Hit-to-lead optimization of novel benzimidazole phenylacetamides as broad spectrum trypanosomacides

Article abstract of DOI:10.1039/d0md00058b

Trypanosoma cruzi and Trypanosoma brucei are the parasitic causative agents of Chagas disease and human African trypanosomiasis (HAT), respectively. The drugs currently used to treat these diseases are not efficacious against all stages and/or parasite sub-species, often displaying side effects. Herein, we report the SAR exploration of a novel hit, 2-(4-chlorophenyl)-N-(1-propyl-1H-benzimidazol-2-yl)acetamide previously identified from high throughput screens against T. cruzi, Trypanosoma brucei brucei and Leishmania donovani. An informative set of analogues was synthesized incorporating key modifications of the scaffold resulting in improved potency whilst the majority of compounds retained low cytotoxicity against H9c2 and HEK293 cell lines. The SAR observed against T. cruzi broadly matches that observed against T.b. brucei, suggesting the possibility for a broad-spectrum candidate. This class of compounds therefore warrants further investigation towards development as a treatment for Chagas disease and HAT. This journal is

Full text of DOI:10.1039/d0md00058b

RSC
Medicinal Chemistry
View Article Online
View Journal
RESEARCH ARTICLE
Hit-to-lead optimization of novel benzimidazole
phenylacetamides as broad spectrum
trypanosomacides†
Cite this: DOI: 10.1039/
d0md00058b
a
b
Nicole McNamara,
Raphael Rahmani,a Melissa L. Sykes,
Vicky M. Avery b and Jonathan Baell
a
*
Trypanosoma cruzi and Trypanosoma brucei are the parasitic causative agents of Chagas disease and
human African trypanosomiasis (HAT), respectively. The drugs currently used to treat these diseases are not
efficacious against all stages and/or parasite sub-species, often displaying side effects. Herein, we report
the SAR exploration of
a novel hit, 2-(4-chlorophenyl)-N-(1-propyl-1H-benzimidazol-2-yl)acetamide
previously identified from high throughput screens against T. cruzi, Trypanosoma brucei brucei and
Leishmania donovani. An informative set of analogues was synthesized incorporating key modifications of
the scaffold resulting in improved potency whilst the majority of compounds retained low cytotoxicity
against H9c2 and HEK293 cell lines. The SAR observed against T. cruzi broadly matches that observed
against T.b. brucei, suggesting the possibility for a broad-spectrum candidate. This class of compounds
therefore warrants further investigation towards development as a treatment for Chagas disease and
HAT.
Received 21st February 2020,
Accepted 12th May 2020
DOI: 10.1039/d0md00058b
rsc.li/medchem
The initial phases of Chagas disease and HAT are
commonly asymptomatic or exhibit non-specific symptoms,
Introduction
Chagas disease (American trypanosomiasis) and human
African trypanosomiasis (HAT, African sleeping sickness) are
parasitic diseases caused by related kinetoplastids.1,2 These
neglected tropical diseases are responsible for significant
morbidity and mortality in Latin America and sub-Saharan
Africa respectively, and affect low socioeconomic areas.1,2
Spread by the vector triatomines or “kissing bugs”, it is
estimated that 6–7 million people are infected with Chagas
disease, leading to approximately 14 000 deaths annually from
clinical manifestations. A further 70 million people live at risk
of infection.3 HAT is spread by infected tsetse flies and the
disease is generally fatal if left untreated. In 2017, HAT
caused approximately 1446 deaths, and a further 65 million
people are estimated at risk of infection.2,4 Whilst there has
been a reduction in reported cases of HAT over time, the
presence of putative animal reservoirs and the inability of
current drugs to treat all disease sub-species may present a
threat of recrudescence of disease.5,6
therefore many patients are not diagnosed or treated.3,4
Chagas disease develops into a chronic phase, whereby
parasites may reside in gut and heart tissue, causing
associated megaesophagus, megacolon and heart disease.7,8
The second stage of HAT is associated with parasites crossing
the blood brain barrier and neurological issues result.9
Unfortunately, these later stages of infection in both diseases
are not easily treated and additionally most current
medications display associated side effects. Collectively, this
highlights the critical need for new efficacious and safe
options for treatment.4,10,11
To identify new compounds for the drug discovery
pipeline for neglected diseases, collaborative efforts between
pharmaceutical companies and academic groups have been
undertaken.
A prominent example of this is a high-
throughput (HTS) screen of 1.8 million “diversity set”
compounds by the Tres Cantos Open Lab foundation (OLF),
supported by GlaxoSmithKline (GSK), the Drugs for Neglected
Diseases Initiative (DNDi) and the Dundee Drug Discovery
unit (DDU). This project identified new compound sets active
against T. cruzi, T.b. brucei and L. donovani.12 This large
diversity set was tested against a recombinant T. cruzi strain
expressing β-galactosidase as an intracellular reporter gene,
in host NIH-3T3 fibroblasts. Toxicity against HepG2 cells,
chemical clustering and intracellular imaging assays in H9c2
rat myocytes were employed to narrow down the compound
a Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences, Monash
University, Parkville, Victoria 3052, Australia. E-mail: jonathan.baell@monash.edu
b Discovery Biology, Griffith Institute for Drug Discovery, Griffith University,
Brisbane Innovation Park, Don Young Road, Nathan, Queensland 4111, Australia
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0md00058b
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.

View Article Online
Research Article
RSC Medicinal Chemistry
set. Those exhibiting selective T. cruzi inhibition over host
cells (pIC50 > 5, SI > 10 NIH-3T3/HepG2) made final cut-offs
with a preference for non CYP51 inhibitors. The Chagas Box
was narrowed down to 222 identified hits. In collaboration
with OLF we initiated SAR exploration of a novel anti-T. cruzi
compound class identified in the GSK Chagas-Box.
Compound (1) was one of the most promising members of
this class, due to its sub-μM activity (0.63 μM) and good
selectivity (SI = 79; Table 1). The hit exhibited 40-fold
selectivity toward T. cruzi in an intracellular imaging assay.12
Additionally, it displayed modest activity against T.b. brucei
(Lister strain 427, IC50 = 2.0 μM) with 50-fold more selectivity
for T.b. brucei over HepG2 cells (Table 1). The activity and
selectivity displayed by this compound presents the
possibility of a broad-spectrum candidate. The hit (1) also
displayed good physicochemical properties, including a low
polar surface area (47 Å2) and a low molecular weight (327 g
mol−1) which would suggest the ability to cross the blood
brain barrier. The mechanism of action is not known.
Although some related alkanoyl-2-aminobenzimidazole
scaffolds with biological activities have been previously
reported,13,14 antiparasitic activity has not previously been
reported. The attractive physicochemical properties, class
novelty for anti-trypanosomal activity, coupled with
anticipated relative ease of synthetic access, provided an
attractive starting point for SAR elaboration.
formation with various phenylacetic acid derivatives to easily
and efficiently access a number of analogues.15–17
Firstly, as the initial hit (1) possessed a para-chloro group,
we began our investigation by focusing on varying the
functionalities at this position to probe the chemical space
(Table 2). Groups were chosen to represent various
properties: hydrophophicity, hydrogen bonding availability,
steric bulk and electronic effects. Investigations began with
the singly substituted analogues 4a–i, which involved smaller
functionalities, to probe fundamental steric, hydrophobic
and electronic effects towards activity. Product yield achieved
was generally quite high, which is expected of para and meta-
substituents, as less hindrance around the carboxylic acid
exists, whilst generally ortho substituents yielded lower
amounts of product.
In parallel, more drastic changes were made to obtain
sharper SAR and further define the chemical space (4j–m).
Replacing the phenyl ring with alkyl chains would determine
if the ring itself was required for activity. In contrast, adding
rings allowed for the exploration of any extended aryl or
hydrophobic binding space. Investigations of the ortho and
meta positions of the phenyl ring were undertaken using a
small diverse range of substituents 4n–x. This was also
completed toward the purpose of correlating group position
and functionality with activity.
N-Alkyl modification. We also explored analogues varying
the N-alkyl chain length, which was also completed using
Scheme 1 to determine the steric and lipophilic effects at this
position (Table 3). Exploration began by shortening the
N-alkyl chain to confirm whether it was essential via the
synthesis of analogues 5a–c. In parallel, the synthesis of
analogues 5d–e was undertaken to explore whether an
increase in lipophilic properties was favoured, either as a
general lipophilic effect or also exploration of favourable
hydrophobic binding space.
Results and discussions
Synthesis
Phenyl modifications. One of the attractive qualities of the
HTS hit 1 (Table 1) is the possibility of swift access to amide
analogues to allow for exploration of the phenylacetamide
substituent. Scheme 1 was devised to access the hit 1 and the
analogues to follow. This started with the regioselective
alkylation of the endocyclic amine with alkyl reagent via an
SN2 mechanism, under mild basic conditions to afford the
N-alkylbenzimidazole-2-amine. This is followed by an amide
Varying the linker chain length. Our next analogue set
focused on studying the optimal chain length of the linker
between the benzimidazole and phenyl groups, also following
Scheme 1. By increasing the chain length, it can be
Table 1 Hit from the GSK HTS Chagas box demonstrating activity against T. cruzi12
IC50 (μM)
SId
T. cruzia
T. cruzib
T. bruceic
0.63
1.26
2.00
79
40
50e
a
b
T. cruzi primary growth inhibition assay performed with NIH-3T3 primary mouse embryonic fibroblast cells host cells. T. cruzi intracellular
c
d
imaging assay performed with H9c2 rat embryonic cardiomyocytes. T. brucei primary growth inhibition assay in complete HMI9 medium. SI
= selectivity index = CC50 mammalian cell line (cytotoxicity)/IC50 trypanosome, cytotoxicity measured with host cell line. Cytotoxicity assay
e
performed with hepatocellular cells (HepG2).
RSC Med. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
RSC Medicinal Chemistry
Research Article
Scheme 1 Synthesis of benzimidazole phenylacetamide analogues with varying phenyl substituents. Reaction conditions and yields: (a) Alkyl-
bromide, KOH, THF, EtOH, 65 °C, 77–95%/alternative conditions for methyl, ethyl halides, KOH, K2CO3, acetone, reflux 46–60%, (b) phenylacetic
acid derivative, EDCI, HOBt, DMF, 7–98%. See Experimental section for individual reaction yields for compounds 1, 4a–x, 5a–e, 6a–d.
determined if increased flexibility, lipophilicity and steric
properties are favourable. The analogues depicted in Table 4
were also chosen based on availability and do not contain
the para-chloro group. They can be compared to the
compound (4a) which contains the same structural
components as the hit, excluding the para-chloro as well as
the hit (1) itself.
Adding heteroatoms to linker. The synthesis of analogues
possessing heteroatoms (Table 4) on the linker portion of the
scaffold was pursued to discover any additional hydrogen
bonding interactions between these compounds and the
binding pocket.
in an attempt to discover any additional interactions with
any putative target binding site. Once again, the functional
groups were chosen to represent a range of properties –
hydrophobicity, hydrogen bonding ability, sterics and
electron withdrawing or donating effects (Table 5).
Scheme 4 starts with the nucleophilic substitution of the
fluoro group with the desired propyl amine. This reaction
occurs cleanly, with relative ease, attributed to the fluoro
acting as an excellent leaving group.22 After nucleophilic
substitution, reduction of the nitro group by hydrogenation
occurred.23,24 Cyclization of the benzenediamine with
cyanogen bromide was employed to form various
benzimidazole derivatives.25,26 This is proceeded by amide
coupling17 as described in Scheme 1 to give the desired
analogues (20a–j) shown in Table 5.
Based on the literature18,19 triphosgene was reacted with
p-chloroaniline (7) to form the isocyanate intermediate (8),
followed by the nucleophilic addition of the benzimidazole
derivative which was shown to successfully yield the urea
analogue (9) (Scheme 2).
Reverse amide linker. The synthesis of the reversed amide
linker (15) was undertaken to investigate its preferred
position within the binding site, exploring whether more
favourable hydrogen bonding interactions can be obtained
(Table 4).
As per Scheme 3, the benzyl alcohol (10) undergoes
nucleophilic addition with trichloroacetonitrile to afford the
benzyltrichloroacetimidate (11).20 As benzyl alcohol is a weak
nucleophile, activation by base and a phase transfer catalyst
was required to allow for reaction to occur in the organic
phase, proceeded by cyclization with o-phenylenediamine
under acidic conditions21 to form the 2-trichloromethyl
substituted benzimidazole (12). This was then converted into
methylbenzimidazole carboxylate (13) using sodium
carbonate in methanol.21 The product then was subjected to
alkylation of the endocyclic amine with bromopropane15 as
described earlier in Scheme 1 to achieve compound (14). The
ester (14) was reacted neat with the commercially available
benzyl amine via nucleophilic acyl substitution, at a high
temperature to accelerate the reaction and to afford the
desired product (15).
Isoquinoline analogues. To explore a more diverse set of
chemical space based around the initial hit (1) scaffold,
quinoline compounds were explored. The introduction of the
six-membered heterocycle in place of the five-membered
imidazole would alter the fit within the binding site along
with its orientation due to
a change in pi stacking
interactions. This would allow for the investigation of steric
and aromatic effects towards activity (Scheme 5).
To access the isoquinoline based analogue 25, the
commercially available isoquinoline-3-amine (21) firstly
underwent
regioselective
electrophilic
aromatic
substitution.27 The bromoisoquinoline (22) was then able to
undergo Stille coupling to form a new C–C bond joining the
allyl group to the 3 position of the isoquinoline along with
generating the bromotributylstannane side product.28
Subsequent
amide
coupling
followed
using
4-chlorophenylacetic acid as described in Scheme 1. The low
yield can be attributed to the steric hindrance of the allyl
chain. Hydrogenation23,24 as described in Scheme 4 occurred
after amide coupling to obtain both allyl analogue (24) and
alkyl analogue (25).
Antitrypanosomal activity. The total 48 compounds
synthesized for this project were screened against T. cruzi and
T.b. brucei to validate the screening hits and generate SAR
results. Assay methods are outlined in the ESI.† Against
Varying
the
benzimidazole
substituents.
Next,
investigations on the left-hand side phenyl ring were explored
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.

View Article Online
Research Article
RSC Medicinal Chemistry
Table 2 SAR surrounding changes to the RHS acyl ring
T. cruzi IC50 H9c2a,b
(μM)
CC50 H9c2a
(μM)
T.b. brucei IC50 HEK-293c,d
(μM)
CC50 HEK-293e
(μM)
CC50 HepG2f
(μM)
Log
P
#
R
1
2.0 1.3
>50
>50
>50
>50
>50
7.1 1.2
4.1 1.3
5.0 0.22
2.5 0.85
14 7.6
>83
>83f
>83
>83
>83
>50
>50
>50
>50
>50
4.7
4.1
4.9
4.3
3.7
4a
4b
4c
4d
4.0 0.09
2.0 0.14
2.5 0.06
32 2.2
4e
4f
40 5.6
3.2 1.8
7.9 2.4
>50
>50
>50
>83
>83
>83
>83f
>50
>50
>50
3.8
4.0
3.7
5.2 1.4
>83
4g
4h
4i
1.0 0.28
1.0 0.09
1.0 0.14
1.3 0.03
>50
10
4.5 1.1
3.2 2.4
3.4 1.5
5.5 1.7
>83
>83
>83f
>83f
>50
>50
>50
25
5.0
5.6
5.8
5.1
4j
>50
>50
4k
4l
13 1.2
2.5 0.53
2.0 0.37
10 0.10
11 2.7
2.6 0.4g
>83f
>50
>50
25
3.9
5.7
4.7
4m
4n
25
20
7.3 2.3
2.3 0.65
14 3.6g
4o
4p
1.6 0.37
3.2 0.07
25 0.02
2.0 0.70
3.6 0.35
>83f
>83f
20
4.9
4.3
>50
>50
4q
4r
2.5 0.29
6.3 1.2
>50
>50
2.6 0.46
>83f
>83
>50
>50
4.6
4.0
>83
RSC Med. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
RSC Medicinal Chemistry
Research Article
Table 2 (continued)
T. cruzi IC50 H9c2a,b
(μM)
CC50 H9c2a
(μM)
T.b. brucei IC50 HEK-293c,d
(μM)
CC50 HEK-293e
(μM)
CC50 HepG2f
(μM)
Log
P
#
R
4s
1.6 0.00
1.0 0.44
2.0 0.05
2.0 0.00
13 0.01
3.6 0.73
2.9 1.5
3.2 0.94
3.6 1.4
15 1.1g
>50
>50
>50
>50
4.7
4.9
4.3
4.6
4t
>50
>83f
4u
4v
>50
>83
>50
14.96 1.14g
4w
4x
4.0 0.83
1.0 0.19
>50
>50
4.3 1.1
>83f
>50
4.0
5.1
2.6 0.88
10 4.7g
20
a
b
Anti T. cruzi activity and toxicity measured in H9c2 host cell lines. Values are means of two experiments. Control compounds for T. cruzi
c
assay. Benznidazole IC50 = 1.58 μM, CC50 = 31.6 0.00 μM. Anti T.b. brucei and cytotoxicity measured in HEK-293 cell lines. (1) >50% activity
at the top concentration tested (83 μM). Values are means of four experiments. Control compounds for T.b. brucei assay. pentamidine IC50
d
=
0.0014 0.0050 μM, CC50 = <50% activity at 0.67 μM, diminazene aceturate IC50 = 0.087 0.031 μM, CC50 = <50% activity at 38.5 μM,
e
puromycin IC50 = 0.038 0.0019 μM, CC50 = 0.23 0.012 μM. >50% activity displayed at 83 μM. Values are the means of two experiments.
f
g
Toxicity measured in HepG2 cell lines. Values are means of two experiments. CC50 value is determined from a sub-efficacious curve, with
the maximum activity from 50–70%. Solubility in the intermediate dilution in medium may have contributed to this effect observed. standard
deviation.
T. cruzi, an intracellular imaging assay was carried out by
GSK as our primary screen and would drive further SAR
studies. This assay tested target compounds against T. cruzi
with rat cardiomyocyte (H9c2) host cells as per Peña et al.12
to determine IC50 values. Cytotoxicity (CC50) was measured
for active compounds using H9c2 host cells, in the reported
image-based assay; as well as HepG2 cells in an Alamar Blue-
based assay to assess viability.12 Resazurin-based viability
assays were carried out by Discovery Biology, at Griffith
University to determine compound activity against T.b. brucei
lister 427 cells, and cytotoxicity against human embryonic
kidney cells (HEK-293) cells. The T.b. brucei assay was
undertaken following the method outlined by Sykes et al.29
with slight modifications to the assay as described by Duffy
et al.30
Potency of acyl substituents. The first series of compounds
investigated (Table 1) explored the substituents on the acyl
ring (right hand side [RHS] as drawn) against T. cruzi. The
investigation involved the significance of the para-choro atom
in the hit as well as the ring itself. The hit 1 was reconfirmed
as active, though found to be less active than originally
reported. The T. cruzi assay shows an electron withdrawing
group of similar size in the para position seems to relatively
maintain activity as seen in the para-bromo and methoxy
groups (4b and f respectively) whilst slight improvements
were observed in the trifluoromethyl and trifluoromethoxy
groups (4h and
i respectively). However, removing the
functionality completely (4a) was shown to decrease activity
(IC50 = 4.0 μM). Moving substituents to the ortho or meta
position of the acyl ring, particularly chloro and bromo
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.

View Article Online
Research Article
RSC Medicinal Chemistry
Table 3 SAR surrounding changes to the N′-alkyl chain
#
R
T. cruzi IC50 H9c2a,b (μM) CC50 H9c2a (μM) T.b. brucei IC50 HEK-293c,d (μM) CC50 HEK-293e (μM) CC50 HepG2f (μM) Log P
5a
H
50
>50
>50
40
13
>50
n.a
>83
>50
>50
>50
>50
>50
3.6
3.8
4.2
5.2
5.6
5b Methyl 3.2 0.07
5c Ethyl
5d Butyl
3.8 2.0
3.1 0.92
4.4 1.4
3.7 1.6
>83
2.5 0.29
1.6 0.48
>83f
>83f
>83f
5e Pentyl 1.0 0.05
a
b
Anti T. cruzi activity and toxicity measured in H9c2 host cell lines. Values are means of two experiments. Control compounds for T. cruzi
c
assay. Benznidazole IC50 = 1.58 μM, CC50 = 31.6 0.00 μM. Anti T.b. brucei and cytotoxicity measured in HEK-293 cell lines. (1) >50% activity
at the top concentration tested (83 μM). Values are means of four experiments. Control compounds for T.b. brucei assay. pentamidine IC50
d
=
0.0014 0.0050 μM, CC50 = <50% activity at 0.67 μM, diminazene aceturate IC50 = 0.087 0.031 μM, CC50 = <50% activity at 38.5 μM,
e
puromycin IC50 = 0.038 0.0019 μM, CC50 = 0.23 0.012 μM. >50% activity displayed at 83 μM. Values are the means of two experiments.
f
Toxicity measured in HepG2 cell lines. Values are means of two experiments. n.a <50% activity. standard deviation.
Table 4 SAR surrounding changes to the linker chain and amide
#
X
T. cruzi IC50 H9c2a,b (μM) CC50 H9c2a (μM) T.b. brucei IC50 HEK-293c,d (μM) CC50 HEK-293e (μM) CC50 HepG2f (μM) Log P
4a CH2
6a No linker 5.0 0.70
6b (CH)2
6c (CH2)2
6d (CH2)3
4.0 0.09
>50
>50
20
>50
16
4.1 1.3
3.8 2.2
2.1 0.92
3.5 1.1
2.4 0.60
5.4 2.7
>83
>83f
>50
25
40
>50
>50
40
4.1
4.1
4.6
4.6
5.0
4.8
4.1
>83f
1.3 0.26
5.0 0.47
1.6 0.15
2.5 0.64
10 3.26
>83f
14 0.1
>83f
9
NH
>50
16
5.2 1.6
>83
15 CH2
>50
a
b
Anti T. cruzi activity and toxicity measured in H9c2 host cell lines. Values are means of two experiments. Control compounds for T.cruzi
c
assay. Benznidazole IC50 = 1.58 μM, CC50 = 31.6 0.00 μM. Anti T.b. brucei and cytotoxicity measured in HEK-293 cell lines. (1) >50% activity
at the top concentration tested (83 μM). Values are means of four experiments. Control compounds for T.b. brucei assay. pentamidine IC50
d
=
0.0014 0.0050 μM, CC50 = <50% activity at 0.67 μM, diminazene aceturate IC50 = 0.087 0.031 μM, CC50 = <50% activity at 38.5 μM,
e
puromycin IC50 = 0.038 0.0019 μM, CC50 = 0.23 0.012 μM. >50% activity displayed at 83 μM. Values are the means of two experiments.
f
Toxicity measured in HepG2 cell lines. Values are means of two experiments. standard deviation.
groups (4n–o, s) was shown to maintain activity. Slight
improvement was observed in the meta-bromo group (4t).
Sharp negative SAR in compounds 4d–e, containing a p-acetyl
and p-hydroxy groups respectively was observed. These
compounds showed a significant loss in activity (32 μM, 40
μM respectively). This suggests the electron donating hydroxy
group (4e) should be avoided. The acyl group of 4d may have
been cleaved to reveal the phenol, causing the poor activity.
This series of selected analogues also showed relatively low
cytotoxicity to the host cell.
changes to the RHS, the SAR is shown to be relatively
shallow, although the cytotoxicity remains reassuringly low.
Though a short alkyl chain (4l) is not tolerated, causing a loss
of activity and selectivity versus the host cell (IC50 = 13 μM,
CC50 = 10 μM, H9c2 cell lines), signifying not all substituents
can exist in this chemical space. Overall, a relatively wide
range of functional groups are tolerated in this position and
could allow for refinement of ADMET properties without
large consequence to anti-T. cruzi activity.
Against T.b. brucei, the hit compound 1 was reconfirmed
as active, with an IC50 of 7.1 μM. As with T. cruzi, a range of
substituents around the ortho, meta and para positions were
tolerated on the RHS phenyl ring. Removing the substituent
The addition of phenyl rings (4j, k, x) along with a long
aliphatic chain (4m) was also shown to maintain activity in
comparison to the hit (1). Even with these more dramatic
RSC Med. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
RSC Medicinal Chemistry
Research Article
correlation can be made between decreasing the N-alkyl
chain to a loss of activity. As seen in compound 5c (N-ethyl),
and then in 5a–b (NH, N-methyl respectively) showing a
complete loss of activity. This suggests that an N-alkyl chain
is required to maintain activity. This assay also showed that
increasing the chain length 5d–e gave
a
marginal
improvement to potency whilst greatly increasing
lipophilicity. This will be discussed further in the LiPE
section of this paper.
Scheme
2 One pot formation of the urea linker 9. Reaction
conditions and yield: (a) triphosgene, sodium bicarbonate, DCM, 0 °C-
rt; (b) 1-propyl-1H-benzoijd]imidazol-2-amine (3a), 44% one pot.
Based on Table 3, a carbon chain at the N-alkyl position is
required to maintain anti-T.b. brucei activity and can tolerate
various lengths.
SAR surrounding linker analogues. By altering the carbon
chain length, it was found that shortening the chain (6a)
correlated to a decrease in anti-T. cruzi activity (Table 4).
Conversely, elongating the chain 6b–d, was shown to increase
activity. Interestingly, the same pattern was observed with
anti-T. brucei activity, where compounds 6b and 6d were
shown to give the largest improvements to activity within the
des-chloro subseries, and were more potent than the closest
des-chloro comparator, 4a. This would suggest an elongated
(4a) was shown to improve activity, dropping the IC50 down to
4.1 μM. In fact, the majority of the analogues surrounding this
series was shown to improve activity. Replacing the 4-Cl with
other halogens at this position was also shown to improve
efficacy, the 4-Br (4b), 4-OCF3 (4i) and 4-F (4c) gave reduced
activities of 5.0 μM, 3.2 μM and 2.5 μM respectively. Moving the
chloro group (4s) to the meta position was found to reduce
activity to 3.6 μM. Other small halogen groups at this position,
such as the 3-F (4u) and 3-Br (4t) were also shown to improve
activity further (IC50 = 3.2 μM, 2.9 μM respectively). Moving the
halogen groups to the ortho position was shown to maintain
activity, with the 2-Br (4o) giving the greatest increase in activity
of this series (IC50 = 2.0 μM).
The addition of phenyl rings (4j, k, x) was also shown to
improve activity (3.4 μM, 5.5 μM, 2.6 μM respectively),
suggesting this system can tolerate larger groups in this
chemical space. In contrast the replacement of the RHS ring
with aliphatic chains (4l and m) was shown to reduce activity
(IC50 = 11 μM, 7.3 μM respectively) suggesting a more rigid
system is necessary for greater anti-T.b. brucei activity.
Additionally, strong electron donating groups (4e) gave a
complete loss of activity and should be avoided. From this
series, a rigid ring with small halogen substituents seen as
most favourable. Further investigation involving these
features is encouraged, though based on the data seen in
Table 1 a wider range of functional groups (rings, weak
electron donating groups) can be tolerated.
chain would be preferable for
a
broad spectrum
trypanosomacidal activity, albeit that close attention should
be paid to limit concomitant increases in lipophilicity.
Introducing a urea linker gave no improvement in anti-T.
cruzi activity. A slight increase in anti-T.b. brucei activity was
observed but is outweighed by the sharp increase in
cytotoxicity. Reversal of the amide (15) was shown to give a
loss of activity.
Benzimidazole substituent (LHS) SAR. With this set of
selected substituents, we were able to see interesting though
differing results between assays. Based on results in Table 5,
against T. cruzi, we were able to see clear changes in SAR. We
observed that adding the 4-Br (20a) decreased potency
against T.cruzi (10 μM), whilst adding halogens to the 5 or 6
positions showed the most improvement to anti T. cruzi
activity out of all the analogues synthesized during this
project. These analogues include compounds 20c–g. Having
halogens in this position could indicate extra non-covalent
interactions between the compound and binding site.
However, cytotoxicity was also seen to increase, particularly
in the chloro-substituents 20c and 20e (IC50 = 0.50 μM, CC50
SAR surrounding changes to the N-alkyl chain. From the
biological results reported in Table 3, against T. cruzi, a direct
Scheme 3 Synthetic pathways of reverse amide (15). Reaction conditions and yield (a) Trichloroacetonitrile, KOH in water, Bu4N+·HSO4−, DCM,
99% (b) o-phenylenediamine, acetic acid, 95% (c) sodium carbonate, MeOH, reflux, 40% (d) bromopropane, K2CO3, DMF, 76% (e) (4-chlorophenyl)
methanamine (20 equiv.), 130 °C neat, 4%.
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.

View Article Online
Research Article
RSC Medicinal Chemistry
Table 5 Benzimidazole substituent (LHS) SAR of isoquinoline analogues
T. cruzi IC50 H9c2a,b
(μM)
CC50 H9c2a
(μM)
T.b. brucei IC50 HEK-293c,d
(μM)
CC50 HEK-293e
(μM)
CC50 HepG2f
(μM)
Log
P
#
R
1
H
4-Br
4-OMe
5-Cl
5-Br
6-Cl
6-Br
6-F
6-Me
7-Br
7-Me
Allyl isoquinoline
Alky lisoquinoline
2.0 1.3
10 0.70
3.2 0.37
0.5 0.05
0.79 0.22
0.5 0.01
0.4 0.04
1.3 0.06
—
>50
>50
>50
7.94
>50
6.3
13 0.15
25 0.15
—
7.1 1.2
19 3.1
>83
>83
54%
>83
84%
>83
>83
>83
14 0.58
>83
>83
>83
>83
—
>50
>50
>50
—
>50
>50
>50
>50
—
4.7
5.5
4.6
5.3
5.5
5.3
5.5
4.9
5.2
5.5
5.2
5.1
5.4
20a
20b
20c
20d
20e
20f
20g
20h
20i
20j
24
4.4 3.3
16 9.9
2.4 1.5
1.5 0.86
0.90 0.29
15 5.3
2.8 1.1
9.4 1.1
>83
—
—
—
1.3 0.23
20 3.3
0.5 0.13
>50
>50
3.2
>50
>50
7.94
25
—
a
b
Anti T. cruzi activity and toxicity measured in H9c2 host cell lines. Values are means of two experiments. Control compounds for T.cruzi
c
assay. Benznidazole IC50 = 1.58 μM, CC50 = 31.6 0.00 μM. Anti T.b. brucei and cytotoxicity measured in HEK-293 cell lines. (1) >50% activity
at the top concentration tested (83 μM). Values are means of four experiments. Control compounds for T.b. brucei assay. pentamidine IC50
d
=
0.0014 0.0050 μM, CC50 = <50% activity at 0.67 μM, diminazene aceturate IC50 = 0.087 0.031 μM, CC50 = <50% activity at 38.5 μM,
e
puromycin IC50 = 0.038 0.0019 μM, CC50 = 0.23 0.012 μM. >50% activity displayed at 83 μM. Values are the means of two experiments.
f
Toxicity measured in HepG2 cell lines. Values are means of two experiments. — indicates that the compound was not tested. standard
deviation.
Scheme 4 Synthesis pathway of substituted benzimidazole targets. Reaction conditions: (a) Propylamine, triethylamine, DMF, under N2, 29–97%
(b) Pt/C or Pd/C, MeOH, H2, 50–99% (c) CNBr, MeOH, 12–99% (d) 4-chlorophenylacetic acid, EDCI, HOBt, DMF, 3–90%. See Experimental section
for individual reaction yields.
Scheme 5 Synthetic pathway of isoquinolines (24–25). Reaction conditions: (a) NBS, NH4OAc, ACN, 80% (b) allyltributylstannane, PdIJPPh3)4, DMF,
reflux, 67% (c) 4-chlorophenylacetic acid, EDCI, HOBt, DMF, 18% (d) Pt/C, MeOH, H2, 97%.
RSC Med. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
RSC Medicinal Chemistry
Research Article
Chart 1 Log P vs. T. cruzi pIC50: LiPE plot. Analogue series 1–5, blue: RHS phenyl modification, green: LHS benzimidazole substitution, yellow-
green: isoquinoline analogues, red: amide linker changes, magenta: N-alkyl modification, “*” two analogues present with same activity and log P.
A: compound 20f, B: 20c, C: 20e, D: 5e, E: 4j, F: 4m. Graph generated with software by Collaborative Drug Discovery Inc. CDD Vault Vision
Module.
= 7.9 μM, IC50 = 0.5 μM, CC50 = 6.3 μM respectively). Based
on this, further exploration at the 5/6 positions may yield
attractive results and is encouraged.
existence of a lipophilic pocket in this chemical space as it is
expected that a sharper increase in activity would be observed
if the lipophilic pocket was present. Instead, the increased
activity could simply be due to an increased ability of a more
lipophilic compound to permeate the cell and be driven into
hydrophobic intracellular environments that could increase
parasite activity and toxicity.
Compound 5e can be calculated to give a LiPE of 0.4,
suggesting the minor improvement in activity was likely not
due to favourable binding site interactions but rather was
achieved due to increased lipophilicity. This could also be
said of compounds 4j–k, m (Table 1, phenyl ring and long
aliphatic chain RHS substituents).
To help understand the relationship between activity and
lipophilicity within our library of analogues, a LiPE plot (-
Chart 1) was incorporated, measuring compound lipophilicity
(x axis) against its potency (y axis). In terms of the LiPE,
compounds 20c–f (halogens added to the LHS ring, plotted
in green) show a larger improvement in anti-Chagas activity
over the majority of our assembled analogues. Though log P
has increased with the addition of halogen substituents, it
would seem these changes may be more worthwhile. Only
one small atom change is required, suggesting this simple
halogen addition in compounds 20c–f may be associated with
favourable binding site interactions. In comparison, the
positive impact that compound set 5b–e and 4j–k, m has on
Alterations to this portion of chemical space gave the
greatest improvements to potency against both T. cruzi and T.b.
brucei. Compounds 20e–g contain halogens at the 6-position of
the LHS ring and were shown to improve activity. Most notably,
compound 20g gave the largest improvement to potency of the
series against T.b. brucei. Interestingly, adding electron
withdrawing groups to the 4-position was also shown to greatly
improve anti-HAT activity.
The introduction of a bulkier, more electron deficient
isoquinoline over the benzimidazole (1) may give a better fit
in the putative binding site by sterics and pi–pi interaction.
Against T. cruzi the saturated propyl chain isoquinoline (25)
showed a relatively large improvement in activity. Despite an
increase in activity, increased cytotoxicity was also observed.
This could be to due to the saturated propyl chain since this
toxicity was not observed with the allyl group. Furthermore,
the allyl group (24) was shown to decrease activity (Table 4),
indicating it was surprisingly unfavourable. Further work is
encouraged to understand this system.
LiPE effect. Lipophilic efficiency was consulted throughout
the exploration of our analogue set to determine whether
changes and improvements to anti-T. cruzi activity are worth
the increase in lipophilicity. Lipophilic efficiency calculations
(LiPE = pIC50 − log P), were used to further measure the
improvement of activity in relation to lipophilicity, where
LiPE measures the quality of a drug using pIC50 (activity) and
the partition coefficient (lipophilicity).31 Compound showing
a high LiPE suggests quality in a compound and that
improvement in activity is not due to lipophilicity alone.31
Compound set 5b–e is an example of the lipophilic effect,
and the increase in carbon chain length from N-ethyl to a
pentyl chain gave a shallow increase in activity (3.2 μM to 1
μM). The SAR based on these compounds cannot imply the
potency is associated increased hydrophobicity
– long
aliphatic chains and aromatic rings – suggesting these
additions to be less productive.
Although there is a general trend towards greater potency
with greater lipophilicity, within the trend there are some
compounds that are relatively more active and less lipophilic
(green spots A, B, C, 20f, c and e respectively) and some that
are relatively less active and more lipophilic (magenta spot D
5e and blue spot E, F 4j, m respectively). The former type of
compound is more attractive as a development prospect. That
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.

View Article Online
Research Article
RSC Medicinal Chemistry
said, the LiPE values of 20f, 20 and 20 in absolute terms is
low (LipE ≤ 1, see LiPE calculations in ESI†) and further
exploration would usefully include a focus on continued
reduction of lipophilicity while improving potency.
thanked for Fellowship support for J. B. The Australian
Translational Medicinal Chemistry Facility (ATMCF) within
Monash Institute of Pharmaceutical Sciences (MIPS),
acknowledges the support of the Australian Government's
National Collaborative Research Infrastructure Strategy (NCRIS)
program via Therapeutic Innovation Australia (TIA).
Conclusions
Our SAR exploration of substituted benzimidazole and
isoquinoline phenyl acetamides of the HTS-hit 1 (Table 1)
against T. cruzi and T.b. brucei has led to identification of a
series of novel compounds with low μM antiparasitic activity,
and which are generally not appreciably cytotoxic. A visual SAR
summary highlighting the significant changes to this scaffold
has been included in the ESI.† The SAR observed against T. cruzi
broadly parallels that observed against T.b. brucei and suggests
that they may be similar, although confirmation of activity on
human-infective sub-species of T. brucei (T.b. rhodesiense and T.
b. gambiense) would be warranted. Based on T. cruzi activity,
particularly compounds 20c–f, 25, future SAR could usefully
focus on altering the LHS benzimidazole substituents and
investigating isoquinoline replacements. Having said this,
aspects of the SAR were relatively shallow, and there was a
general lipophilic effect observed, suggesting that further
optimization may be relatively challenging. This is important
information for the anti-kinetoplastid drug discovery arena,
because high throughput phenotypic screening has led to the
advent of overwhelming numbers of active compounds being
deposited in public databases. It is vital for this drug discovery
community to be able to assess active chemotype and rank
accordingly in terms of those that may be more challenging to
develop and those that may be less challenging to develop.
References
1 World Health Organization, Chagas disease (American
trypanosomiasis), 2018 [cited 2018 April]. Available from:
http://www.who.int/news-room/fact-sheets/detail/chagas-
disease-(american-trypanosomiasis).
2 World Health Organization, Trypanosomiasis, human
African (sleeping sickness), 2018 [cited 2018 December 12].
Available from: https://www.who.int/news-room/fact-sheets/
detail/trypanosomiasis-human-african-(sleeping-sickness).
3 DNDI, About Chagas Disease. [cited 2018 October]. Available
from: https://www.dndi.org/diseases-projects/chagas/.
4 DNDi, About Sleeping Sickness – DNDi 2018 [cited 2018
December 12]. Available from: https://www.dndi.org/
diseases-projects/hat/.
5 C. H. Baker and S. C. Welburn, The Long Wait for a New
Drug for Human African Trypanosomiasis, Trends Parasitol.,
2018, 34(10), 818–827.
6 P. Büscher, J.-M. Bart, M. Boelaert, B. Bucheton, G. Cecchi
and N. Chitnis, et al. Do Cryptic Reservoirs Threaten
Gambiense-Sleeping Sickness Elimination?, Trends Parasitol.,
2018, 34(3), 197–207.
7 A. Velasco and C. A. Morillo, Chagas heart disease: A contemporary
review, Journal of Nuclear Cardiology, 2020, 27(2), 445–451.
8 N. M. Matsuda, S. M. Miller and P. R. B. Evora, The chronic
gastrointestinal manifestations of Chagas disease, Clinics,
2009, 64(12), 1219–1224.
9 World Health Organization, WHO | Symptoms, diagnosis
and treatment 2016 [updated 2016-08-05 15:41:52June 2019].
Available from: https://www.who.int/trypanosomiasis_
african/diagnosis/en/.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
The authors would sincerely like to thank GlaxoSmithKline,
Tres Cantos, Spain for allowing for us to investigate hit
compounds taken from their HTS. They generously continued
to add our library of analogues to their screening plates and
kindly assessed their biological activity. This work was
supported by funding from a Griffith University Postdoctoral
Fellowship (GUPF) awarded to MLS. The authors would like to
thank Elouise Gaylord for assisting with the T.b. brucei and
HEK293 assays undertaken in Discovery Biology. The authors
would like to sincerely thank DNDi and their donors for
financial support of this work. For the work described in this
paper, DNDi received financial support from the following
donors: the Swiss Agency for Development and Cooperation
(SDC), Switzerland; the Directorate General for International
Cooperation (DGIS), The Netherlands; (DFID), UK. The donors
had no role in study design, data collection and analysis,
decision to publish, or preparation of the manuscript. This
work was also supported by funding from The National Health
and Medical Research Council of Australia (NHMRC) is
10 J. A. Castro and E. G. Diaz de Toranzo, Toxic effects of
nifurtimox and benznidazole, two drugs used against
American trypanosomiasis (Chagas' disease), Biomed.
Environ. Sci., 1988, 1(1), 19–33.
11 World Health Organization, Research Priorities for Chagas
Disease,
Human
African
Trypanosomiasis
and
Leishmaniasis, Technical report series 975, 2012, [Available
from: http://www.who.int/iris/handle/10665/77472.
12 I. Peña, M. Pilar Manzano, J. Cantizani, A. Kessler, J. Alonso-
Padilla and A. I. Bardera, et al. New Compound Sets
Identified from High Throughput Phenotypic Screening
Against Three Kinetoplastid Parasites: An Open Resource,
Sci. Rep., 2015, 5, 8771.
13 S. Vidavalur, M. B. Gajula, R. Tadikonda, M. Nakka, S. Dega
and S. K. Yadav, et al. PTSA catalyzed straightforward
protocol for the synthesis of 2-(N-acyl)aminobenzimidazoles
and 2-(N-acyl)aminobenzothiazoles in PEG, Tetrahedron Lett.,
2014, 55(16), 2691–2694.
RSC Med. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
RSC Medicinal Chemistry
Research Article
14 X. He, S. K. Lakkaraju, M. Hanscom, Z. Zhao, J. Wu and B.
Stoica, et al. Acyl-2-aminobenzimidazoles: A novel class of
neuroprotective agents targeting mGluR5, Bioorg. Med.
Chem., 2015, 23(9), 2211–2220.
15 X. Guida, H. Jianhua and L. Xiaomin, Synthesis and QSAR
studies of novel 1-substituted-2-aminobenzimidazoles
derivatives, Eur. J. Med. Chem., 2006, 41(9), 1080–1083.
16 L. V. Lafrance, J. D. Leber, L. Mei and S. K. Verma,
Serotonin Transporter Affinity within
N-Aryloxyethylindolylalkylamines, J. Med. Chem., 2004, 47(15),
3823–3842.
a
Class of
24 A. Ramanathan and L. S. Jimenez, Reductive Dehalogenation
of Aryl Bromides and Chlorides and Their Use as Aryl
Blocking Groups, Synthesis, 2010, 2010(02), 217–220.
25 K. Garg, Y. Bansal, G. Bansal and R. K. Goel, Design,
synthesis, and PASS-assisted evaluation of novel
Benzimidazolecarboxamides
WO2010126922A1, 2010.
17 H. Takasugi, T. Terasawa, Y. Inoue, H. Nakamura, A.
Nagayoshi, H. Ohtake, Y. Furukawa, M. Mikami, K. Hinoue
and M. Ohtsubo, Preparation of pyridinyl-substituted
as
inhibitors
of
FAK,
2-substituted
benzimidazole
derivatives
as
potent
anthelmintics, Med. Chem. Res., 2014, 23(5), 2690–2697.
26 M. C. Jetter, M. J. Macielag, M. Xia and X. Xu, Benzimidazole
derivatives useful as transient receptor potential melastatin
subfamily type 8 channel modulators, WO2010144680A1,
2010.
benzamides
as
Apo
B
secretion
inhibitors,
WO2002028835A1, 2002.
27 G. Van Baelen, G. L. F. Lemière, R. A. Dommisse and
B. U. W. Maes, Synthesis of 5-methyl-5H-pyrroloij2,3-
c]quinoline and 4-methyl-4H-pyrroloij2,3-c]isoquinoline: two
new unnatural D-ring stripped isomers of the cryptolepine
series, ARKIVOC, 2009, 2009(6), 174–182.
28 S. B. Hoyt, C. London, D. Ok and W. H. Parsons, Preparation
of benzazepine amino acids as sodium channel blockers,
WO2007145922A2, 2007.
29 M. L. Sykes, J. B. Baell, M. Kaiser, E. Chatelain, S. R.
Moawad and D. Ganame, et al. Identification of
Compounds with Anti-Proliferative Activity against
Trypanosoma brucei brucei Strain 427 by a Whole Cell
Viability Based HTS Campaign, PLoS Neglected Trop. Dis.,
2012, 6(11), e1896.
30 S. Duffy, M. L. Sykes, A. J. Jones, T. B. Shelper, M.
Simpson and R. Lang, et al. Screening the Medicines for
Malaria Venture Pathogen Box across Multiple Pathogens
Reclassifies Starting Points for Open-Source Drug
Discovery, Antimicrob. Agents Chemother., 2017, 61(9),
e00379-17.
18 C. E. Brotherton-Pleiss, Z. Chen, S. D. Erickson, K. Kim, H.
Li, A. J. Lovey, Y. Qian, S.-S. So, P. M. Wovkulich and L. Yi,
Preparation of substituted carbamate compounds as
transient receptor potential (TRP) channel antagonists,
WO2014060341A1, 2014.
19 X. Geng and C. Wang, Rhenium-catalyzed C–H
aminocarbonylation of azobenzenes with isocyanates, Org.
Biomol. Chem., 2015, 13(28), 7619–7623.
20 C. Zanato, Approaches to the Total Synthesis of Dictyostatin
and Synthesis of epi-Dictyostatins, PhD thesis, Universita
Degli Studi di Milano, 2010.
21 J. R. Allen, J. J. Chen, M. J. Frohn, E. Hu, Q. Liu, A. J.
Pickrell, S. Rumfelt, R. M. Rzasa and W. Zhong, Preparation
of nitrogen heterocyclic compounds useful as 3′,5′-cyclic
nucleotide-specific phosphodiesterase 10 (PDE10) inhibitors,
WO2011143365A1, 2011.
22 P. Zhang, E. A. Terefenko, J. Bray, D. Deecher, A. Fensome
and J. Harrison, et al. 1- or 3-(3-Amino-2-hydroxy-1-phenyl
propyl)-1,3-dihydro-2H-benzimidazol-2-ones: Potent, Selective,
and Orally Efficacious Norepinephrine Reuptake Inhibitors,
J. Med. Chem., 2009, 52(18), 5703–5711.
23 R. E. Mewshaw, D. Zhou, P. Zhou, X. Shi, G. Hornby and T.
Spangler, et al. Studies toward the Discovery of the Next
Generation of Antidepressants. 3. Dual 5-HT1A and
31 M. P. Edwards and D. A. Price, Chapter 23 – Role of
Physicochemical Properties and Ligand Lipophilicity
Efficiency in Addressing Drug Safety Risks, In: Annual
Reports in Medicinal Chemistry, ed. J. E. Macor, Academic
Press, 2010, vol. 45, pp. 380–391.
This journal is © The Royal Society of Chemistry 2020
RSC Med. Chem.