Structure-Reactivity Relationships on Substrates and Inhibitors of the Lysine Deacylase Sirtuin 2 from Schistosoma mansoni (SmSirt2)

Article abstract of DOI:10.1021/acs.jmedchem.9b00638

The only drug currently available for treatment of the neglected disease Schistosomiasis is Praziquantel, and the possible emergence of resistance makes research on novel therapeutic agents necessary and urgent. To this end, the targeting of Schistosoma m

Full text of DOI:10.1021/acs.jmedchem.9b00638

Subscriber access provided by CARLETON UNIVERSITY
Article
Structure-Reactivity Relationships on Substrates and Inhibitors of
the Lysine Deacylase Sirtuin 2 from Schistosoma Mansoni (SmSirt2)
Daria Monaldi, Dante Rotili, Julien Lancelot, Martin Marek, Nathalie Wössner, Alessia Lucidi, Daniela
Tomaselli, Elizabeth Ramos-Morales, Christophe Romier, Raymond Pierce, Antonello Mai, and Manfred Jung
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00638 • Publication Date (Web): 08 Sep 2019
Downloaded from pubs.acs.org on September 8, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure-reactivity relationships on substrates and inhibitors of the lysine deacylase
Sirtuin 2 from Schistosoma mansoni (SmSirt2)
1
2*
3
4
1
Daria Monaldi , Dante Rotili , Julien Lancelot , Martin Marek , Nathalie Wössner , Alessia
2
2
4
4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Lucidi , Daniela Tomaselli , Elizabeth Ramos-Morales , Christophe Romier , Raymond J.
3
2*
1*
Pierce , Antonello Mai and Manfred Jung
1Institute of Pharmaceutical Sciences, University of Freiburg, Albertstr. 25, 79104, Freiburg,
Germany.
2Dipartimento di Chimica e Tecnologie del Farmaco “Sapienza” Università di Roma, 00185
Rome, Italy
3Université de Lille, CNRS, Inserm, CHU Lille, Institut Pasteur de Lille, U1019 - UMR 8204 -
CIIL - Centre d’Infection et d’Immunité de Lille, 59000 Lille, France
4Département de Biologie Structurale Intégrative, Institut de Génétique et Biologie Moléculaire
et Cellulaire (IGBMC), Université de Strasbourg (UDS), CNRS, INSERM, 67404 Illkirch Cedex,
France
*
corresponding authors, E-mail : dante.rotili@uniroma1.it, antonello.mai@uniroma1.it,
manfred.jung@pharmazie.uni-freiburg.de
ABSTRACT
The only drug currently available for treatment of the neglected disease Schistosomiasis is
Praziquantel, and the possible emergence of resistance makes research on novel therapeutic
agents necessary and urgent. To this end, the targeting of Schistosoma mansoni epigenetic
enzymes, which regulate the parasitic life cycle, emerged as promising approach. Due to the
strong effects of human Sirtuin inhibitors on parasite survival and reproduction, Schistosoma
sirtuins were postulated as potential therapeutic targets. In vitro testing of synthetic substrates of
S. mansoni Sirtuin 2 (SmSirt2) and kinetic experiments on a myristoylated peptide demonstrated
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
lysine long chain deacylation as an intrinsic SmSirt2 activity in addition to its known deacetylase
activity for the first time. Focused in vitro screening of the GSK Kinetobox library and structure-
activity relationships (SAR) of identified hits, led to the first SmSirt2 inhibitors with activity in
the low micromolar range. Several SmSirt2 inhibitors showed potency against both larval
schistosomes (viability) and adult worms (pairing, egg laying) in culture without general toxicity
to human cancer cells.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
2

Page 3 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Schistosomiasis is a neglected tropical disease, affecting millions of people in tropical and sub-
1
tropical countries and causing more than 300000 deaths per year. One of its major causative
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
agents is the blood fluke Schistosoma mansoni, characterized by a complex life cycle, where the
parasite is developed through four morphologically distinct forms, with two of them,
schistosomula and adult worms, present in the final human host. Praziquantel is so far the gold
standard for the treatment of schistosomiasis, showing several positive features like low cost,
2
3
high efficacy and low toxicity, that made possible its use for mass treatment campaigns. Long
term mass treatment with Praziquantel has led to a reduction in mortality and morbidity in
endemic areas,4,5 but also to a reduction in treatment efficacy and, in some cases, to the isolation
of resistant strains.6,7,8,9 This aspect, in association with the drug inactivity on larval stages and its
unknown mechanism of action,2,7 render relying on only one drug inadequate in the long term
and highlights the need for the development of novel therapeutic agents. Different approaches
1
0
11
have been already studied, for example the use of artemisinins and benzodiazepines, but none
of them is similar to Praziquantel in terms of efficacy, safety and cost.
1
2
The publication of the Schistosoma mansoni genome sequence made the research and validation
of novel therapeutic targets a particularly promising paradigm for the development of novel
antischistosomal drugs. A drug repurposing approach,13 based on targeting orthologues of
proteins already targeted in other pathologies,10 has demonstrated potential for developing
compounds able to selectively inhibit parasitic enzymes without affecting the correspondent
human isoforms. Moreover, since schistosomes can be considered similar to cancer cells in terms
1
4
of intensive metabolic activity and invisibility to host immune system, our interest is focused on
histone deacetylase enzymes (HDACs), which are able to remove acetyl groups from lysine
residues of histones and other proteins and are already known cancer drug targets with clinically
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
approved drugs. The essentiality of several histone deacetylases for growth and survival has
already been demonstrated in a variety of parasite genera, including Plasmodium, Leishmania,
Trypanosoma and Schistosoma.15 The complex life cycles of these parasites are subject to
complex epigenetic regulation and the selective inhibition of enzymes involved in these
processes, including HDACs, represents a valid therapeutic strategy. For example, in the cases of
Plasmodium falciparum and Trypanosoma brucei several molecules display selective in vitro and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
+
15
in vivo inhibitory activity against parasitic Zn -dependent HDACs (class I and II). This
approach has also led to the development of selective inhibitors of S. mansoni HDAC8, including
a
mercaptoacetamide analogue of suberoylanilide hydroxamic acid (SAHA),16
1
7
18
alkylhydroxamates and benzoylhydroxamates, characterized by submicro or nanomolar IC
5
0
values for SmHDAC8 with good selectivity over hHDAC1, hHDAC6 and, in some cases, also
over hHDAC8. Some of these compounds also had low micromolar EC50 values for killing
schistosome larvae and abolished pairing stability and egg laying in adult worms.16-18
+
The situation is different for NAD -dependent HDACs (class III, Sirtuins), where, despite their
evident potential as human anticancer and metabolic disease targets,19 only one compound
(Selisistat) has reached clinical trials so far as a potential treatment for Huntington´s disease. In
parasites, sirtuin isoforms have been identified in Plasmodium falciparum (PfSir2A and
PfSir2B),20,21 in Trypanosoma cruzi (TcSir2rp1 and TcSir2rp3)22,23 and in Leishmania (LmSir2rp1
and Lisir2rp1),24,25 but most in vitro tested inhibitors showed modest activity and/or lack of
selectivity.15 An exception is provided by bisnaphthalimidopropyl derivatives that showed
significant activity in vitro and in mice chronically infected with Leishmania infantum.26
In 2013 Lancelot et al. published the identification and characterisation of five Schistosoma
mansoni sirtuins (SmSirt1, SmSirt2, SmSirt5, SmSirt6 and SmSirt7) as orthologues of their
2
7
respective mammalian counterpart isoforms. Furthermore, it was demonstrated that Sirtinol and
ACS Paragon Plus Environment
4

Page 5 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Salermide, known inhibitors of human Sirtuin 1 and 2 (hSirt1 and hSirt2),28,29,30 in addition to
being inducers of selective apoptosis in cancer cell lines,29,31 and showing protective effects in a
muscular dystrophy model in nematodes,32 have pro-apoptotic effects in schistosome larvae
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(schistosomula), through DNA fragmentation, and markedly reduce pairing stability and egg
production in adult worms. These features support the potential of Schistosoma sirtuins as drug
targets for the development of novel and selective antischisosomal drugs.
Beyond their deacetylase activity, human sirtuins are also implicated in the removal of short,
medium and long fatty acyl groups from lysine residues of histones and non-histones
proteins.33,34,35,36,37,38 Lysine acylation has been identified as a posttranslational modification and
it is strongly connected to regulation of metabolism. In fact, metabolic intermediates are used for
this process and enzymes implicated in energy pathways are, in many cases, subject to these
modifications.36,39,40 Although deacylation of acyllysine is a common feature for all mammalian
sirtuins, each isoform is characterised by a different pattern of specificity and efficiency for
deacylation that may be quite distinct from the deacetylase activity.41 In parasites, studies
regarding lysine deacylation are lacking so far with the exception of the medium and long fatty
4
2
acyl chain removal by PfSir2A in Plasmodium falciparum. In a previous study, we established a
homogeneous in vitro assay for the determination of SmSirt2 deacetylase activity which uses the
readily available Z-(Ac)Lys-AMC (ZMAL, 1) and represents an optimal tool for cost efficient
high throughput campaigns.43 In order to further characterize the function of SmSirt2, we
performed and report here the analysis of SmSirt2 lysine deacylation activity by the use of both
lysine-derived small molecule and oligopeptidic substrates.
Moreover, with the aim of finding novel and selective drug-like inhibitors of SmSirt2, we present
an extensive structure-activity relationship study concerning novel hits identified by an in vitro
4
4
screening of the Kinetobox library, provided by GSK. This library is constituted by compounds
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
that were shown to be potent and specific inhibitors of growth of Leishmania donovani,
Trypanosoma cruzi and Trypanosoma brucei with low human cellular cytotoxicity.44 The
kinetoplastid parasites Leishmania sp. and Trypanosoma sp. are characterized by complex
morphological changes and an involvement of epigenetic regulation during their life cycle, and
we postulated that the screening of this diverse library would be a good starting point for our
study on S. mansoni aimed at identifying novel chemical entities able to interfere with parasite
growth. S. mansoni is phylogenetically very distant from kinetoplastids, but has similar dynamic
phenotypic changes through the different life stages, some of them implying epigenetic
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
5
modifications. This inhibitor collection is freely available and provides a set of 592 compounds
with diverse structural features, potentially providing novel chemical space for chemically yet
uncharted targets. Here we present new hits for SmSirt2 (as well as hSirt2) and present initial
SAR to further characterize ligand affinity and specificity to SmSirt2. In particular, we can show
for the first time that SmSirt2 is a druggable target with selectivity over hSirt2. Moreover, several
of the characterized SmSirt2 inhibitors were active against both schistosomula larvae and adult
worms in culture. Some of the negative controls showed activity on Schistosoma as well, but they
were also toxic against human cancer cells in culture. In contrast, selective SmSirt2 inhibitors did
not show toxicity in mammalian cells, further supporting the use of SmSirt2 as a valuable drug
target in schistosomes.
ACS Paragon Plus Environment
6

Page 7 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RESULTS
Short-, medium-, and long-chain deacylation activity of SmSirt2. To extend the biochemical
characterization of SmSirt2 activity, we studied its ability to deacylate long chain fatty acids from
the -amino group of lysine substrates. We synthesized seven analogues of the SmSirt2 substrate
ZMAL (Z-(Ac)Lys-AMC) 1, ie 3a (Z-(But)Lys-AMC), 3b (Z-(Hex)Lys-AMC), 3c (Z-(Oct)Lys-
AMC), 3d (Z-(Dec)Lys-AMC), 3e (Z-(Lau)Lys-AMC), 3f (Z-(Myr)Lys-AMC) and 3g (Z-
(Pal)Lys-AMC), by replacing its acetyl group with short, medium and long acyl chains. Substrate
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
was synthesized according to published procedures.46 For the preparation of 3a-g, the
appropriate acyl chlorides (commercially available for 3a-d,g, and synthesized for 3e and 3f by
reaction of the corresponding lauric or myristic acid with thionyl chloride) were treated with
Z-Lys-OH leading to the formation of the ɛ-acyl-Z-Lys-OH 2a-g, which were converted into the
desired substrates 3a-g by treatment with phosphorus oxychloride and 7-amino-4-
methylcoumarin (AMC) (Scheme 1).
Scheme 1. Synthesis of 3a-ga
O
O
O
H
H
b
R
N
a
R
N
N
O
O
OH
R
Cl
H
O
O
NH
O
O
NH
H N
O
O
2
O
O
3
2
R = (CH ) CH (a), (CH ) CH (b), (CH ) CH (c), (CH ) CH (d),
2 2
3
2 4
3
2 6
3
2 8
3
(
CH2) CH (e), (CH ) CH (f), (CH ) CH (g)
10 3 2 12 3 2 14 3
aReagents and conditions: a) Z-Lys-OH, 1 M sodium hydroxide, water, rt, 20-45 min; b) AMC
(structure shown), phosphoryl chloride, dry pyridine, -15 ⁰C, 40 min to 3h.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Then we used the acyl-lysine substrates 3a-g, in comparison to the acetylated 1, in the
4
3
homogeneous fluorescence based assay, in order to evaluate the ability of SmSirt2 to catalyse
their conversion into free lysine substrate. Compounds 1, 3a-g were tested at 10.5 µM with one
4
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
time point measurement, according to a published procedure for 1. As shown in Figure 1A
(Table S1 in Supporting Information), the measured substrate conversion increases with
increasing chain length from acetyl to hexanoyl analogues (see 1, 3a and 3b), then exhibit an
opposite trend from the octanoyl to the palmitoyl analogue (from 3c to 3g). When tested with
hSirt2 (Figure 1B, Table S2), 1 and 3a-g showed a somewhat discontinuous pattern of
conversion, with 3a (butyryl) and 3b (hexanoyl) as the best substrates. Unfortunately, solubility
issues did not allow the measurement of K values for any of the ZMAL analogues in order to
m
obtain a more quantitative overview, however we performed a quantitative analysis on peptidic
substrates (see below).
Figure 1. In vitro conversion of 1 (acetyl) and 3a-g (see Scheme 1) by SmSirt2 (A) and hSirt2
(B). A pure AMC sample is measured to simulate a (hypothetical) conversion of 100% for
comparison. Errors are represented as standard deviation of the mean (SD, n=3).
ACS Paragon Plus Environment
8

Page 9 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
Nicotinamide (NA) is a physiological sirtuin inhibitor which, once released from NAD , stays in
a subpocket of the enzyme (pocket C) and, by a rebound mechanism, can block the enzymatic
activity.47 In hSirt2 the lysine acyl chain pocket is in a close contact with the C pocket,
determining the possibility for NA potency to be influenced by the substrate acyl chain length.34
To study the effect of short-, medium- and long-chain lysine substrates on the relative NA
potency for SmSirt2 inhibition, we measured IC50 values for NA in presence of our ZMAL
analogues 1, 3a-g. As shown in Figure 2 (Table S3), the IC values of NA are similar using
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
substrates up to the decanoyl group (3c), after which they increase with the growth of the acyl
3
4
chain length. Since an opposite trend has been published for acylated peptides and hSirt2, we
can hypothesize that, despite the structural identiy27,48 between the schistosome and the human
Sirt2, these two enzymes are characterized by different kinetics. It could be possible that the
putative structural differences in the active site/C pocket between hSirt2 and smSirt2 can explain
the obtained results, but due to a lack of structural data, this is yet speculative.
Figure 2. IC values of NA against SmSirt2 in the presence of the acetylated substrate 1 and the
5
0
acylated ZMAL analogues 3a-g (see Scheme 1). Errors are represented as standard error of the
mean (SEM, n=3). IC values are reported in Table S3 (Supporting information).
5
0
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In order to gain a more quantitative insight into the interplay between SmSirt2 and its acylated
substrates, we performed a kinetic analysis through the use of a HPLC-based assay using more
soluble peptides, not tagged with a fluorescent label to rule out potential artifactual effects. Km,
kcat and k /K of SmSirt2 were measured in the presence of an acetylated (PSDKac) and a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
cat
m
4
9
myristoylated peptide (PSDKmyr) derived from α-tubulin which is a Sirt2 substrate. As shown
in Figure 3, the acyl chain elongation in PSDKmyr determined an increase of affinity for SmSirt2
without affecting the enzymatic turnover, with consequent threefold higher catalytic efficiency in
the presence of PSDKmyr than with PSDKac (Table 1).
+
50
Moreover, since in hSirt2 the acylated substrate binds the enzyme before NAD , the length of
+
the acyl group could influence the NAD binding affinity for SmSirt2 and have, more generally,
+
+
an influence on the NAD reaction. Consequently, the K values for NAD were also analyzed in
m
the presence of PSDKac and PSDKmyr, respectively. As shown in Figure 4, the extension of the
+
peptide chain length increases the binding affinity of NAD to SmSirt2 and the catalytic turnover.
In fact, a threefold higher kcat is the reason for a better catalytic efficiency in presence of
PSDKmyr than with PSDKac (Table 1). There was no inhibition of ZMAL conversion by
PSDKac in a concentration of 300 µM (data not shown).
ACS Paragon Plus Environment
1
0

Page 11 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 3. Michaelis-Menten plots for SmSirt2 dependent deacetylation (A) and demyristoylation
(B). Error bars indicate standard error of the mean (n ≥ 2).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
Figure 4. Michaelis-Menten plots for NAD with PSDKac (A) and with PSDKmyr (B) in
SmSirt2. Error bars indicate standard error of the mean (n ≥ 2).
+
Table 1. Kinetic parameters for PSDKac, PSDKmyr and NAD for SmSirt2.
a
kcat (min-1)a
-1 -1 a
K (μM)
m
k /K (s M )
cat m
(31.8 ± 2.4) 10-2
(33.8 ± 1.3) 10-2
(10.3 ± 0.6) 10-2
(1.3 ± 0.3) 102
(3.7 ± 0.9) 102
(5.1 ± 1.8) 101
PSDKac
40.9 ± 12.2
15.1 ± 2.5
33.9 ± 8.5
13.3 ± 2.3
37.5 ± 12.7
75.8 ± 8.1
9.3 ±1.9
PSDKmyr
+
NAD (PSDKac)
+
(32.3 ± 1.3) 10-2
(50.6 ± 4.9) 10-4
(71.3 ± 3.2) 10-4
(51.8 ± 1.6) 10-4
(4.1 ± 0.9) 102
(1.3 ± 0.4) 102
(9.4 ± 4.0) 101
(5.6 ± 0.9) 102
NAD (PSDKmyr)
ZMML
+
NAD (ZMAL)
+
NAD (ZMML)
aReported as ± SE. K for ZMAL (59 µM) taken from lit43
m
To further characterize the deacylation reactivity, we compared the potency of selected reference
5
1
52
hSirt2 deacetylation inhibitors (NA, SirReal1 and 2, AEM2 ) with their ability to inhibit the
demyristoylation reaction. As reported in Table 2, while in hSirt2 these compounds show higher
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibitory potency for deacetylation than for demyristoylation, they did not inhibit either reaction
catalysed by SmSirt2, with the only exception of NA.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Activity of hSirt2 standard inhibitors with 1 or 3f as substrates.
SmSirt2
IC [µM] or % inhibition
hSirt2
a
a
IC [µM] or % inhibition
5
0
50
Compound
at 20 µM
at 20 µM
1
3f
1
3f
NA
23.1 ± 1.8
24.2 ± 0.8 %
< 15 %
420 ± 11.6
< 15 %
< 15 %
< 15 %
49.8 ± 4.6b
8.7 ± 0.5 %
< 15 %
SirReal1
SirReal2
AEM2
3.7 ± 0.8c
0.4 ± 0.1c
< 15 %
< 15 %
2.5 ± 0.2b
42.8 ± 3.0%
a
b
53
c
51
IC are reported as ± SEM (n=3), for inhibition % ± SD (n = 3). Reference . Reference .
5
0
4
3
SmSirt2 inhibition. Using the in vitro assay described above we initially screened the FDA-
approved drugs library (provided by Enzo Life Sciences (ELS) AG, Switzerland) that had already
been tested on schistosomula and adult worms.54,55 Idebenone (4) emerged as an overlapping
compound that was known to block the growth of schistosomula, showed good SmSirt2 inhibition
5
6
potency and selectivity over hSirt2 (Figure 5). Although Idebenone probably has a pleiotropic
mode of action, this lent credence to our approach to identify druggable molecules from a diverse
set of compounds that also show antischistosomal activity. We next proceeded with the GSK
Kinetobox library, which was initially tested at the fixed assay concentration of 25 µM. Since we
were looking for compounds with an IC in the low micromolar range, we focused our attention
5
0
only on compounds with inhibitory potency higher than 50% at that concentration. Furthermore,
to test whether hits quench the AMC signal, inhibit the detection enzyme trypsin or are
ACS Paragon Plus Environment
1
2

Page 13 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
autofluorescent compounds, all promising candidates were subjected to counter-analysis in order
to exclude false positives and false negatives (data not shown). Compounds showing no assay
interference and a dose-dependent activity were then tested on hSirt2 to evaluate their selectivity.
Among the 592 tested compounds, we identified three hits characterized by potency in the low
micromolar range: TCMDC-143159 (5), TCMDC-143362 (6) and TCMDC-143295 (7) (Figure
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
). Due to reported toxicity of the anabasine ring,57 present in 5, we decided initially to
resynthesize 6 and 7 to confirm their activity, and to modify their structures in order to obtain
analogues with improved activity and selectivity profiles. Since the inhibition of hSirt2 is an
undesired feature in new antischistosomials, but might identify new scaffolds for optimization of
human sirtuin inhibitors we followed the lead of 6 as an unselective compound to some extent but
focused mostly on the selective compound 7.
A
O
N
N
O
PZQ
B
O
O
H3CO
H3CO
Cl
N
N
O
H
N
OH
4
5
SmSirt2 IC50 = 1.9 ± 0.3 M
hSirt2 IC50 = 15.4 ± 2.1 M
SmSirt2 IC50 = 18.2 ± 3.7 M
hSirt2 inh. less than 15% at 50 M
N
NH2
O
NH
O
N
N
N
O
H2N
N
N
N
N
F
O
6
7
SmSirt2 IC50 = 14.0 ± 2.0 M
hSirt2 IC50 = 3.6 ± 1.0 M
SmSirt2 IC50 = 23.7 ± 9.6 M
hSirt2 inh. 21.9 ± 2.5% at 25 M
2
4.1 ± 1.4% at 50 M
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 5. A) Structure of antischistosomal treatment standard Praziquantel (PZQ)
B)Structures and in vitro data of idebenone (4) and hits from the Kinetobox library (5-7). For IC50
values, errors are represented as ± SEM (n = 3), while for inhibition % as ± SD (n = 3).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Synthesis and initial SAR of 1,2,4-oxadiazolyl compound 6.
As previously cited, Feldmann et al. recently published the crystal structure of hSirt2 in complex
with myristoylated peptides, showing the placement of the acyl chain in a hydrophobic pocket
and the conservation of this localization during the entire catalytic reaction.34,51 Since this could
also be the case in SmSirt2, after establishing its synthesis, initial optimization of 6 was attempted
with the replacement of the acetyl group with a longer fatty acyl group, such as an octanoyl (11)
or a decanoyl (12) chain, in order to potentially address such a pocket. The synthesis of 6 and its
analogues 11 and 12 is outlined in Scheme 2. A nucleophilic substitution between the
commercially available 2-(piperidin-3-yl) acetic acid ethyl ester and 3-fluoro-2-methylbenzyl
bromide was performed in order to obtain the intermediate 8. The reaction between the requisite
acyl chlorides and 4-aminobenzonitrile yielded the 4-cyanoanilides 9b and 9c which, together
with the commercially available 4-cyanoacetanilide 9a, were treated with hydroxylamine
hydrochloride to give the benzamidoximes 10a-c. Subsequent condensation between 10a,b and 8
afforded the required compounds 6 and 11, and the reaction between 10c and the acyl chloride of
8
, obtained by hydrolysis of the ethyl ester and subsequent chlorination, gave the final compound
12.
When tested against SmSirt2 and hSirt2 at 25 µM, the hit 6 was confirmed to have inhibitory
activity from the newly synthesized sample whereas the octanoyl and decanoyl analogues 11 and
1
2 were practically inactive against both parasite and human enzymes (6: SmSirt2 IC = 14.5 ±
50
ACS Paragon Plus Environment
1
4

Page 15 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
0
.6 µM; hSirt2 IC = 8.0 ± 1.1 µM; 11: SmSirt2 inh. 30.5 ± 3.5%; hSirt2 inh. <15%. 12: SmSirt2
5
0
inh. <15%; Sirt2 inh. <15%).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthesis of compounds 6, 11 and 12.a
O
O
OEt
OEt
a
N
d (for 10a)
e (for 10b)
or f (for 10c)
O
N
NH
N
F
NH
O
N
R
8
NH2
CN
OH
F
O
O
O
N
b
c
R
Cl
R
N
H
R
N
6, 11, 12
H
R = CH (a), (CH ) CH (b),
3
2 6
3
9
10
(CH ) CH (c)
R = (CH ) CH ,
2 8
3
2
6
3
R = CH3 (
CH ) CH (c)
a), (CH ) CH (b), R = CH (a), (CH ) CH (b),
3 2 6 3
(
CH ) CH
2 6 3
2
8
3
(
(CH2)8CH3 (c)
2 8
3
a
Reagents and conditions: (a) 3-fluoro-2-methylbenzylbromide, triethylamine, toluene, 0 °C →
rt, 18 h; (b) 4-aminobenzonitrile, triethylamine, dry DCM, 0 °C → rt, 2 h, N ; (c) hydroxylamine
2
hydrochloride, sodium carbonate, water/ethanol, reflux, 6-8 h; (d) 8, 10a, potassium carbonate,
pyridine, reflux 8h, then rt 72 h; (e) 8, 10b, potassium carbonate, pyridine, microwave, 180 °C,
1
0 min, 300 Watt, then reflux 43 h and finally rt 12 h; (f) from 8: i) 1M lithium hydroxide,
o
ethanol, 20 h, rt; ii) thionyl chloride, DMF, dry DCM, -15 C → r.t., 3h, N ; iii) 10c,
2
o
triethylamine, DMF, dry DCM, rt, 22 h, then 150 C 3 h, N .
2
Synthesis and SAR of pyrimidopyrimidine 7.
To work on the structure of hit 7, we started with a fragment-based approach with the
development of analogues with a simplified structure, in order to identify the substructure(s)
responsible of the inhibitory activity. In particular, since the 2,4,7-triaminopyrimidopyrimidine
+
moiety of 7 (“part A”, Figure 6) could mimic the adenosine part of NAD , we synthesized a first
generation of analogues keeping this portion fixed and introducing several modifications at the
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Nα (Figure 6). More precisely, we purchased parent pyrimido[4,5-d]pyrimidine-2,4,7-triamine 13
and prepared derivatives where the Nα is a secondary (14, 18 and 19), tertiary (20-23), or inserted
in a cyclic amine (15-17). While 14-17 are characterized by small amine substitution at the C7
7
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
position, the N -benzylpyrimido[4,5-d]pyrimidine-2,4,7-triamines 18-23 represent the most
substituted compounds of the series, with changes at the Nα, Cβ and C4 benzyl positions (“part
B”, Figure 6).
NH2
N
NH2
N
NH2
N
NH2
N
N
N
N
N
N
N
N
N
N
HN
H2N
N
NH2
H2N
N
N
H2N
N
N
2
H N
N
N
O
H
13
14
15
16
46k
NH2
NH2
N
1
8
R = OMe; R1 = R2 = H
NC
2
N
N
N
N
R1
19 R = R2 = H; R1 = Me;
N
2
2
0
1
R = R1 = H; R2 = Me
R = H; R1 = R2 = Me
O
H2N
N
N
N
H2N
N
N
H N
N
N
N
R2
18-23
22 R = OMe, R1 = R2 = Me
R = Ph, R1 = R2 = Me
17
R
O
23
47m
fragments and simplified
analogues
"
part A"
NH2
N
NH2
N
NH2
N
N
R1
"part B"
N
N
R1
slight
simplification
"part C"
R
N
N
SAR studies
H2N
N
N

N


2
H N
N
N
O
R2
H2N
N
N
O
X
2
4-26
O
27-29,33,41
27 R = OMe; R1 = Et; X = O
24 R = H; R1 = R2 = Me
R = OMe; R1 = Me; R2 = H
R = OMe; R1 = H; R2 = Me
7
2
2
5
6
28 R = OMe; R1 = i-Pr; X = O
29 R = OMe; R1 = Ph; X = O
33 R = 3,5-(OMe)2; R1 = Ph; X = O
SAR studies
41
R = 3,5-(OMe)2; R1 = Ph; X = S
NH2
N
N
N
H2N
N
N
Y
X
30-32,34-40,42-44
3
3
0
1
X = O; Y = 3-OMe-Ph
X = O; Y = 3,4-(OMe)2-Ph
38 X = O; Y = 4-OMe-1-Naphth
39 X = O; Y = 4-OMe-Bn
40 X = S; Y = 4-OMe-Ph
32 X = O; Y = 3,5-(OMe)2-Ph
3
3
4
5
X = O; Y = 3,4,5-(OMe)3-Ph 42 X = S; Y = 3,5-(OMe)2-Ph
X = O; Y = 4-OCF3-Ph
43 X = NH; Y = 4-OMe-Ph
36 X = O; Y = 4-SMe-Ph
X = O; Y = 4-OBn-Ph
44 X = NHCO; Y = 4-OMe-Ph
37
Figure 6. Overview of analogues of 7: fragment-based approach and SAR studies.
As shown in Table 3 (see below), compounds 13-22 show complete loss of activity on SmSirt2
and remain inactive on the human enzyme. Among the simplified analogues, only 23, bearing a
large phenyl substituent at the right part of the molecule (4-biphenyl portion, “part C”, Figure 6),
displayed low SmSirt2 inhibition. As a consequence, we designed a second generation of
ACS Paragon Plus Environment
1
6

Page 17 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
analogues in which small focused modifications were applied, without leading to an excessive
alteration of the prototype structure, but maintaining a high similarity to it. Specifically, we
focused on the Nα, Cβ and Cγ positions of 7 (Figure 6) through the removal of the methyl or
methoxy groups (24-26). We increased the steric hindrance of the Cβ substituent (“part B”,
Figure 6), by replacing the methyl with an ethyl, isopropyl, or phenyl group (27-29, 33, 41), and
we explored the effect of replacement of the 4-methoxyphenoxy portion of 7 (“part C”, Figure 6)
with other substituted phenoxy, 1-naphthyloxy, benzyloxy, phenylthio, aniline and benzanilide
moieties (30-44).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chemistry. The pyrimido[4,5-d]pyrimidine-2,4,7-triamine 13 is commercially available. The final
compounds 14-44 were prepared by reaction between the commercial 4-amino-2-
bromopirimidine-5-carbonitrile and the appropriate amines 46 in anhydrous 2-methoxyethanol, in
the presence of triethylamine at 80 °C. The obtained pyrimidine intermediates 47 were converted
into the desired compounds 14-44 through condensation reactions performed at 150 °C with the
guanidine free base in dry 2-methoxyethanol (Scheme 3). The amines 46a-h used for the
synthesis of the pyrimidine intermediates 47a-h are commercially available and were purchased
from vendors. The amines 46i-f', useful synthons for the synthesis of the final compounds 22-44,
were prepared by reaction of the corresponding phenyl ketones or aldehydes 45a-w with 7 M
ammonia (for secondary amines) or 2 M methylamine (for tertiary amines) solution in anhydrous
methanol in the presence of titanium isopropoxide at room temperature, followed by the addition,
after the formation of the imine intermediates, of the reducing agent sodium borohydride at 0 °C
under nitrogen atmosphere (Scheme 4A). The phenyl ketones/aldehydes 45a-e are commercially
available. The ketones 45f-q,s-u were synthesized by reaction between the appropriate aromatic
alcohols or thiols and the 1-(4-fluorophenyl)ethan-1-one, -propan-1-one, -2-methylpropan-1-one,
and -(phenyl)methanone in the presence of anhydrous potassium carbonate in dry DMF at 175 °C
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
Scheme 4B). The 1-(4-((4-methoxybenzyl)oxy)phenyl)ethan-1-one 45r,58 the 1-(4-((4-
methoxyphenyl)amino)phenyl)ethan-1-one 45v,59 and the N-(4-acetylphenyl)-4-methoxy-
6
0
benzamide 45w were prepared according to the literature. For the structures of the intermediates
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
5-47 see Supporting Information.
Scheme 3. Synthesis of compounds 7, 14-44.a
N
NH2
N
H2N
NC
4
6a
N
N
b
N
N
N
a
H N
2
N
N
H
H N
2
N
N
H
47a
14
NH2
N
HN
NC
H N
N
N
N
4
6b
b
N
N
H N
2
N
N
2
a
47b
15
HN
NH2
NC
NC
H N
N
46c
N
b
N
N
a
H N
2
N
N
N
Br
H N
2
N
N
N
2
47c
16
HN
N
NH2
NC
N
46d
b
N
N
H N
2
N
N
a
H N
2
N
N
N
47d
N
R1
17
N
HN
R2
NH2
R
NC
46e-f'
b
N
R1
N
N
R1
a
H N
2
N
N
H N
2
N
N
N
R2
R2
R
R
47e-f'
7, 18-44
R = 4-OMe; 4-Ph; 3-OMe-, 4-OMe-, 4-OCF -, 4-SMe-, 4-OBn-, 3,4-(OMe) -, 3,5-(OMe) -, 3,4,5-(OMe) -PhO;
4
R = H, Me, Et, Ph; R = H, Me
3
2
2
3
-OMe-1-naphthyl-O, 4-OMe-BnO; 4-OMe-, 3,5-(OMe) -PhS; 4-OMe-PhNH; 4-OMe-PhCONH;
2
1
2
aReagents and conditions: (a) triethylamine, dry 2-methoxyethanol, 80 °C, 2.5-8 h; (b) guanidine
free base, dry 2-methoxyethanol, 150 °C, 1.5-4.5 h.
ACS Paragon Plus Environment
1
8

Page 19 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 4. Synthesis of the intermediate compounds 45 and 46.a
A
R2
O
HN
R1
R1
a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R
R
45a-w
46i-f'
R = 4-OMe; 4-Ph; 4-PhO; 3-OMe-, 4-OMe-, 4-OCF -, 4-SMe-, 4-OBn-, 3,4-(OMe) -, 3,5-(OMe) -, 3,4,5-(OMe) -
3
2
2
3
PhO; 4-OMe-1-naphthyl-O; 4-OMe-BnO; 4-OMe-, 3,5-(OMe) -PhS; 4-OMe-PhNH; 4-OMe-PhCONH;
2
R = H, Me, Et, Ph; R = H, Me
1
2
B
O
O
R1
R
b or c
R
R1
XH
+
X
F
45f-q,s-u
R = 3-OMe, 4-OMe, 4-OCF , 4-SMe, 4-OBn, 3,4-(OMe) , 3,5-(OMe) , 3,4,5-(OMe) ;
3
2
2
3
X = O, S; R = Me, Et, Ph
1
aReagents and conditions: (a) 2 M methylamine in methanol, titanium isopropoxide, dry THF, N2,
5-6 h, rt, then sodium borohydride, N , 2 h, rt; (b) 7 M ammonia in methanol, titanium
2
isopropoxide, N , 5 h, rt, then sodium borohydride, N , 2 h, rt; c) anhydrous potassium carbonate,
2
2
dry DMF, 175 °C, 5 h;
All the above compounds were tested in vitro against SmSirt243 and hSirt2, to study their
selectivity for the parasitic enzyme (Table 3).56 An amine and a pyrimidine synthetic
intermediates (46k and 47m, see Figure 6 and Table S4 for their structures) were also included in
the list, to confirm the importance of the intact pyrimido[4,5-d]pyrimidine-2,4,7-triamine scaffold
for the inhibiting activity. The percentages of inhibition at 25 µM or the IC values of the new
5
0
compounds tested against SmSirt2 as well as hSirt2 are reported in Table 3.
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. In vitro inhibition of SmSirt2 and hSirt2 by 7, 13-44, 46k, and 47m.
NH2
N
N
R1
H N
2
N
N
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R2
R
SmSirt2a
hSirt2a
%
inhibition at
compd
R
R1
R2
% inhibition at 25
2
5 µM or IC
5
0
µM or IC [µM]
5
0
[µM]
O
7
Me
Me
2
3.7 ± 9.6
21.9 ± 2.5%
<15%
O
NH2
N
N
1
1
3
4
<15%
<15%
H N
N
N
N
NH2
2
NH2
N
N
N
23.2 ± 0.8%
<15%
H N
N
N
H
2
NH2
N
N
N
CH3
15
<15%
<15%
H N
N
N
2
NH2
N
N
1
1
6
7
<15%
H2N
N
N
N
NH2
N
N
N
H2N
N
N
<15%
<15%
N
1
1
8
9
-OMe
H
H
H
<15%
<15%
<15%
<15%
<15%
<15%
<15%
<15%
Me
H
H
20
H
Me
Me
2
1
H
Me
ACS Paragon Plus Environment
2
0

Page 21 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
2
2
3
-OMe
-Ph
Me
Me
Me
Me
<15%
<15%
24.8 ± 2.2% <15%
27.6 ± 2.8% <15%
2
2
2
2
2
4
5
6
7
8
Me
Me
H
Me
H
O
O
O
O
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
O
O
O
O
37.5 ± 5.6% 21.0 ± 3.4%
Me
Me
Me
<15%
<15%
Et
12.8 ± 0.8
27.7 ± 3.8
36.7 ± 7.7
57.4 ± 5.2%
i-Pr
2
3
3
9
0
1
Ph
Me
Me
Me
2.34 ± 0.2b
23.1 ± 1.4
44.7 ± 4.4
22.1 ± 3.4%
33.8 ± 0.4%
35.3 ± 4.9%
O
O
Me
Me
O
O
O
O
O
3
3
2
3
Me
Ph
Me
Me
Me
12.5 ± 1.1
3.3 ± 0.2
30.8 ± 3.0
7.4 ± 3.9%
29.6 ± 1.9%
29.2 ± 3.9%
O
O
O
O
O
O
O
O
3
3
4
5
Me
O
O
O
O
CF3
Me
Me
Me
Me
20.3 ± 2.1% 37.5 ± 4.8%
S
36
49± 1.1%
<15%
O
3
7
Me
Me
40.3 ± 5.2% 70.4± 4.2%
O
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
3
8
Me
Me
18.4 ± 0.9% 30.6 ± 0.2%
46.2 ± 4.3% 61.7± 5.3%
O
O
S
O
3
4
9
0
Me
Me
Me
Me
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
14.9 ± 0.9
4.3 ± 0.4
13.3 ± 1.6
O
4
4
1
2
Me
Me
27.9 ± 1.8 %
S
O
O
Ph
Me
Me
2.0 ± 0.1
21.5 ± 3.1
S
O
O
43
Me
65.1 ± 7.2
40.8 ± 4.9%
N
H
O
H
N
4
4
4
Me
Me
41.9 ± 2.2% <15%
O
CH3
6k
HN
CH3
<15%
<15%
<15%
<15%
O
NC
N
OCH3
4
7m
H N
N
N
2
CH3
O
a
b
IC ± SEM are reported (n=3), while inhibition % are reported as ± SD (n = 3). maximum of
5
0
observed inhibition: 60%.
As stated above, none of the fragments or simplified molecules 13-22 showed inhibitory activity
on either the parasite or human enzymes. Only the derivatives carrying the 1-([1,1'-biphenyl]-4-
yl)ethyl (23) or 1-(4-phenoxyphenyl)ethyl (24) substituent at Nα (Figure 6) partially retained
inhibition against SmSirt2, demonstrating the crucial role of the substituted phenoxyphenyl or
ACS Paragon Plus Environment
2
2

Page 23 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
,1'-biphenyl portion at the right side of the molecule (“part C”, Figure 6). The removal of the
methyl group at either the 7 Nα (see 25) or Cβ position (see 26) (Figure 6) decreased or totally
abolished the SmSirt2 inhibitory activity, respectively. In contrast, the insertion at Cβ of groups
bigger than methyl [ethyl (27), isopropyl (28), and phenyl (29)], generally improved the SmSirt2
inhibitory potency up to 10-fold, reaching with 29 the single-digit micromolar level (IC = 2.34
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
µM). Interestingly, 29 is more selective than 7 for SmSirt2 over hSirt2 as judged by IC values,
50
but it reaches only a maximum of inhibition of 60 % which complicates the interpretation of the
selectivity data. The isopropyl derivative 28 was an exception, displaying similar inhibitory
potency as 7 against SmSirt2, and higher potency against hSirt2.
At the “part C” of the molecule (4-methoxyphenoxy moiety, Figure 6), no particular increase in
activity or selectivity was obtained with the shift of the methoxy group from para to meta
position (30), and the introduction of 3,4-dimethoxy (31) or 3,4,5-trimethoxy (34) groups at the
phenoxy portion, as well as the replacement of the 4-methoxy with a 4-trifluoromethoxy (35), 4-
methylthio (36), or 4-benzyloxy (37) group reduced the inhibitory potency against SmSirt2.
However, the insertion of two methoxy groups at 3,5 position of the phenoxy moiety (32) led to
2-fold increase of potency against SmSirt2 and improved selectivity over hSirt2. In this last
compound, the further replacement of the methyl group at Cβ with a phenyl ring (33) determined
an additional increase of potency against SmSirt2 (IC50 = 3.3 µM, 7-fold higher potency
compared to 7), confirming the positive SAR about Cβ substitution with large groups. The
replacement of the 4-methoxyphenoxy moiety of 7 with the larger 4-methoxy-1-naphthyloxy (38)
or the longer 4-methoxybenzyloxy (39) group led to a decrease in potency against SmSirt2 and, in
the case of 39, improved hSirt2 inhibition. Again in the 4-methoxyphenoxy group, the isosteric
change oxygen-sulphur atom led to the 4-methoxyphenylthio analogue 40, which produced a 1.6-
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
fold enhancement of SmSirt2 inhibition (IC = 14.9 µM), combined with improved inhibition of
5
0
the human enzyme (IC50 = 13.3 µM). The further change from the 4-methoxy to the 3,5-
dimethoxy substitution at the phenylthio moiety of 40 provided 41, that showed improved
potency and selectivity against SmSirt2 (SmSirt2 IC = 4.3 µM, hSirt2 inhibition = 27.9 % at 25
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
µM). Moreover, the combination of the positive SAR of the series with the replacement of the Cβ
methyl of 41 with the Cβ phenyl group gave 42, the most potent compound of the series vs
SmSirt2 (IC = 2 µM) and 10-fold selective over the human counterpart hSirt2 (IC = 21.5 µM).
5
0
50
The replacement of the 4-methoxyphenoxy portion with a 4-methoxyaniline (43) or 4-
methoxybenzamide (44) group led to a decrease in potency against SmSirt2. Finally, the total
absence of inhibitory activity against both parasitic and human enzymes by the intermediates 46k
and 47m confirmed the importance of the intact pyrimido[4,5-d]pyrimidine-2,4,7-triamine
scaffold (“part A”, Figure 6) for the inhibitory activity of such compounds. The inhibiting
activity of selected compounds 33, 36, 41, 42 and 44 was also evaluated in the presence of the
myristoylated substrate 3f instead of the acetylated 1. As reported in Table 4, in many cases no
significant inhibition was measured in the presence of SmSirt2 and hSirt2 with the exception of
3
3, 41 and 42, which show an inhibitory potency for the hSirt2 catalyzed demyristoylation in the
low micromolar range. This may present a new starting point for the development of human
Sirtuin inhibitors with a selectivity for long chain acyl removal, as so far a preference, if any, has
been observed only for deacetylation.
Table 4. Inhibitory activity of selected compounds 33, 36, 41, 42, and 44 against SmSirt2
and hSirt2 with myristoylated 3f as substrate.
SmSirt2 (3f)
hSirt2 (3f)
Compound
a
a
IC [µM] or %
IC [µM] or %
5
0
50
ACS Paragon Plus Environment
2
4

Page 25 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibition at 20
µM
inhibition at 20
µM
7
16.8 ± 1.14b
<15%
<15%
33
13.1 ± 2.2
38.1 ± 5.2%
25.1 ± 4.1
15.6 ± 1.9
<15%
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
4
4
4
6
1
2
4
<15%
15%
<15%
<15%
a
b
IC values are reported as ± SEM (n=3), while inhibition % as ± SD (n = 3). maximum of
50
inhibition: 45 %.
We also tested compounds 26, 33, 39 and 41 on human isotypes Sirt1 and 3 regarding
deacetylation. All compounds showed less than 15% inhibition at 25 µM (data not shown) while
51
5
00 µM of nicotinamide led to complete inhibition as expected. Evenmore, hSirt2 is the closest
4
8
human homologue to SmSirt2 (see also Table S5, data not shown for hSirt1, 3).
In vitro effects on schistosomula and adult worms.
In order to evaluate whether these optimized compounds have effects on the parasite, 29, 32, 33,
4
1 and 42 together with the prototype 7 were tested on schistosomula and adult worms. Two
potential negative controls 26 and 39 were also included in the series. As shown in Figure S6, all
tested compounds reduce the viability of cultured schistosomula at 10 and 20 µM, including 26
and 39, which show low or no activity on the recombinant enzyme (Table 3).
Experiments on adult worms report a similar trend (Figure 7), where again the SmSirt2 inhibitors
as well as the potential negative controls reduce worm pairing and egg laying already after 24
hours of culture. Whilst the possibility that the enzyme inhibitors do exert their effects via
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibition of SmSirt2 cannot be ruled out, the strong effects that 26 and 39 also have on parasite
viability and reproduction indicate that the effects of compounds shown in Figure 7 on the
parasite may be at least in part due to the modulation of other targets (off-target effects).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7. Effects of 7 and its analogues on adult worms (AW) pairing at 10 µM and 20 µM (A
and B respectively). Assay used: microscopy examination. Results after 24, 48 and 72 hours are
represented with black, white and chequered bars respectively. Effect of these compounds on egg
laying (C). Compounds 26 and 39 do not block SmSirt2 in-vitro.
To get some further insight, we measured selected compounds 7, 26, 29, 32, 33, 39 and 41 for
toxicity on human cultured cancer cells using a MTS viability assay. In this assay, the compounds
ACS Paragon Plus Environment
2
6

Page 27 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
that are inactive on SmSirt2 but active on schistosomes also show toxicity on the human cells
while the SmSirt2 inhibitors with activity on the worms do not show this general toxicity (see
Figure 8).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. Growth inhibition of HL-60 acute myeloid leukaemia cells by compound 7 and
selected analogues at 10 µM assay concentration.
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 95
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DISCUSSION AND CONCLUSION
The identification of a robust deacylase activity, supported by kinetic investigations, expands the
range of possible biological activities relevant for SmSirt2. The high demyristoylation efficiency
of this enzyme suggests the merit of undertaking future investigations on deacylation inhibition,
in addition to deacetylation inhibition, as a new strategy to kill both schistosomula and adult
worms. For inhibitors of hSirt2 with a thiomyristoylated lysine core (also called TM), it has been
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
proposed that a dual deacetylase/demyristoylase activity is beneficial for cellular potency in
cancer cells and selectivity over non-cancer cells.61
With the aim of finding potent and selective inhibitors of SmSirt2, we performed screening of the
GSK-kinetobox library, composed of 592 compounds with proven activity against cultured
kinetoplastids, namely Leishmania donovani, Trypanosoma cruzi and Trypanosoma brucei. From
an initial test we identified 5, 6 and 7 as valuable hits for further optimization. As outlined above,
we focused our attention on 6 and 7. Regarding 6, the attempt to potentially target the enzymatic
(possibly extended) C pocket through an octanoylated and decanoylated analogue, led to a loss in
activity. Extensive SAR studies performed on 7 highlighted the crucial role of the presence of the
intact pyrimido[4,5-d]pyrimidine-2,4,7-triamine (“part A” of the molecules), disubstituted at N7
(Nα in Figure 6) with a methyl and a substituted 1-phenylethyl moiety (“part B”), further carrying
a substituted phenoxy/phenylthio group at its C4 position (“part C”), recognized as crucial to
elicit SmSirt2 inhibitory potency and selectivity over hSirt2 (Figure 6). In particular, the presence
at “part C” of the 4-methoxyphenoxy, 3,5-dimethoxyphenoxy and 3,5-dimethoxyphenylthio
portion, combined with the the phenyl group at Cβ position (“part B”) led to the most potent
derivatives 29, 33, 41, and 42 with IC values in the single-digit micromolar range for SmSirt2.
5
0
Experiments with 7 and its analogues on schistosomula and adult worms showed strong activity
which was also present in enzymatically inactive compounds. However, while compounds 26 and
ACS Paragon Plus Environment
2
8

Page 29 of 95
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
9, inactive on SmSirt2, also showed unspecific toxicity to human cancer cells, the selective
SmSirt2 inhibitors 7, 29, 32, 33 and 41 were non-toxic to human cancer cells. Generally, the
compounds show good properties in terms of druglikeness, e.g. logP values are between 2.7 (32)
and 4.6 (42) (see Table S7). Molecular weight is mostly above 400 but only up to 525 Da
maximum.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Thus, for the first time, we can show that SmSirt2 can be drugged with selectivity over the human
isotype (especially 29, 33 and 41) and the inhibitors block Schistosoma growth without general
toxicity to human cells. This can be used as starting point for further optimization studies. In
addition, leads for new inhibitors of hSirt2 have been identified, interestingly some with a
preference for demyristoylation over deacetylation.
ACS Paragon Plus Environment
2
9