6-5
imidazole-phenyl-thiazole scaffold, respectively.
RESULTS AND DISCUSSION
Chemistry. The proposed proteomimetic structures 2 and 3 incorporate, in addition to the
hydrophobic side chain of Ile, mimics of the charged and polar side-chains of residues Lys
and Gln. This area of -helix mimicry for PPI modulation has not been as extensively
explored as that involving hydrophobic or aromatic replacements. The present study
represents an example of the synthetic difficulties encountered in the application of
previously described methodologies to our particular side-chains. Thus, for the
preparation of the target compounds, alternative routes and/or extensive optimization of
reaction conditions were required.
Synthesis of pyrrolopyrimidine derivatives 2a-i (Figure 1). The pyrrolopyrimidine scaffold
designed by H-S. Lim et al.32,33 is attractive due to its synthetic accessibility via a solid-
phase synthesis, its increased conformational rigidity and its favorable physical properties
8
ACS Paragon Plus Environment

Page 9 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
such as water solubility and cell permeability. The synthesis of target pyrrolopyrimidine
2a bearing the R , R , and R substituents that mimic the Lys, Gln and Ile residues,
1
2
3
respectively, was first attempted following the described solid-phase methodology.32
Thus, introduction of the first two functionalities (R and R substituents) (Scheme 1) was
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
1
2
followed by on-resin cyclization to generate the pyrrolopyrimidine ring, and finally the
nucleophilic substitution of a sulfone group with amines to introduce the R substituent
3
(Scheme 1). The reactions were monitored by HPLC-MS after cleavage of small quantities
of the resin. Thus, bromo acetylation of the amine group on Rink amide resin (Scheme 1)
followed by displacement of the bromide with N-Boc-1,4-butanediamine (Lys mimetic R1
substituent) yielded the peptoidyl-resin intermediate 5. Repeating the same procedure
with the 3-aminopropanamide Gln mimetic (R substituent) provided the Boc-protected
2
dimeric peptoid resin 6 quantitatively. Nucleophilic substitution of 6 with commercially
available 4,6-dichloro-2-methylthio-5-formylaldehyde (7) afforded aldehyde intermediate
8
. Dimethylamination followed by base-induced cyclization of the resin-bound aldehyde,
in the presence of DBU in DMF and MeOH at 90 °C, furnished the pyrrolopyrimidine
peptidyl-resin 9 with no traces of starting material or iminium intermediates. Next,
oxidation of the thioether group of 9 with meta-chloroperbenzoic acid, in the presence of
NaHCO , provided sulfone intermediate 10 in excellent conversions by HPLC-MS
3
(Scheme 1). However, unexpectedly, all attempts of nucleophilic substitution of the
sulfone intermediate 10 with sec-butylamine (Ile mimetic R substituent) in the presence of
3
DIEA in NMP at 170 °C to give the final compound 2a failed. Different temperatures (150-
190 °C) and bases (DIEA, TEA, DBU) were used but only complex mixtures of
9
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 10 of 66
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
decomposition products or unreacted starting material were detected by HPLC-MS after
final standard TFA cleavage of the resin. It seems that the high-temperature conditions
required for the nucleophilic substitution reaction with amines in resin 10 is not
compatible with the presence of our selected charged/polar substituents R and R .
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
2
Scheme 1. Initial attempts of synthesis of pyrrolopyrimidine 2a following reported methoda
1
0
ACS Paragon Plus Environment

Page 11 of 66
ACS Infectious Diseases
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
Next, we devised an alternative strategy (Scheme 2) using a combination of solution and
modified solid-phase methodologies originally described by H-S. Lim et al.32,33 This
involves the initial introduction of the R hydrophobic substituent in solution, through the
3
nucleophilic substitution of commercially available 2,4,6-trichloropyrimidine-5-
carbaldehyde 11 with the appropriate primary amines, followed by a second nucleophilic
substitution reaction with the previously obtained resin-bound peptoid 6, amination and
final cyclization (Scheme 2). Thus, treatment of 11 with sec-butyl amine in the presence of
DIEA at -78 °C for 3 h gave a mixture of the regioisomers substituted at 2- or 4-position
12a (minor) and 13a (major) which were easily separated by chromatography (Scheme 2).
All the attempts to increase the 12a/13a ratio by using higher temperatures, longer reaction
1
1
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 12 of 66
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
times or other bases failed since side-products such as disubstituted derivatives or
decomposition products were detected. Then, the resin-bound dimeric peptoid 6 was
reacted with 4,6-dichloro-2-sec-butylamine-pyrimidine-5-carbaldehyde 12a to afford
aldehyde intermediate 14a. Amination of 14a with dimethylamine in NMP at 60 °C,
subsequent cyclization with DBU at 90 °C and final acidolytic TFA cleavage from the resin
furnished the desired proteomimetic 2a in 22% overall yield after purification on a Biotage
using a reverse-phase column. Similarly, target pyrrolopyrimidines 2b-h, bearing a variety
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
of aliphatic and aromatic groups at R substituent, were obtained from the corresponding
3
2-substituted-4,6-dichloropyrimidine carbaldehydes 12b-h and Boc-protected peptoidyl
resin 6 in 10-57% overall yields after chromatographic purification. Finally, the guanidine
analogue 2i was prepared following a similar procedure from the Alloc-protected
peptoidyl-resin 6’ and 4,6-dichloro-2-naphthylethylamine-pyrimidine-5-carbaldehyde 12f
followed by dimethylamination and cyclization to give the Alloc-protected resin 16’f.
Next, the allyl group was removed with Pd(PPh ) and PhSiH as an allyl scavenger and
3
4
3
the free primary amine group was transformed into guanidine by treatment with di-Boc-
triflylguanidine in the presence of TEA. Final cleavage from the resin furnished the target
guanidine pyrrolopyrimidine 2i in low overall yield after flash purification on the Biotage.
1
2
ACS Paragon Plus Environment

Page 13 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
Scheme 2. Synthesis of intermediates 12a-g (in solution) and pyrrolopyrimidines 2a-i (solid-
phase)a
a
b
NHsec-butyl
NHiso-butyl
e
f
NH(CH ) Ph
2 3
O
Cl
O
Cl
N
O
Cl
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
NH(CH2)2Naphthyl
NHCH2Biphenyl
a
H
N
+
H
N
H
N
c
g
Cl
Cl
Cl
N
R3
R2
N
Cl
NHCH Ph
2
11
12a-g, 12h*
13a-g
*commercially
N
d
NH(CH ) Ph h*
2 2
available
NHR1
NHR1
NHR1
O
H
Cl
O
N
N
O
H
N
O
O
N
1
2a-h
1
2a-h
N
N
N
HN
N
R3
HN
N
N
R3
NH
c
b
NH
O
O
O
O
O
NH2
NH2
O
NH2
6
6
R = Boc
’ R = Alloc
14a-h R = Boc
14’f R = Alloc
15a-h R = Boc
1
15’f R = Alloc
1
1
1
1
1
d
NH
O
R HN
1
N
e
N
2a-i
N
or f, g, e for 2i
O
N
N
R3
NH
2
i R =
1
NH2
O
NH2
1
6a-h R = Boc
1
16’f R = Alloc
1
a
Reagents and conditions: (a) R NH , DIEA, dry DCM, -78 °C; (b) 12a-h, TEA, THF, rt; (c) NMe , NMP, 60 °C; (d)
2
2
2
DBU,DMF, MeOH, 90 °C; (e) TFA, TIPS, H O, rt; (f) Pd(PPh ) , PhSiH , DCM, rt; (g) 1,3-Di-Boc-2-
2
3 4
3
(trifluoromethylsulfonyl)guanidine, TEA, DCM, rt
Synthesis of 5-6-5 imidazole-phenyl-thiazole derivatives 3a-f (Figure 1). The 5-6-5 imidazole-
28
phenyl-thiazole scaffold described by Hamilton and co-workers replaces the phenyl end
units of the original terphenyl scaffold with water-soluble five-membered heterocyclic
groups. The reported synthetic route, in solution, is flexible and allows the introduction of
a variety of R -R substituents (including mimics of charged and polar side-chains) via
1
3
high-yielding nucleophilic aromatic substitution (SN ) of aryl fluorides with alcohols,
Ar
1
3
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 14 of 66
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
Ullman coupling and Hantsch thiazole syntheses.28 Initial attempts of SNAr of
commercially available 4-bromo-2-fluorobenzonitrile (17) with 3-hydroxypropanamide (R2
substituent to mimic the Gln residue) using KHMDS as the base, failed to provide the
desired compound 18 but instead gave phenol 19 (Scheme 3). The formation of this
compound could be explained by abstraction of the acidic alpha hydrogen adjacent to the
carboxamide group in the presence of a strong base (KHDMS) followed by intramolecular
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
elimination of the phenol 19 and acrylamide. Similar side-reactions have been reported
with aryl ethers in the presence of a strong base.37
Scheme 3. SNAr with aryl fluoride 17 or 21 and alcohols to introduce the Gln mimetic side-
chaina
Although this elimination reaction was partly reduced by the use of a weaker base such
as Cs CO to give the desired compound 18 in 30% yield, the base-sensitive nature of the
2
3
compound and the unsatisfactory yield for an initial step of the synthetic pathway
prompted us to explore other Gln mimetics (R ) with no acidic protons to avoid this side
2
reaction. Thus, reaction of 17 with 1-(2-hydroxyethyl)imidazolin-2-one in the presence of
1
4
ACS Paragon Plus Environment

Page 15 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
KHDMS in dry THF provided the desired compound 20 in much higher yields (71%)
(Scheme 3). Similarly, the iodide derivative 22 was obtained in 83% yield from
commercially available 4-iodo-2-fluorobenzonitrile (21) and the cyclic alcohol derivative.
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
Our next step involved incorporation of a Lys mimetic (R substituent) onto an
1
imidazole ring and subsequent Ullman coupling with bromide 20 (Scheme 4). Imidazole
24 was easily synthesized by reductive amination of commercially available imidazole-5-
carboxaldehyde (23) with N-Boc-ethylenediamine using NaBH in 79% yield.
4
Scheme 4. Synthesis of imidazole-phenyl-thiazoles 3a-f and -bromoketones 32-35a
N
N
N
Boc
H
Boc
N
N
N
Ullmann
coupling
N
NHBoc
NH
N
NHBoc
NH
N
NHBoc
NH
O
O
O
b
c
d
22
N
N
N
O
O
O
CN
CN
S
NH2
N
H
N
25
2
6
27
N
NHBoc
H
24
Hantzsch
synthesis
h
a
N
N
Boc
CHO
N
N
NHBoc
NH
N
H
Synthesis of -bromoketones
23
O
j
O
3a-f
N
O
Br
R
S
N
32 R=isobutyl
33 R=neopentyl
R3
34 R=(E) CH=CH-Naphthyl
35 R=(E) CH=CH-Biphenyl
36a R = isobutyl
3
3
6b R = neopentyl
3
g
36c R = CH Ph
e
3
2
36d R = (CH ) Ph
O
3
2 2
O
3
6e R = (E) CH=CH-Naphthyl
3
3
6f R = (E) CH=CH-Biphenyl
Ar
30 Ar=Naphthyl
3
R
i
3
7e R = (CH ) Naphthyl
7f R = (CH ) Biphenyl
28 R=isobutyl
29 R=neopentyl 31 Ar=Biphenyl
3
2 2
3
3
2 2
f)
ArCHO
a
Reagents and conditions: (a) i) H N(CH ) NHBoc, MeOH, rt; ii) NaBH , rt; (b) imidazole 24, N,N-dimethylglycine,
2
2 2
4
Cs CO , Cu O, DMSO, 85 °C, MW; (c) (Boc) O, TEA, MeOH, rt; (d) (NH ) S, DMF, 80 °C; (e) 28 and 29, Br ,
2
3
2
2
4 2
2
MeOH, 0 °C; (f) ArCHO, acetone, NaOH 10%, MeOH, rt; (g) 30 or 31, NBS, p-TsOH, acetonitrile, rt; (h)
i
commercially available BrCH CO(CH ) Ph (n=1,2) or synthetic -bromoketones 32-35, PrOH, 70 °C; (i) H (ballon),
2
2 n
2
Pd/C 20%, THF/MeOH 1:1, rt; (j) TFA, DCM, rt
1
5
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 16 of 66
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
Attempts were made to couple bromide 20 with imidazole 24 via the C-N Ullman-type
2
8
conditions described by Hamilton et al, but no coupling product was observed. Different
3
8
reaction conditions modifying a variety of reaction parameters such as the copper source
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
(
CuI and Cu O), different N-O, N-N and O-O bidentate ligands (e.g. L-proline, N,N-
2
dimethylglycine, 4,7-dimethoxy-1,10-phenantroline and dipivaloylmethane), different
bases (K CO and Cs CO ), temperatures or microwave assistance were all unsuccessful.
2
3
2
3
However, the desired coupling product 25 was succesfully obtained from the more
reactive aryl iodide derivative 22 (Scheme 4). Thus, treatment of 22 with imidazole 24 in
3
9
the presence of Cu O, N,N-dimethylglycine in DMSO at 85 °C, under microwave
2
irradiation for 2 h, furnished the desired coupling product 25 in 87% yield after
chromatographic purification. To avoid side reactions in the following steps of the
synthesis, protection of the secondary amine with tert-butoxycarbonyl group (Boc), using
Boc O in TEA at room temperature, afforded the di-Boc protected derivative 26 in
2
excellent yield after purification. Subsequent treatment with aqueous 20% ammonium
o
sulfide at 80 C for 4 h furnished the thioamide 27 in quantitative yield. Finally, Hantsch
thiazole synthesis from appropriate -bromomethylketones and the thioamide 27
facilitated late-stage attachment of a variety of alkyl and aryl R substituents onto the 1,3-
3
thiazole ring to mimic the hydrophobic side-chain of Ile residue. The synthesis of non-
commercially available -bromoketone intermediates 32-35 was performed by
bromination of the corresponding methylketones using two different methodologies. For
aliphatic methylketones 28 and 29, bearing one of the alpha carbonyl positions hindered
by a bulky group, reaction with Br at 0 °C allows the regioselective halogenation of the
2
1
6
ACS Paragon Plus Environment

Page 17 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
end-terminal position providing compounds 32 and 33 in 68% and 52% yield, respectively.
When both -carbonyl positions were not differentiated for halogenation (i.e. aromatic
ethyl methyl ketones), the strategy consisted of the use of the corresponding /-
unsatured precursors 30 and 31, previously synthesized by Claisen-Schmidt condensation
of arylaldehydes and acetone, in the presence of NaOH in good yields. Subsequent
regioselective bromination of the -methyl group of 30 and 31 with N-bromosuccinimide
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
4
0
in p-TsOH afforded the desired /-unsaturated aryl -bromomethyl ketones 34 and 35
in 46% and 42% yields, respectively after chromatographic purification. Next, Hantzsch
thiazole synthesis of thioamide 27 with commercially available or previously synthesized

-bromoketones 32-35, heating for 4 h, gave "decorated" thiazoles 36a-f in good to
excellent yields (69-95%). In the case of /-unsatured derivatives 36e,f an additional step
was necessary to reduce the double bond (H , Pd/C) so as to obtain the saturated arylethyl
2
analogues 37e,f in 70% and 80% yields. Finally, treatment of di-Boc-protected
intermediates (36a-d and 37e,f) with TFA in DCM for 2-3 h at room temperature furnished
the target proteomimetics 3a-f quantitatively.
Proteomimetics based on the imidazole-phenyl-thiazole scaffold 3a-f exhibited more
potent TryR inhibition and leishmanicidal activity than did the pyrrolopyrimidine
analogues 2a-i (see below). We next measured the aqueous termodynamic solubilities of
the most active compounds 3a-f at 25 °C after 24 h (Supplemental Table S1). All the
compounds within this series, in consonance with the predicted negative values of cLogP
(range from -0.7 to -3.2), showed excellent water solubility (≥ 25 mg/mL) including 3e and
3f, which bear bulky aromatic R substituents (naphthyl or biphenyl) on the thiazole
3
1
7
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 18 of 66
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
moiety. According to the United States Pharmacopeia solubility criteria (USP 30/NF 25),
the compounds can be classified as very soluble since <1 parts of solvent are required for
one part of solute. Interestingly, the biologically more active imidazole-phenyl-thiazole
derivatives 3a-f displayed intrinsic fluorescence that could be used for imaging and
localization of the compounds inside the parasites. Compound 3d exhibited an absorption
spectrum with a maximum at 320 nm and a single emission band at 364 nm in the blue
region of the visible spectrum (Supplemental Figure S1).
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
Biological and structural results.
4
1
Enzymatic assays. The effect of all the compounds on the oxidoreductase activity and
14
dimeric status of Li-TryR was evaluated using prototype peptide 1 and mepacrine as
positive controls. According to the results shown in Table 1, compounds 2a-i (based on a
pyrrolopyrimidine scaffold) are significantly less potent than their imidazole-phenyl-
thiazole counterparts. In the pyrrolopyrimidine series, modifications at the hydrophobic
R substituent (Ile mimetic) are relevant for TryR inhibitory activity. Thus, the presence of
3
an alkyl or a benzylamino substituent at R or a pyrrolidinyl (2a-c and 2h) leads to inactive
3
compounds. Interestingly, replacement of the benzyl group at R by a phenylethyl or a
3
phenylpropyl group (2d and 2e), and especially the use of large polycyclic aromatic rings
(i. e. naphthyl 2f or biphenyl 2g), resulted in the best compounds of this series with
moderate micromolar IC values. Replacement of the positively charged amino group at
5
0
the R substituent (Lys mimetic) by a guanidinium group in naphthyl derivatives (2f vs 2i)
1
slightly increased the TryR inhibitory activity.
1
8
ACS Paragon Plus Environment

Page 19 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
In the imidazole-phenyl-thiazole series, all the compounds (3a-f) showed Li-TryR
inhibitory activity irrespectively of the alkyl or aromatic nature of the R substituent. As
3
observed in the pyrrolopyrimidine series, introduction of a bulky hydrophobic aromatic
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
group at R clearly enhances Li-TryR inhibition. The naphthyl and biphenyl analogues 3e
3
and 3f, with IC50 values in the low micromolar range, were the most potent Li-TryR
inhibitors of the entire series.
However, the activity of the proteomimetic compounds as disruptors of Li-TryR
dimerization is much lower than that observed for the prototype peptide 1 (Table 1).
Interestingly, only the most potent inhibitors of redox activity, i.e. 2i, 3e and 3f, showed a
moderate effect on the enzyme dimerization status at 20 µM (14%, 26% and 32% reduction,
respectively).
Table 1: IC ± SEM values for compounds 2a-i and 3a-f in the Li-TryR oxidoreductase
5
0
activity and dimerization assays.
NHR1
TFA
H2
N
N
N
NH3 TFA
NH
O
N
O
H2N
N
O
N
N
O
N
N
R3
S
N
O
NH2
R3
2
a-i
3a-f
Dimerization
inhibition at
20 µM b
R1
TFA)
IC activity
50
Comp
R3
µM)a
(
(
/
/
peptide 1
1.2 ± 0.2
95 %
0 %
/
Mepacrine
/
12.9 ± 0.5
1
9
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 20 of 66
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
a
H
H
H
H
NH-sec-butyl
No inhibition
No inhibition
No inhibition
100.2 ± 13.5
94.6 ± 6.4
0 %
2
b
NH-isobutyl
0 %
0 %
0 %
0 %
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
NH-CH Ph
2
2
c
2
d
NH-(CH ) Ph
2
2
2
e
-NH-(CH ) Ph
2
3
H
H
NH-
CH ) Naphthyl
2
f
52.2 ± 1.8
0 %
(
2
2
2
g
H
H
NH-CH Biphenyl
51.7 ± 5.7
0 %
0 %
2
2
h
Pyrrolidinyl
No inhibition
(
C=NH)
NH2
NH(CH ) -
2
2
2
i
39.9 ± 6.8
14 %
Naphthyl
3
a
H
H
H
H
H
isobutyl
51.2 ± 4.6
51.3 ± 6.1
32.8 ± 2.3
22.4 ± 2.2
5.1 ± 0.4
8.6 ± 1.4
0 %
0 %
3
b
neopentyl
3
c
CH -Ph
0 %
2
3
d
(CH ) -Ph
0 %
2 2
3e
(CH ) -Naphthyl
26 %
32 %
2
2
3
f
(CH ) -Biphenyl
2
2
H
a
41
Enzymatic activity according to a modified assay described by Hamilton et al. Results are
representative of three independent experiments each performed in triplicate. Dimer quantitation
assay (ELISA).
b
1
4
2
0
ACS Paragon Plus Environment

Page 21 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
Crystal structure of Li-TryR in complex with 2f. Attempts to co-crystallize the most
potent TryR inhibitors 3e and 3f with the enzyme failed due to quasi-complete protein
precipitation. Incubation of Li-TryR with 3e and 3f (concentration of the inhibitors ranging
from equimolar to 10-fold the molarity of TryR, for periods of 2 to 16 h, 4 ºC) resulted in
strong protein precipitation and no formation of crystals, in agreement with the moderate
effect of both inhibitors as dimerization disruptors.
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
To study the potential binding site of 2a-i compounds in Li-TryR, soaking experiments
were performed with Li-TryR crystals using the 2f ligand, which does not disrupt the TryR
dimer (Table 1). The crystal structure of the Li-TryR:2f complex, solved at 3.3 Å resolution
(Supplemental Table S2 and Figure 2), displayed a nice electron density for 2f
(Supplemental Figure S2). The final model contains a TryR functional dimer consisting of
two subunits related by a non-crystallographic two-fold axis. Interestingly, only chain B
presents one 2f molecule bound to its active site. The 2f ligand is placed close to the
catalytic Cys residues (Cys52 and Cys57) and to the FAD cofactor (Figure 2B). The central
pyrrolopyrimidine core makes a stacking interaction with Trp21 (Figure 2B) at the so-
called polyamine-binding site, where its natural substrate Try usually binds, and where a
4
2
mepacrine inhibitor has also been previously localized. Substituens R and R also play
1
2
an important role in the stabilization of 2f at Li-TryR by establishing different polar
interactions with the protein. Thus, the Glu18 side chain and backbone atoms of Leu17
residues establish H-bond interactions with the amide moiety and keto groups in R1,
respectively. Another residue involved is Ser109, which interacts with the amide group of
2
1
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 22 of 66
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
the glutamine mimetic in R . Moreover, the lysine mimetic substituent forms ion-dipole
2
interactions with water molecules positioned in the active site of Li-TryR. The naphthalene
ring in R protrudes out of the cavity and its electron density is poorly defined, indicating
3
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
high mobility in this region.
The ligand-protein interaction does not seem to dramatically alter the structure of the
monomer as both independent monomers, chains B and A (with and without ligand
respectively), are very similar (rmsd of 0.42Å). Changes observed between the backbone of
both chains (Supplemental Figure S3A) are concentrated in different loops and at the N-
terminal tail. 2f interaction at the active site only provokes a re-orientation of side chains
of those residues directly involved in its stabilization (Supplemental Figure S3B).
Figure 2. Three-dimensional structure of the Li-TryR:2f complex. (A) Molecular surface of the Li-
TryR dimer in complex with 2f (cyan spheres). The enzyme’s backbone is depicted as a cartoon
and its cofactor FAD in orange sticks. Each monomer is colored differently (grey for chain B and
magenta for chain A). (B) Close-up view of the binding site of 2f. The ligand prevents
trypanothione from fitting into the cavity. Polar interactions are represented by dashed lines.
Relevant amino acids are depicted as capped sticks and labeled. Water molecules represented as
red spheres.
2
2
ACS Paragon Plus Environment

Page 23 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
When the crystal structure of Li-TryR in complex with 2f is best-fit superimposed onto
published structures of TryR from Trypanosoma brucei (Tb-TryR), either covalently bonded
to
cyclohexylpyrrolidine inhibitors RD0 and RD7 bound to the so-called ‘mepacrine site’
Figure S4B), the only common interaction observed to all inhibitors involves the side
a
quinacrine mustard (Supplemental Figure S4A) or in complex with
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
4
3
(
chain of Trp21 in Li-TryR. The fact that all of them occupy the active site strongly suggests
that they all prevent the binding/exit of Try.
Computational studies. Pharmacophore for pyrrolopyrimidine compounds 2. The fact
that in our Li-TryR:2f co-crystal structure the naphthalene moiety is projected towards the
solvent makes it difficult to extract SAR features for activity in the pyrrolopyrimidine
series 2. Moreover, it is known that the large active site of this enzyme allows a multitude
of different orientations for small-molecule ligands. This fact hampers the rationalization
4
4
of the activity data on the basis of a single crystal structure for one complex. For this
reason, a pharmacophore model for compounds 2a-2i (considering only 2f, 2g and 2i as
active) was constructed based on four features: aromatic (AR), hydrophobic (H),
hydrogen-bond acceptor (HBA) and hydrogen-bond donor (HBD). The model showed
(Figure S5) that the presence of a bulky aromatic group in the H+AR region this position is
critical for Li-TryR dimer disruption and enzyme inhibition.
MD simulations for imidazole-phenyl-thiazole 3f bound at the dimer interface. We run
classical MD simulations in explicit solvent for a Li-TryR monomer with 3f, the most
potent compound of series 3, bound at the dimerization interface (Figure 3A). As the
2
3
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 24 of 66
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
reference system, we simulated the 13-mer 1a (TRL14), with capping groups at both N- and
C-termini and a Lys residue at position 2 (Figure 3B).14,16 Both systems were stabilized
after a few ns of MD simulation and the root-mean squared deviation (RMSd) evolved
softly along the 100-ns trajectory (Figure 3C). The interactions of the imidazolone of 3f
with the loop where Ile458 and Val460 are located were quite stable along the MD
simulation and of the same order as those found for the Q5-containing peptide TRL14 (1a).
The van der Waals contributions stand out as the main component of the drug:monomer
binding energies in both cases (Figure S6). Hydrophobic residues Val460 and Ile437 are the
two residues that contribute the most to the total binding energy of 3f, closely followed by
Phe454, Thr457 and Leu468 (Figure S6). The biphenyl group of this molecule occupies a
large hydrophobic pocket lined by the residues shown as spheres in Figure 3A. The size of
this hydrophobic site can account for the fact that derivatives 3a-d, which lack large and
bulky aromatic groups, do not behave as enzyme dimerization disruptors. Finally, the
direct hydrogen bonds between the side-chain carboxylate of Asp432 and either the
ammonium group on the imidazole side chain of 3f or the K2 of 1a are not constant along
the trajectory (Figure 3C). This is because the ligands’ positively charged groups establish
alternative salt bridging interactions with Glu466 and Glu436, respectively.
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
ACS Paragon Plus Environment

Page 25 of 66
ACS Infectious Diseases
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 3. Comparative results from molecular dynamics (MD) simulations of Li-TryR:1a
and Li-TryR:3f complexes. (A) Representative structure of 3f (shown as sticks and balls; C
atoms colored in yellow) bound at the dimerization interface of a Li-TryR monomer (C
atoms colored in grey). (B) Representative structure of the complex formed between a Li-
TryR monomer and peptide 1a (TRL14) (1a) with a Lys at position 2 (PKIIQSVGICMKM,
C atoms colored in yellow). (C) Evolution of the root-mean-square deviation (RMSd, Å) of
bound 3f or 1a (top) and time evolution of the distances (Å) separating 3f from the
backbone oxygen of Ile458, backbone nitrogen of Val460 (middle) and the carboxylate
oxygen of Asp432 (bottom) along 100 ns of unrestrained MD simulation.
Taken together, our results point to at least two different binding sites for the
compounds studied here. One, as shown in the crystal structure solved, is at the active site
and may contribute to binding and subsequent enzyme inhibition through direct
competition with Try. A second one should be located at the dimerization interface but we
2
5
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 26 of 66
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
have been unable to detect it so far by X-ray crystallography because the monomeric form
of the enzyme is unstable and does not form good crystals and/or spoils existing ones. The
results obtained for series 3 strongly suggest that the presence of a bulky substituent at R3
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
is highly relevant for binding to this second site and for the ligand to behave as a
dimerization disruptor.
Antileishmanial activity in cell culture. All the compounds were tested against L.
infantum axenic amastigotes and promastigotes using miltefosine and the R -conjugate of
9
the prototype peptide as positive controls. Neither the pyrrolopyrimidine compounds 2a-i
nor the imidazole-phenyl-thiazole analogues 3a-d exhibited any significant
antileishmanial activity at the maximum concentration assayed (25 M). On the contrary,
the most potent Li-TryR inhibitors 3e and 3f, for which a modest dimer disruption activity
was also observed in vitro, displayed potent leishmanicidal activity (Table 2). Significantly,
the activity against amastigotes of peptidomimetic 3f was found to be similar to that
observed for our peptidic prototype and for the reference drug miltefosine. The toxicity
against the human cell line THP-1 was also found to be similar between 3f and the R9-
conjugated peptide 1 and only slightly higher than that observed for miltefosine.
Cell penetration of the antileishmanicidal imidazole-phenyl-thiazole compounds was
demonstrated by direct visualization of their intrinsic fluorescence inside L. infantum
promastigotes (Figure 4).
2
6
ACS Paragon Plus Environment

Page 27 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
Bright field
Blue fluorescence
Merge
3e
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
3f
Figure 4. Fluorescence microscopy images of L. infantum promastigotes treated with 25
µM of compounds 3e and 3f for 1 h.
Compounds 3e and 3f were also able to cross the macrophage plasma and
phagolysosomal membranes, as indicated by their capacity to decrease the parasite load in
amastigote-infected THP-1 cells (EC values of 21.5 µM and 14.5 µM respectively).
5
0
A good correlation between the TryR inhibitory activity values and the in cellulo results
is observed, which supports the idea that this enzyme may be the main molecular target of
the imidazole-phenyl-thiazole analogues in L. infantum parasites.
Table 2: EC ± SEM values for the most active compounds 3e and 3f on L. infantum
promastigotes, amastigotes and cytotoxic activity in THP-1 cells.a
5
0
EC50
promastigote
s (µM)
EC50
amastigotes
(µM)
EC50
THP-1
(µM)
Comp.
R3
2
7
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 28 of 66
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
Miltefosin
e
/
2 ± 0.1
29.9 ± 1.5
12.6 ± 3.5
4
7.6 ± 0.6
R9-
conjugated
peptide 1
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
/
4.6 ± 0.2
3.5 ± 0.9
(
CH ) -
Naphthyl
2
2
3e
12.8 ± 0.7
5.3 ± 0.3
12.8 ± 1.3
5.3 ± 0.2
26.5 ± 0.9
(CH2)2-
Biphenyl
3
f
14.2 ± 14.2
a
EC = Half-maximal effective concentration. Results are representative of three independent
5
0
experiments each performed in triplicate.
CONCLUSION
We believe that the development of drugs able to inhibit dimerization of functionally
obligate homodimeric enzymes is an attractive goal. Recently, we succesfully developed
peptide-based dimerization inhibitors of the homodimeric TryR from L. infantum that were
derived from an -helix located in the interfacial domain. However, conjugation to cell-
penetrating peptides was required to kill the parasites in cell culture. With the aim of
identifying small-molecule Li-TryR dimerization inhibitors with better drug-like
properties, we then devised two -helical proteomimetics, the pyrrolopyirimidine and the
imidazo-phenyl-thiazole scaffolds, which we decorated with surrogates of the side-chains
of key amino acids present in the prototype peptide. The pyrrolopyrimidine analogues
behaved as moderate inhibitors of the oxidorreductase activity of Li-TryR but did not
2
8
ACS Paragon Plus Environment

Page 29 of 66
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
show any relevant enzyme disruption capacity and did not kill the parasites. The crystal
structure of Li-TryR with pyrrolopyrimidine 2f revealed that this compound binds to the
"mepacrine binding site" of the large active site and represents a novel class of bicyclic and
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
heterocyclic TryR inhibitor. On the other hand, the use of the imidazo-phenyl-thiazole
scaffold allowed us to identify 3e and 3f as the first non-peptidic Li-TryR dissociative
inhibitors that display in vitro leishmanicidal activities similar to those of our previous
peptide-based compounds. The nice correlation between enzyme inhibition and
antileishmanial activity against both extra- and intracellular L. infantum forms suggests
that the compounds probably act by inhibiting Li-TryR. Comparative MD simulations of
compound 3f and the linear peptide prototype 1a each bound at the dimerization interface
of a Li-TryR monomer provide support to the pharmacophore hypothesis that the
presence of a bulky aromatic group is essential for the ability of these molecules to disrupt
enzyme dimerization. Studies are currently underway on the more promising imidazo-
phenyl-thiazole class of small molecules for further optimization and development in our
search for novel and safer anti-leishmanial agents.
EXPERIMENTAL SECTION
Chemistry. Melting points were obtained on a Mettler Toledo M170 apparatus and are
uncorrected. IR spectra were obtained on
a
Perkin-Elmer Spectrum One
spectrophotometer. Microanalyses were obtained with a Heraeus CHN-O-RAPID
instrument and the analytical results were within 0.4% of the theoretical values.
Electrospray mass spectra were measured on a quadropole mass spectrometer equipped
2
9
ACS Paragon Plus Environment