3,6-Diamino-4-(2-halophenyl)-2-benzoylthieno[2,3- b ]pyridine-5- carbonitriles are selective inhibitors of plasmodium falciparum glycogen synthase kinase-3

Article abstract of DOI:10.1021/jm301575n

Plasmodium falciparum is the infective agent responsible for malaria tropica. The glycogen synthase kinase-3 of the parasite (PfGSK-3) was suggested as a potential biological target for novel antimalarial drugs. Starting from hit structures identified in a high-throughput screening campaign, 3,6-diamino-4-(2-halophenyl)-2-benzoylthieno[2,3-b]pyridine-5-carbonitriles were discovered as a new class of PfGSK-3 inhibitors. Being less active on GSK-3 homologues of other species, the title compounds showed selectivity in favor of PfGSK-3. Taking into account the X-ray structure of a related molecule in complex with human GSK-3 (HsGSK-3), a model was computed for the comparison of inhibitor complexes with the plasmodial and human enzymes. It was found that subtle differences in the ATP-binding pockets are responsible for the observed PfGSK-3 vs HsGSK-3 selectivity. Representatives of the title compound class exhibited micromolar IC₅₀ values against P. falciparum erythrocyte stage parasites. These results suggest that inhibitors of PfGSK-3 could be developed as potential antimalarial drugs.

Full text of DOI:10.1021/jm301575n

Article
pubs.acs.org/jmc
3,6-Diamino-4-(2-halophenyl)-2-benzoylthieno[2,3‑b]pyridine-5-
carbonitriles Are Selective Inhibitors of Plasmodium falciparum
Glycogen Synthase Kinase‑3
Wiebke Fugel,† Anselm Erich Oberholzer,‡ Bernhard Gschloessl,§ Ron Dzikowski,∥ Narkiss Pressburger,∥
Lutz Preu,† Laurence H. Pearl,⊥ Blandine Baratte,# Morgane Ratin,# Ilya Okun,∇ Christian Doerig,○
Sebastian Kruggel,◆ Thomas Lemcke,◆ Laurent Meijer,*,#,¶ and Conrad Kunick*,†
†Institut fur Medizinische und Pharmazeutische Chemie, Technische Universitat Braunschweig, Beethovenstrasse 55, 38106
̈
̈
Braunschweig, Germany
‡Structural Biology Community Laenggasse (sbcl), 3000 Bern, Switzerland
§Centre de Biologie pour la Gestion des Populations (CBGP), UMR1062, Institut National de la Recherche Agronomique (INRA),
Campus International de Baillarguet, CS 30016, 34988 Montferrier-sur-Lez, France
∥Department of Microbiology & Molecular Genetics, The Kuvin Center for the Study of Infectious and Tropical Diseases, Institute
for Medical Research Israel-Canada (IMRIC), The Hebrew University-Hadassah Medical School, Jerusalem 91120, Israel
⊥Genome Damage and Stability Centre, School of Life Sciences, University of Sussex, Brighton BN1 9RH, United Kingdom
#Protein Phosphorylation & Human Disease Group, CNRS USR 3151, Station Biologique, Place Georges Teissier, 29680 Roscoff,
France
∇Research & Development, ChemDiv, Inc., 6605 Nancy Ridge Drive, San Diego, California 92121, United States
○Department of Microbiology, School of Biomedical Sciences, Monash University, Wellington Road, Clayton, Victoria 3800,
Australia
◆Institut fur Pharmazie, Universitat Hamburg, Bundesstrasse 45, 20146 Hamburg, Germany
̈
̈
¶ManRos Therapeutics, Perharidy Research Center, 29680 Roscoff, Bretagne, France
S
* Supporting Information
ABSTRACT: Plasmodium falciparum is the infective agent responsible for malaria
tropica. The glycogen synthase kinase-3 of the parasite (PfGSK-3) was suggested as
a potential biological target for novel antimalarial drugs. Starting from hit structures
identified in a high-throughput screening campaign, 3,6-diamino-4-(2-halophenyl)-
2-benzoylthieno[2,3-b]pyridine-5-carbonitriles were discovered as a new class of
Pf GSK-3 inhibitors. Being less active on GSK-3 homologues of other species, the
title compounds showed selectivity in favor of Pf GSK-3. Taking into account the X-
ray structure of a related molecule in complex with human GSK-3 (HsGSK-3), a
model was computed for the comparison of inhibitor complexes with the
plasmodial and human enzymes. It was found that subtle differences in the ATP-
binding pockets are responsible for the observed Pf GSK-3 vs HsGSK-3 selectivity.
Representatives of the title compound class exhibited micromolar IC50 values
against P. falciparum erythrocyte stage parasites. These results suggest that
inhibitors of Pf GSK-3 could be developed as potential antimalarial drugs.
While older medications such as chloroquine or pyrimethamine
INTRODUCTION
■
have become ineffective against P. falciparum due to resistance in
most endemic regions, the current standard therapy consists of
an artemisinin derivative combined with a second antiplasmodial
drug.2,3 Recent reports of P. falciparum decreased susceptibility
against artemisinin from the Cambodia−Thailand border are
alarming.1,4 Further propagation of artemisinin resistance could
compromise the progress achieved in the fight against malaria in
Malaria is one of the most severe infectious diseases worldwide,
affecting mainly the poor population in developing countries.
Young children in tropical African countries are particularly
threatened by Plasmodium falciparum, the parasite causing the
most virulent form of malaria. Although the use of insecticide-
treated mosquito nets has resulted in a significant reduction in
the number of infections in several countries, there were still 225
million malaria cases reported in 2009, 781 000 of which ended
fatally.1 A major problem of malaria therapy is the limited
number of suitable drugs and the development of resistance.
Received: October 25, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Sequence analysis of GSK-3 orthologues in 71 species. Phylogenetic tree according to the GSK-3 sequence. The set of 71 GSK-3 proteins is
composed of sequences presented in the orthoMCL database (http://www.orthomcl.org)36 and orthologues which were recovered by bidirectional
BLAST (blastp) analyses of the GSK3 human isoforms against the specific species reference protein set. Multiple alignments were created with ClustalW
(version 2.0.10)37 and were manually improved afterward. The phylogenetic tree itself was built with the maximum likelihood approach implemented in
PHYML (version 3.0).38,39 As amino acid substitution, the LG model40 was chosen. To obtain confidence estimates in the tree topologies, a bootstrap
analysis with 1000 was applied. The consensus tree was visualized with the application Forester (http://www.phylosoft.org/forester). Color codes are as
given in the Supporting Information, Figure S2. Plasmodium GSK-3 is given in bold.
recent years. As a consequence, there is an urgent need for
additional antimalarial drugs which ideally should act on novel
biological targets and/or should represent novel chemotypes. In
a recent study published by Gamo et al., a collection of nearly 2
million compounds from the GlaxoSmithKline inventory were
screened against in vitro cultivated P. falciparum, more than 13
000 compounds of which inhibited the growth of parasites in a 2
μM concentration.5 Interestingly, the fraction of molecules
presumably acting by inhibiting protein kinases was considerably
higher among the hit compound group compared to all screened
compounds. The authors concluded that protein kinase
inhibitors might be a rich source of antimalarial drug leads in
the future.4−6 The selective inhibition of microbial protein
kinases has repeatedly been suggested as a therapeutic
strategy.7−15 However, of the published examples of rationally
developed inhibitors against Plasmodium protein kinases (e.g.,
Pfnek-1,16,17 Pf PK7,18,19 Pfmrk,20−25 and PfCDPK126), none,
to our best knowledge, have entered a clinical evaluation stage. In
a previous paper the P. falciparum glycogen synthase kinase-3
(Pf GSK-3) was characterized.27 Although its biological functions
are not yet identified, Pf GSK-3 was demonstrated as an essential
enzyme for completion of the asexual erythrocytic cycle of the
parasite28 and thus suggested as a putative antimalarial drug
target.27 In a model using Plasmodium berghei-infected mice, it
was recently shown that the GSK-3 inhibitor lithium suppressed
erythrocytic parasitemia and protected the animals against
malaria.29 In a recent paper, the identification of inhibitors of
the GSK-3 homologue of Trypanosoma brucei, the parasite
causing African sleeping sickness, was reported.30 Although none
of the identified compounds showed absolute selectivity for
TbGSK-3 against a panel of approximately 40 human kinases,
some showed antitrypanosomal activity in submicromolar
concentrations and a reasonable safety window when tested on
human MRC-5 cells. The authors concluded that these
B
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
compounds are promising starting points for an antitrypanoso-
mal drug development program.30 Aiming at the development of
new antimalarial drugs, the campaign reported here was directed
toward identification of small molecular PfGSK-3 inhibitors
displaying a certain degree of selectivity over mammalian
PfGSK-3 orthologues.
Chart 1. Hits 1 and 2 from the Initial Test Run, Parent 3-
Amino-4-arylthieno[2,3-b]pyridine Scaffold 3, and Hit
Structure 4 Identified from the Focused ChemDiv Library
a
Based on Scaffold 3
RESULTS
■
Identification of 3-Amino-4-arylthieno[2,3-b]pyridines
as PfGSK-3 Inhibitors by High-Throughput Screening. A
preliminary study reported the interaction of both plasmodial
and mammalian GSK-3 with a selection of established kinase
inhibitors.27 Indeed, some entities (e.g., indirubin-3′-monoxime,
hymenialdisine) showed Pf GSK-3 inhibition, however without
selectivity against the porcine GSK-3 homologue (SsGSK-3).
This was found to also be the case for other reference kinase
inhibitors (Supplementary Table S1, Supporting Information).
In contrast, several members of the paullone family31−33 of GSK-
3 inhibitors showed a strong selectivity for inhibition of the
mammalian GSK-3. Although the latter finding was disappoint-
ing at first sight, we concluded that the mammalian and the
plasmodial homologues are sufficiently divergent to enable
selectivity. Furthermore, a meticulous comparison of the ATP-
binding site architectures of PfGSK-3 and its human orthologue
by bioinformatic methods revealed differences that warranted a
search for PfGSK-3-selective chemotypes.34,35 Given the strong
conservation of GSK-3 across evolution, we analyzed a selection
of 71 GSK-3 orthologues of various species (Figure 1;
Supplementary Table S2, Supporting Information). The results
showed that GSK-3 orthologues tended to group into clusters of
their taxonomic origin such as metazoa, fungi, and green plants.
The only exceptions were GSK-3 proteins of Entamoeba
histolytica (Amoebozoa) and Encephalitozoon cuniculi (fungi),
which were arranged with the Apicomplexa group. The analyzed
GSK-3 proteins to all five Plasmodium species cluster separately
from GSK-3 orthologues, suggesting that inhibitors specific for
GSK-3 from Plasmodium species might be identified. In addition,
the recent formal demonstration that PfGSK-3 is absolutely
required for the survival of the erythrocytic stage of P. falciparum
validates Pf GSK-3 as a potential therapeutic target.28
We therefore set up a high-throughput screening (HTS) in
which 10 480 compounds (about half of which were kinase
directed) were evaluated at a concentration of 10 μM against
recombinant Pf GSK-3, which was previously shown to be active
in in vitro kinase assays.27 Only 18 molecular entities were found
to be active (>50% inhibition at 10 μM), representing a rather
modest hit rate of 0.17%. Compounds displaying more than 50%
inhibition were re-evaluated by determination of IC50 values
against PfGSK-3 and against the native mammalian (porcine)
SsGSK-3. Interestingly, five of these inhibitors shared a common
heterocyclic ring system as the parent scaffold, namely, the 3-
amino-4-arylthieno[2,3-b]pyridine motif 3. Two examples, 1 and
2, of these initial hits are depicted in Chart 1.
a
Kinase inhibitory activities (IC50, μM): (1) Pf GSK-3, 3.0; SsGSK-3,
7.0; (2) PfGSK-3, 9.0; SsGSK-3, >10.0; [4 (resynthesized material)]
Pf GSK-3, 1.61; SsGSK-3, >100.0.
Chart 2. 3-Amino-4-arylthieno[2,3-b]pyridines 5−8 Designed
as Analogues of Hit Structure 4
a
a
For substituents R1 and R2, refer to Table 1.
namely, 5, in which the substitution pattern at the phenyl rings of
hit compound 4 is varied, 6, in which, by insertion of a nitrogen
atom, the aromatic ketone moiety is changed to a carboxamide
(analogous to hit compound 1), 7, in which the benzoyl
substituent is replaced by a cyano group (crossover design based
on 2 and 4), and 8, in which the amino and cyano groups at the
pyridine ring are replaced by annulation of a cyclohexene ring
(crossover design based on 2 and 4).
At the start of the synthesis program, the hit structure 4 was
compared with the commercially available analogue 5dd (Table
1) distinguished only by the lack of the o-iodo substituent present
in 4. Since 5dd was completely devoid of activity against either
Pf GSK-3 or SsGSK-3, we (1) concluded that the mentioned
ortho substitution was important for kinase inhibitory activity
and (2) because iodo substituents are undesirable in drug
molecules, paid special attention to a replacement of this element
in the projected series 5−8.
We then tested a small focused library of 427 compounds
related to 3 for PfGSK-3 inhibition. Although the majority of the
compounds were inactive, some showed selective activity for
PfGSK-3 with IC50 values in the micromolar concentration
range, as exemplified by congener 4.41
Stimulated by these preliminary results, we initialized a
program for molecular modification of 4 with the aim of gaining
information on structure−activity relationships and of increasing
the Pf GSK-3 inhibitory activity and selectivity. In this context we
synthesized the four groups of derivatives illustrated in Chart 2,
Chemistry. The precursor molecules 12−20 were synthe-
sized from suitable aromatic aldehydes 9, malonodinitrile 10, and
cyanothioacetamide 11 in a modified one-pot procedure
analogous to the method described by Sharanin et al.42 (Scheme
1).
Subsequently, the desired 3-amino-4-(2-haloaryl)thieno[2,3-
b]pyridines 5−7 were prepared employing a successive one-pot
S-alkylation/Thorpe−Ziegler cyclization methodology.43−45
Initially, the 2-thioxo-1,2-dihydropyridines 12−20 were dis-
solved in DMF, deprotonated by 1 equiv of potassium hydroxide,
C
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. Structures and Kinase Inhibitory Activities of 3-Amino-4-arylthieno[2,3-b]pyridines 5−8
ab
IC50(PfGSK-3) (μM)
ab
,
,
compd
R1
R2
IC50(SsGSK-3) (μM)
5a
5b
5c
5d
5e
5f
2-MeOPh
2-MeOPh
2-MeOPh
2-MeOPh
2-MeOPh
2-MeOPh
2-MeOPh
2-MeOPh
2-IPh
Ph
6
>100
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>100
>10
>10
>100
>10
>10
>10
>10
>10
>10
>10
>10
>10
>100
>100
>100
>10
>10
>10
>10
>10
1.3
4-ClPh
4-MeOPh
4-MePh
4-CF3Ph
3-ClPh
2-ClPh
4-CNPh
Ph
>10
>10
>10
>10
8.5
5g
5h
5i
>10
>10
2.4
5j
2-IPh
4-MeOPh
4-MePh
4-CF3Ph
3-ClPh
2-ClPh
3,4-Cl2Ph
4-ClPh
Ph
1.7
5k
5l
2-IPh
3.2
2-IPh
>10
0.91
>10
70
5m
5n
5o
5p
5q
5r
2-IPh
2-IPh
2-IPh
2-EtOPh
2-BrPhe
2-BrPhe
2-BrPhe
2-BrPhe
2-thienyl
2-ClPhe
2-ClPhe
2-ClPhe
2-ClPhe
2-ClPhe
2-ClPhe
2-ClPhe
2-FPhe
Ph
>10
1.1
4-ClPh
3-ClPh
3,4-Cl2Ph
3-ClPh
3-ClPh
3,4-Cl2Ph
3-MeOPh
3-CF3−Ph
3-CNPh
3-BrPh
3-FPh
>100
0.61
3.8
5s
5t
5u
5v
5w
5x
5y
5z
5aa
5bb
5cc
5dd
6a
6b
6c
7a
7b
7c
7d
7e
7f
>10
0.48
2.7
1.7
2.2
0.5
0.91
>100
10
3-ClPh
4-ClPh
4-ClPh
4-ClPh
4-ClPh
>100
>10
>10
>10
>10
>10
0.8
2-IPh
2-BrPh
2-ClPh
2-MeOPh
2-IPh
2-BrPh
2-ClPh
2-EtOPh
2-MePh
2-thienyl
3-indolyl
2-IPh
0.13
>10
0.51
0.18
1.6
0.4
>10
2
7g
7h
8a
8b
2.3
3.5
Ph
>10
>10
>10
>10
2-IPh
4-ClPh
a
IC50 values were calculated from dose−response curves. The highest inhibitor concentration used in the assays was either 10 or 100 μM.
b
Recombinant PfGSK-3 and native, affinity-purified SsGSK-3 were used in the studies. All data points for construction of dose−response curves were
recorded in triplicate. Typically, the standard deviation of single data points was below 10%.
of the haloalkane and the 2-thioxo-1,2-dihydropyridine,
completion of the reaction at room temperature required
between 30 min and 6 h (Scheme 2).
Scheme 1. Synthesis of 2-Thioxo-1,2-dihydropyridines 12−
a
20
Following a modified protocol originally reported by
Elnagdi,46 the 3-amino-4-aryl-5,6,7,8-tetrahydrothieno[2,3-b]-
quinolines 8a,b were prepared in a two-step procedure from
cyclohexanone (24) and 2-cyano-3-(2-iodophenyl)-2-propene-
thioamide (25). The resulting 4-(2-iodophenyl)-2-thioxo-
1,2,5,6,7,8-hexahydroquinoline-3-carbonitrile (26) underwent
Thorpe−Ziegler ring closure after S-alkylation by suitable
phenacyl bromides 21 and treatment with alkali (Scheme 3).
Biological Evaluation and Discussion. The compounds
listed in Table 1 were evaluated in vitro by measuring the IC50
a
Reagents and conditions: (i) EtOH, piperidine, reflux, 3 h, 21−61%.
and then S-alkylated by means of a suitable haloalkane, 21, 22, or
23. The resulting thioether was not isolated, but cyclized by
addition of a second equivalent of base. Depending on the nature
D
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
modes in the enzyme, refuting the rationale for the combination
of structure elements. On the other hand, the replacement of the
cyclohexene annulation displayed by hit structure 2 with the 6-
amino-5-cyano pattern of hit structure 4 led to derivative 7g,
which exhibited marked but unselective PfGSK-3/SsGSK-3
inhibition. A similar unselective GSK-3 inhibitory activity was
found with the analogues 7c, 7d, and 7f, which display a
comparably small ortho substituent at the phenyl ring. In
contrast, 7a, 7b, and 7e, which are distinguished by larger ortho
substituents, remained inactive.
Scheme 2. Synthesis of 3,6-Diamino-4-arylthieno[2,3-
b]pyridine-5-carbonitriles 5−7
a
For an explanation of the observed structure−activity
relationships within the family of thieno[2,3-b]pyridines, a
comparison of the molecular inhibitor/kinase interactions was
desirable. Unfortunately, all attempts to generate crystals of
Pf GSK-3 in complex with one of the inhibitors of series 5 failed.
Therefore, the analogue 7d showing the highest SsGSK-3
inhibition was cocrystallized with the human enzyme (Figure 1).
Hexagonal crystals of the (pTyr216)-GSK3β−7d complex
showed poor diffraction quality. (pTyr216)-GSK3β and 7d
were cocrystallized with an axin-derived peptide [axin(383−
401)]47 to find a crystal form with better diffraction properties.
Crystals of the (pTyr216)-GSK3β−axin(383−401)−7d com-
plex belonged to the orthorhombic space group P212121, with
one copy of the (pTyr216)-GSK3β−axin(383−401)−7d
complex per crystallographic asymmetric unit.
a
Reagents and conditions: (i) (1) DMF, 1 equiv of KOH, rt, 30 min,
(2) DMF, 1 additional equiv of KOH, rt, TLC monitoring, 15−91%.
Scheme 3. Synthesis of 3-Amino-4-aryl-5,6,7,8-
tetrahydrothieno[2,3-b]quinolines 8a,b
a
The crystal structure of the (pTyr216)-GSK3β−axin(383−
401)−7d complex was determined by molecular replacement
using the structure of the (pTyr216)-GSK3β−axin(383−401)
complex47 as a search model and refined at 2.64 Å resolution.
Statistics for data collection and refinement are given in Table 2.
The overall structure of (pTyr216)-GSK3β−axin(383−401)−
7d is basically identical to the (pTyr216)-GSK3β−axin(383−
401) structure, with the N-terminal lobe of GSK3β predom-
a
Reagents and conditions: (i) 1,4-dioxane, piperidine, 80 °C, 2−4 h,
TLC monitoring, 37−43%; (ii) (1) DMF, 1 equiv of KOH, rt, 30 min,
(2) DMF, 1 additional equiv of KOH, rt, TLC monitoring, 45−49%.
Table 2. Crystal Structure Data and Refinement Statistics for
GSK-3β−7d
values on recombinant PfGSK-3 and on native porcine SsGSK-3
in a radiometric protein kinase assay. The finding that 5u was
inactive on PfGSK-3 confirmed the observation that in series 5
an ortho substituent at the 4-aryl ring is mandatory for Pf GSK-3
inhibition. The size of this ortho substituent appears to be
important, since large 2-alkoxy substituents (5a−5h, 5p) or the
small fluorine substituent (5cc) were detrimental for Pf GSK-3
inhibitory activity. The halogen substituents iodine, bromine,
and chlorine were best suited in the ortho position with a
tendency in favor of chlorine. Regarding substitution at the
benzoyl ring, all congeners showing submicromolar Pf GSK-3
inhibition (5m, 5s, 5v, 5z, 5aa) displayed a substituent in the 3-
position, highlighting a preference for this structural pattern. All
derivatives of series 5 showing activity on Pf GSK-3 were inactive
on the mammalian homologue, thus corroborating the direction
of selectivity found with hit compound 4.
The representatives of the small series 6a−6c deviate from the
hit structure 4 by insertion of an additional nitrogen atom
between the thieno[2,3-b]pyridine and the benzoyl elements.
The fact that 6 are completely inactive on Pf GSK-3 may originate
from a close space limitation of the pocket accommodating the
benzoyl substituent. An obvious structural crossover design
strategy based on the hit compounds 2 and 4 was realized in
series 7 and 8. When the 2-iodo-substituted 4-aryl substituent
and the cyclohexene annulation at the thieno[2,3-b]pyridine
scaffold were combined, the resulting congeners 8a and 8b were
found to be completely inactive. A reasonable explanation for this
disappointing result would be that 2 and 4 have different binding
space group
P212121
unit cell params a, b, c (Å)
73.15, 76.16, 86.54
1
a
no. of molecules/asu
Data Collection (XDS)
beamline
I02
wavelength (Å)
resolution range
0.9700
b
45.05−2.64 (2.80−2.64)
57827 (14488)
98.8 (97.0)
6.9 (86.0)
no. of observations (unique)
completeness (%)
c
Rsym (%)
I/σ(I)
18.0 (2.3)
Refinement (Phenix)
resolution range (Å)
45.05−2.64
13612
869
no. reflns in working set
no. of reflns in test set
no. of non-hydrogen atoms
no. of solvent water molecules
R/Rfree (%)
3029
70
20.1/24.2
0.003
RMSD(bond length) (Å)
RMSD(bond angle) (deg)
0.750
a
b
asu = asymmetric unit. The values in parentheses of the resolution
range, completeness, Rsym, and I/σ(I) correspond to the outermost
c
resolution shell. Rsym = ∑hkl∑j|I(hkl;j) − ⟨I(hkl)⟩|/(∑hkl∑j⟨I(hkl)⟩),
where I(hkl;j) is the jth measurement of the intensity of the unique
reflection (hkl) and ⟨I(hkl)⟩ is the mean over all symmetry-related
measurements.
E
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
inantly consisting of β-sheets and a predominantly α-helical C-
terminal lobe (Figure 2A). The axin(383−401) peptide binds as
the gatekeeper (gk) amino acid. A further hydrogen bond
network is mediated by a water molecule located between the
inhibitors’s pyridine nitrogen atom and the backbone carbonyl
oxygen of Asp133 (gk + 1) on one hand and the NH of Val135 on
the other hand. Eventually, the 3-amino-2-cyano motif of the
ligand shapes a pair of hydrogen bonds directed to the ε-amino
group of the conserved Lys85 and to the side chain of Asp200 of
the DFG motif. The 4-aryl substituent is oriented orthogonally to
the plane of the heterocyclic scaffold, making hydrophobic
contacts to Gln185 and Asn186 at the bottom of the ATP-
binding pocket (not depicted in Figure 3 for the sake of clarity).
The 2-chloro substituent is directed toward the roof of the
pocket, where larger substituents (such as iodo or ethoxy) cannot
be accommodated, which explains the lack of activity by 7a, 7b,
and 7e.
For structure comparison, congener 5v was docked into the
ATP-binding site of a Pf GSK-3 homology model48 we have
reported recently and to HsGSK-3 (PDB code 1j1b)49 using the
FlexX module of LeadIT (Figure 4). Docking poses closely
Figure 2. (A) Ribbon diagram showing the overall fold of the
(pTyr216)-GSK3β−axin(383−401)−7d complex. α-Helices are col-
ored in blue and β-sheets in green. The axin peptide is colored in red.
The pTyr216 and 7d molecules are shown as stick models with atomic
colors as follows: oxygen, red; nitrogen, blue; carbon, light gray; sulfur,
yellow; chlorine, green. (B) A 2Fo − Fc composite annealed omit
electron density map of 7d depicted in chicken wire representation
(marine) contoured at 1.0 standard deviation above the mean. Atomic
colors are as follows: oxygen, red; nitrogen, blue; carbon, light gray;
sulfur, yellow; chlorine, green.
a single amphipathic α-helix into a hydrophobic surface channel
on the C-terminal helical domain (Figure 2A). As a prominent
difference, the glycine-rich loop (AA62−70) is shifted toward the
active site, resulting in a more closed active site conformation. A
2Fo − Fc composite annealed omit map shows clear electron
density for compound 7d (Figure 2B).
It was found that 7d occupies the ATP-binding pocket of the
enzyme, representing the characteristic binding mode of small
molecular kinase inhibitors to the host protein (Figure 3). Within
the pocket, the flat annulated heterocyclic ring system is rendered
into position by an ensemble of H-bonds to amino acid residues
of the protein. The first H-bond is formed between the inhibitor’s
6-amino group and the backbone carbonyl oxygen of Val135,
which represents the amino acid three positions downstream of
Figure 4. Docking of 5v (carbon scaffold shown in orange) to Pf GSK-3
(left, protein shown in light pink) and HsGSK-3 (right, X-ray crystal
structure 1J1B (PDB), protein shown in light blue). Upper row:
Illustration of the halogen-binding groove at the bottom of the binding
site. The o-halogen substituent penetrates the protein surface of the
human binding site (right, halogen surface depicted as a mesh), while it
fits almost perfectly to the protein surface of the plasmodial enzyme
(left). Lower row: Hydrogen-bonding network that anchors the ligand
in the binding site. The distance between the heterocyclic sulfur and the
respective gatekeeper residue and the hydrophobic contact area are
depicted (surface areas of the interacting atoms are shown as meshes).
Images were drawn with Pymol 1.4.1 (PyMOL Molecular Graphics
System, Schrodinger, LLC).
̈
related to the 7d−HsGSK-3 complex were generated with 5v for
each target. The 6-amino group of 5v is making a direct H-bond
to the hinge residue Ile160/Val135 and via a water molecule to
Glu158/Asp133. The ε-amino group of the conserved Lys 108/
Lys85 acts as a hydrogen bond donor to the carbonyl oxygen of
the inhibitor 5v. Both aromatic substituents are perpendicular to
the heterocyclic scaffold, but in contrast to the 7d−HsGSK-3
complex, the ortho substituent of the phenyl ring of 5v is
accommodated in a shallow groove at the bottom of the pocket,
which was identified by molecular interaction field calculations as
a potential interaction point for halogen substituents in the
plasmodial binding site (refer to Figure 4).34 Due to the bulky 2-
aroyl substituent, the heterocyclic plane of 5v is rotated about
Figure 3. Binding mode of compound 7d (carbon scaffold shown in
green) in HsGSK-3 (X-ray crystal structure). The red spheres
correspond to crystallographically observed water molecules, one of
which mediates the contact between the ligand and Asp133 (the gk + 1
amino acid). Key binding site residues are depicted as sticks.
F
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
30° clockwise and kept in position by a slightly different
hydrogen bond network to the hinge residues and the Lys108/85
compared to the position of 7d in the binding site. In the
plasmodial enzyme this leads to a hydrophobic interaction
between the gatekeeper side chain (Met157) and the thiophene
sulfur of compounds 5. In the human enzyme such an interaction
is not established, because the gatekeeper side chain (Leu132) is
located further away from the sulfur atom of ligands 5. This
selectivity phenomenon is not observed with representatives of
the 7 series, because these analogues demand less space below
the glycine-rich loop and are therefore accommodated equally
well in the human and plasmodial enzymes. In contrast, the
compounds of the 5 series are much better accommodated in the
plasmodial kinase than in the human one. This assumption is
supported by the score values for the respective binding poses
calculated by the Hyde scoring function. For compound 5v the
value is about 10 kJ/mol higher in the plasmodial enzyme
compared to the human one.
Table 3. Inhibition of GSK-3 Orthologues of Different Species
by 4 and Selected Congeners 5 (IC50, μM)
a b
,
species
Plasmodium
falciparum
4
5a
5m
5s
5v
5z
0.5
1.61
6
0.91
0.61
0.48
Plasmodium knowlesi 15.5
95.0
2.75
1.45
2.45
6.5
Toxoplasma gondii
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
Cryptosporidium
parvum
Trypanosoma cruzi
Trypanosoma brucei
Leishmania major
>100
>100
>100
>100
>100
>100
80
>100
>100
>100
70
>100
>100
>100
44
>100
>100
>100
72
>100
>100
>100
100
Leishmania donovani 80
Dictyostelium
discoideum
>100
>100
>100
80
>10
>100
Sus scrofa (GSK-3α/ >100
β)
>100
>100
>100
>100
>100
50
5.6
>10
>100
3.3
22
Homo sapiens (GSK- >100
3α)
>100
4.0
>25
8.1
The inactive compounds 6 could not be docked to either the
plasmodial or the human GSK-3, which is in perfect agreement
with our binding mode assumption. Derivatives 8 are inactive
because they lack the 6-amino group present in 5, which is
important as a H-bond anchoring group toward the gk + 3
residue of the kinase hinge region.
Homo sapiens (GSK- >100
3β)
a
IC50 values were calculated from dose−response curves. The highest
b
inhibitor concentration used in the assays was 10, 25, or 100 μM. All
data points were recorded in triplicate. Typically, the standard
deviation of the single points was below 10%.
GSK-3 is a widely conserved enzyme in eukaryotic organisms
(Figure 1). It has been pointed out before that selective GSK-3
inhibitors might also be useful to treat infections by other
parasites besides P. falciparum.30,35,50,51 Compounds known as or
derived from GSK-3 inhibitors have been studied as anti-
trypanosomal51 and antileishmanial52 agents. In this context, we
explored the inhibitory activity of the congeners 4, 5a, 5m, 5s, 5v,
and 5z on the GSK-3 homologues of other pathogenic
microorganisms belonging to the phylum Apicomplexa, namely,
of Plasmodium knowlesi, Toxoplasma gondii, Cryptosporidium
parvum, Trypanosoma cruzi, T. brucei, Leishmania major, and
Leishmania donovani. To further estimate the selectivity of the
compounds, we also included the GSK-3 homologues of the
slime mold Dictyostelium discoideum and the two human GSK-3α
and GSK-3β isoforms. The results given in Table 3 show that 4
and 5 indeed selectively inhibit Pf GSK-3. The only other
parasitic orthologues inhibited by micromolar inhibitor concen-
trations were GSK-3 of P. knowlesi, the closest organism to P.
falciparum, and GSK-3 of L. donovani, the infectious agent
causing visceral leishmaniasis (kala azar). The GSK-3 ortho-
logues of the other parasites belonging to the order
Trypanosomatida (L. major, T. cruzi, T. brucei) remained
unaffected up to 100 μM. The same insensitivity was observed
with both GSK-3's of the parasites T. gondii and C. parvum.
Similar to the native porcine enzyme, the recombinant human
GSK-3α isoform was not inhibited by congeners 5 listed in Table
3. In contrast, 5s, 5v, and 5z inhibited the human GSK-3β
isoform in one-digit micromolar concentrations. Nevertheless,
also for these compounds there is still a selectivity gap of 1 order
of magnitude for inhibition of the plasmodial and human
enzymes. Interestingly, the compounds also inhibited the GSK-3
homologue of another species of Plasmodium, P. knowlesi. This
suggests that the selectivity may extend to GSK-3 orthologues of
other Plasmodium species, which are also responsible for other
forms of malaria, namely, P. vivax, P. ovale, P. malariae, and P.
knowlesi.
compound 5v in broad panels of human protein kinases
established in the University of Dundee screening platform.53
Kinase inhibition activity was determined with 1 and 10 μM
concentrations of the test compound. None of the 77 kinases was
inhibited by the 1 μM compound concentration by more than
50%. The 10 μM concentration of 5v inhibited only 12 kinases by
more than 50% (BRSK2, BTK, CAMK1, CHK2, ERK8, IKKε,
IR-HIS, MAPKAP-K2, MKK1, p38α MAPK, P38 β MAPK,
PAK4) and only 1 additional kinase (PKB α) by more than 80%
(refer to the Supporting Information for details), indicating a
high selectivity in favor of Pf GSK-3 versus diverse human kinases
(Supplementary Table S3).
To complement the selectivity studies, 5v was screened on the
large-scale DiscoverX KinomeScan panel. This interaction assay
provides an overall view of the affinity of a compound for any of
402 kinases.54,55 A semiquantitative scoring of this primary
screen was provided (Figure 5). The results showed only low
interaction with a few kinases (Figure 5; Supplementary Table
S4, Supporting Information).
For evaluation of antiparasitic activity, eight congeners of
series 5 showing potent Pf GSK-3 inhibition were tested in a
luminescence-based assay against erythrocytic stages of P.
falciparum parasites. For an initial activity assessment, parasites
expressing luciferase were incubated for 72 h with a 15 μM
concentration of the test compounds and luminescence was
measured after cell lysis. Compounds showing more than 80%
inhibition of parasite viability were then characterized by
determination of EC50 values in a similar experiment. The
results given in Table 4 show that all test compounds inhibited
the parasites at 15 μM. Compounds 5v, 5w, 5y, and 5aa
exhaustively decreased viability at 15 μM and exhibited single-
digit micromolar EC50 values. Although there is a gap of roughly
1 order of magnitude between concentrations for enzyme
inhibition and parasite inhibition, the results demonstrate that
Pf GSK-3 could serve as an exploitable biological target for the
development of antimalarial agents. In terms of structure−
activity relationships, the results show that a 2-chloro substituent
at the phenyl ring and a halogen-containing substituent at the 3-
Considering the selectivity data across the GSK-3 orthologues
of different species on one hand and overall bioactivity in the
compound class on the other hand, we decided to evaluate
G
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
halophenyl)-2-benzoylthieno[2,3-b]pyridine-5-carbonitrile
motif. The compounds of highest potency inhibit PfGSK-3 in
submicromolar concentrations and exhibit selectivity versus
GSK-3 orthologues of other species and versus other human
protein kinases. A structure model providing an informative basis
for the Pf GSK-3/HsGSK-3 selectivity of 5 was generated by
docking studies based on a Pf GSK-3 homology model, taking
into account the X-ray structure of the related molecule 7d in
complex with human GSK-3. Selected congeners 5 showed in
vitro antiplasmodial activity against erythrocyte stages of P.
falciparum parasites in single-digit micromolar concentrations.
The results warrant further development of the structure class
toward antimalarial drugs.
EXPERIMENTAL PROCEDURES
■
Synthetic Chemistry. Starting materials were purchased from the
suppliers indicated below and were used without further purification,
unless stated otherwise: aromatic aldehydes 9, cyanothioacetamide
(11), phenacyl bromides 21, 2-chloro-N-(4-chlorophenyl)acetamide
(22), and chloroacetonitrile (23) were from Acros Organics, Geel,
Belgium, and 2-fluorobenzaldehyde and malonodinitrile (10) were from
Sigma-Aldrich, Steinheim, Germany. Prior to use, liquid aromatic
aldehydes 9 (unsubstituted benzaldehyde, 2-chlorobenzaldehyde, 2-
bromobenzaldehyde, 2-ethoxybenzaldehyde, 2-methylbenzaldehyde,
thiophene-2-carbaldehyde) were purified by extraction with aqueous
saturated sodium carbonate solution to remove acidic impurities.
Melting points (mp's) were determined on an electric variable heater
(Electrothermal IA 9100) in open glass capillaries. IR spectra were
recorded as KBr disks on a Thermo Nicolet FT-IR 200 (remark
regarding IR spectra: being part of a vinylogic amide element which may
be involved in H-bonding, the carbonyl CO valence vibration band of
compounds 5, 6, and 8 is strongly shifted to low wavenumbers and
Figure 5. Protein kinase selectivity of compound 5v tested in a 402
human kinase interaction assay (KinomeScan). Compound 5v was
tested at a 10 μM final concentration in the kinase interaction panel. A
semiquantitative scoring of this primary screen was estimated. This
score relates to the probability of a hit rather than strict affinity. Scores of
>10, between 1 and 10, and <1 indicate the probability of being a false
positive, which is <20%, <10%, and <5%, respectively. The results are
presented as a TREEspot kinase interaction map (top). Raw data (all
402 values) are provided in the Supporting Information, Table S4.
1
overlaps in most cases the aromatic CC vibration band). H NMR
spectra and 13C NMR spectra were recorded on Bruker Avance DRX-
400 and Bruker Avance II-600 instruments (NMR Laboratories of the
Chemical Institutes of the Technische Universitat Braunschweig) with
̈
Table 4. Antiplasmodial Activity of Selected PfGSK-3
an internal standard of tetramethylsilane, and signals are given in parts
per million (δ scale). Elemental analyses were determined on a CE
Instruments FlashEA 1112 elemental analyzer (Thermo Quest). Mass
spectra were recorded on a Finnigan-MAT 90 double-focused sector
field mass spectrometer. Accurate measurements were conducted
according to the peakmatch method using perfluorokerosene (PFK)
as an internal mass reference. EI-MS: ionization energy 70 eV
(Department of Mass Spectrometry of the Chemical Institutes of the
Inhibitors
a
b
compd
inhibition at 15 μM (%)
EC50 (μM)
nt
5m
5s
65.4 ( 2.0)
77.1 ( 4.6)
100.0 ( 0.0)
100.0 ( 0.0)
63.5 ( 4.3)
100.0 ( 0.0)
30.3 ( 3.6)
100.0 ( 0.0)
0.0 ( 9.6)
nt
5v
5.5 (4.2−7.2)
3.7 (2.7−5.0)
5w
5x
nt
Technische Universitat Braunschweig). TLC: Polygram Sil G/UV
,
̈
254
5y
5.6 (4.3−7.2)
Macherey-Nagel, 40 × 80 mm, visualization by UV illumination (254
and 366 nm). Column chromatography: silica gel 60 (Merck), column
width 2 cm, column height 10 cm unless stated otherwise. Purity was
determined by HPLC using the following devices and settings: Elite
LaChrom (Merck/Hitachi), pump L-2130, autosampler L-2200, diode
array detector (DAD) L-2450, organizer box L-2000, column Merck
LiChroCART 125-4, LiChrosphere 100, RP 18, 5 μm, flow rate 1.000
mL/min, isocratic, volume of injection 10 μL, detection (DAD) at 254
and 280 nm, AUC-%-method; time of detection 15 min, net retention
time (tN), dead time (tm) related to DMSO. Preparation of H2O +
(Et3NH)2SO4 buffer (pH 2.5) for HPLC: Triethylamine (20.0 mL) and
sodium hydroxide (242 mg) were dissolved in water (980 mL). The
solution was adjusted to pH 2.5 by addition of sulfuric acid. Absorption
maxima (λmax) were extracted from the spectra recorded by the DAD in
the HPLC peak maxima (software EZ Chrom Elite Client/server
version 3.1.3). All compounds employed in biological tests were used in
≥95% purity. Synthetic procedures and structure characterization data
for the following compounds are available in the Supporting
Information: 12−16, 18−20, 5a−5u, 5w−5cc, 6a−6c, 7a−7h, 8a, 8b.
General Procedure A for the Preparation of 6-Amino-4-aryl-2-
thioxo-1,2-dihydropyridine-3,5-dicarbonitriles 12−20. A suitable
aromatic aldehyde, 9 (1.00 mmol), malonodinitrile (10; 1.00 mmol),
and cyanothioacetamide (11; 1.00 mmol) were dissolved in EtOH (4
5z
nt
5aa
7.5 (4.6−12.3)
control (DMSO)
nt
a
Luminescence assay (luciferase) after 72 h of incubation of P.
falciparum NF54LUC parasites with compounds at 15 μM. SEM
values are given in parentheses. EC50 values were determined for
compounds showing inhibition rates above 80% and were calculated
from dose−response curves. Single data points were measured after 48
h of incubation with test compounds. Confidence intervals are given in
parentheses. nt = not tested.
b
position of the benzoyl ring appear favorable for antiplasmodial
activity.
CONCLUSION
■
We have identified and developed a novel class of selective
inhibitors of Pf GSK-3, an enzyme that was suggested as a suitable
target for antimalarial drugs. Initial hits were identified in an HTS
campaign from a commercial compound library. Representatives
of the new inhibitor family 5 share the 3,6-diamino-4-(2-
H
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
mL). Piperidine (50 μL) was added, and the mixture was refluxed for 3 h.
After evaporation of the solvent in vacuo, glacial acetic acid (six drops),
water (10 mL), and CH2Cl2 (2 mL) were added, and the mixture was
shaken for 1 min. A precipitate was formed either immediately or after
storage overnight at 2−5 °C. The precipitate was filtered off with suction
and used for the following synthetic steps without further purification.
6-Amino-4-(2-chlorophenyl)-2-thioxo-1,2-dihydropyridine-3,5-di-
carbonitrile (17). Preparation according to general procedure A from 2-
chlorobenzaldehyde, 10, and 11 yielded a yellow powder (57%). Mp:
177−180 °C (lit.56 180−182 °C).
5 mL of LB medium (tryptone, 10 g/L; yeast extract, 5 g/L; NaCl, 10 g/
L; pH 7.5) with ampicillin (100 μg/mL) and incubated at 37 °C
overnight. The culture was then transferred into 250 mL of LB medium
with ampicillin (100 μg/mL) in a 1 L bottle and cultured at 37 °C under
constant shaking (250 rpm) to an OD600 of ∼0.5. Expression was
induced with 0.5 mL of 20% arabinose (0.02% final concentration) for 4
h at 37 °C under constant shaking.
Recombinant E. coli BL21-(DE3)-pRARE2 containing expression
plasmid pET15-MHL CpGSK3 was streaked on an LB broth plate in the
presence of ampicillin/chloramphenicol (100 and 34 μg/mL). A single
colony was inoculated into 5 mL of LB medium with ampicillin/
chloramphenicol (100 and 34 μg/mL, respectively) and incubated at 37
°C overnight. Then the culture was transferred into 250 mL of LB
medium with ampicillin/chloramphenicol (100 and 34 μg/mL,
respectively) in a 1 L bottle, cultured at 37 °C under shaking (250
rpm) to an OD600 of ∼0.5, cooled to 15 °C, and induced with 0.5 mM
isopropyl 1-thio-D-galactopyranoside overnight at 15 °C under constant
shaking.
General Procedure B for the Synthesis of 3,6-Diamino-2-aroyl-4-
arylthieno[2,3-b]pyridine-5-carbonitriles 5, 3,6-Diamino-4-aryl-N-
(4-chlorophenyl)-5-cyanothieno[2,3-b]pyridine-2-carboxamides 6,
3,6-Diamino-4-arylthieno[2,3-b]pyridine-2,5-dicarbonitriles 7, and
(3-Amino-4-aryl-5,6,7,8-tetrahydrothieno[2,3-b]quinolin-2-yl)(aryl)-
methanones 8. A suitable precursor compound, 12−20 or 26 (0.400
mmol), was dissolved in DMF (0.5 mL). After addition of 10% aqueous
KOH (224 μL), the mixture was stirred for 1 min. A suitable haloalkane,
21, 22, or 23 (0.400 mmol), was added, and stirring was continued for
30 min at room temperature. After addition of a second portion of 10%
aqueous KOH (224 μL), stirring was continued at room temperature.
The reaction was monitored by TLC. Completion of the reaction was
detected after the reaction time specified in the distinct procedure
descriptions. Upon addition of water (5 mL), a precipitate formed which
was filtered off with suction and washed successively with water, EtOH,
and petroleum ether. Purification was accomplished by column
chromatography or crystallization from the indicated solvents.
3,6-Diamino-2-(3-chlorobenzoyl)-4-(2-chlorophenyl)thieno[2,3-
b]pyridine-5-carbonitrile (5v). Preparation was according to general
procedure B from 17 and 2-bromo-1-(3-chlorophenyl)ethanone,
reaction time 3.5 h. Crystallization from EtOH yielded an orange
powder (37%). Mp: 289−290 °C. IR (KBr): 3477 and 3362 cm−1 (NH),
2216 cm−1 (CN). 1H NMR (DMSO-d6, 600 MHz): δ (ppm) = 6.69
(br s, 2 H, NH2), 7.54−7.56 (m, 1 H, ArH), 7.62−7.64 (m, 1 H, ArH),
7.65−7.67 (m, 4 H, ArH), 7.69−7.72 (m, 1 H, ArH), 7.76−7.91 (m, 3 H,
NH2 and ArH overlapping). 13C NMR (DMSO-d6, 150.9 MHz): δ
(ppm) = 125.7, 126.9, 128.6, 130.0, 130.4, 130.6, 130.8, 132.5 (tertiary
C); 90.9, 99.6, 112.4, 114.8, 131.1, 131.8, 133.4, 142.7, 150.8, 151.2,
159.3, 166.4, 185.9 (quaternary C). Anal. (C21H12Cl2N4OS) C, H, N, S.
HPLC: 98.0% at 254 nm and 98.0% at 280 nm, tN = 5.39 min, tM = 1.03
min (ACN/H2O, 55:45), λmax = 328 nm.
Cultures were harvested by centrifugation at 9000 rpm for 10 min.
Bacterial pellets were stored at −80 °C.
Extraction and Purification of Recombinant (His)6-Tagged
Pf GSK3, PkGSK3, TgGSK3, and CpGSK3 and Purification of
Mammalian GSK-3. Extraction. Pellets from 250 mL E. coli cultures
were resuspended to approximately 10 mL in lysis buffer (50 mM
HEPES, pH 7.5, 500 mM NaCl, 5 mM imidazole, and 5% glycerol) with
the addition of protease inhibitors (Complete EDTA-free inhibitor
cocktail from Roche). Resuspended pellets stored at −80 °C were
thawed overnight at 4 °C the day before purification. Prior to lysis, each
pellet from a 250 mL culture was pretreated with lysozyme, 300 μg/mL
final (or 0.5% CHAPS), and 500 units of DNase I for 30 min at room
temperature. Cells were disrupted by potterization and sonication, and
the cell lysate was centrifuged at 24 000 rpm for 30 min at 4 °C.
Purification. Recombinant Pf GSK3, PkGSK3, TgGSK3, and
CpGSK3 proteins were purified by immobilized metal affinity
chromatography by a batch procedure on 500 μL (for 10 mL of lysate)
of nickel affinity resin (Ni−NTA (nitrilotriacetic acid) agarose beads,
Qiagen). The beads were washed four times with 10 mL of washing
buffer (50 mM HEPES, pH 7.5, 500 mM NaCl, 30 mM imidazole, 5%
glycerol). The lysates were incubated with the beads for 2 h at 4 °C on a
rotating wheel. After centrifugation (1500 rpm, 2 min at 4 °C) the beads
were washed four times with 1 volume of wash buffer. Proteins were
then eluted successively three times with respectively 500 μL (first
elution), 300 μL (second elution), and 200 μL (third elution) of elution
buffer (50 mM HEPES, pH 7.5, 500 mM NaCl, 250 mM imidazole, 5%
glycerol). The protein identity and purity were evaluated by SDS−
PAGE analysis. Samples were stored at −80 °C.
Kinase Expression and Activity Assays. GSK-3 Orthologue
Clones. The PkGSK3 (P. knowlesi GSK3) gene was synthezised by
GenCust and cloned into the expression vector plasmid pBAD-Thio-
TOPO (Invitrogen).
The TgGSK3 (To. gondii GSK3) gene was generated by PCR from To.
gondii cDNA using a 5′-primer, ATGCCGGACCCGCAGTACGA-
TCC (F-Toxog), and a 3′-primer, GCCACGGTTGTTTGCACTGG-
Mammalian GSK-3. GSK-3α/β was purified from porcine brain by
affinity chromatography on immobilized axin.57
CG (R-Toxog), and provided by Maryse Lebrun (Jean-Franco̧ is
Kinase Assays. The activity of the recombinant kinases was measured
by incubating the proteins with 40 μM GS-1 peptide substrate
(YRRAAVPPSPSLSRHSSPHQpSEDEEE, where pS stands for phos-
phorylated serine; Proteogenix, Oberhausbergen, France) in buffer A,
pH 7.5 (10 mM MgCl2, 1 mM EGTA, 1 mM DTT, 25 mM Tris/HCl,
pH 7.5, 50 μg/mL heparin, 0.15 mg/mL BSA), in the presence of 15 μM
γ-33P-ATP in a final volume of 30 μL. After 30 min of incubation at 30 °C
except for DdGSK3 (30 min at room temperature), 25 μL of the reaction
was spotted onto Whatman P81 phosphocellulose paper. The filters
were washed five times (for at least 5 min each time) in a 1% phosphoric
acid solution. The wet filters were counted in the presence of 1 mL of
ACS (Amersham) scintillation fluid. Blank values were subtracted and
activities calculated as picomoles of phosphate incorporated during a 30
min incubation. The activities are usually expressed as a percentage of
the maximal activity, i.e., in the absence of inhibitors. Controls were
performed with appropriate dilutions of dimethyl sulfoxide. For
molecules showing inhibitory activity at 10 μM, dose−response curves
were performed to calculate the IC50 value.
Dubremetz laboratory, Montpellier, France). The gene was cloned
into the expression vector pBAD-Thio-TOPO (Invitrogen) by direct
insertion of the Taq polymerase amplified PCR product.
The CpGSK3 (C. parvum GSK3) gene, kindly provided by Dr.
Raymond Hui (Structural Genomics Consortium, Toronto, Canada),
was cloned into the expression vector pET15-MHL.
GSK-3 orthologues from T. cruzi and T. brucei were kindly tested by
Dr. Wesley C. Van Voohis (Department of Medicine, University of
Washington, Seattle, WA).
GSK-3 orthologues from L. major and L. donovani were kindly tested
by Dr. Evangelia Xingi (Pasteur Institute, Athens, Greece).
D. discoideum GSK-3 was kindly provided by Dr. Adrian J. Harwood
and Dr. W. Jonathan Ryves (Cardiff School of Biosciences, Cardiff
University, Cardiff, U.K.).
Nt-GST-tagged human GSK-3α and GSK-3β were obtained from
BPS Bioscience.
Expression of Recombinant (His)6-Tagged Pf GSK3, PkGSK3,
TgGSK3, and CpGSK3. Recombinant proteins were expressed in
Escherichia coli as previously described.27 Recombinant E. coli TOP10
containing expression plasmids pBAD-Thio-TOPO PfGSK3, PkGSK3,
and TgGSK3 were streaked on an LB (Luria−Bertani) broth plate in the
presence of ampicillin (100 μg/mL). A single colony was inoculated into
Compound Libraries. The compound library for high-throughput
screening was obtained through the WHO. It comprised compounds
from ChemDiv (5000 kinase inhibitory scaffolds plus a collection of
5000 random compounds) and Biospecs (480 products). The focused
I
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
library was supplied by ChemDiv. It comprised 427 compounds
containing the thieno[2,3-b]pyridine parent scaffold.
ASSOCIATED CONTENT
* Supporting Information
■
S
Crystallography and Molecular Modeling. GSK-3 Crystal
Structures. The human (pTyr216)-GSK3β−axin(383−401) complex
was prepared as previously described.47 Samples of the (pTyr216)-
GSK3β−axin(383−401) complex [6 mg/mL GSK3b + 0.37 mg/mL
axin(383−401)] were incubated on ice with 500 μM 7d for 1 h prior to
crystallization.
Crystals of the (pTyr216)-GSK3β−axin(383−401)−7d complex
were obtained within 6 weeks by the hanging drop vapor diffusion
method mixing 1 μL of complex with 1 μL of reservoir solution
containing 0.1 M 2-(N-morpholino)ethanesulfonic acid (MES), pH 6.5,
and 12% (w/v) PEG 20 000 at 20 °C. Crystals belong to the
orthorhombic space group P212121 with cell dimensions of a = 73.15 Å, b
= 76.16 Å, and c = 86.54 Å and one copy of the (pTyr216)-GSK3β−
axin(383−401)−7d complex per crystallographic asymmetric unit.
X-ray data were collected at beamline I02 at the Diamond Light
Source (Oxfordshire, U.K.) using an ADSC Quantum Q315r detector.
Processing was performed using the program XDS.58
The structure was solved by molecular replacement with the program
PHASER59 employing the (pTyr216)-GSK3β−axin(383−401) com-
plex47 (PDB code 1O9U) as a search model. Bound ligand in the active
site was identified by the difference Fourier method. Structure
refinement and model building were performed using iterative cycles
of phenix.refine60 and Coot.61 Crystallographic statistics are given in
Table 2. The coordinates were deposited in the RCSB Protein Data
Bank62 with accession number 3ZDI. 2Fo − Fc composite annealed omit
electron density maps were calculated with CNS.63 Figures were
prepared with PyMOL (www.pymol.org).
Docking. Docking studies were accomplished by the FlexX module of
LeadIT 2.0.2 (BiosolveIT GmbH, St. Augustin, Germany, 2011).
Receptors were prepared by the standard setup routine using the 1J1B
PDB file49 for the human enzyme and a corresponding PfGSK-3
model.48 Basic and acidic binding site side chains were protonated or
deprotonated, respectively. Two water molecules observed in the X-ray
structure of 7d were merged into the binding sites and defined as freely
rotatable and displaceable during the docking procedure. Ligands were
constructed with SYBYL X1.3 and minimized using AM1 charges and
the Tripos force field. The single interaction scan (SIS) mode of
FlexX64−67 was used for base fragment placement, and the obtained
poses were rescored by the new Hyde scoring function.68
in Vitro Antimalarial Activity Assay. Erythrocytic stages of P.
falciparum parasites stably expressing the luciferase gene by the hrp2
promoter from a chromosomal locus (NF54:LUC) were used. These
parasites constitutively express high levels of luciferase. Cultures of 20
mL total volume were grown in 75 mL tissue culture flasks (Costar
brand, NUNC, Denmark) as previously described.69 Upon reaching 3%
parasitemia, the medium was aspirated and 100 μL of 25% hematocrit
cultured parasites was transferred with a multichannel pipet to a 96-well
sterile plate (100 μL in each well). Drugs diluted in complete RPMI-
1640 containing 1% DMSO (10 μM, 100 μL/well) were added to each
well. Each drug was tested in triplicate. As controls three wells of
(NF54:LUC) in regular medium with no drug and another three wells in
which (NF54:LUC) is exposed only to the 1% DMSO were inoculated.
Three additional wells were inoculated with NF54 “wild-type” parasites
that do not express luciferase and were used as “blanks”. The plate was
incubated in a sealed culture chamber and incubated for 48 h (37 °C,
90% N2, 5% CO2, and 5% O2). After 48 h, 100 μL of the medium was
removed and the red blood cells were lysed with the addition of lysis
buffer of the Bright-Glo luciferase assay system (100 μL/well, Promega,
MT) and incubation at room temperature for 5 min. A 100 μL volume
was transferred into an opaque 96-well flat-bottom plate (Costar brand,
NUNC, Denmark) for the luciferase assay, and the rest was frozen at
−20 °C. A 100 μL volume of the Bright-Glo substrate was added to each
well, and the luminescence was measured after 5 min using a microplate
reader (Fluoroskan Ascent, Thermo Labsystems, Finland). After
screening the activity of the compounds at 15 μM, titrations were
carried out to determine the IC50 of the most active compounds as
described above.
Pf GSK-3 and GSK-3α/β sensitivity to a panel of kinase
inhibitors (Biomol, Supplementary Table S1), accession codes
of GSK-3 orthologues used in this study (Supplementary Table
S2), results of kinase inhibition by 5v in the University of Dundee
kinase selectivity panel (77 kinases, Supplementary Table S3),
results of kinase inhibition by 5v in the DiscoveRx kinase
selectivity panel (402 kinases, Supplementary Table S4), details
for the synthesis and spectroscopic data of compounds 12−16,
18−20, 5a−5u, 5w−5cc, 6a−6c, 7a−7h, 8a, and 8b, and HPLC
purity data of all compounds employed in biological tests. This
material is available free of charge via the Internet at http://pubs.
acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: meijer@manros-therapeutics.com (L.M.); c.kunick@
tu-braunschweig.de (C.K.). Phone: +33-608605834 (L.M.);
+49-5313912754 (C.K.). Fax: +33-298725528 (L.M.); +49-
5313912799 (C.K.).
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was funded in part by the Commission of the
European Communities (Contracts LSHB-CT-2004-503467
and Health-F3-2008-223414 to L.M. and C.K., FP7 HEALTH-
2007-2.3.4-1 LEISHDRUG to L.M., and HEALTH, LSH-2003-
2.3.0_2 ANTIMAL to L.M.). This joint project was financially
supported by the State of Lower-Saxony, Hannover, Germany
(to R.D., N.P., and C.K). S.K. was supported by a Ph.D. grant
from the University of Hamburg, Germany. We are thankful to
Evangelia Xingi for performing the Leishmania GSK-3 assays, to
Wesley C. Van Voohis for assaying the Trypanosoma GSK-3, to
Maryse Lebrun and Jean-Franco̧ is Dubremetz for providing the
TgGSK-3 cDNA, to Adrian J. Harwood and W. Jonathan Ryves
for providing DdGSK-3, and to Raymond Hui for the CpGSK-3
cDNA.
ABBREVIATIONS USED
■
ACN, acetonitrile; DAD, diode array detector; DFG, aspartate−
phenylalanine−glycine sequence; EGTA, ethylene glycol tetra-
acetic acid; EWG, electron-withdrawing group; gk, gatekeeper;
GSK-3, glycogen synthase kinase-3; HEPES, N-(2-
hydroxyethyl)piperazine-N′-2-ethanesulfonic acid; LB, Luria−
Bertani; MES, 2-(N-morpholino)ethanesulfonic acid;
Pf CDPK1, Plasmodium falciparum calcium-dependent protein
kinase 1; PFK, perfluorokerosene; Pfmrk, Plasmodium falciparum
MO15-related protein kinase; Pfnek-1, Plasmodium falciparum
NIMA-related protein kinase; PfPK7, Plasmodium falciparum
protein kinase 7; RPMI, Roswell Park Memorial Institute
REFERENCES
■
(1) WHO World Malaria Report 2010. http://www.who.int/malaria/
world_malaria_report_2010/worldmalariareport2010.pdf (accessed
June 8, 2011).
(2) Schlitzer, M. Malaria chemotherapeutics part I: History of
antimalarial drug development, currently used therapeutics, and drugs
in clinical development. ChemMedChem 2007, 2, 944−986.
(3) Schlitzer, M. Antimalarial drugsWhat is in use and what is in the
pipeline. Arch. Pharm. Chem. Life Sci. 2008, 341, 149−163.
J
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(4) Fidock, D. A. Priming the antimalaria pipeline. Nature 2010, 465,
297−298.
(25) Caridha, D.; Kathcart, A. K.; Jirage, D.; Waters, N. C. Activity of
substituted thiophene sulfonamides against malarial and mammalian
cyclin dependent protein kinases. Bioorg. Med. Chem. Lett. 2010, 20,
3863−3867.
(26) Lemercier, G.; Fernandez-Montalvan, A.; Shaw, J. P.; Kugelstadt,
D.; Bomke, J.; Domostoj, M.; Schwarz, M. K.; Scheer, A.; Kappes, B.;
Leroy, D. Identification and characterization of novel small molecules as
potent inhibitors of the plasmodial calcium-dependent protein kinase.
Biochemistry 2009, 48, 6379−6389.
(27) Droucheau, E.; Primot, A.; Thomas, V.; Mattei, D.; Knockaert,
M.; Richardson, C.; Sallicandro, P.; Alano, P.; Jafarshad, A.; Baratte, B.;
Kunick, C.; Parzy, D.; Pearl, L.; Doerig, C.; Meijer, L. Plasmodium
falciparum glycogen synthase kinase-3: Molecular model, expression,
intracellular localisation and selective inhibitors. Biochim. Biophys. Acta
2004, 1697, 181−196.
(28) Solyakov, L.; Halbert, J.; Alam, M. M.; Semblat, J. P.; Dorin-
Semblat, D.; Reininger, L.; Bottrill, A. R.; Mistry, S.; Abdi, A.; Fennell,
C.; Holland, Z.; Demarta, C.; Bouza, Y.; Sicard, A.; Nivez, M. P.;
Eschenlauer, S.; Lama, T.; Thomas, D. C.; Sharma, P.; Agarwal, S.; Kern,
S.; Pradel, G.; Graciotti, M.; Tobin, A. B.; Doerig, C. Global kinomic and
phospho-proteomic analyses of the human malaria parasite Plasmodium
falciparum. Nat. Commun. 2011, 2, 565.
(5) Gamo, F. J.; Sanz, L. M.; Vidal, J.; de Cozar, C.; Alvarez, E.;
Lavandera, J. L.; Vanderwall, D. E.; Green, D. V.; Kumar, V.; Hasan, S.;
Brown, J. R.; Peishoff, C. E.; Cardon, L. R.; Garcia-Bustos, J. F.
Thousands of chemical starting points for antimalarial lead
identification. Nature 2010, 465, 305−310.
(6) Schlitzer, M.; Ortmann, R. Feeding the antimalarial pipeline.
ChemMedChem 2010, 5, 1837−1840.
(7) Canduri, F.; Perez, P. C.; Caceres, R. A.; de Azevedo, W. F. J.
Protein kinases as targets for antiparasitic chemotherapy drugs. Curr.
Drug Targets 2007, 8, 389−398.
(8) Doerig, C. Protein kinases as targets for anti-parasitic chemo-
therapy. Biochim. Biophys. Acta 2004, 1697, 155−168.
(9) Doerig, C.; Meijer, L. Antimalarial drug discovery: Targeting
protein kinases. Expert Opin. Ther. Targets 2007, 11, 279−290.
(10) Doerig, C.; Meijer, L.; Mottram, J. C. Protein kinases as drug
targets in parasitic protozoa. Trends Parasitol. 2002, 18, 366−371.
́
́
(11) Lawton, P.; Sarciron, M.-E.; Petavy, A.-F. Chemotherapeutic
targets for antiparasitic therapy. Drugs Future 2007, 31, 793−809.
(12) Jirage, D.; Keenan, S. M.; Waters, N. C. Exploring novel targets for
antimalarial drug discovery: Plasmodial protein kinases. Infect. Disord.:
Drug Targets 2010, 10, 134−146.
(29) Zakaria, N. A.; Embi, N.; Sidek, H. M. Suppression of Plasmodium
berghei parasitemia by LiCl in an animal infection model. Trop. Biomed.
2010, 27, 624−631.
(13) Doerig, C.; Billker, O.; Haystead, T.; Sharma, P.; Tobin, A. B.;
Waters, N. C. Protein kinases of malaria parasites: An update. Trends
Parasitol. 2008, 24, 570−577.
(14) Ward, P.; Equinet, L.; Packer, J.; Doerig, C. Protein kinases of the
human malaria parasite Plasmodium falciparum: The kinome of a
divergent eukaryote. BMC Genomics 2004, 5, 19.
(15) Doerig, C.; Abdi, A.; Bland, N.; Eschenlauer, S.; Dorin-Semblat,
D.; Fennell, C.; Halbert, J.; Holland, Z.; Nivez, M.-P.; Semblat, J.-P.;
Sicard, A.; Reininger, L. Malaria: Targeting parasite and host cell
kinomes. Biochim. Biophys. Acta 2010, 1804, 604−612.
(16) Dorin, D.; Le Roch, K.; Sallicandro, P.; Alano, P.; Parzy, D.;
Poullet, P.; Meijer, L.; Doerig, C. Pfnek-1, a NIMA-related kinase from
the human malaria parasite Plasmodium falciparum. Eur. J. Biochem.
2001, 268, 2600−2608.
(17) Laurent, D.; Jullian, V.; Parenty, A.; Knibiehler, M.; Dorin, D.;
Schmitt, S.; Lozach, O.; Lebouvier, N.; Frostin, M.; Alby, F.; Maurel, S.;
Doerig, C.; Meijer, L.; Sauvain, M. Antimalarial potential of
xestoquinone, a protein kinase inhibitor from a Vanuatu marine sponge
Xestospongia sp. Bioorg. Med. Chem. 2006, 14, 4477−4482.
(18) Bouloc, N.; Large, J. M.; Smiljanic, E.; Whalley, D.; Ansell, K. H.;
Edlin, C. D.; Bryans, J. S. Synthesis and in vitro evaluation of
imidazopyridazines as novel inhibitors of the malarial kinase PfPK7.
Bioorg. Med. Chem. Lett. 2008, 18, 5294−5298.
(19) Merckx, A.; Echalier, A.; Langford, K.; Sicard, A.; Langsley, G.
Structures of P. falciparum protein kinase 7 identify an activation motif
and leads for inhibitor design. Structure 2008, 16, 228−238.
(20) Geyer, J. A.; Keenan, S. M.; Woodard, C. L.; Thompson, P. A.;
Gerena, L.; Nichols, D. A.; Gutteridge, C. E.; Waters, N. C. Selective
inhibition of Pfmrk, a Plasmodium falciparum CDK by antimalarial 1,3-
diaryl-2-propenones. Bioorg. Med. Chem. Lett. 2009, 19, 1982−1985.
(21) Woodard, C. L.; Li, Z.; Kathcart, A. K.; Terrell, J.; Gerena, L.;
Lopez-Sanchez, M.; Kyle, D. E.; Bhattacharjee, A. K.; Nichols, D. A.;
Ellis, W.; Prigge, S. T.; Geyer, J. A.; Waters, N. C. Oxindole-based
compounds are selective inhibitors of Plasmodium falciparum cyclin
dependent protein kinases. J. Med. Chem. 2003, 46, 3877−3882.
(22) Xiao, Z.; Waters, N. C.; Woodard, C. L.; Li, Z.; Li, P.-K. Design
and synthesis of Pfmrk inhibitors as potential antimalarial agents. Bioorg.
Med. Chem. Lett. 2001, 11, 2875−2878.
(30) Oduor, R. O.; Ojo, K. K.; Williams, G. P.; Bertelli, F.; Mills, J.;
Maes, L.; Pryde, D. C.; Parkinson, T.; Van Voorhis, W. C.; Holler, T. P.
Trypanosoma brucei glycogen synthase kinase-3, a target for anti-
trypanosomal drug development: A public-private partnership to
identify novel leads. PLoS Neglected Trop. Dis. 2011, 5, e1017.
(31) Pies, T.; Schaper, K.-J.; Leost, M.; Zaharevitz, D. W.; Gussio, R.;
Meijer, L.; Kunick, C. CDK1-inhibitory activity of paullones depends on
electronic properties of 9-substituents. Arch. Pharm. (Weinheim, Ger.)
2004, 337, 486−492.
(32) Wieking, K.; Knockaert, M.; Leost, M.; Zaharevitz, D. W.; Meijer,
L.; Kunick, C. Synthesis of paullones with aminoalkyl side chains. Arch.
Pharm. (Weinheim, Ger.) 2002, 335, 311−317.
(33) Xie, X.; Lemcke, T.; Gussio, R.; Zaharevitz, D. W.; Leost, M.;
Meijer, L.; Kunick, C. Epoxide-containing side chains enhance
antiproliferative activity of paullones. Eur. J. Med. Chem. 2005, 40,
655−661.
(34) Kruggel, S.; Lemcke, T. Comparative investigation of the ATP-
binding site of human and plasmodial glycogen synthase kinase-3. QSAR
Comb. Sci. 2009, 28, 885−890.
(35) Osolodkin, D. I.; Zakharevich, N. V.; Palyulin, V. A.; Danilenko, V.
N.; Zefirov, N. S. Bioinformatic analysis of glycogen synthase kinase 3:
Human versus parasite kinases. Parasitology 2011, 138, 725−735.
(36) Chen, C.; Mackey, A. J.; Stoeckert, J., C. J.; Roos, D. S.
OrthoMCL-DB: Querying a comprehensive multi-species collection of
ortholog groups. Nucleic Acids Res. 2006, 34 (Suppl. 1), D363−368.
(37) Thompson, J. D.; Higgins, D. G.; Gibson, T. J. CLUSTAL W:
Improving the sensitivity of progressive multiple sequence alignment
through sequence weighting, position-specific gap penalties and weight
matrix choice. Nucleic Acids Res. 1994, 22, 4673−4680.
(38) Guindon, S.; Dufayard, J. F.; Lefort, V.; Anisimova, M.; Hordijk,
W.; Gascuel, O. New algorithms and methods to estimate maximum-
likelihood phylogenies: Assessing the performance of PhyML 3.0. Syst.
Biol. 2010, 59, 307−321.
(39) Guindon, S.; Gascuel, O. A simple, fast and accurate algorithm to
estimate large phylogenies by maximum likelihood. Syst. Biol. 2003, 52,
696−704.
(40) Le, S. Q.; Gascuel, O. An improved general amino acid
replacement matrix. Mol. Biol. Evol. 2008, 25, 1307−1320.
(41) Sharanin, Y. A.; Krivokolysko, S. G.; Dyachenko, V. D. Cyclization
reactions of nitriles. LIV. Synthesis and properties of 6-amino-4-aryl-3,5-
dicyanopyridin-2(1H)-ones, the corresponding thiones, the pyridylide-
nemalononitriles and their hydrogenated analogs. Russ. J. Org. Chem.
1994, 30, 620−626.
(23) Keenan, S. M.; Geyer, J. A.; Welsh, W. J.; Prigge, S. T.; Waters, N.
C. Rational inhibitor design and iterative screening in the identification
of selective plasmodial cyclin dependent kinase inhibitors. Comb. Chem.
High Throughput Screening 2005, 8, 27−38.
(24) Woodard, C. L.; Keenan, S. M.; Gerena, L.; Welsh, W. J.; Geyer, J.
A.; Waters, N. C. Evaluation of broad spectrum protein kinase inhibitors
to probe the architecture of the malarial cyclin dependent protein kinase
Pfmrk. Bioorg. Med. Chem. Lett. 2007, 17, 4961−4966.
K
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(42) Sharanin, Y. A.; Shestopalov, A. M.; Litvinov, V. P.; Klokol, G. V.;
Mortikov, V. Y.; Demerkov, A. S. Cyclization reactions of nitriles. XV.
Synthesis and reactions of chalcogen-containing 6-amino-3,5-dicyano-
pyridines. Russ. J. Org. Chem. 1988, 24, 771−776.
(43) Granik, V. G.; Kadushkin, A. V.; Liebscher, J. Synthesis of amino
derivatives of five-membered heterocycles by Thorpe-Ziegler cycliza-
tion. Adv. Heterocycl. Chem. 1998, 72, 79−125.
(44) Litvinov, V.; Dotsenko, V.; Krivokolysko, S. The chemistry of
thienopyridines. Adv. Heterocycl. Chem. 2007, 93, 117−178.
(45) Litvinov, V. P.; Dotsenko, V. V.; Krivokolysko, S. G.
Thienopyridines: Synthesis, properties, and biological activity. Russ.
Chem. Bull., Int. Ed. 2005, 54, 864−904.
(46) Elnagdi, M. H.; Abdelrazek, F. M.; Ibrahim, N. S.; Erian, A. W.
Studies on alkyl heteroaromatic compounds. The reactivity of alkyl
polyfunctionally substituted azines towards electrophilic reagents.
Tetrahedron 1989, 45, 3597−3604.
(59) McCoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.; Winn, M.
D.; Storoni, L. C.; Read, R. J. Phaser crystallographic software. J. Appl.
Crystallogr. 2007, 40, 658−674.
(60) Adams, P. D.; Afonine, P. V.; Bunkoczi, G.; Chen, V. B.; Davis, I.
W.; Echols, N.; Headd, J. J.; Hung, L. W.; Kapral, G. J.; Grosse-
Kunstleve, R. W.; McCoy, A. J.; Moriarty, N. W.; Oeffner, R.; Read, R. J.;
Richardson, D. C.; Richardson, J. S.; Terwilligere, T. C.; Zwarta, P. H.
PHENIX: A comprehensive Python-based system for macromolecular
structure solution. Acta Crystallogr. 2010, D66, 213−221.
(61) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K. Features and
development of Coot. Acta Crystallogr. 2010, D66, 486−501.
(62) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.;
Weissig, H.; Shindyalov, I. N.; Bourne, P. E. The Protein Data Bank.
Nucleic Acids Res. 2000, 28, 235−242.
(63) Brunger, A. T. Version 1.2 of the Crystallography and NMR
system. Nat. Protoc. 2007, 2, 2728−2733.
(64) Rarey, M.; Kramer, B.; Lengauer, T. Multiple automatic base
selection: Protein-ligand docking based on incremental construction
without manual intervention. J. Comput.-Aided Mol. Des. 1997, 11, 369−
384.
(65) Rarey, M.; Kramer, B.; Lengauer, T. The particle concept: Placing
discrete water molecules during protein-ligand docking predictions.
Proteins 1999, 34, 17−28.
(66) Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. A fast flexible
docking method using an incremental construction algorithm. J. Mol.
Biol. 1996, 261, 470−489.
(67) Rarey, M.; Wefing, S.; Lengauer, T. Placement of medium-sized
molecular fragments into active sites of proteins. J. Comput.-Aided Mol.
Des. 1996, 10, 41−54.
(47) Dajani, R.; Fraser, E.; Roe, S. M.; Yeo, M.; Good, V. M.;
Thompson, V.; Dale, T. C.; Pearl, L. H. Structural basis for recruitment
of glycogen synthase kinase 3β to the axin-APC scaffold complex.
EMBO J. 2003, 22, 494−501.
(48) Kruggel, S.; Lemcke, T. Generation and evaluation of a homology
model of PfGSK-3. Arch. Pharm. Chem. Life Sci. 2009, 342, 327−332.
(49) Aoki, M.; Yokota, T.; Sugiura, I.; Sasaki, C.; Hasegawa, T.;
Okumura, C.; Ishiguro, K.; Kohno, T.; Sugio, S.; Matsuzaki, T.
Structural insight into nucleotide recognition in tau-protein kinase I/
glycogen synthase kinase 3 β. Acta Crystallogr. 2004, D60, 439−446.
(50) Ojo, K.; Arakaki, T.; Napuli, A.; Inampudi, K.; Keyloun, K.;
Zhang, L.; Hol, W.; Verlinde, C.; Merritt, E.; Van Voorhis, W. Structure
determination of glycogen synthase kinase-3 from Leishmania major and
comparative inhibitor structure-activity relationships with Trypanosoma
brucei GSK-3. Mol. Biochem. Parasitol. 2011, 176, 98−108.
(51) Ojo, K. K.; Gillespie, J. R.; Riechers, A. J.; Napuli, A. J.; Verlinde,
C. L.; Buckner, F. S.; Gelb, M. H.; Domostoj, M. M.; Wells, S. J.; Scheer,
A.; Wells, T. N.; Van Voorhis, W. C. Glycogen synthase kinase 3 is a
potential drug target for African trypanosomiasis therapy. Antimicrob.
Agents Chemother. 2008, 52, 3710−3717.
(68) Reulecke, I.; Lange, G.; Albrecht, J.; Klein, R.; Rarey, M. Towards
an integrated description of hydrogen bonding and dehydration:
Decreasing false positives in virtual screening with the HYDE scoring
function. ChemMedChem 2008, 3, 885−897.
(69) Dzikowski, R.; Frank, M.; Deitsch, K. Mutually exclusive
expression of virulence genes by malaria parasites is regulated
independently of antigen production. PLoS Pathog. 2006, 2, e22.
(52) Reichwald, C.; Shimony, O.; Dunkel, U.; Sacerdoti-Sierra, N.;
Jaffe, C. L.; Kunick, C. 2-(3-Aryl-3-oxopropen-1-yl)-9-tert-butylpaul-
lones: A new antileishmanial chemotype. J. Med. Chem. 2008, 51, 659−
665.
(53) Bain, J.; Plater, L.; Elliott, M.; Shpiro, N.; Hastie, C. J.;
McLauchlan, H.; Klevernic, I.; Arthur, J. S.; Alessi, D. R.; Cohen, P. The
selectivity of protein kinase inhibitors: A further update. Biochem. J.
2007, 408, 297−315.
(54) Fabian, M. A.; Biggs, W. H.; Treiber, D. K.; Atteridge, C. E.;
Azimioara, M. D.; Benedetti, M. G.; Carter, T. A.; Ciceri, P.; Edeen, P.
T.; Floyd, M.; Ford, J. M.; Galvin, M.; Gerlach, J. L.; Grotzfeld, R. M.;
́
Herrgard, S.; Insko, D. E.; Insko, M. A.; Lai, A. G.; Lelias, J. M.; Mehta, S.
A.; Milanov, Z. V.; Velasco, A. M.; Wodicka, L. M.; Patel, H. K.;
Zarrinkar, P. P.; Lockhart, D. J. A small molecule-kinase interaction map
for clinical kinase inhibitors. Nat. Biotechnol. 2005, 23, 329−336.
(55) Karaman, M. W.; Herrgard, S.; Treiber, D. K.; Gallant, P.;
Atteridge, C. E.; Campbell, B. T.; Chan, K. W.; Ciceri, P.; Davis, M. I.;
Edeen, P. T.; Faraoni, R.; Floyd, M.; Hunt, J. P.; Lockhart, D. J.; Milanov,
Z. V.; Morrison, M. J.; Pallares, G.; Patel, H. K.; Pritchard, S.; Wodicka,
L. M.; Zarrinkar, P. P. A quantitative analysis of kinase inhibitor
selectivity. Nat. Biotechnol. 2008, 26, 127−132.
(56) Sharanin, Y. A.; Krivokolysko, S. G.; Dyachenko, V. D. Cyclization
reactions of nitriles. LIV. Synthesis and properties of 6-amino-4-aryl-3,5-
dicyanopyridin-2(1H)-ones, the corresponding thiones, the pyridylide-
nemalononitriles and their hydrogenated analogs. Russ. J. Org. Chem.
1994, 30, 620−626.
(57) Primot, A.; Baratte, B.; Gompel, M.; Borgne, A.; Liabeuf, S.;
Romette, J. L.; Costantini, F.; Meijer, L. Purification of GSK-3 by affinity
chromatography on immobilised axin. Protein Expression Purif. 2000, 20,
394−404.
(58) Kabsch, W. Software XDS for image rotation, recognition and
crystal symmetry assignment. Acta Crystallogr. 2010, D66, 125−132.
L
dx.doi.org/10.1021/jm301575n | J. Med. Chem. XXXX, XXX, XXX−XXX