Anti-schistosomal activities of quinoxaline-containing compounds: From hit identification to lead optimisation

Article abstract of DOI:10.1016/j.ejmech.2021.113823

Schistosomiasis is a neglected disease of poverty that is caused by infection with blood fluke species contained within the genus Schistosoma. For the last 40 years, control of schistosomiasis in endemic regions has predominantly been facilitated by administration of a single drug, praziquantel. Due to limitations in this mono-chemotherapeutic approach for sustaining schistosomiasis control into the future, alternative anti-schistosomal compounds are increasingly being sought by the drug discovery community. Herein, we describe a multi-pronged, integrated strategy that led to the identification and further exploration of the quinoxaline core as a promising anti-schistosomal scaffold. Firstly, phenotypic screening of commercially available small molecules resulted in the identification of a moderately active hit compound against Schistosoma mansoni (1, EC₅₀ = 4.59 μM on schistosomula). Secondary exploration of the chemical space around compound 1 led to the identification of a quinoxaline-core containing, non-genotoxic lead (compound 22). Compound 22 demonstrated substantially improved activities on both intra-mammalian (EC₅₀ = 0.44 μM, 0.20 μM and 84.7 nM, on schistosomula, juvenile and adult worms, respectively) and intra-molluscan (sporocyst) S. mansoni lifecycle stages. Further medicinal chemistry optimisation of compound 22, resulting in the generation of 20 additional analogues, improved our understanding of the structure-activity relationship and resulted in considerable improvements in both anti-schistosome potency and selectivity (e.g. compound 30; EC₅₀ = 2.59 nM on adult worms; selectivity index compared to the HepG2 cell line = 348). Some derivatives of compound 22 (e.g. 31 and 33) also demonstrated significant activity against the two other medically important species, Schistosoma haematobium and Schistosoma japonicum. Further optimisation of this class of anti-schistosomal is ongoing and could lead to the development of an urgently needed alternative to praziquantel for assisting in schistosomiasis elimination strategies.

Full text of DOI:10.1016/j.ejmech.2021.113823

European Journal of Medicinal Chemistry 226 (2021) 113823
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Anti-schistosomal activities of quinoxaline-containing compounds:
From hit identification to lead optimisation
a
b
b
b
c
Gilda Padalino , Nelly El-Sakkary , Lawrence J. Liu , Chenxi Liu , Danielle S.G. Harte ,
c
d
a
a
Rachel E. Barnes , Edward Sayers , Josephine Forde-Thomas , Helen Whiteland ,
Marcella Bassetto , Salvatore Ferla , George Johnson , Arwyn T. Jones ,
Conor R. Caffrey , Iain Chalmers , Andrea Brancale , Karl F. Hoffmann
Institute of Biological, Environmental and Rural Sciences (IBERS), Aberystwyth University, Aberystwyth, SY23 3DA, United Kingdom
e
c
c
d
b
a
d
a, *
a
b
Center for Discovery and Innovation in Parasitic Diseases (CDIPD), Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San
Diego, La Jolla, CA 92093, USA
c
Swansea University Medical School, Swansea University, Swansea, SA2 8PP, United Kingdom
School of Pharmacy and Pharmaceutical Sciences, Cardiff University, Redwood Building, King Edward VII Avenue, Cardiff, CF10 3NB, United Kingdom
Department of Chemistry, College of Science and Engineering, Swansea University, Swansea, SA2 8PP, United Kingdom
d
e
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 July 2021
Received in revised form
Schistosomiasis is a neglected disease of poverty that is caused by infection with blood fluke species
contained within the genus Schistosoma. For the last 40 years, control of schistosomiasis in endemic
regions has predominantly been facilitated by administration of a single drug, praziquantel. Due to
limitations in this mono-chemotherapeutic approach for sustaining schistosomiasis control into the
future, alternative anti-schistosomal compounds are increasingly being sought by the drug discovery
community. Herein, we describe a multi-pronged, integrated strategy that led to the identification and
further exploration of the quinoxaline core as a promising anti-schistosomal scaffold.
13 August 2021
Accepted 27 August 2021
Available online 10 September 2021
Keywords:
Firstly, phenotypic screening of commercially available small molecules resulted in the identification of
Schistosomiasis
Quinoxaline
SAR
a moderately active hit compound against Schistosoma mansoni (1, EC50 ¼ 4.59
mM on schistosomula).
Secondary exploration of the chemical space around compound 1 led to the identification of a
Drug discovery
quinoxaline-core containing, non-genotoxic lead (compound 22). Compound 22 demonstrated sub-
stantially improved activities on both intra-mammalian (EC50 ¼ 0.44
mM, 0.20 mM and 84.7 nM, on
schistosomula, juvenile and adult worms, respectively) and intra-molluscan (sporocyst) S. mansoni
lifecycle stages. Further medicinal chemistry optimisation of compound 22, resulting in the generation of
20 additional analogues, improved our understanding of the structure-activity relationship and resulted
in considerable improvements in both anti-schistosome potency and selectivity (e.g. compound 30;
EC50 ¼ 2.59 nM on adult worms; selectivity index compared to the HepG2 cell line ¼ 348). Some de-
rivatives of compound 22 (e.g. 31 and 33) also demonstrated significant activity against the two other
medically important species, Schistosoma haematobium and Schistosoma japonicum. Further optimisation
of this class of anti-schistosomal is ongoing and could lead to the development of an urgently needed
alternative to praziquantel for assisting in schistosomiasis elimination strategies.
©
2021 The Author(s). Published by Elsevier Masson SAS. This is an open access article under the CC BY
license (http://creativecommons.org/licenses/by/4.0/).
1. Introduction
[1,2]. It is currently listed in the World Health Organisation's road
map for ‘Ending the Neglect to Attain the Sustainable Development
The neglected tropical disease (NTD) schistosomiasis is the
second most debilitating human parasitic disease after malaria
Goals’ with an ambition to eliminate schistosomiasis as a public
health problem in 78 endemic countries by 2030 [3]. In the absence
of a clinically-approved vaccine, the complete (over) reliance on a
single chemotherapy (praziquantel, PZQ) to treat schistosomiasis
represents a serious challenge in reaching this ambitious objective.
Therefore, there is an urgent need to identify novel anti-
*
Corresponding author.
E-mail address: krh@aber.ac.uk (K.F. Hoffmann).
https://doi.org/10.1016/j.ejmech.2021.113823
223-5234/© 2021 The Author(s). Published by Elsevier Masson SAS. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
0

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
schistosomals as an alternative to PZQ, should drug insensitive
schistosomes develop [4], or to be used in combination with PZQ, to
improve upon some of PZQ's limitations [5,6] (e.g. ineffectiveness
against juvenile stage schistosomes [7]). Towards this goal, several
drug discovery strategies have been applied in the pursuit of
identifying novel chemotherapeutic agents and include (e.g.) drug
repositioning [8e10], ‘piggybacking’ approaches [11e13], machine
learning [14] and the de novo design of compounds using either
ligand- or target-based molecular modelling approaches [15e19].
Regardless of the strategy used to initiate drug discovery for
schistosomiasis, next-generation anti-schistosomals should be
prioritised for their effectiveness in killing schistosome lifecycle
stages found in the human host (e.g. schistosomula, juvenile worms
and adult worms) as well as their ability to inhibit the production of
eggs, which drives both pathogenesis and transmission of schis-
tosomiasis [20]. Compounds should also demonstrate broad anti-
schistosomal activity against human-infective species endemic to
both New (e.g. South America) and Old (e.g. Africa) Worlds (Schis-
tosoma mansoni) [21], species responsible for most pathology as
well as additionally linked to bladder cancer (Schistosoma haema-
tobium) [22,23] and species predominantly responsible for schis-
tosomiasis in Asia (Schistosoma japonicum) [24]. Finally, anti-
schistosomal compounds should show a greater selectivity for the
parasite (when compared to the definitive host) and not display
overt genotoxicity. While other criteria (e.g. cost of goods, synthesis
steps, compound stability/shelf life, oral delivery, safe in children/
pregnant women/breast-feeding mothers) are being actively dis-
cussed within the schistosomiasis drug-discovery community, the
considerations listed above are often used as go/no-go decision
points in early-stage hit discovery and hit to lead optimisation
projects.
also significantly inhibited the production of in vitro laid eggs
(IVLEs) up until 12.5
mM; reductions in IVLEs were also found at the
lowest concentration tested (6.25
mM), although not significant.
2.2. Secondary screens and prioritisation of lead compound for
follow-up studies
Aiming to characterise related compounds with increased anti-
schistosomal potencies, additional small molecules were sought for
secondary screening based on their structural similarity to com-
pound 1 (similarity threshold 50%) and commercial availability
(from Specs, availability >0 mg at the time this study was per-
formed). Structural diversity of these compounds was focused
broadly around three regions of compound 1: the central
condensed scaffold, functionalisation of the C6 position and sub-
stituents on the N-aromatic ring (Fig. S2). As a result, an additional
23 small molecules (Table S2), containing the commercially-
available, structurally-related modifications were identified and
purchased from the Specs library (from which compound 1 was
originally obtained for the primary screen).
An initial evaluation of these compounds (in addition to com-
pound 1) was performed on the schistosomula stage at 10 and
50
m
M (replicate primary screens for compound/concentration
0
point and Z scores are reported in Table S3). A total of 17/24
compounds were hits at 50
11,16,17,19, 20, 22 and 23; Table 1) retained activity at 10
m
M; however, only 9 compounds (9, 10,
M. Based
M) were selected for
m
on these findings, the 9 hit compounds (at 10
m
0
further titrations (Fig. 2; Z scores of these dose-response titration
screens reported in Table S3). Among the compounds screened,
three (9, 10 and 16; Table 1) showed no improvement in anti-
schistosomal activity compared to the initial hit compound (1).
Analysis of the dose-response curves of the remaining 6 com-
pounds (compounds 11, 17, 19, 20, 22 and 23; Table 1) highlighted a
substantial improvement in potency over compound 1 on both
phenotype and motility metrics (Fig. 2A and B, respectively).
Further investigation of these 9 compounds (compared to
compound 1) aimed to quantify their effect on the adult stage of the
In this study, as part of our ongoing small-molecule screening
activities, we present the discovery of a novel anti-schistosomal hit
compound (1) that led to the identification of a more active,
quinoxaline-containing compound 22. As this compound addi-
tionally demonstrated favourable cytotoxicity and genotoxicity
characteristics, optimisation around the central quinoxaline scaf-
fold was performed resulting in analogues displaying substantial
improvements in both anti-parasitic activity and selectivity. Pro-
gression of these leads represents a promising starting point for the
development of urgently needed, next-generation anti-
schistosomals.
parasite (Fig. 3). Each compound was initially tested at 12.5
which we identified as a sublethal concentration for compound 1
(Fig. 1C) and then at 6.25 and 3.15 M. While all compounds
mM,
m
demonstrated a dose-dependent effect on adult worm motility, six
(11, 16, 19, 20, 22 and 23) were particularly active at 12.5 and
2
. Results and discussion
6.25 mM. Compounds 11, 19 and 22 caused complete cessation of
2.1. Primary screen and hit compound identification
worm mobility even at the lowest concentration tested (3.15
resulting in the most active compounds within the series.
mM),
During phenotypic screening of a pre-plated small molecule li-
brary (the Specs pre-plated fragment-based library, provided as 96
deep well plates with 80 compounds each [25]), compound 1 was
The biological evaluation of these 24 compounds highlights
three main series based on their effects on phenotype and motility
of S. mansoni schistosomula (where the most complete dataset has
been collected, Table 1). Firstly, six chemicals (2, 3, 4, 8, 14 and 21)
did not have any effect on the parasite (even at the higher con-
centration tested). A weak effect on both metrics was recorded for
the second group of compounds (including 5, 6, 7, 12, 13, 15, 18 and
24). Based on the calculated EC50 values, three quinoxaline de-
rivatives (9, 10 and 16) showed a comparable anti-schistosomal
identified as a hit at both 50 and 10
mM on S. mansoni schistosomula
(
Fig. 1A and Fig. S1) after 72 h of co-incubation (number of repli-
0
cates performed and Z scores reported in Table S1). Further dose-
response titration on this intra-mammalian larval stage (Fig. 1B)
showed that this compound retained activity on the parasite down
to 5
m
M (EC50 of 4.72 and 4.47
m
M for phenotype and motility,
0
respectively; Z scores reported in Table S1). Translation of com-
pound 1's anti-schistosomula activity on adult male and female
worm pairs was subsequently investigated (Fig. 1C). At the highest
activity to the hit compound (1, EC50 ¼ 4.72 and 4.47
phenotype and motility, respectively). The remaining derivatives
(11, 17, 19 and 20) had an EC50 value between 1 and 2 M (same
mM on
m
concentrations tested (50 and 25
m
M), compound 1 had a lethal
range as praziquantel, PZQ). Above all, compound 22 was the most
potent compound with an EC50 in the high nanomolar range
effect on adults (absence of body/gut movement, plate detachment
and severe damage to the tegument). At lower compound con-
centrations, worm motility began to recover with individuals dis-
(EC50 ¼ 0.39
mM), comparable to the gold-containing anti-schisto-
somal compound, Auranofin (AUR) [26,27]. These anti-
schistosomula results broadly match the same activity trends
observed for adult worms (e.g. Fig. 3).
playing normal motility at 6.25
mM (i.e. no different than worms co-
cultured in media containing only the DMSO solvent). Compound 1
2

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Fig. 1. Anti-schistosomal activities of compound 1. (A) - Mechanically-transformed schistosomula were incubated for 72 h with a pre-plated library of small molecules (the Specs
pre-plated fragment-based library). Here is shown only a subset of 80 small molecules of this library relevant to this study. Each compound was tested at 10 and 50
m
M (each of
0
them in duplicate). One of the four representative screens (primary screens) is shown here (barcode 0255, Z scores are reported in Table S1). Compounds with activity on both
schistosomula phenotype and motility are shown within the ‘Hit Zone’ (delineated by the dotted red lines, defining - 0.15 and - 0.35 as threshold anti-schistosomula values for
phenotype and motility scores, respectively). The chemical structure of compound 1 is shown here. The grey circles represent the remaining compounds not displaying any sig-
nificant activity. (B) - Compound 1 was screened against mechanically-transformed schistosomula during four repeat dose-response titrations (0.625, 1.25, 2.5, 5 and 10 mM) with
average phenotype (P) and motility (M) scores reported as a heat map. The range of scores is shown as a gradient: dark blue indicates the most positive effect of the compound on
0
either phenotype or motility, light blue represents progressively reduced compound efficacy and white represents minimal/no effect on either phenotype or motility. Z scores for
these four repeat titrations are reported in Table S1. (C) - A dose response titration (50e6.25 mM) of compound 1 was additionally performed on adult S. mansoni worms. After 72 h,
worm motility (black bars) and egg production (orange squares) was quantified. Each titration was performed in three independent screens. A Kruskal-Wallis ANOVA followed by
Dunn's multiple comparisons test was performed (comparing treatments to DMSO control). *** represents p < 0.0002.
2
.3. Preliminary structure-activity relationship (SAR) observations
when the nitro group in C6 position was removed (23 vs 10,
assuming the methyl/ethyl ester did not significantly affect the
in vitro activity), or when replaced by other groups such as a
methoxy group (19 vs 6; Fig. S3B). Despite the limited number of
Comparing the anti-schistosomula activities of these com-
pounds, some trends in structure-activity relationship (SAR) were
found to be associated with four distinctive structural features: (1)
the central scaffold (organosulfur compounds containing one (as
with compounds 2, 3 and 4) or two sulphur atoms (as with com-
pound 1)- or a quinoxaline core - remaining compounds); (2)
different substitutions at the C6 position of the central scaffold; (3)
the linker between the central scaffold and the aromatic rings; (4)
the different substitutions of the aromatic rings (Fig. S2). Interest-
ingly, all compounds that contained both nitrogen and sulphur
heteroatoms on the central scaffold were inactive on the parasite
compounds available, preliminary data suggested
a positive
contribution of the N-linker between the quinoxaline core and the
aromatic rings on the anti-schistosomal activity (11 vs 12 and 15 vs
14; Fig. S3C).
The effect of different substituents and different patterns of
substitution on the two aromatic rings linked to the central scaffold
was investigated next (Fig. S3D). This analysis showed that func-
tionalisation in the meta position (position 3 of the aromatic ring)
alone (19 vs 20; Fig. S3D) or together with a second group in the
para position (position 4 of the aromatic ring, such as with com-
pounds 15 vs 16; Fig. S3D) resulted in moderate to good anti-
schistosomula activity, particularly when a halogen substituent
was introduced (16 vs 17). However, when a group was introduced
(
2, 3 and 4; Table 1). The introduction of the quinoxaline scaffold
induced a consistent increase in activity (exemplified with com-
parison between compounds 11 vs 4, 9 vs 3 and 15 vs 2; Fig. S3A).
We also identified another general trend in activity decreasing
3

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Table 1
Anti-schistosomula activities of compounds 1e24.
Activity on schistosomula
(
EC50, mM - 72 h)
Compound
1
Phenotype
4.72
Motility
Average
4.59
4.47
(
4.04e5.19)
(3.98e5.45)
2
3
4
5
6
7
8
9
>50
>50
>50
>10
>10
>10
>50
5.92
ND
ND
ND
ND
ND
ND
ND
5.39
4.87
(
4.37e7.47)
5.12
2.82e7.42)
1.38
1.12e1.78)
(3.38e6.79)
5.83
(4.40e7.26)
1.12
1
1
0
1
5.47
1.25
(
(
(0.99e1.25)
12
13
14
15
16
>10
>10
>50
>10
5.42
ND
ND
ND
ND
4.55
3.69
(
3.27e7.57)
2.88
2.32e3.44)
(1.69e5.69)
2.45
(0.65e4.25)
17
2.66
(
1
1
8
9
>10
1.89
ND
1.62
1.35
(
1.57e2.28)
1.62
1.36e1.95)
(1.06e1.76)
1.23
(1.01e1.54)
20
1.42
(
2
2
1
2
>50
0.40
ND
0.39
0.38
Fig. 2. Compound 1 related analogues demonstrate variable anti-schistosomula ac-
tivities. A total of 24 compounds (hit compound 1 and 23 structural analogues) were
(
0.25e0.55)
1.23
0.27e2.19)
(0.11e0.49)
1.44
(0.54e2.34)
2
2
3
4
1.33
tested on schistosomula for 72 h at 50
which retained activity at 10 M. These 9 compounds were further titrated (for 72 h)
on schistosomula (0.625, 1.25, 2.5, 5 and 10 M). Two independent dose response ti-
mM resulting in the identification of 17 hits, 9 of
(
m
>10
0.37
ND
0.41
m
AUR
0.45
(0.42e0.47)
0.88
trations were performed, and each compound concentration was evaluated in dupli-
(
0.36e0.39)
1.14
1.10e1.17)
0
cate (Z scores for each screen are reported in Table S3). The mean average on
PZQ
1.01
phenotype and motility is here represented as heat maps (shades defined in Fig. 1) in
panels (A) and (B), respectively. Scores of - 0.15 and - 0.35 are defined as threshold
anti-schistosomula values for phenotype and motility, respectively (shown in red on
the heat map).
(
(0.79e0.97)
Schistosomula EC50 (phenotype and motility metrics as well as the mean of these
two metrics) values of compound 1 and its 23 derivatives were calculated based on
two dose response titrations (0.313e10 mM). Similarly, EC50 values were calculated
for Auranofin (AUR) and praziquantel (PZQ), included here for comparison. Where
no effect was seen on schistosomula at the primary tested concentrations (10 and 50
m
1
M), the EC50 was said to be higher than 50
m
M. The EC50 was said to be higher than
M, but not at 10 M. All the data
schistosomal compounds.
0
m
M if the compound was defined as a hit at 50
m
m
were analysed using GraphPad Prism. ND: Not Determined (where titration was not
performed so EC50 value was not available and arithmetic average was not calcu-
lated). The 95% CI (Confidence Interval) values are reported in brackets.
2
.4. Further anti-schistosomal investigations of the prioritised lead
compound (22)
in the ortho position (8 vs 9; Fig. S3D, assuming equal contribution
of methoxy- or ethoxy-group on the biological activity), a decrease
in anthelmintic activity was observed. Overall, these initial in-
vestigations provided insights into the SAR profile of this com-
pound class (summarised in Fig. S3E) and identified the 2,3-
bis(phenylamino)quinoxalines as promising candidates for further
anti-schistosomal development.
Due to its strong anti-schistosomula (EC50 ¼ 0.39
m
M, Table 1)
and adult worm (motility was completely inhibited at 3.15
mM,
Fig. 3) activities, compound 22 was further titrated against adult
worms and assessed for juvenile worm and miracidia-sporocyst
transformation inhibition actions (Fig. 4).
Initially, the potency of compound 22 on adult worms was
directly compared to PZQ. Here, a dose-response titration of both
To our knowledge, the anti-schistosomal activity of the qui-
noxaline core was first demonstrated during the evaluation of the
Medicines for Malaria Venture (MMV) malaria box on S. mansoni
PZQ and compound 22 (from 0.0095 mM to 10 mM) demonstrated
that both compounds induced a comparable effect on worm
motility (Fig. 4A). Oviposition was similarly inhibited by both
treatments with IVLEs being produced only in cultures containing
nanomolar concentrations of either compound (PZQ performed
slightly better than compound 22; Fig. 4B). When next titrated on
juvenile worms (Fig. 4C), compound 22 induced parasite death
[
2 3
28]. As part of a ‘piggybacking’ approach, a N ,N -bis(4-
bromophenyl)quinoxaline-2,3-diamine derivative (MMV007224)
had in vitro activity in the sub micromolar range and moderate
in vivo activity. While a different approach was applied herein, our
results expand upon these initial observations and further confirm
the utility of the quinoxaline core (containing diverse functional
groups) as a potential component of next generation anti-
down to 1.25
mM (supported by propidium iodide, PI, staining,
Figs. S4A and B). At lower concentrations (down to 0.156
mM),
compound 22 led to increased granularity (Fig. S4C) and reduced
movement (V1 in Supplementary Data), but not lethality
4

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Fig. 3. Dose response titration of compound 1-related analogues on adult S. mansoni worms. The 9 compounds (identified as hits at 10
adult worms at 12.5 M (sub-lethal concentration for 1, here reported as the reference) and two lower concentrations (3.15 and 6.25 M). Each titration was performed in duplicate
across two independent screens. The effect on schistosome motility was quantified using WormassayGP2. The bar chart shows the average worm movements recorded during the
two independent screens, in comparison to the controls (0.625% DMSO and 10 M PZQ in 0.625% DMSO). A Kruskal-Wallis ANOVA followed by Dunn's multiple comparisons test
mM on schistosomula) were screened on
m
m
m
was performed to compare each population mean to DMSO mean. * and ** represent p < 0.0332 and p < 0.0021, respectively.
Fig. 4. Compound 22 is active on both intra-mammalian and intra-molluscan S. mansoni life cycle stages. (A) - A dose response titration (0.0095e10 mM) of compound 22 and PZQ
was performed to assess their comparable potencies on S. mansoni adult worms. The bar chart shows the average worm movements recorded by WomassayGP2 in two independent
screens (each screen contained two technical replicates). The average of each compound was shown in comparison to the standard controls (0.625% DMSO and PZQ 10 M). (B) -
m
After 72 h, eggs were enumerated in both compound 22 and PZQ treatment groups and reported in the scatter chart. For each concentration tested, the mean of the egg count and
the standard error across the two biological and technical replicates were represented on the graph. (C) - Juvenile S. mansoni worms (3 weeks post infection; n ¼ 10e25) were
cultured in different concentrations of compound 22 (0.156, 0.312, 0.625, 1.25, 2.50, 5 and 10
m
M in 1.25% DMSO) and motility scored (0 ¼ dead, 4 ¼ normal movement) at 72 h.
Control parasites (n ¼ 10e25) include those co-cultivated in the presence of 1.25% DMSO and PZQ and AUR (both 15
and 50 M) on miracidia transformation was registered in terms of % fully transformed sporocysts enumerated after 48 h incubation. A Kruskal-Wallis ANOVA followed by Dunn's
multiple comparisons test was performed to compare each population mean to DMSO mean. **, ***, **** represent p < 0.0021, p < 0.0002, p < 0.0001, respectively.
mM in 1.25% DMSO). (D) The effect of compound 22 (0.5, 2, 5, 10
m
(
supported by a reduced PI uptake; Fig. S4). This compound showed
declined to about 20% when the compound was tested at 2 and
0.5 M. Collectively these data demonstrate the broad-spectrum
an EC50 of 0.201 M (0.118e0.27195% CI). Finally, compound 22 also
m
m
demonstrated a concentration-dependent inhibition to miracidial
transformation (Fig. 4D). This compound was lethal to this parasite
activities (e.g. decreases in motility, oviposition and viability) of
compound 22 on diverse (intra-mammalian and intra-molluscan)
S. mansoni lifecycle stages.
stage at the higher concentrations tested (10 and 50
miracidia to sporocyst transformation was significantly reduced by
0%, compared to DMSO. Miracidial transformation inhibition
mM); at 5 mM,
5
5

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
2
.5. Evaluation of compound 22 induced genotoxicity and
p53 content to discriminate levels resulting from increased DNA
damage compared to levels normally observed in replicating cells
cytotoxicity
[
36]. The fold change of both p53 and ɣH2AX biomarkers in TK6
Assessment of compound genotoxicity/cytotoxicity is a key
cells when exposed to different concentrations of compound 22 did
not exceed the criteria for a positive result (Fig. 5B and C; cut off
values of 1.5 for ɣH2AX and p53 [30,37e39]). In contrast, the pos-
itive (Methyl Methanesulphonate - MMS - and Carbendazim - Crbz
- as representative clastogenic and aneugenic agents, respectively
[36,40]) and negative (solvent or vehicle) controls behaved as ex-
pected according to Takeiri et al. [40]. From these data, we
concluded that compound 22 induced little or no DNA damage and,
at these concentrations, does not induce genotoxicity.
stage in drug development and can be performed at early or late
stages of the pipeline [29e31]. Whereas compound-associated
cytotoxicity can be mitigated by medicinal chemistry and does
not necessarily stop progression of a series, compound-associated
genotoxicity represents a red flag and could halt progression
[
29,32,33]. Therefore, in vitro assays were conducted to assess
whether compound 22 (or its metabolites) had the potential to
cause DNA damage in a surrogate human cell line (TK6 cells; Fig. 5).
Alteration to the chromosomal segregation machinery (chromo-
somal loss and chromosomal breaks; aneugenicity and clastoge-
nicity, respectively) was first evaluated via micronuclei (MNi)
formation in cells that have undergone mitosis (Fig. 5). Based on
preliminary 24 h cytotoxicity data generated for compound 22
Follow-on cell-cycle analysis (Fig. 5D) revealed a G2/M phase
arrest induced by compound 22 at 2 and 4 mM (mirroring the RCG
results in Fig. 5A). These findings, along with low levels of MNi,
ɣH2AX and p53 indicated a cell cycle effect most likely due to
compound 22-induced cytotoxicity (e.g. apoptosis or necrosis) at
these concentrations. During the genotoxicity studies, we observed
that compound 22 displayed autofluorescence properties and was
almost exclusively localised to the cytoplasm of TK6 cells (Fig. S5B).
These results were confirmed in three additional human cell lines
(HeLa, human cervical epithelial; MDA-MB-231, human breast
epithelial; HepG2, human liver epithelial) (Fig. S6). Compound 22,
thus, accumulates in the cytoplasm and displays some cytotoxic,
but no genotoxic, properties; medicinal chemical modifications of
this lead was next pursued to obtain more potent and less cytotoxic
quinoxaline analogues.
(
CC50 ¼ 9.36
and 4 M) were assessed on TK6 cells in addition to negative
DMSO, used as solvent for compound preparation) and positive
Carbendazim, Crbz) controls (Fig. 5A). The number of MNi did not
mM; Fig. S5A), three sub-CC50 concentrations (0.5, 2
m
(
(
change significantly (compared to DMSO treated cells) across the
three assayed concentrations of compound 22. However, there was
a slight decrease at the higher concentration tested (4 mM). This
evidence is in line with the increased mitotic arrest at this higher
dose as shown by the Cell Count Relative Cell Growth (RCG) data.
Two additional biomarkers of genotoxicity were also used in this
investigation: ɣH2AX (phosphorylated histone variant H2AX at
serine 139) to detect DNA double-strand breaks [34,35] and nuclear
Fig. 5. Compound 22 exhibits some cytotoxic, but no genotoxic activities in TK6 cells. (A) e Quantification of %MNi induction in TK6 cells treated with three doses of compound 22
0.5, 2 and 4 M) compared to the negative (1% DMSO; vehicle) and positive (carbendazim; Crbz, 8.4 M) controls, after 1.5e2 cell-cycles (24 h). MNi were scored from the
(
m
m
mononucleated images collected on the ImageStream Mark II using the DeepFlow neural network algorithm as described in Wills et al. [77]. Graph shows the mean value of %
MNi ± SD (n ¼ 3, except for Crbz and the highest concentration tested of compound 22; n ¼ 2). Relative Cell Growth (RCG, as %) is also displayed. (B) - ɣH2AX observed in TK6 cells
co-cultivated with either 0.5, 2 or 4 mM compound 22 compared to the positive (45.4 mM MMS and 8.4 mM Crbz) and negative (1% DMSO; vehicle) controls. The dotted line
represents the fold change cut off value of 1.5. (C) - p53 measured in the same six TK6 cell/compound co-cultures described in (B). The dotted line represents the fold change cut off
value (1.5). For (B) and (C), each experimental point is shown (n ¼ 3, except for Crbz and the highest concentration tested of compound 22; n ¼ 2). Error bars represent SD. For (A),
(
B) and (C) any significant difference compared to the DMSO control was quantified by a one tailed Dunnett's test (*, ** and **** representing p < 0.0332, p < 0.0021 and p < 0.0001,
respectively). (D) Cell-cycle analyses conducted with compound 22. The percentage of TK6 cells in each cell-cycle phase (G1, S and G2/M) is shown for compound 22 treated cells as
well as the positive and negative controls (n ¼ 3). For each phase, any significant difference compared to the DMSO control was indicated as *, ***, **** representing p < 0.0332,
p < 0.0002, p < 0.0001, respectively (by two tailed Dunnett's test).
6

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
2
.6. Medicinal chemistry optimisation of compound 22
contained within compound 22a was performed with the appro-
priate acyl derivative and an excess of anhydrous pyridine in
Dichloromethane (DCM) generating the derivatives 22c-22g in
45e95% yield (remaining Series C analogues; Scheme 1).
Based on the biological data described thus far and the identi-
fication of compound 22 as a potent lead, we next explored the
impact of chemical structure modifications on anti-schistosomal
activity aiming to obtain more potent and less cytotoxic quinoxa-
line analogues. The lead optimisation approach aimed to first
explore the SAR around the aryl moiety, second to investigate the
influence of the N-linker and third to functionalise the C6 position
of the central scaffold. Towards this goal, three compound series
2.7. Biological activities of the compound 22 analogues
All newly synthesised compounds (22, 22a-g, 25e38 - 22
compounds in total) were initially screened against 72 h-cultured
schistosomula at 10 and 50 M concentrations (Fig. 6). In these
m
(
Series A, B and C, clustered based on their chemical similarity)
screens, the starting synthesis material (2,3-dichloro-6-
nitroquinoxaline) and its derivative (6-nitro-2,3-dihydroxy-qui-
noxaline) were also included to fully explore the SAR around the
quinoxaline scaffold. These 24 compounds were screened alongside
controls (AUR and DMSO) and 21 of these (87.5% hit rate) demon-
were identified leading to the creation of 21 compounds (re-syn-
thesis of compound 22 and 20 additional analogues, Scheme 1). In
brief, 2,3-bis(phenylamino)-quinoxalines were obtained from the
reaction of 2,3-dichloro-6-nitroquinoxaline with different
substituted anilines (resynthesis of compound 22 and synthesis of
compound 23e35, Series A) or phenyl-alkyl amines (compound 36
and 37, Series B) as previously described [41,42]. The yields of the
reactions reported in Scheme 1 show that the derivatives were
obtained in varying amounts reflecting the different reactivity of
the anilines and phenyl-alkyl amines used in the reaction.
strated activity at 50
mM (Fig. 6A). At the tested concentration
(
50 M), all compounds affected both phenotype and motility of the
m
schistosomula except for compound 22b; this compound affected
schistosomula motility, but not phenotype. Motility scores for three
compounds (22a, 25 and 38) were comparable to the positive
controls (AUR treated schistosomula), although phenotype scores
2 3
Furthermore, an analogue of compound 29 (N ,N -bis(4-
were not as severe. At the lower concentration (10
75% hit rate) of these compounds remained active (Fig. 6B).
To further assess the potency of the 18 hits found to be active at
M, a dose-response titration (0.313e10 M) was conducted on
mM), only 18
bromophenyl)quinoxaline-2,3-diamine, here labelled as com-
pound 38) was synthesised by a similar synthetic route (starting
from 2,3-dichloro-quinoxaline, Scheme 1) as this chemical
(
10
m
m
(
MMV007224) was previously identified during anti-schistosomal
schistosomula (Table 2 and Table S4). This extensive investigation
identified three main bioactive groups: a smaller group (including
screening of the Medicines for Malaria Venture (MMV) malaria
box collection [28]. Therefore, compound 38 was suitable for
comparing the analogues created in this study. In preparation for
the synthesis of Series C compounds, the classical catalytic hydro-
genation of the 6-nitro-substituted quinoxaline 22 into the amino
derivative (compound 22a) was performed [43,44]. A microwave-
assisted reaction was subsequently applied to the 6-amino-
substituted quinoxaline and dibromo butane in acetonitrile for the
synthesis of compound 22b (a Series C compound). The further
synthesis of N-acyl compound 22 derivatives was carried out as
previously described [45]. Functionalisation of the amine group
2
2a, 22c, 36, 38) with an EC50 > 2
improved activity (with 1 M < EC50 < 2
chemicals with potency in the high nanomolar range
EC50 < 1 M). Collectively, the in vitro screens highlighted three
m
M; a second group having
m
m
M) and a final set of 8
a
(
m
compounds (30, 31, 33) with highly improved anti-schistosomula
activities when compared to lead compound 22. For comparison,
the positive controls AUR and PZQ demonstrated EC50 values
against schistosomula of 0.41 and 1.01
ment with previous studies [28,46e48]).
To determine if these analogues also displayed activity against
mM, respectively (in agree-
Scheme 1. Synthesis of the 2,3-bis(phenylamino) quinoxaline series. The scheme shows the re-synthesis of compound 22 and the synthesis of 20 related derivatives (grouped by
structural similarity in Series A, B and C - highlighted with blue, green and magenta boxes, respectively). Compound 38 (same as MMV007224, previously identified during the anti-
schistosomal screening of the Medicines for Malaria Venture (MMV) malaria box collection) was synthesised as reference for the compounds synthesised and tested in this study.
ꢀ
Reagents and conditions: (i) different substituted anilines (compounds 22e35) or phenyl-alkyl amines (36 and 37), anhydrous DMSO, 130 C, 30 min; (ii) H
2
, cat. Pd/C, AcOEt, rt, 2h;
ꢀ
ꢀ
(
iii) Br(CH
2
)
2
Br, K
2
3
CO , CH
3
CN, MWI (300 W), 150 C, 15 min; (iv) R
2
COCl, anhydrous Pyr, anhydrous DCM, 0 C / rt, 1 h. Compound yield was reported in brackets.
7

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Fig. 6. Anti-schistosomal effects of newly synthesised compounds. The effect of each compound at 50 (A) and 10
m
M (B) on 72 h cultured schistosomula was assessed in duplicate in
three independent screens. Positive (Auranofin, AUR) and negative (DMSO) controls were included in each screen. Here the effect on the phenotype and the motility of the parasite
(
phenotype e on the left - and motility e on the right - score) for each compound was shown as an average score from the different biological replicates. In total, four screens were
0
performed and the Z calculated for both phenotype and motility obtained for each screen was reported in Table S4. The hit boundaries for both motility (- 0.35) and phenotype (-
0
.15) are shown in yellow (dotted line). All compounds showing a score lower than both reference values were considered hits (highlighted in orange). Error bars were included to
represent the standard error for the average score of the different screens. In both (A) and (B), the entries ‘dichloro-quinoxaline’ and ‘dihydroxy-quinoxaline’ represent 2,3-Dichloro-
-nitroquinoxaline and its hydroxy derivative (6-nitro-2,3-dihydroxyquinoxaline), respectively.
6
Table 2
Anti-schistosomal and HepG2 cytotoxic activities of the newly synthesised quinoxaline derivatives.
Activity on schistosomula
Activity on adult worms
HepG2 cells
Selectivity Index (SI)
(EC50
,
mM - 72 h)
(EC50
,
m
M - 72 h)
(CC50
, mM)
Compound
P
M
Average
Motility
24 h
72 h
Somula
Adult worms
2
6
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
,3-dichloro-6-nitroquinoxaline
-nitro-2,3-dihydroxy-quinoxaline
2
>10
>50
0.47
3.51
>10
3.04
1.02
>50
1.91
>50
1.22
0.42
>10
2.14
0.78
0.34
0.34
0.89
0.15
0.37
0.63
3.19
2.20
3.67
0.37
1.14
>10
>50
0.42
3.25
>10
2.79
1.80
>50
2.00
>50
1.30
0.39
>10
1.40
0.65
0.43
0.28
0.54
0.15
0.43
0.69
2.64
1.47
1.21
0.45
0.88
ND
ND
ND
ND
27.51
ND
ND
0.26
ND
ND
4.59
2.43
ND
5.69
ND
0.58
0.13
ND
ND
ND
0.59
ND
ND
1.57
1.72
ND
2.91
ND
0.46
0.32
ND
ND
ND
3.11
ND
ND
6.52
7.65
ND
ND
ND
136.40
7.40
25.11
ND
8.60
64.75
ND
14.84
ND
6.76
13.05
ND
10.16
8.72
5.85
8.80
7.75
27.52
10.07
5.64
11.67
4.40
7.12
5.07
142.8
ꢂ3
0.44
3.38
ND
2.91
1.41
ND
1.95
ND
1.26
0.41
ND
1.77
0.71
0.38
0.31
0.72
0.15
0.40
0.66
2.91
1.84
2.44
0.41
1.01
84.7 ꢁ 10
2a
2b
2c
2d
2e
2f
2g
5
6
7
8
9
0
1
2
3
4
5
6
7
ND
ND
ꢂ3
ꢂ3
703.5 ꢁ 10
318.3 ꢁ 10
ND
ND
ND
11.99 ꢁ 10
20.31 ꢁ 10
ND
ꢂ3
ꢂ3
47.99
6.33
ND
ꢂ3
ꢂ3
ꢂ3
ꢂ3
ꢂ3
ꢂ3
381.6 ꢁ 10
ND
ND
ND
122 ꢁ 10
3.97
0.90
0.08
0.67
0.10
ND
0.43
5.08
4.52
4.64
2.94
101.05
5.55
2.36
0.26
0.94
0.66
ND
0.65
1.74
2.47
1.90
7.17
100.05
32.52
347.59
10.21
60.99
17.87
ND
9.73
ND
23.20
28.65
ND
2.59 ꢁ 10
7.7 ꢁ 10
11.01 ꢁ 10
5.60 ꢁ 10
6.47 ꢁ 10
ꢂ3
ꢂ3
44 ꢁ 10
ND
ꢂ3
ꢂ3
195 ꢁ 10
162 ꢁ 10
ND
106.90 ꢁ 10
8
AUR
PZQ
ꢂ3
945.31
Schistosomula EC50 (Phenotype - P - and Motility - M) values of the new compounds were calculated based on three replicate titrations (0.313e10
m
M). Similarly, EC50 values
were calculated for Auranofin (AUR) and praziquantel (PZQ), included here for comparison. Where no effect was seen on schistosomula at the tested concentrations (10 and 50
M), the EC50 was said to be higher than 50 M. The EC50 was said to be higher than 10 M if the compound was a defined hit at 50 M, but not at 10 M. An average
schistosomula EC50 (Phenotype and Motility) values is also provided. Adult worm EC50 values of the new compounds were calculated based on three replicate titrations
0.0095e5 M). Each titration was performed in duplicate and repeated. Worm movement was recorded with WormassayGP2. Praziquantel and auranofin (10 M in 0.625%
DMSO; positive controls) as well as DSMO (0.625%, negative control) were also included in the schistosomula and adult worm screens. Each compound was tested against the
human HepG2 cell line at the concentration range spanning 200 Me0.01 M (three technical replicates each). The average CC50 values of each compound were calculated
m
m
m
m
m
(
m
m
m
m
from three or two replicates after 24 or 72 h co-incubation, respectively. All data were analysed using GraphPad Prism. The Selectivity Index (SI) of the compounds on
schistosomula and adult worms was calculated based on 72 h CC50 values and schistosomula and adult worms EC50 values, respectively. ND: not determined (where titration
was not performed so EC50 value was not available and arithmetic average/selectivity index was not calculated). Refer to Table S7 for the 95% CI (Confidence Interval) values.
8

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
adults, a dose response titration on adult worm pairs was per-
formed with 15 of the most potent anti-schistosomula compounds
(at 100 nM) against the urinary blood fluke S. haematobium. Similar
to their effects on S. japonicum, compounds 31 and 33 demon-
strated the best activity on S. haematobium motility (Fig. 8B, V3,
Table S6). Of the two remaining compounds, 22 was mainly inactive
and 30 had relatively mild effects on worm movement and plate
attachment. (Fig. 8B, Table S6). For both S. japonicum and
S. haematobium, compounds 31 and 33 were the most active in
terms of severity of phenotypic response and decrease in worm
motility as well as being fast-acting compounds (onset time within
24 h). In addition, S. japonicum appeared to be more susceptible to
these two compounds than S. haematobium. When we cross-
compared the activity of these compounds on S. japonicum and
S. haematobium species, the trend of activity was broadly consistent
with the biological data collected on S. mansoni (Table 2). Thus,
compounds 31 and 33 were the most active against all three spe-
cies, whereas compound 35 was the least active. Interestingly, the
lead compound 22 was not as potent (at 100 nM) on
S. haematobium when compared to either S. japonicum or
S. mansoni. While we have yet to explain this discrepancy, differ-
ences in activities between the schistosome species might be
influenced by factors such as membrane composition, intra-worm
compound stability/compound metabolism or sequence varia-
tions within the active site of the putative (yet to be identified)
target.
(
Table 2). Here, all the tested compounds displayed anthelmintic
activity with 6 compounds (22c, 22d, 28, 29, 37 and 38) displaying
EC50 values in the high to mid-nanomolar range (less potent than
compound 22). The other compounds (25, 26, 30, 31, 32, 33, 34 and
3
5) showed a marked increase in activity when compared to the
lead compound 22, with compounds 30, 31, 32, 33 and 34 repre-
senting the most potent. Notably, while not the most potent com-
pound in the series, the adult worm activity registered for
compound 38 was comparable to previous reports [28] and similar
to praziquantel itself [28,46]. Reassuringly, the re-synthesised
compound 22 showed identical anti-schistosomal activities as the
original commercially sourced compound on both schistosomula
(
0.51
mM - 95% CI 0.43 to 0.59 - and 0.61 mM - 95% CI 0.53 to 0.72,
respectively) and adult worms (73.14 nM - 95% CI 66.74e79.54- and
6
8.50 nM - 95% CI 63.50e73.50, respectively; Fig. S7).
The anthelmintic activity of these compounds was also assessed
in terms of IVLE production (Fig. 7). While most compounds
inhibited oviposition at low concentrations (low nanomolar range),
some were not as effective (e.g. 22a, 22c, 22d, 28, 29 and 38) as
compound 22. Interestingly, compounds 33 and 34 were even more
potent than PZQ (and compound 22) in affecting this important
aspect of schistosome lifecycle maintenance and pathology devel-
opment (Fig. 7).
Based on the S. mansoni adult worm activities (Table 2), the
seven most active compounds (22, 25, 26, 30, 31, 33 and 35) were
then screened against adult S. japonicum male worms (Fig. 8). Here,
a single-point concentration (100 nM, based on EC50 values calcu-
lated on S. mansoni adult worms) screen was performed to assess
the effect of these compounds on worm motility at four different
time-points (5, 24, 48 and 72 h). On S. japonicum, compounds 31
and 33 decreased parasite motility as early as 24 h post treatment
and this remained the case for the entire 72 h of the assay (Fig. 8A,
V2). Based on phenotypic descriptors, these compounds were
essentially lethal after 72 h (Table S5). Among the remaining
compounds, the quinoxaline derivatives 22, 25, 26 and 30 were not
lethal over the 3-day incubation, but they all decreased motility
starting at 48 h. Conversely, compound 35 showed no appreciable
activity against S. japonicum at the selected concentration. A subset
of these compounds (22, 30, 31 and 33) was subsequently screened
After examining the newly synthesised compounds’ activities
against two lifecycle stages of S. mansoni found in the mammalian
host (schistosomula and adults) and worms of the two other
medically important species S. haematobium and S. japonicum, we
next investigated their potential in blocking miracidia to sporocyst
transition in S. mansoni (Fig. 9). Based on the dose-dependent effect
on miracidial transformation observed with compound 22 (Fig. 4),
16 compounds (the lead compound 22 and 15 analogues) were
subsequently screened to assess their activity on miracidial trans-
formation (Fig. 9). The first screen was performed at 5 mM as this
concentration was not lethal to miracidia cultured in the presence
of the lead compound 22 (Fig. 9A). At this concentration, 11 com-
pounds had a lethal effect on the parasite and, therefore, were more
active than the lead compound. A second screen at 2 mM was then
performed with these 11 compounds in comparison to the same
concentration of compound 22 (Fig. 9B). Here, 9 compounds killed
the parasite and completely inhibited transformation to sporocyst.
Fig. 7. Comparative anti-fecundity activity of compound 22 and its structural analogues. Each compound was screened on S. mansoni adult worms and the effect of each compound
on egg production (after 72 h) was recorded at each concentration tested (0.0024e10 M in 0.625% DMSO v/v). The egg count derived from adult worms co-cultured in DMSO
0.625% v/v) was included as a negative control. Data were shown as mean average across the replicates and presented as a heat map (white to blue gradient scale; white e lowest
egg count, dark blue e highest egg count).
m
(
9

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Fig. 8. Compound 22 analogues are active against S. japonicum and S. haematobium adults. (A) - Seven compounds (22, 25, 26, 30, 31, 33 and 35) were screened against S. japonicum.
Each compound was tested at 100 nM in triplicate (except for compounds 25 and 33, tested in duplicate). (B) - Four compounds (22, 30, 31 and 33) were screened against
S. haematobium. Each compound was tested at 100 nM in triplicate. The negative control (DMSO) was included in each experimental screen. Parasite worm movement was recorded
at 5, 24, 48 and 72 h by WormAssay. The biological results were represented as the percent reduction in compound mediated motility compared to DMSO controls and displayed as
a bar chart (average motility across the replicates þ SD). A one-way ANOVA followed by Dunnett's multiple comparisons test was performed to compare each treatment to DMSO. *,
**, ***, **** represent p < 0.0332, p < 0.0021, p < 0.0002, p < 0.0001, respectively.
However, only 4 compounds were lethal for miracidia at the lowest
concentration tested (0.5 M, Fig. 9C). In conclusion, these three
screens led to the identification of compounds with an increased
potency on miracidia transformation when compared to the lead
compound. Four compounds (including compound 33) also caused
parasite death at the lowest concentration tested.
As compound 22 demonstrated some cytotoxicity (Fig. 5 and
Fig. S5), the cytotoxic activity of these analogues was next explored
on human HepG2 cells by prioritising those which showed the
most potent anti-schistosomal activity (EC50 on schistosomula
toxic among this family of compounds having the lowest CC50
values (5.85, 5.64 and 4.40 M respectively). Nine compounds (22,
22c, 22f, 21, 28, 31, 32, 34 and 36) showed an intermediate CC50
between 6 and 20 M. In contrast, compounds 22a, 22d, 28 and 33
m
m
m
showed low toxicity. The cytotoxicity was also evaluated after 72 h
incubation and the results broadly recapitulated the trend seen
during the 24 h screens (Table 2). When compared to EC50 values
derived from schistosomula and adult schistosome screens at 72 h,
a selectivity index (SI) for some of the newly synthesised com-
pounds was subsequently calculated (Table 2). The analysis of these
values highlighted that the lead optimisation of compound 22 led
to the identification of compounds with a much-improved anti-
schistosomal activity/(surrogate) host safety ratio (compounds 22c,
22d, 25, 26, 29, 30, 31, 32, 33, 35, 37 and 38).
below 10
mM). Each compound was initially tested for 24 h in a dose
response titration (0.01e200
mM) with the average CC50 of each
compound reported in Table 2. As shown, most compounds
induced some degree of cytotoxicity against the HepG2 cell line
(
including compound 22, which confirms results presented in Fig. 5
During the preparation of this manuscript, an anti-schistosomal
evaluation on a small subset of compounds structurally related to
and Fig. S5A). Compounds 30, 35 and 37 were found to be the most
10

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Fig. 9. Structural analogues of compound 22 block miracidia to primary sporocyst transformation. (A) - All 16 compounds (at 5
m
M in 1% DMSO) were assessed for their ability to
M were
M screen. Each treatment was set up
block miracidial to sporocyst transformation and compared to compound 22 (at 5 M in 1% DMSO) and DMSO (negative control, 1% DMSO). (B) - All compounds lethal at 5
m
m
then screened at 2
m
M. (C) - Another screen at 0.5
m
M was performed to assess the activity of the most potent compounds derived from the 2
m
ꢀ
in triplicate and parasites were cultured in CBSS at a controlled temperature of 26 C (in the dark). A Kruskal-Wallis ANOVA followed by Dunn's multiple comparisons test was
performed to compare each population mean to DMSO mean. *, **, ***, **** represent p < 0.0332, p < 0.0021, p < 0.0002, p < 0.0001, respectively.
the quinoxaline derivatives presented in this study (particularly 22,
2c, 25, 26, 29 and 30) was published [49]. Starting from a previ-
ously described dianilinoquinoxaline (MMV007224) [28], this
recent study focused on the investigation of other structurally-
related analogues. While some biological discrepancies were
observed between our study and that conducted by Debbert et al.
activity since 2,3-dichloro-6-nitroquinoxaline or its dihydroxy de-
rivative (6-nitro-2,3-dihydroxy-quinoxaline) had poor to moderate
activity against schistosomula.
2
With regards to the linker between the quinoxaline central core
and the aromatic ring, we previously observed that the dia-
nilinoquinoxalines were more active than the 2,3-bis(arylthio)- or
diphenoxyquinoxalines (Fig. S3C). However, this feature had not
been identified previously [49] and this observation, derived from a
limited number of available compounds, will likely require further
investigation. Among the dianilinoquinoxalines, we also gathered
some preliminary observations around the length of N-linker. This
linker seemed to be quite a flexible feature since the N-methylene
derivative (36) and N-ethyl derivative (37) showed similar activity
to the parental compound 19 (Table 3). The analysis of HepG2
cytotoxicity added some additional information regarding the N-
linker (Series B, Table 3). In contrast to the anti-schistosomal ac-
tivities, the N-linker length seemed to affect the cytotoxicity of the
compounds. In fact, compound 19 displayed moderate toxicity
[
[
49], likely due to differentially-employed experimental processes
50], broadly overlapping anti - S. mansoni activities were found
with these six compounds. However, we herein have added a
substantial body of additional work that more broadly explored the
anti-schistosomal activities of both dianilinoquinoxalines and 6-
nitroquinoxalines (or 6-acyl derivatives); these supporting data
include activities against S. mansoni juveniles (compound 22 only)
and sporocysts (compounds 22, 22a, 22c, 22d, 22f, 25, 26, 28, 30, 31,
3
2
2, 33, 34, 35, 36 and 37) as well as adult S. japonicum (compounds
2, 25, 26, 30, 31, 33 and 35) and S. haematobium worms (com-
pounds 22, 30, 31 and 33). Using our collective anthelmintic and
cytotoxic information gathered in this study, and informed by
others [49], an expansion of the preliminary SAR (introduced in
Fig. S3) around the quinoxaline analogues is now possible.
(CC50 ¼ 12.59
(compound 36, CC50 ¼ 11.67
toxicity which increased instead with the introduction of the N-
M). To conclude, this re-
mM). The introduction of the N-methylene linker
mM) did not significantly affect the
ethyl linker (compound 37, CC50 ¼ 4.40
m
3
. Structure-activity relationship (SAR) studies on the anti-
gion of the molecules could be more critical for host toxicity than
for anthelmintic activities.
schistosomal quinoxaline derivatives
We next expanded our investigation to include a different set of
R-groups on the aromatic rings and to assess the effect of these
groups on both parasite and HepG2 cells (Series A, Table 4). Firstly,
we deduced that the di-chloro substitution was not essential for the
activity as one of the two synthesised mono-chloro-substituted
Our first objective assessed whether the quinoxaline scaffold
contributed to the in vitro anti-schistosomal activity and then
subsequently explored the effects of the linker between the central
quinoxaline core and the aromatic rings. As shown in Table 3, the
quinoxaline moiety was not solely responsible for the anti-parasitic
11

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Table 3
SAR study around the central quinoxaline scaffold.
Anti-schistosomal activity EC50a
(m
M)
Adult
Schistosomula
Cytotoxic activity
c
CC50
(mM)
Compound
R
1
R
2
Phenotype
Motility
Motility
24 h
72 h
2
6
2
,3-dichloro-6-nitroquinoxaline
-nitro-2,3-dihydroxy-quinoxaline
2
Cl
OH
Cl
OH
>10
>50
0.47
>10
>50
0.42
N.D.b
N.D.b
0.085
27.51
136.4
7.4
N.D.b
N.D.b
0.26
19
1.89
1.35
N.D.b
12.59
N.D.b
Series B
36
3.19
2.20
2.64
1.47
N.D.b
0.195
11.67
4.40
5.08
4.52
37
a
5
0% Effective concentration, or compound concentration required to inhibit parasite movement by 50%;
N.D.: not determined.
0% Inhibitory concentration, or compound concentration required to inhibit cell proliferation by 50%.
b
c
5
Table 4
SAR study of Series A derivatives.
Anti-schistosomal activity EC50a
Cytotoxic activity
c
CC50
(mM)
Schistosomula (
m
M)
Adult (nM)
Compound
R
1
Phenotype
Motility
Motility
24 h
72 h
Series A
22
3,4-diCl
m-Cl
p-Cl
2,3-diCH
p-F
p-Br
0.47
1.22
0.42
>10
2.14
1.06
0.34
0.34
0.89
0.15
0.37
0.63
0.42
1.30
0.39
>10
1.40
0.83
0.43
0.28
0.54
0.15
0.43
0.69
84.7
11.99
20.31
N.D.b
381.60
122
2.59
7.70
11.01
5.60
6.47
44
7.40
6.76
13.05
N.D.b
10.16
8.72
5.85
8.80
7.75
0.26
0.58
0.13
N.D.b
N.D.b
3.97
0.90
0.08
0.67
0.10
N.D.b
0.43
25
26
27
28
29
30
31
32
33
34
35
3
m-CF
3
p-CF
3
3-CF
3
, 4-F
2-F, 5-CF
3-F, 5-CF
3-CF , 4-OCH
3 3
3
3
27.52
10.07
5.64
a
5
0% Effective concentration, or compound concentration required to inhibit parasite movement by 50%;
N.D.: not determined;
0% Inhibitory concentration, or compound concentration required to inhibit cell proliferation by 50%.
b
c
5
compounds (26) showed similar anti-schistosomal potency as the
lead compound (22). Moreover, there is a consistent difference in
activity between the two structural isomers (25 and 26) with a
preference for the para substitution (26) in schistosomula; inter-
estingly, an opposite trend was observed with adult worms
activity (especially against the adult stage). The positive contribu-
tion of this trifluoromethyl substituent was verified by the
increased anthelmintic potency found for compounds 32, 33 and 34
compared to compound 28 (p-fluoro derivative) on both schisto-
somula and adult stages of the parasite; similar conclusions can be
drawn for compound 35 where the trifluoromethyl group was
combined with a methoxy group. Cytotoxic analysis of the N-aro-
matic derivatives was instead more complicated; no consistent
trend was observed except for noticing an increased toxicity asso-
ciated with improvement in anti-schistosomal activity (the only
exception being compound 33). Moreover, the differential effect of
meta/para substitution at the two cytotoxicity endpoints was
generally observed (higher toxicity of meta and para substitution at
24 and 72 h, respectively). Obviously, a longer co-incubation of the
compounds resulted in an increased compound-induced toxicity on
HepG2 cells (CC50 24 h vs 72 h).
(
preferred meta substitution). As described in the preliminary SAR
investigations (Fig. S3), the meta and para substitution was
preferred to the ortho position. The anti-schistosomal results ob-
tained with the compound 22 analogues further supported this
statement. For example, compound 20 (with only one methyl group
in the meta position, Fig. S3) displayed good anti-schistosomula
activity (EC50 ¼ 1.42
methyl group in the ortho position of the aromatic ring (27)
decreased anti-schistosomula activity (EC50 > 10 M, Table 4). The
mM) whereas the introduction of another
m
substitution of the aromatic ring with a trifluoromethyl group led
to compounds 30 and 31 showing increased anti-schistosomal
12

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Table 5
SAR study of Series C derivatives.
Anti-schistosomal activity EC50a
Cytotoxic activity
c
CC50
(mM)
Schistosomula (
mM)
Adult (nM)
Compound
R
1
Phenotype
Motility
Motility
84.7
24 h
72 h
0.26
Series C
22
NO
NH
2
0.47
3.51
>10
0.42
3.25
>10
7.40
b
b
22c
2
N.D.
25.11
N.D.
22b
N.D.b
N.D.b
N.D.b
22c
3.04
2.79
703.50
8.60
4.59
2
2
2d
2e
1.02
1.80
318.30
N.D.b
64.75
N.D.b
2.43
>50
>50
N.D.b
2
2f
1.91
2.00
N.D.b
N.D.b
14.84
N.D.b
5.69
2
2g
>50
>50
N.D.b
a
5
0% Effective concentration, or compound concentration required to inhibit parasite movement by 50%;
N.D.: not determined;
0% Inhibitory concentration, or compound concentration required to inhibit cell proliferation by 50%.
b
c
5
Subsequently, the effect of a small number of C6 position
4. Conclusions
functionalisation on the central scaffold was analysed (Table 5). In
fact, the preliminary SAR studies (Fig. S3B) identified another
general trend with in vitro activity decreasing when the nitro group
in the C6 position was removed. Similar observations were reported
in Debbert et al. [49], despite this effect not being consistent across
all sets of analogues presented in their study. Firstly, the positive
contribution of the nitro group is evident when comparing the two
bromo-derivatives (compounds 29 and 38, EC50 ¼ 0.71 and
This study presents an extensive investigation around the anti-
schistosomal activity of the quinoxaline scaffold. Firstly, phenotypic
screening of commercially available small molecules led to the
identification of the hit compound 1 (EC50 ¼ 4.59
mM on schisto-
somula). Further exploration around the chemical space of this
compound led to the identification of the lead compound 22 with a
more promising anti-schistosomal profile (EC50 ¼ 0.44
mM and
2
.44
mM, respectively - Table 2) with a 4-fold decrease of anti-
EC50 ¼ 84.7 nM for schistosomula and adult stages, respectively)
and calculated selectivity indices (SI) of 0.59 and 3.11 for schisto-
somula and adult stages, respectively. This lead compound under-
went extensive medicinal chemistry optimisation, which resulted
in the creation of 21 analogues. The in vitro profiling of these
schistosomal activity when the nitro-group was removed (com-
pound 38).
As presented here, the reduction of the nitro group (in lead
compound 22) to an amino derivative (22a) led to a decrease in
anti-schistosomal activity of approximately 10-fold. Low anti-
schistosomal activity was also identified for compound 22b
where the amino group was incorporated into a cyclic ring. Among
the acyl derivatives, the introduction of a long alkyl chain (22d) did
not substantially decrease parasite activity (against both schisto-
somula and adult worms). However, a short alkyl chain did
decrease activity (compound 22c). Compound 22e and 22g
demonstrated the greatest loss in potency and this was associated
with its cycloalkane ring. The incorporation of additional aromatic
rings in the alkyl chain (22f and 22g) influenced the anti-
schistosomal activity with only compound 22f (containing a het-
eroaromatic ring) maintaining moderate potency. As far as the
cytotoxicity profile, improvements in cytotoxicity with the nitro-
amine conversion mirrored other data on nitro-containing com-
pounds [51]. In fact, an improved safety profile was registered for all
the derivatives of Series C when compared to the lead compound
compounds highlighted
a substantial improvement in anti-
schistosomal potency, particularly on the adult stage (with EC50
in the low nanomolar range). Despite the observed cytotoxicity on
human cell lines, some compounds reported in this study had an
improved safety profile compared to the initial lead compound
(particularly compound 25, 30 and 32 with selectivity indices (SI)
on adult worms of 47.99, 347.59 and 60.99, respectively). More
interestingly and without literature precedent, we additionally
reported activity against S. mansoni juveniles (compound 22) and
miracidia/sporocyst lifecycle stages as well as S. japonicum and
S. haematobium adults for a subset of the quinoxaline derivatives
presented in this study. Based on these collective data, compounds
31 and 33 are extremely promising due to the demonstrated broad-
species activity against the three medically important schistosome
species.
Overall, these findings reveal the great potential of quinoxaline-
containing compounds as drug candidates for schistosomiasis with
the identification of more potent/selective compounds in this study
representing valid pre-clinical candidates. Towards this end, the
2
2. All these data should improve the SAR around the quinoxaline
scaffold (Fig. S8), which would direct further chemical optimisation
of this series in the future.
2,3-dianilinoquinoxaline derivative (MMV007224) had significant
in vivo efficacy in a murine model of schistosomiasis with worm
13

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
burden reductions between 53.4% (four per oral doses of 100 mg/
kg, delivered every 4 h) and 40.8% (single per oral dose of 400 mg/
kg) [28]. While the in vivo efficacy of MMV007224 was not as high
as expected based on in vitro potencies, this discrepancy could be
explained by protein binding effects. Poor pharmacokinetic prop-
erties and in vivo efficacies of related quinoxalines against
S. mansoni were also found in another recent study [49]. Therefore,
we cannot exclude some challenges in translating in vivo efficacies
of compounds 25, 30, 31, 32 and 33 despite demonstrating more
potent in vitro activities when compared to existing anti-
schistosomal quinoxaline analogues reported to date.
For this reason, we are currently working on improving the
pharmacokinetic properties of these compounds as well as
exploring different formulations and dosing regimens. A combi-
nation of these factors may facilitate better translation of their
excellent in vitro activity to an in vivo murine model of schistoso-
miasis. Alongside this, we are confident that mechanism of action
studies [52] would help identify the target of these compounds and
perhaps lead to the further development of more potent and se-
lective quinoxaline-containing, anti-schistosomal derivatives.
13C NMR) can be found in the Supplementary Information.
2 n
5.1.1. Standard procedure for preparation of N-(CH ) aromatic
quinoxaline analogues (Series A and B, compounds 22e38, n ¼ 0, 1,
2)
To
a
stirring solution of purified 2,3-Dichloro-6-
nitroquinoxaline (1 equivalent - 1 eq) in anhydrous DMSO under
a nitrogen atmosphere, the appropriate aniline or phenyl-alkyl
amine (5 eq) was added. The reaction mixture was left stirring at
ꢀ
130 C for 30 min. The reaction was monitored by TLC with n-
hexane/EtOAc (different proportions of the mobile phase constit-
uents were used depending on the specific analogue being syn-
thesised). After completion of the reaction, the mixture was diluted
with EtOAc and poured into ice water. The aqueous phase was
extracted with EtOAc (3 ꢁ 15 ml), the combined organic phase was
then washed with 6 N hydrochloric acid (4 ꢁ 5 ml) and with brine
(1 ꢁ 10 ml). The organic phase was then dried over anhydrous
2 4
Na SO , filtered and reduced to dryness to give a crude product. The
crude product was purified by flash chromatography and eluted
with different proportions of n-hexane/EtOAc depending on the
specific analogue being synthesised.
2
3
5
. Experimental section
N ,
N -bis(3,4-dichlorophenyl)-6-nitroquinoxaline-2,3-diamine
1
(
22): red powder, 36% yield. H NMR (500 MHz, DMSO‑d
6
)
d
9.52
5
.1. Chemistry
(s, 2H, 2 x NH), 8.34e8.18 (m, 3H, 3 x ArH), 8.11 (d, J ¼ 8.9 Hz, 1H,
13
ArH), 7.87 (m, 2H, 2 x ArH), 7.65 (dd, J ¼ 8.8, 6.9 Hz, 3H, 3 x ArH);
C
The starting material (2,3-Dichloro-6-nitroquinoxaline, product
NMR (125 MHz, DMSO) 139.65 (ArC), 130.95 (ArC), 130.57 (3 x
d
code 49489, purity 98%) and its hydroxy derivative (6-nitro-2,3-
dihydroxyquinoxaline, product code 657239, purity 96%) were ac-
quired from Fluorochem and Sigma, respectively. All solvents and
reagents were used as supplied from Sigma-Aldrich, Fluorochem or
other commercial sources without further purification or treat-
ment. All reactions were monitored by Thin Layer Chromatography
ArCH), 124.98 (ArC), 124.50 (ArC), 122.26 (ArCH), 121.66 (ArCH),
121.13 (ArCH), 120.82 (ArCH), 120.61 (ArCH), 119.87 (ArCH); UPLC-
þ
þ
MS: Rt (Retention Time): 2.53 min, MS (ESI) : 496.08 [MþH] .
The spectral data are in accordance with those reported in the
literature [41,42,49].
2
3
N , N -bis(3,4-dichlorophenyl)quinoxaline-2,3,6-triamine (22a): A
mixture of compound 22 (0.24 g, 0.48 mmol) and 10% Pd/C (0.25 g)
was stirred under hydrogen atmosphere at room temperature in
EtOAc (24 ml) for 8 h. After completion of the reaction, it was
filtered through Celite. The filtrate was dried over anhydrous
(
(
TLC) on commercially available Merck Kieselgel 60 F254 plates
105554, Merck). TLC was performed with n-hexane/EtOAc (Ethyl
Acetate) as the mobile phase; different proportions of the mobile
phase constituents were used depending on the specific analogue
being synthesised. Separated components were visualised using
ultraviolet light (245 and 366 nm). The compounds were purified
by flash column chromatography (using the eluents indicated) on
an Interchim PuriFlash 430 using high performance silica gel par-
2 4
Na SO , filtered and reduced to dryness to give a crude product. The
crude product was purified by automated flash column chroma-
tography eluting with n-hexane:EtOAc 100:0 v/v increasing to
50:50 v/v in 15 CV. The final product 22a was obtained at 50% yield
1
13
1
ticle size 50-
m
m cartridges. H and C were recorded on a Bruker
as a yellow powder. H NMR (500 MHz, DMSO‑d
6
):
d
9.06 (s, 1H,
Avance III HD spectrometer operating (500 and 125 MHz respec-
NH), 8.88 (s, 1H, NH), 8.36 (d, J ¼ 2.5 Hz, 1H, ArH), 8.14 (d, J ¼ 2.6 Hz,
1H, ArH), 7.78 (dd, J ¼ 8.9, 2.6 Hz, 1H, ArH), 7.70 (dd, J ¼ 8.9, 2.6 Hz,
1H, ArH), 7.61 (d, J ¼ 8.8 Hz, 1H, ArH), 7.56 (d, J ¼ 8.8 Hz, 1H, ArH),
7.34 (d, J ¼ 8.7 Hz, 1H, ArH), 6.83 (dd, J ¼ 8.8, 2.5 Hz, 1H, ArH), 6.72
ꢀ
tively, at 25 C) in DMSO‑d
6
solution. Spectra were calibrated to the
1
residual signal of the deuterated solvent (
d
¼ 2.50 and 39.52, for H
13
and C NMR, respectively). Chemical shifts were given in parts
d
13
per million (ppm) and rounded to two decimal places. The
following abbreviations were used in the NMR signals assignment:
s for singlet, br s for broad singlet, d for doublet, t for triplet, q for
quartet, m for multiple. Coupling constants (J) were measured in
Hertz (Hz) and rounded to one decimal place. The purity of syn-
thesised compounds was determined by ultra-performance liquid
chromatography-mass spectrometry (UPLC-UV-MS) analysis using
a Waters UPLC system with both Diode Array detection and Elec-
trospray (ESI)þ MS detection. The following conditions were
(d, J ¼ 2.4 Hz, 1H, ArH), 5.42 (s, 2H, NH
2
); C NMR (125 MHz,
DMSO)
d
147.98 (ArC), 141.44 (ArC), 140.98(ArC), 140.83 (ArC),
137.91(ArC), 137.21 (ArC), 130.76 (ArC), 130.37 (2 x ArCH), 128.38
(ArC), 126.31 (ArCH), 122.97 (ArC), 122.12 (ArC), 120.63 (ArCH),
119.83 (ArCH), 119.74 (ArCH), 118.98 (ArCH), 116.86 (ArCH), 105.85
þ
þ
(ArCH); UPLC-MS: Rt: 2.66 min, MS (ESI) : 466.04 [MþH] .
2
3
N , N -bis(3,4-dichlorophenyl)-6-(pyrrolidin-1-yl)quinoxaline-
2,3-diamine (22b): To a mixture of compound 22b (0.03 g,
0.06 mmol, 1 eq) in CH
3
CN (2 ml), dibromobutane (0.03 g,
applied: Waters Acquity UPLC BEH C18 1.7
m
m 2.1 ꢁ 50 mm column,
0.14 mmol, 2.2 eq) and anhydrous K
2
CO
3
(0.01 g, 0.07 mmol, 1.1 eq)
ꢀ
ꢀ
0
.5 ml/min, column temperature 40 C; mobile phase was LC-MS
was added. The reaction mixture was heated at 150 C for 15 min.
After reaction completion, the mixture was filtered with EtOAc
(20 ml). The resultant organic phase was washed with saturated
2
grade H O containing 0.1% formic acid (A) and LC-MS grade
MeCN (acetonitrile) containing 0.1% formic acid (B); sample
diluent: MeCN; sample concentration: 1 g/ml; injection volume:
l, gradient 90% eluent A (0.1 min), 90%e0% eluent A (1.5 min), 0%
m
aqueous NaHCO
then dried over anhydrous Na
3
(3 ꢁ 5 ml) and brine (1 ꢁ 10 ml). The organic layer
2
m
2
SO , filtered and reduced to dryness
4
eluent A (1.4 min), 90% eluent A (0.1 min) (method 1). All com-
pounds tested in the biological assays showed >95% purity. For each
compound series (Series A, B and C), the general synthetic method
is reported together with the characterisation of one representative
to give a crude product. The crude product was purified by auto-
mated flash column chromatography eluting with n-hexane:EtOAc
100:0 v/v increasing to 80:20 v/v in 10 CV. The final product 22b
1
was obtained at 28% yield as a brown powder. H NMR (500 MHz,
1
compound. All other compound characterisations (UPLC-MS, H,
DMSO‑d
6
):
d
9.09 (s, 1H, NH), 8.93 (s, 1H, NH), 8.23 (d, J ¼ 2.6 Hz, 1H,
14

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
ArH), 8.15 (d, J ¼ 2.5 Hz, 1H, ArH), 7.90 (dd, J ¼ 8.9, 2.5 Hz, 1H, ArH),
5.2.3. Preparation of S. mansoni schistosomula for compound
screening
7
.75 (dd, J ¼ 8.9, 2.5 Hz, 1H, ArH), 7.63 (d, J ¼ 8.8 Hz, 1H, ArH), 7.58
(
2
d, J ¼ 8.8 Hz, 1H, ArH), 7.47 (d, J ¼ 8.9 Hz, 1H, ArH), 6.90 (dd, J ¼ 9.0,
Cercariae derived from the NMRI (National Medical Research
Institute) Puerto Rican strain (PR-1) of S. mansoni were shed from
two Biomphalaria glabrata strains, the NMRI albino and pigmented
outbred strains, under intensified lighting conditions (1 h of incu-
.6 Hz,1H, ArH), 6.57 (d, J ¼ 2.6 Hz,1H, ArH), 3.35 (t, J ¼ 6.6 Hz, 4H, 2
13
2 2
x CH ), 2.04e1.96 (m, 4H, 2 x CH ); C NMR (125 MHz, DMSO)
d
145.75 (ArC), 141.34 (ArC), 141.20 (ArC), 140.63 (ArC), 138.11 (ArC),
36.89 (ArC), 131.13 (ArC), 131.04 (ArC), 130.72 (ArCH), 130.58
ꢀ
1
bation at 26 C) [53]. Cercariae were then mechanically trans-
(
1
ArCH), 128.61(ArC), 126.51 (ArCH), 124.09 (ArC), 122.91 (ArC),
21.69 (ArCH), 120.62 (ArCH), 120.43 (ArCH), 119.49 (ArCH), 114.87
formed into schistosomula as previously described [54].
High-throughput screening (HTS) of selected compounds was
conducted as previously described [18,55]. Briefly, newly trans-
formed schistosomula were distributed in a 384-well tissue culture
plate (PerkinElmer, cat 6007460, seeding density of 120 parasites/
well). Each compound (at a single concentration or two-fold titra-
tions) was transferred into individual wells alongside the positive
(
ArCH), 105.12 (ArCH), 48.97 (2 x CH
2
), 25.01 (2 x CH
2
); UPLC-MS:
þ
þ
Rt: 2.71 min, MS (ESI) : 518.05 [M ꢂ H] .
5
.1.2. Standard procedure for preparation of N-acyl derivatives of
compound 22 (Series C, compounds 22c-22g)
and negative controls (AUR at 10 mM final concentration in 0.625%
To a solution of compound 22a (1 eq) dissolved in anhydrous
DCM, anhydrous pyridine (3.6 eq) was added under nitrogen. The
appropriate acyl derivate (1.1 eq) was added dropwise to the above
solution cooled to 0 C in an ice-bath under nitrogen atmosphere.
The resulting mixture was stirred at room temperature for 1 h. After
completion, the reaction mixture was diluted with DCM and
quenched with saturated aqueous sodium bicarbonate (NaHCO
The organic phase was washed first with saturated aqueous
NaHCO
(3 ꢁ 5 ml) and brine (1 ꢁ 10 ml). The organic layer was
then dried over anhydrous Na SO , filtered and reduced to dryness
to give a crude product. The crude product was purified by flash
chromatography eluting with n-hexane/EtOAc in different pro-
portions depending on the specific analogue being synthesised.
DMSO and 0.625% DMSO, respectively). Schistosomula/compound
ꢀ
co-cultures were then incubated at 37 C for 72 h in a humidified
ꢀ
2
atmosphere containing 5% CO . Following the co-culture period,
automatic assessment of compound-induced schistosomula
motility and phenotypic changes were conducted with the high-
content imaging platform Roboworm using the image analysis
model previously described by Paveley et al. [56].
3
).
3
5.2.4. S. mansoni juvenile and adult worm culture and in vitro
2
4
compound screening
Mice (Mus musculus HsdOLa:TO - Tuck Ordinary; Envigo, UK)
were used for the mammalian-specific propagation of the
S. mansoni life cycle. Experimental animals were infected with 4000
or 180 S. mansoni cercariae/mouse and perfused three- or seven-
weeks post-infection to obtain S. mansoni juvenile or adult
worms, respectively [57]. Preparation steps following perfusion
have been described previously [55]. Briefly, the parasite material
was washed three times with pre-warmed adult worm media
(DMEM (Gibco, Paisley, UK) supplemented with 10% (v/v) Fetal Calf
N-(2,3-bis((3,4-dichlorophenyl)amino)quinoxalin-6-yl)acetamide
1
(
22c): pale-yellow powder, 52% yield.
DMSO‑d ):
NH), 8.31 (d, J ¼ 2.5 Hz, 1H, ArH), 8.24 (d, J ¼ 2.6 Hz, 1H, ArH), 8.00
H NMR (500 MHz,
6
d 10.11 (s, 1H, NHeC]O), 9.23 (s, 1H, NH), 9.15 (s, 1H,
(
d, J ¼ 2.1 Hz, 1H, ArH), 7.81 (ddd, J ¼ 9.3, 6.9, 2.5 Hz, 2H, 2 x ArH),
7
2
1
1
(
1
.63 (dd, J ¼ 18.7, 8.8 Hz, 2H, 2 x ArH), 7.58e7.50 (m, 2H, 2 x ArH),
13
.09 (s, 3H, CH
41.15 (ArC), 140.64 (ArC), 140.43 (ArC), 139.82 (ArC), 137.46 (ArC),
36.30 (ArC), 132.18 (ArC), 130.79 (2 x ArC), 130.47 (ArCH), 130.43
ArCH), 125.78 (ArCH), 123.52 (ArC), 123.20 (ArC), 121.04 (ArCH),
20.78 (ArCH), 120.05 (ArCH), 119.81 (ArCH), 118.78 (ArCH), 113.96
3
); C NMR (125 MHz, DMSO): d 168.38 (C]O),
Serum (FCS, Gibco, Paisley, UK), 1% (v/v)
L-glutamine (Gibco, Paisley,
UK) and an antibiotic mixture (150 Units/ml penicillin and 150
ml streptomycin; Gibco, UK)) to remove residual host contamina-
mg/
tion [8] and incubated in a humidified environment containing 5%
ꢀ
þ
CO
2
at 37 C for at least 1 h.
(
[
ArCH), 24.12 (CH
3
); UPLC-MS: Rt: 2.19 min, MS (ESI) : 508.02
þ
Juvenile worms were distributed in a 96-well tissue culture
plate (n ¼ 20e25 individuals/well) containing 200 l/well. Com-
pounds were added to each treatment well at a concentration range
of 10e0.156 M. Negative (1.25% DMSO) and positive (15 M PZQ
and 15 M AUR in 1.25% DMSO) controls were also included in each
MþH] . The spectral data are in accordance with those reported in
m
the literature [49].
m
m
5
5
.2. Biological evaluation
m
drug screen. Treated juvenile worms were incubated for 72 h in a
.2.1. Ethics statement
All procedures performed on mice (for provision of S. mansoni)
ꢀ
humidified environment containing 5% CO
2
at 37 C. Compound-
induced phenotype and motility effects were quantified using an
adapted version of the WHO-TDR scoring system (in summary:
adhered to the United Kingdom Home Office Animals (Scientific
Procedures) Act of 1986 (project license PPL 40/3700 and
P3B8C46FD) as well as the European Union Animals Directive 2010/
0
¼ dead, 1 ¼ movement of the suckers only and slight contraction
of the body, 2 ¼ movement at the anterior and posterior regions
only, 3 ¼ full body movement but sluggish and 4 ¼ normal
movement) [58]. Additionally, differential uptake of propidium
iodide (PI, P1304MP, Sigma-Aldrich, UK) was used to quantify
worm viability as previously described [11,59]. In brief, after 15 min
6
3/EU and were approved by Aberystwyth University's Animal
Welfare and Ethical Review Body (AWERB). The use of hamsters
and mice (for provision of S. haematobium and S. japonicum,
respectively) was approved by the Institutional Animal Care and
Use Committee of the University of California San Diego.
incubation with 2 mg/ml PI, the parasite cultures were imaged un-
der both bright-field and fluorescent settings (PI detection, 562 and
624 excitation and emission wavelengths, respectively) using an
ImageXpressXL high content imager (Molecular Devices, UK).
Enumeration of PI positive and negative juvenile worms was
manually performed across all parasites within the well and the
percentage of PI positive parasite was calculated per well.
For the evaluation of anti-schistosomal activity on the adult
stage of the parasite, 7-wk old worms were transferred into 48 well
tissue culture plates (1e3 worm pairs/well, in duplicate). The
worms were dosed with compounds at final concentrations
5
.2.2. Compound handling
Prior to the biological assays, each compound was dissolved in
dimethyl sulfoxide (DMSO, 276855, Sigma-Aldrich, UK) at a stock
concentration of 10 mM and a working concentration of 1.6 mM (or
lower, if required). Small aliquots of both stock and working solu-
ꢀ
tions were stored at ꢂ20 C prior to use. Similar preparations were
followed for auranofin (AUR, A6733, Sigma-Aldrich, UK) and pra-
ziquantel (PZQ, P4668, Sigma-Aldrich, UK) positive controls used in
the schistosomula and adult worm screens, respectively.
15

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
spanning 0.78e50
m
M (in 0.5% DMSO). DMSO (0.5%) and prazi-
maximum value of 4. Descriptors for severe phenotypes, specif-
ically, death, degeneracy or damage to the surface tegument, were
given the maximum value of 4. The phenotypic characterisation of
compound effects focused on male worms due to the greater range
of their motility characteristics that include flexing, bending,
stretching, sucker walking and adherence, compared to the more
sedentary female worms. In addition to descriptors, WormAssay
[61], as adapted for the schistosome, was used to measure average
worm motility per well [72,73].
quantel (10 M in 0.5% DMSO) were also included as negative and
m
positive control treatments. Treated adult worms were incubated
for 72 h in a humidified environment containing 5% CO
ꢀ
2
at 37 C.
Parasite motility after compound treatment was assessed by a
digital image processing-based system (WormassayGP2) [60],
modified after Wormassay [61,62]. Following compound-parasite
co-incubation, eggs were collected, fixed in formalin (10% v/v
formaldehyde) and enumerated aiming to collect further infor-
mation on any compound-induced effects on parasite fecundity and
IVLE production.
5.3. HepG2 cell culture and MTT assay
5
.2.5. S. mansoni miracidia-sporocyst transformation and in vitro
The Human Caucasian Hepatocyte Carcinoma (HepG2) cell line
(85011430, Sigma Aldrich) was used to perform cytotoxicity tests of
compounds used in this study. Initially the cells were handled
under sterile conditions and grown at 37 C in a humidified at-
2
mosphere containing 5% CO . Cytotoxicity was evaluated by way of
compound screening
Following hepatic portal vein perfusion of TO mice, experi-
mentally infected 7 weeks previously, the mouse livers were
collected to isolate parasite eggs. Briefly, the infected tissues were
homogenised in double saline solution (1.7% w/v NaCl) using a
ꢀ
a 3-(4, 5- dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide
(MTT) assay as previously described [8,11,74]. The cells were
counted in a dual chamber counting slide using the BIORAD TC-10
Automated cell counter, seeded in a 96- well plate (Fisher Scientific,
Waring blender. The homogenates were passed through a 0.45 mm
filter to retain egg material and the resulting filtered solution was
collected in a volumetric flask. Miracidia hatching was induced by
exposure of eggs to light in artificial river water (e.g. Lepple water)
4
Loughborough, UK) at a density of 2 ꢁ 10 cells/well and incubated
ꢀ
[
18,63,64]. The resulting miracidial suspension was collected,
in an atmosphere of 5% CO
2
at 37 C. Following a 24 h incubation,
washed with Chernin's balanced salt solution (CBSS) [64] and
enumerated prior to being used for in vitro miracidia to sporocyst
screens, as previously described [18]. Briefly, each well of a 24-well
cells were incubated with the compounds at seven different con-
centrations (1, 10, 20, 50, 75, 100 and 200
m
M) for a further 24 h in a
ꢀ
CO
2
incubator at 37 C. Alternatively, when a lower seeding density
3
culture plate was loaded with 500
volume of compound added to each well to give the final concen-
tration for the screen (usually a titration at 0.5, 2, 5, 10 and 50
ml of CBSS and the relevant
(5 ꢁ 10 cells/well) was used, cells were dosed at 24 h and incu-
ꢀ
bated in a CO
2
incubator at 37 C for 72 h (the incubation time used
mM
for both schistosomula and adult worm screens). In both cases,
control cell treatments included DMSO (1.25% v/v; negative con-
trol) and Triton X-100 (1% v/v, Sigma-Aldrich; positive control).
Each compound was tested in triplicate and each experiment was
performed either three (for 24 h timepoint) or two times (for 72 h
timepoint). The absorbance readings (at 570 nm) of the soluble
formazan crystals were recorded using PolarStar Omega plate
reader (BMG-Labtech). Dose response curves were generated by
non-linear regression analysis in GraphPad Prism 7.02 (GraphPad
Software Inc., San Diego, CA, USA), after log concentration trans-
formation and normalization of the absorbance readings of each
concentration point. The cytotoxic concentration inhibiting 50% cell
growth (CC50) was derived at 24 and 72 h. The latter was used to
calculate the selectivity index (SI) of the compounds presented
herein.
was performed, unless differently stated). Each treatment was set
up in duplicate and parasites cultured in CBSS with 1% DMSO were
ꢀ
set up as negative controls. After 48 h incubation at 26 C, dead,
fully transformed and partially transformed miracidia were
enumerated as previously described [18,64e66].
5.2.6. In vitro compound screening of S. haematobium and
S. japonicum adults
Adult S. haematobium worms (Egyptian isolate) were recovered
from male Golden Syrian hamsters infected with 600 cercariae 20
weeks earlier. Adult S. japonicum worms (Philippine isolate) were
recovered from female Swiss Webster mice infected with 100
cercariae five weeks earlier.
For each species, parasite recovery involved reverse perfusion of
the hepatic portal and mesenteric system [67,68] using DMEM
(
Fisher Scientific). Parasites were washed six times with Basch 169
5.4. Genotoxicity assessments
medium [69] supplemented with 100 U/ml penicillin and 100 g/
m
ml streptomycin before manually distributing approximately 5
males per well (on occasion 1 or 2 females might also be present) of
a 24-well plate (Corning Costar 3524). The volume in each well was
made up to approximately 1 ml using the same medium supple-
The in vitro micronucleus (MN) assay was employed for the
assessment of chromosome damage. Human, p53 competent,
lymphoblastoid TK6 cells (Cat. No. 95111735) were used in this
study and obtained from the European Collection of Authenticated
Cell Cultures (ECACC) [75]). RPMI 1640 (Gibco, Paisley, UK) culture
media supplemented with 1% penicillin/streptomycin (100 U/ml
mented with 4% Fetal Bovine Serum (FBS, Gibco, Paisley, UK) and
ꢀ
the worms were left to acclimate overnight at 37 C in a 5% CO
2
environment. The next morning, a final concentration of 100 nM
compound (in 1 L DMSO) was added to each well and the final
Penicillin and 100
serum (Gibco, Paisley, UK), 1%
m
g/ml Streptomycin), 10% heat inactivated horse
-glutamine (Gibco, Paisley, UK) and
m
L
volume made up to 2 ml with Basch 169 medium supplemented
with antibiotics and FBS. Each compound was tested in triplicate
hygromycine-B was used for TK6 cell culture [76]. Cells were
incubated at 37 C in a humidified atmosphere of 5% (v/v) CO2. A
ꢀ
(
except for some instances tested in duplicate). Incubations were
preliminary cell toxicity assessment of compound 22 was per-
formed to define the most suitable working concentration range
ꢀ
maintained at 37 C in a 5% CO
2
environment and phenotypic
changes recorded at 5, 24, 48 and 72 h.
(0.5, 2 and 4 mM, derived from dilutions in DMSO) based on Cell
Using a Zeiss Axiovert A1 inverted microscope, the male para-
site's phenotypic changes in shape, density and motility were
recorded using a constrained nomenclature of simple descriptors
Count Relative Cell Growth (RCG) and toxicity not exceeding 70%.
Once cells reached confluency, sub-cultures were established.
5
Approximately 2 ꢁ 10 TK6 cells/ml were placed in a series of
[
68,70,71]. To allow for the partially quantitative comparisons of
sterile vented tissue culture flasks (Fisher brand) and treated for
1.5e2 cell-cycle period (24 h) with no recovery with compound 22,
compound effects, each descriptor was typically given a value of 1
and these were summed to generate a ‘severity score’ with a
the negative control (1% DMSO) or the positive controls (45.4 mM
16

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Methyl methanesulphonate (MMS) and 8.4
in DMSO; supplied from Sigma-Aldrich - CAS numbers 66-27-3 and
m
M Carbendazim (Crbz)
10% FBS. Cells were washed three times in phenol red free DMEM
and imaged using a Leica SP5 laser scanning confocal microscope.
Fluorescence intensity was measured following 405 nm exci-
tation by sequentially imaging at 9 nm wavelength intervals be-
tween 420 nm and 780 nm using a 10 nm bandwidth. A final
fluorescence readout was produced by quantifying the images
produced in ImageJ.
ꢀ
10605-21-7, respectively). All incubation steps occurred at 37 C in
2
a humidified atmosphere containing 5% CO .
After the treatment period, cultures were centrifuged at 200ꢁg
for 5 min at room temperature, the supernatant was discarded, and
the pellet re-suspended in pre-warmed phosphate-buffered saline
(
2
(
PBS). Subsequently, the PBS was removed via centrifugation at
00ꢁg and the pellet was re-suspended in BD FACS™ lysis solution
CAS 349202) enabling cell fixation and membrane permeabiliza-
HeLa, MDA-MB-231 and HepG2 cells were incubated with 5 mM
of compound 22 in DMEM containing 10% serum for 60 min before
washing and imaging in phenol red free DMEM by laser scanning
confocal microscopy. Non-treated cells were also analysed as
controls.
tion. Post fixation, cells were centrifuged for 5 min at 200ꢁg, su-
pernatant discarded and subsequently stained for ɣH2AX (using
Alexa Fluor® 488 Mouse anti-H2AX (pS139), Clone N1-431, Mouse
BALB/c IgG1,
k, Cat. No. 560445, BD Biosciences) and p53 (using PE
5.6. Statistics
anti-p53 Antibody, Clone DO-7, Mouse IgG2b, Cat. No. 645805,
BioLegend), as described [36]. Following that, cells were counter
stained with DRAQ5™ (Cat. No. 564902, BD Biosciences) or left in
PBS (as negative fluorescence controls). Sample incubation time for
DRAQ5™ staining was a minimum of 20 min at room temperature
under agitation. After all incubation periods were completed,
samples were washed with PBS. Stained cells were then acquired
on the ImageStreamX Mark II® imaging platform (the experiment
was repeated three times).
Statistical analyses of parasitological assays were performed
using a nonparametric two-way ANOVA followed by Least Signifi-
cant Difference post-hoc correction (Kruskal-Wallis ANOVA fol-
lowed by Dunn's multiple comparisons test). During genotoxicity
studies, Bartlett and Shapiro-Wilk tests were used to confirm ho-
mogenous variance and normal distribution for p53 and ɣH2AX
(p ꢃ 0.05) datasets allowing for parametric data assessment using
the Dunnett's test. All Statistical analyses were conducted using
GraphPad Prism 7.02 (GraphPad Software Inc., San Diego, CA, USA).
For MNi assessment, of the 14,000e15,000 viable cells per
experiment, about 4000 cell images were assessed except the
positive control (Crbz) where 1600 cells were examined. Data
analysis was performed as described [77]. Briefly, individual images
of cell populations were exported as compensated image files (.cif)
and the image files were then converted to grayscale, three 8-bit
channel TIF files (nuclear fluorescence, brightfield and darkfield
channels). The three individual channels were renormalised (max-
min) and cropped to 64 ꢁ 64 pixel size. Automated scoring and
classification of the TIF files was performed using the DeepFlow
neural network. The deep neural network was previously trained
on a ground truth image set representing a wide range of cell
phenotypes that arose from TK6 cells treated with carbendazim
and MMS as described in Wills et al. [77]. The TIF file images from
this study were inputted into the trained DeepFlow convolutional
neural network. TIF files of mononucleated cells and mono-
nucleated cells with MN with a percentage confidence of 70% and
above were extracted and MN response (expressed as %MNi) was
calculated using the following formula: (mononucleated cells with
MN/mononucleated cells) x 100.
Author contributions
Conceived and designed the experiments: GP, AB, KFH. Per-
formed the experiments: GP, MB, SF (compound synthesis and
characterisation), GP (all S. mansoni screening and cell toxicity
study), NE, LJL, CL (S. japonicum and S. haematobium screening),
DSGH, REB (genotoxicity study), ES (confocal microscopy). Re-
sources: JFT, HW (parasite material). Manuscript preparation: GP,
KFH. Manuscript revision: GP, NE, LJL, CL, DSGH, JFT, HW, MB, SF,
ATJ, CRC, IC, AB, KFH.
Funding sources
KFH, GP and AB thank the Welsh Government, Life Sciences
Research Network Wales scheme and the Wellcome Trust (107475/
Z/15/Z) for financially supporting this project. The CDIPD team ac-
knowledges the Bill and Melinda Gates Foundation (OPP1171488)
for support. Hamsters and mice infected with S. haematobium and
S. japonicum, respectively, were provided by the National Institute
of Allergy and Infectious Diseases (NIAID) Schistosomiasis Resource
Centre of the Biomedical Research Institute (Rockville, MD, USA)
through the National Institutes of Health (NIH)-NIAID Contract
HHSN272201700014I for distribution through BEI Resources.
The full 14,000e15,000 cell population was assessed for the
biomarkers ɣH2AX and p53 alongside compound 22 auto-
fluorescence. Gating parameters for ɣH2AX and p53 were deter-
mined based on the cell populations of unstained PBS controls and
stained DMSO controls compared to positive control samples.
These same gates were then applied for the batch processing of 22-
treated cell populations.
Declaration of competing interest
Percentage of cells identified at each fluorescence channel were
recorded in three independent experiments and data were then
transformed as fold changes. A cut off for a negative vs a positive
ɣH2AX and p53 response over the control is defined as a 1.5-fold
change [30,37,39]. The MN response was considered positive
based on a statistically significant response compared to that of
DMSO control [78]. Where MN positive response was observed and
biomarker response exceed fold change cut offs an indication of
Mode of Action (MoA) may be inferred [36,37].
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
Acknowledgments
We thank past and current members of the Hoffmann group and
Miss Julie Hirst for contributions to S. mansoni lifecycle mainte-
nance. S.F. is supported by the S e^ r Cymru II programme, which is
5
.5. Laser scanning confocal microscopy on mammalian cells
part-funded by Swansea University and the European Regional
Development Fund through the Welsh Government. We acknowl-
edge the contribution of Professor Paul Rees in the development of
the Deep Flow neural network used for the interpretation of the
cytotoxicity studies.
To investigate the compound fluorescence emission, a pre-
liminary lambda scan was performed on HeLa cells treated for 6 h
with 5 M of compound 22 in DMEM (Fisher Scientific) containing
m
17

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
Appendix A. Supplementary data
[24] A. Almoghrabi, O. Mzaik, B. Attar, Schistosoma japonicum associated with
colorectal cancer, ACG case reports journal 8 (2021), e00572.
[
[
25] http://www.specs.net/.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2021.113823.
26] F. Angelucci, A.A. Sayed, D.L. Williams, G. Boumis, M. Brunori,
D. Dimastrogiovanni, A.E. Miele, F. Pauly, A. Bellelli, Inhibition of Schistosoma
mansoni thioredoxin-glutathione reductase by auranofin: structural and ki-
netic aspects, J. Biol. Chem. 284 (2009) 28977e28985.
References
[27] A.N. Kuntz, E. Davioud-Charvet, A.A. Sayed, L.L. Califf, J. Dessolin, E.S. Arn eꢀ r,
D.L. Williams, Thioredoxin glutathione reductase from Schistosoma mansoni:
an essential parasite enzyme and a key drug target, PLoS Med. 4 (2007), e206.
[28] K. Ingram-Sieber, N. Cowan, G. Panic, M. Vargas, N.R. Mansour, Q.D. Bickle,
T.N.C. Wells, T. Spangenberg, J. Keiser, Orally active antischistosomal early
leads identified from the open access malaria box, PLoS Neglected Trop. Dis. 8
(2014), e2610.
[29] P.A. White, A.S. Long, G.E. Johnson, Quantitative interpretation of genetic
toxicity dose-response data for risk assessment and regulatory decision-
making: current status and emerging priorities, Environ. Mol. Mutagen. 61
(2020) 66e83.
[
[
[
1] M.L. Nelwan, Schistosomiasis: life cycle, diagnosis, and control, Curr. Ther. Res.
1 (2019) 5e9.
2] D.G. Colley, A.L. Bustinduy, W.E. Secor, C.H. King, Human schistosomiasis,
Lancet 383 (2014) 2253e2264.
3] W.H. Organization, Ending the Neglect to Attain the Sustainable Development
Goals: a Global Strategy on Water, Sanitation and Hygiene to Combat
Neglected Tropical Diseases, 2021, pp. 2021e2030.
9
[4] D. Cioli, L. Pica-Mattoccia, A. Basso, A. Guidi, Schistosomiasis control: prazi-
quantel forever? Mol. Biochem. Parasitol. 195 (2014) 23e29.
[
5] N. Vale, M.J. Gouveia, G. Rinaldi, P. Brindley, F. G €a rtner, J.M.C. Correia da Costa,
[30] L. Khoury, D. Zalko, M. Audebert, Complementarity of phosphorylated his-
tones H2AX and H3 quantification in different cell lines for genotoxicity
screening, Arch. Toxicol. 90 (2016) 1983e1995.
Praziquantel for Schistosomiasis, Single Drug Revisited Metabolism, Mode of
Action and Resistance, 2017.
[
[
6] C.R. Caffrey, Chemotherapy of schistosomiasis: present and future, Curr. Opin.
Chem. Biol. 11 (2007) 433e439.
7] Y.-S. Liang, J.-R. Dai, A. Ning, D.-B. Yu, X.-J. Xu, Y.-C. Zhu, G.C. Coles, Suscep-
tibility of Schistosoma japonicum to praziquantel in China, Trop. Med. Int.
Health 6 (2001) 707e714.
[31] https://pubmed.ncbi.nlm.nih.gov/.
[32] D. Clayson, D. Grant, The assessment of mutagenicity: health Protection
Branch mutagenicity guidelines, Environ. Mol. Mutagen. 21 (1993) 15e37.
[33] R.H. Heflich, G.E. Johnson, A. Zeller, F. Marchetti, G.R. Douglas, K.L. Witt,
B.B. Gollapudi, P.A. White, Mutation as a toxicological endpoint for regulatory
decision-making, Environ. Mol. Mutagen. 61 (2020) 34e41.
[
8] K.C.L. Whatley, G. Padalino, H. Whiteland, K.K. Geyer, B.J. Hulme,
I.W. Chalmers, J. Forde-Thomas, S. Ferla, A. Brancale, K.F. Hoffmann, The
repositioning of epigenetic probes/inhibitors identifies new anti-schistosomal
lead compounds and chemotherapeutic targets, PLoS Neglected Trop. Dis. 13
[34] M. Audebert, A. Riu, C. Jacques, A. Hillenweck, E.L. Jamin, D. Zalko, J.P. Cravedi,
Use of the gH2AX assay for assessing the genotoxicity of polycyclic aromatic
hydrocarbons in human cell lines, Toxicol. Lett. 199 (2010) 182e192.
[35] S.M. Bryce, D.T. Bernacki, J.C. Bemis, S.D. Dertinger, Genotoxic mode of action
predictions from a multiplexed flow cytometric assay and a machine learning
approach, Environ. Mol. Mutagen. 57 (2016) 171e189.
[36] S.D. Dertinger, A.R. Kraynak, R.P. Wheeldon, D.T. Bernacki, S.M. Bryce, N. Hall,
J.C. Bemis, S.M. Galloway, P.A. Escobar, G.E. Johnson, Predictions of genotoxic
potential, mode of action, molecular targets, and potency via a tiered multi-
flow® assay data analysis strategy, Environ. Mol. Mutagen. 60 (2019)
513e533.
(
2019), e0007693.
[
9] F. Dubois, S. Caby, F. Oger, C. Cosseau, M. Capron, C. Grunau, C. Dissous,
R.J. Pierce, Histone deacetylase inhibitors induce apoptosis, histone hyper-
acetylation and up-regulation of gene transcription in Schistosoma mansoni,
Mol. Biochem. Parasitol. 168 (2009) 7e15.
[
10] J. Lobo-Silva, F.J. Cabral, M.S. Amaral, P.A. Miyasato, R.P. de Freitas,
A.S.A. Pereira, M.I. Khouri, M.M.F. Barbosa, P.I.P. Ramos, L.C.C. Leite, O.A. Asojo,
E. Nakano, S. Verjovski-Almeida, L.P. Farias, The antischistosomal potential of
GSK-J4, an H3K27 demethylase inhibitor: insights from molecular modeling,
transcriptomics and in vitro assays, Parasites Vectors 13 (2020) 140.
[37] S.M. Bryce, D.T. Bernacki, J.C. Bemis, R.A. Spellman, M.E. Engel, M. Schuler,
E. Lorge, P.T. Heikkinen, U. Hemmann, V. Thybaud, S. Wilde, N. Queisser,
A. Sutter, A. Zeller, M. Gu
eꢀ rard, D. Kirkland, S.D. Dertinger, Interlaboratory
evaluation of a multiplexed high information content in vitro genotoxicity
assay, Environ. Mol. Mutagen. 58 (2017) 146e161.
[
11] H. Whiteland, A. Crusco, L.W. Bloemberg, J. Tibble-Howlings, J. Forde-Thomas,
A. Coghlan, P.J. Murphy, K.F. Hoffmann, Quorum sensing N-Acyl homoserine
lactones are a new class of anti-schistosomal, PLoS Neglected Trop. Dis. 14
(
2020), e0008630.
[38] A. Wilson, P. Grabowski, J. Elloway, S. Ling, J. Stott, A. Doherty, Transforming
early pharmaceutical assessment of genotoxicity: applying statistical learning
to a high throughput, multi end point in vitro micronucleus assay, Sci. Rep. 11
(2021) 2535.
[
12] A. Crusco, H. Whiteland, R. Baptista, J.E. Forde-Thomas, M. Beckmann,
L.A.J. Mur, R.J. Nash, A.D. Westwell, K.F. Hoffmann, Antischistosomal proper-
ties of sclareol and its heck-coupled derivatives: design, synthesis, biological
evaluation, and untargeted metabolomics, ACS Infect. Dis.
188e1199.
5
(2019)
[39] D.J. Smart, K.P. Ahmedi, J.S. Harvey, A.M. Lynch, Genotoxicity screening via the
1
gH2AX by flow assay, Mutat. Res. Fund Mol. Mech. Mutagen 715 (2011)
25e31.
[40] A. Takeiri, K. Matsuzaki, S. Motoyama, M. Yano, A. Harada, C. Katoh, K. Tanaka,
[
13] C. Dissous, C.G. Grevelding, Piggy-backing the concept of cancer drugs for
schistosomiasis treatment: a tangible perspective? Trends Parasitol. 27 (2011)
5
9e66.
M. Mishima, High-content imaging analyses of gH2AX-foci and micronuclei in
[
14] K.M. Zorn, S. Sun, C.L. McConnon, K. Ma, E.K. Chen, D.H. Foil, T.R. Lane, L.J. Liu,
N. El-Sakkary, D.E. Skinner, S. Ekins, C.R. Caffrey, A machine learning strategy
for drug discovery identifies anti-schistosomal small molecules, ACS Infect.
Dis. 7 (2021) 406e420.
TK6 cells elucidated genotoxicity of chemicals and their clastogenic/aneugenic
mode of action, Gene Environ. : the official journal of the Japanese Environ-
mental Mutagen Society 41 (2019) 4.
[41] S.S. Kher, M. Penzo, S. Fulle, J.P. Ebejer, P.W. Finn, M.J. Blackman, A. Jirgensons,
Quinoxaline-based inhibitors of malarial protease PfSUB1, Chem. Heterocycl.
Compd. 50 (2015) 1457e1463.
[42] H. Jonathan, Preparation of amides as farmesoid X receptor modulators, in:
PCT Int. Appl., 2002.
[43] S.S. Kulkarni, S. Singh, J.R. Shah, W.-K. Low, T.T. Talele, Synthesis and SAR
optimization of quinazolin-4(3H)-ones as poly(ADP-ribose)polymerase-1 in-
hibitors, Eur. J. Med. Chem. 50 (2012) 264e273.
[44] I.D. Entwistle, A.E. Jackson, R.A.W. Johnstone, R.P. Telford, Reduction of nitro-
compounds, Journal of the Chemical Society, Perkin Transactions 1 (1977)
443e444.
[
[
[
15] L.G. Ferreira, G. Oliva, A.D. Andricopulo, Target-based molecular modeling
strategies for schistosomiasis drug discovery, Future Med. Chem. 7 (2015)
753e764.
16] A.C. Mafud, L.G. Ferreira, Y.P. Mascarenhas, A.D. Andricopulo, J. de Moraes,
Discovery of novel antischistosomal agents by molecular modeling ap-
proaches, Trends Parasitol. 32 (2016) 874e886.
17] S. Kannan, J. Melesina, A.-T. Hauser, A. Chakrabarti, T. Heimburg,
K. Schmidtkunz, A. Walter, M. Marek, R. Pierce, C. Romier, M. Jung, W. Sippl,
Discovery of Inhibitors of Schistosoma Mansoni HDAC8 by Combining Ho-
mology Modeling, Virtual Screening and In Vitro Validation, 2014.
[
[
[
[
18] G. Padalino, I.W. Chalmers, A. Brancale, K.F. Hoffmann, Identification of 6-
[45] M.-Z. Wang, H. Xu, S.-J. Yu, Q. Feng, S.-H. Wang, Z.-M. Li, Synthesis and
fungicidal activity of novel aminophenazine-1-carboxylate derivatives,
J. Agric. Food Chem. 58 (2010) 3651e3660.
[46] K. Ingram, I.A. Yaremenko, I.B. Krylov, L. Hofer, A.O. Terent’ev, J. Keiser,
Identification of antischistosomal leads by evaluating bridged 1,2,4,5-
tetraoxanes, alphaperoxides, and tricyclic monoperoxides, J. Med. Chem. 55
(2012) 8700e8711.
[47] F. Ronketti, A.V. Ramana, X. Chao-Ming, L. Pica-Mattoccia, D. Cioli, M.H. Todd,
Praziquantel derivatives I: modification of the aromatic ring, Bioorg. Med.
Chem. Lett 17 (2007) 4154e4157.
(
piperazin-1-yl)-1,3,5-triazine as a chemical scaffold with broad anti-
schistosomal activities, Wellcome open research 5 (2020) 169.
19] C.R. Caffrey, N. El-Sakkary, P. M €a der, R. Krieg, K. Becker, M. Schlitzer,
D.H. Drewry, J.L. Vennerstrom, C.G. Grevelding, Drug discovery and develop-
ment for schistosomiasis, in: Neglected Tropical Diseases, 2019, pp. 187e225.
20] A.H. Costain, A.S. MacDonald, H.H. Smits, Schistosome egg migration: mech-
anisms, pathogenesis and host immune responses, Front. Immunol. 9 (2018)
3042, 3042.
21] J.A. Ojo, S.A. Adedokun, A.A. Akindele, A.B. Olorunfemi, O.A. Otutu,
T.A. Ojurongbe, B.N. Thomas, T.P. Velavan, O. Ojurongbe, Prevalence of uro-
genital and intestinal schistosomiasis among school children in South-west
Nigeria, PLoS Neglected Trop. Dis. 15 (2021), e0009628.
[48] B.J. Neves, R.F. Dantas, M.R. Senger, W.C.G. Valente, J.d.M. Rezende-Neto,
W.T. Chaves, L. Kamentsky, A. Carpenter, F.P. Silva-Junior, C.H. Andrade, The
antidepressant drug paroxetine as a new lead candidate in schistosome drug
discovery, MedChemComm 7 (2016) 1176e1182.
[
[
22] L.L. Santos, J. Santos, M.J. Gouveia, C. Bernardo, C. Lopes, G. Rinaldi,
P.J. Brindley, J.M.C.d. Costa, Urogenital schistosomiasisdhistory, pathogenesis,
and bladder cancer, J. Clin. Med. 10 (2021) 205.
23] M.C. Botelho, H. Alves, J. Richter, Halting Schistosoma haematobium-
associated bladder cancer, International journal of cancer management 10
[49] S.L. Debbert, M.J. Hintz, C.J. Bell, K.R. Earl, G.E. Forsythe, C. H
a€ berli, J. Keiser,
Activities of quinoxaline, nitroquinoxaline, and [1,2,4]Triazolo[4,3-a]qui-
noxaline analogs of MMV007204 against schistosoma mansoni, Antimicrob.
Agents Chemother. 65 (2021) e01370, 01320.
(
2017).
[50] T.B. Duguet, A. Glebov, A. Hussain, S. Kulkarni, I. Mochalkin, T.G. Geary,
18

G. Padalino, N. El-Sakkary, L.J. Liu et al.
European Journal of Medicinal Chemistry 226 (2021) 113823
M. Rashid, T. Spangenberg, P. Ribeiro, Identification of annotated bioactive
molecules that impair motility of the blood fluke Schistosoma mansoni, Int. J.
Parasitol.: Drugs and Drug Resistance 13 (2020) 73e88.
[65] A.S. Taft, F.A. Norante, T.P. Yoshino, The identification of inhibitors of Schis-
tosoma mansoni miracidial transformation by incorporating a medium-
throughput small-molecule screen, Exp. Parasitol. 125 (2010) 84e94.
[66] A. Azzi, C. Cosseau, C. Grunau, Schistosoma mansoni: developmental arrest of
miracidia treated with histone deacetylase inhibitors, Exp. Parasitol. 121
(2009) 288e291.
[
[
51] M. Orlandi, D. Brenna, R. Harms, S. Jost, M. Benaglia, Recent developments in
the reduction of aromatic and aliphatic nitro compounds to amines, Org.
Process Res. Dev. 22 (2018) 430e445.
52] M.G. Thomas, M. De Rycker, R.J. Wall, D. Spinks, O. Epemolu, S. Manthri,
S. Norval, M. Osuna-Cabello, S. Patterson, J. Riley, F.R.C. Simeons,
L. Stojanovski, J. Thomas, S. Thompson, C. Naylor, J.M. Fiandor, P.G. Wyatt,
M. Marco, S. Wyllie, K.D. Read, T.J. Miles, I.H. Gilbert, Identification and opti-
mization of a series of 8-hydroxy naphthyridines with potent in vitro anti-
leishmanial activity: initial SAR and assessment of in vivo activity, J. Med.
Chem. 63 (2020) 9523e9539.
[67] R.H. Duvall, W.B. DeWitt, An improved perfusion technique for recovering
adult schistosomes from laboratory animals, Am. J. Trop. Med. Hyg. 16 (1967)
483e486.
[68] M.-H. Abdulla, D.S. Ruelas, B. Wolff, J. Snedecor, K.-C. Lim, F. Xu, A.R. Renslo,
J. Williams, J.H. McKerrow, C.R. Caffrey, Drug discovery for schistosomiasis: hit
and lead compounds identified in a library of known drugs by medium-
throughput phenotypic screening, PLoS Neglected Trop. Dis. 3 (2009) e478.
[69] P.F. Basch, Cultivation of Schistosoma mansoni in vitro. I. Establishment of
cultures from cercariae and development until pairing, J. Parasitol. (1981)
179e185.
[
53] K.K. Geyer, U.H. Niazi, D. Duval, C. Cosseau, C. Tomlinson, I.W. Chalmers,
M.T. Swain, D.J. Cutress, U. Bickham-Wright, S.E. Munshi, C. Grunau,
T.P. Yoshino, K.F. Hoffmann, The Biomphalaria glabrata DNA methylation
machinery displays spatial tissue expression, is differentially active in distinct
snail populations and is modulated by interactions with Schistosoma man-
soni, PLoS Neglected Trop. Dis. 11 (2017), e0005246.
[70] T. Long, L. Rojo-Arreola, D. Shi, N. El-Sakkary, K. Jarnagin, F. Rock, M. Meewan,
A.A. Rasc oꢀ n Jr., L. Lin, K.A. Cunningham, G.A. Lemieux, L. Podust, R. Abagyan,
K. Ashrafi, J.H. McKerrow, C.R. Caffrey, Phenotypic, chemical and functional
characterization of cyclic nucleotide phosphodiesterase 4 (PDE4) as a poten-
tial anthelmintic drug target, PLoS Neglected Trop. Dis. 11 (2017), e0005680.
[71] T. Long, R.J. Neitz, R. Beasley, C. Kalyanaraman, B.M. Suzuki, M.P. Jacobson,
C. Dissous, J.H. McKerrow, D.H. Drewry, W.J. Zuercher, R. Singh, C.R. Caffrey,
Structure-bioactivity relationship for benzimidazole thiophene inhibitors of
polo-like kinase 1 (PLK1), a potential drug target in schistosoma mansoni,
PLoS Neglected Trop. Dis. 10 (2016), e0004356.
[54] D.G. Colley, S.K. Wikel, Schistosoma mansoni: simplified method for the
production of schistosomules, Exp. Parasitol. 35 (1974) 44e51.
[
55] H.L. Whiteland, A. Chakroborty, J.E. Forde-Thomas, A. Crusco, A. Cookson,
J. Hollinshead, C.A. Fenn, B. Bartholomew, P.A. Holdsworth, M. Fisher, An Abies
procera-derived tetracyclic triterpene containing a steroid-like nucleus core
and a lactone side chain attenuates in vitro survival of both Fasciola hepatica
and Schistosoma mansoni, Int. J. Parasitol.: Drugs and Drug Resistance 8
(
2018) 465e474.
[72] J.C. Weeks, W.M. Roberts, C. Leasure, B.M. Suzuki, K.J. Robinson, H. Currey,
P. Wangchuk, R.M. Eichenberger, A.D. Saxton, T.D. Bird, B.C. Kraemer,
A. Loukas, J.M. Hawdon, C.R. Caffrey, N.F. Liachko, Sertraline, paroxetine, and
chlorpromazine are rapidly acting anthelmintic drugs capable of clinical
repurposing, Sci. Rep. 8 (2018) 975, 975.
[
56] R. Paveley, N. Mansour, I. Hallyburton, L. Bleicher, E. Alex, I. Mikic, A. Guidi, I. H
Gilbert, A. L Hopkins, Q. D Bickle, Whole organism high-content screening by
label-free, image-based bayesian classification for parasitic diseases, PLoS
Neglected Trop. Dis. 6 (2012), e1762.
[
[
[
[
57] S. Smithers, R. Terry, The infection of laboratory hosts with cercariae of
Schistosoma mansoni and the recovery of the adult worms, Parasitology 55
[73] A.R. Wolfe, R.J. Neitz, M. Burlingame, B.M. Suzuki, K.C. Lim, M. Scheideler,
D.L. Nelson, L.Z. Benet, C.R. Caffrey, TPT sulfonate, a single, oral dose schis-
tosomicidal prodrug: in vivo efficacy, disposition and metabolic profiling, Int.
J. Parasitol.: Drugs and Drug Resistance 8 (2018) 571e586.
[74] A. Crusco, C. Bordoni, A. Chakroborty, K.C.L. Whatley, H. Whiteland,
A.D. Westwell, K.F. Hoffmann, Design, synthesis and anthelmintic activity of
7-keto-sempervirol analogues, Eur. J. Med. Chem. 152 (2018) 87e100.
[75] R.F. Branda, J.P. O'Neill, E.M. Brooks, L.M. Trombley, J.A. Nicklas, The effect of
folate deficiency on the cytotoxic and mutagenic responses to ethyl meth-
anesulfonate in human lymphoblastoid cell lines that differ in p53 status,
Mutat. Res. 473 (2001) 51e71.
(
1965) 695e700.
58] B. Ramirez, Q. Bickle, F. Yousif, F. Fakorede, M.-A. Mouries, S. Nwaka, Schis-
tosomes: challenges in compound screening, Expet Opin. Drug Discov. 2
(
2007) S53eS61.
59] E. Peak, I.W. Chalmers, K.F. Hoffmann, Development and validation of a
quantitative, high-throughput, fluorescent-based bioassay to detect schisto-
soma viability, PLoS Neglected Trop. Dis. 4 (2010) e759.
60] J. Wang, C. Paz, G. Padalino, A. Coghlan, Z. Lu, I. Gradinaru, J.N.R. Collins,
M. Berriman, K.F. Hoffmann, J.J. Collins, Large-scale RNAi screening uncovers
therapeutic targets in the parasite Schistosoma mansoni, Science 369 (2020)
[76] J.R. Verma, D.S.G. Harte, U.-K. Shah, H. Summers, C.A. Thornton, S.H. Doak,
G.J.S. Jenkins, P. Rees, J.W. Wills, G.E. Johnson, Investigating FlowSight® im-
aging flow cytometry as a platform to assess chemically induced micronuclei
using human lymphoblastoid cells in vitro, Mutagenesis 33 (2018) 283e289.
[77] J.W. Wills, J.R. Verma, B.J. Rees, D.S.G. Harte, Q. Haxhiraj, C.M. Barnes,
R. Barnes, M.A. Rodrigues, M. Doan, A. Filby, R.E. Hewitt, C.A. Thornton,
J.G. Cronin, J.D. Kenny, R. Buckley, A.M. Lynch, A.E. Carpenter, H.D. Summers,
G. Johnson, P. Rees, Inter-laboratory Automation of the in Vitro Micronucleus
Assay Using Imaging Flow Cytometry and Deep Learning, bioRxiv, 2021,
2021.2005.2005.442619.
1649e1653.
[
61] C. Marcellino, J. Gut, K.C. Lim, R. Singh, J. McKerrow, J. Sakanari, WormAssay,
A novel computer application for whole-plate motion-based screening of
macroscopic parasites, PLoS Neglected Trop. Dis. 6 (2012), e1494.
[
62] B. Storey, C. Marcellino, M. Miller, M. Maclean, E. Mostafa, S. Howell,
J. Sakanari, A. Wolstenholme, R. Kaplan, Utilization of computer processed
high definition video imaging for measuring motility of microscopic nema-
tode stages on a quantitative scale: “The Worminator”, Int. J. Parasitol.: Drugs
and Drug Resistance 4 (2014) 233e243.
[
63] F. Lewis, J. Coligan, A. Kruisbeek, D. Margulies, E. Shevach, W. Strober, Animal
models for infectious diseases, Curr. Protoc. Im. 19 (1998) 19, 11.
64] D. Roquis, A. Taudt, K.K. Geyer, G. Padalino, K.F. Hoffmann, N. Holroyd,
M. Berriman, B. Aliaga, C. Chaparro, C. Grunau, R.d.C. Augusto, Histone
methylation changes are required for life cycle progression in the human
parasite Schistosoma mansoni, PLoS Pathog. 14 (2018), e1007066.
[78] G.E. Johnson, L.G. Soeteman-Hern aꢀ ndez, B.B. Gollapudi, O.G. Bodger,
K.L. Dearfield, R.H. Heflich, J.G. Hixon, D.P. Lovell, J.T. MacGregor,
L.H. Pottenger, C.M. Thompson, L. Abraham, V. Thybaud, J.Y. Tanir, E. Zeiger,
J. van Benthem, P.A. White, Derivation of point of departure (PoD) estimates in
genetic toxicology studies and their potential applications in risk assessment,
Environ. Mol. Mutagen. 55 (2014) 609e623.
[
19