Amit Mahindra1, Omar Janha2, Kopano Mapesa1, Ana Sanchez-Azqueta2, Mahmood M Alam3, Alfred Amambua-Ngwa4, Davis C Nwakanma4, Andrew B Tobin2, Andrew G Jamieson1. 1. School of Chemistry, University of Glasgow, Joseph Black Building, University Avenue, Glasgow G12 8QQ, U.K. 2. Centre for Translational Pharmacology, Institute of Molecular Cell and Systems Biology, University of Glasgow, Davidson Building, Glasgow G12 8QQ, U.K. 3. Wellcome Centre for Integrative Parasitology and Centre for Translational Pharmacology, Institute of Infection Immunity and Inflammation, University of Glasgow, Glasgow G12 8TA, U.K. 4. MRC Unit The Gambia at LSHTM, Atlantic Boulevard, Fajara, P. O. Box 273, Banjul, The Gambia.
Abstract
The protein kinase PfCLK3 plays a critical role in the regulation of malarial parasite RNA splicing and is essential for the survival of blood stage Plasmodium falciparum. We recently validated PfCLK3 as a drug target in malaria that offers prophylactic, transmission blocking, and curative potential. Herein, we describe the synthesis of our initial hit TCMDC-135051 (1) and efforts to establish a structure-activity relationship with a 7-azaindole-based series. A total of 14 analogues were assessed in a time-resolved fluorescence energy transfer assay against the full-length recombinant protein kinase PfCLK3, and 11 analogues were further assessed in asexual 3D7 (chloroquine-sensitive) strains of P. falciparum parasites. SAR relating to rings A and B was established. These data together with analysis of activity against parasites collected from patients in the field suggest that TCMDC-135051 (1) is a promising lead compound for the development of new antimalarials with a novel mechanism of action targeting PfCLK3.
The protein kinase PfCLK3 plays a critical role in the regulation of malarial parasite RNA splicing and is essential for the survival of blood stage Plasmodium falciparum. We recently validated PfCLK3 as a drug target in malaria that offers prophylactic, transmission blocking, and curative potential. Herein, we describe the synthesis of our initial hit TCMDC-135051 (1) and efforts to establish a structure-activity relationship with a 7-azaindole-based series. A total of 14 analogues were assessed in a time-resolved fluorescence energy transfer assay against the full-length recombinant protein kinase PfCLK3, and 11 analogues were further assessed in asexual 3D7 (chloroquine-sensitive) strains of P. falciparum parasites. SAR relating to rings A and B was established. These data together with analysis of activity against parasites collected from patients in the field suggest that TCMDC-135051 (1) is a promising lead compound for the development of new antimalarials with a novel mechanism of action targeting PfCLK3.
In the global fight
against malaria, the distribution of over 0.5
billion insecticide impregnated bed nets since 2015, together with
artemisinin-based combination therapies, has contributed to 20 million
fewer cases of malaria being reported in 2017 compared to 2010.[1] Despite this success, the rate of reduction in
malaria has stalled with no significant reduction seen over the last
3 years.[1] This, together with evidence
of the emergence of resistance to both artemisinin[2,3] and
partner drugs including piperaquine and mefloquine,[4,5] means
that there is a danger that the progress made in the reduction of
malaria will be reversed unless new medicines with novel mechanisms
of action are discovered. To address this, we have been testing the
notion that targeting malaria parasite protein kinases, known to be
essential for the survival of the parasite, might offer a novel strategy
for the development of next-generation antimalarials.[6,7]In 2011, we determined that 36 of the 65 eukaryotic protein
kinases
are essential for the survival of the blood stage of the most virulent
species of human malaria, Plasmodium falciparum.[8] Among these protein kinases was PfCLK3 (PF3D7_1114700), one of the four
members of the cyclin-dependent like protein kinase family (PfCLK1-4). The plasmodium CLK-family are closely related
to the mammalian CLK family and the serine–arginine-rich protein
kinase (SRPK) family,[9] both of which are
crucial mediators of multiple phosphorylation events on splicing factors
necessary for the correct assembly and catalytic activity of spliceosomes.[10]In vitro studies showed that PfCLK3 can phosphorylate parasite SR proteins,[11] indicating that PfCLK3, together
with other members of the PfCLK family,[12] plays a role in the processing of parasite RNA.[7] We reasoned therefore that inhibitors of PfCLK3
might have parasiticidal activity at any stage of the parasite life
cycle where RNA splicing played an essential role.Screening
of ∼25,000 compounds, including all 13,533 compounds
of the Tres cantos antimalarial set, resulted in the discovery of
TCMDC-135051 (1, Figure ), a compound with nanomolar (nM) activity against PfCLK3 in vitro kinase assays and submicromolar
(μM) parasiticidal activity in asexual blood stage P. falciparum parasites (Figure ).[13] Subsequent
studies revealed that TCMDC-135051 rapidly killed P.
falciparum at the trophozoite to schizont stages as
well as preventing the development of stage V gametocytes and inhibiting
the development of liver stage parasites. TCMDC-135051 also showed
equivalent in vitro kinase activity at CLK3 from P. falciparum, Plasmodium berghei (mouse malaria), Plasmodium vivax (human malaria), and Plasmodium knowlesi (monkey and human malaria).[13] Our recent
studies have therefore validated PfCLK3 as a target
with the potential to deliver a curative treatment, be transmission
blocking, and act as a prophylactic target.[13]
Figure 1
Structure
and biological profile of hit PfCLK3 inhibitor TCMDC-135051, 1.
Structure
and biological profile of hit PfCLK3 inhibitor TCMDC-135051, 1.Key structural features of 1 include a core 7-azaindole
scaffold with aromatic rings in the 2- and 4-positions (ring A and
B, respectively). These aromatic rings are further substituted with
various functional groups, including a tertiary amine (ring A) and
carboxylic acid (ring B), resulting in a zwitterionic compound at
physiological pH.7-Azaindoles are a widely studied pharmacophore
incorporated into
several therapeutic agents.[14] Kinases are
the predominate target with the 7-azaindole moiety generally interacting
at the ATP binding site within the kinase hinge region.[15,16] An interaction occurs between the kinase hinge region peptide backbone
and the azaindole via two H-bonds; the first involving
the pyridine nitrogen lone pair and peptide backbone NH, and the second
between the pyrrole NH and peptide backbone carbonyl. The resulting
H-bond acceptor and donor bidentate 7-azaindole interaction with the
hinge region of the kinase can occur in the more common “normal”
orientation or the “flipped” orientation with the 2-position
of the 7-azaindole projected out of the hinge region into the solvent-exposed
space.[17,18]A number of small molecule kinase
7-azaindole inhibitors have progressed
to different stages of clinical trials.[19] A potential drug candidate GSK1070916 is being developed as an aurora
kinase (Ser/Thr protein kinases family) inhibitor and has reached
human clinical trials (Figure A).[20] The core scaffold is a 7-azaindole
with aromatic substituents in the 2- and 4-positions. An X-ray crystal
structure of the molecule:aurora kinase complex revealed a flipped
hinge region binding mechanism, with 2-aryl projecting out of the
hinge region into solvent and 4-aryl bound within the ribose pocket.[20]
Figure 2
(A) Aurora kinase inhibitor GSK1070916, (B) 1-p-chlorobenzyl-7-azaindole-3-α-piperidylmethanol has
in vivo
activity against P. berghei, (C) PfHsp90
inhibitor IND31119, and (D) drug candidates incorporating a core 7-azaindole
scaffold, vemurafenib, PLX3397, and AZD5363.
(A) Aurora kinase inhibitor GSK1070916, (B) 1-p-chlorobenzyl-7-azaindole-3-α-piperidylmethanol has
in vivo
activity against P. berghei, (C) PfHsp90
inhibitor IND31119, and (D) drug candidates incorporating a core 7-azaindole
scaffold, vemurafenib, PLX3397, and AZD5363.Substructure analysis using Scifinder revealed several 7-azaindole
analogues described as antimalarials. In 1972, Verbiscar reported
the first example of a 7-azaindole with antimalarial activity.[21] 1-p-Chlorobenzyl-7-azaindole-3-α-piperidylmethanol
displayed antimalarial activity in mice against P.
berghei(Figure B). More recently, Picard et al. reported 7-azaindole compounds
that were identified from an in silico structure-based
drug screen (1,013,483 compounds, Chembridge library).[22] Compound IND31119 (Figure C) binds to the recombinant N-terminal domain
of PfHsp90 and shows selectivity over human Hsp90
and PfHsp90 mutants.[22]In addition to antimalarial activity, the 7-azaindole scaffold
has been identified as having a wide variety of biological applications,
including antitumor activity, and acts as an HIV-1 inhibitor in infected
patients.[23,24] The 7-azaindole motif can be regarded as
a privileged scaffold in medicinal chemistry as it is found in three
clinical candidates (vemurafenib, PLX3397, and AZD5363) (Figure D), which suggests
its usefulness for developing novel pharmaceuticals.[25−27]To investigate the structure–activity relationship
(SAR)
of hit compound 1, a series of analogues were prepared
through varying different substituents on ring A and ring B (Figure ). The N-diethyl amine group (ring A) was initially replaced with different
amine substituents to investigate lipophilicity and solubility (Figure , SAR1). Next, we
replaced the methoxy moiety (ring A) with hydroxyl or hydrogen to
modify polarity and to explore the role of the methoxy group on activity
(Figure , SAR2). The
isopropyl substituent (ring B) was replaced with other small hydrophobic
substituents to probe noncoplanar conformations that could potentially
lower the energy of crystal packing and consequently improve aqueous
solubility and log P (Figure , SAR3). Finally, we exchanged the carboxylic
acid group (ring B) with other substituents to investigate its role
in binding, to potentially improve metabolic stability, and to explore
the effect of increased lipophilic character (Figure , SAR4).TCMDC-135051 is a promising
hit compound for a medicinal chemistry
program to develop as a preclinical lead that meets many of the criteria
set by the Medicines for Malaria Venture (capable of rapidly clearing
the parasite, has multistage potency, and killing multiple parasite
species with action as a transmission blocker).[28,29] Here, we describe the synthetic route to TCMDC-135051 and determine
a SAR that will be key for lead development.
Chemistry
To investigate
the effect of the N-diethyl group of ring A on antimalarial
activity, analogues
of 4-(2-(5-((diethylamino)methyl)-2-methoxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
acid (TCMDC-135051) 1, 8a–c were
prepared, as shown in Scheme . Protection of 4-bromo-7-azaindole 2 was achieved
using p-toluenesulfonyl chloride under basic conditions
to provide N-tosyl-7-azaindole 3. Regioselective
iodination of 3 using lithium diisopropylamide (LDA)
and iodine in tetrahydrofuran (THF) at −78 °C provided
2-iodoazaindole 4, which was subsequently subjected to
Suzuki coupling with (5-formyl-2-methoxyphenyl)boronic acid to give
2-aryl substituted azaindole 5. Reductive amination of
the aldehyde functionality of 5 was performed with various
amines in the presence of sodium triacetoxyborohydride to give amines 6a–d in excellent yields. Next, the N-tosyl protecting group was removed under basic conditions at reflux
for 18 h to yield 7a–d. Finally, the desired analogues 1 and 8a–c were obtained by Suzuki cross-coupling
with 2-isopropyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoic
acid (see the Supporting Information for
boronate ester synthesis).
Scheme 1
Synthesis of TCMDC-135051 1 and Analogues 8a–c & 9
Reagents and conditions: (i)
TsCl, NaH, THF, 0 °C, 2 h; (ii) LDA, I2, THF, −78
°C, 3 h; (iii) (5-formyl-2-methoxyphenyl)boronic acid, Pd(PPh3)4, Na2CO3, 1,4-dioxane,
110 °C, 12 h; (iv) amine, NaBH(AcO)3, 1,4-dioxane,
20 °C, 12 h; (v) CH3OH, K2CO3, 55 °C, 18 h; (vi) 2-isopropyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoic
acid, Pd(dppf)Cl2·CH2Cl2, Na2CO3, 1,4-dioxane, 110 °C, 0.5 h, MW; and (vii)
SOCl2, CH3CH2OH, reflux, 18 h.
Synthesis of TCMDC-135051 1 and Analogues 8a–c & 9
Reagents and conditions: (i)
TsCl, NaH, THF, 0 °C, 2 h; (ii) LDA, I2, THF, −78
°C, 3 h; (iii) (5-formyl-2-methoxyphenyl)boronic acid, Pd(PPh3)4, Na2CO3, 1,4-dioxane,
110 °C, 12 h; (iv) amine, NaBH(AcO)3, 1,4-dioxane,
20 °C, 12 h; (v) CH3OH, K2CO3, 55 °C, 18 h; (vi) 2-isopropyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoic
acid, Pd(dppf)Cl2·CH2Cl2, Na2CO3, 1,4-dioxane, 110 °C, 0.5 h, MW; and (vii)
SOCl2, CH3CH2OH, reflux, 18 h.Compound 9 was prepared, as described
in Scheme . The carboxylic
acid of 1 was converted to its corresponding ethyl ester
by refluxing in thionyl chloride and ethanol for 18 h.The synthetic
route outlined in Scheme was modified to prepare primary amine 12. The
first step of this synthetic route involved coupling
2-iodo-azaindole 4 with the appropriate boronate ester
under Suzuki conditions to yield nitrile 10 in 87% yield
(Scheme ). Tosyl deprotection
was then achieved using K2CO3, followed by a
Suzuki coupling at the 4-bromo-azaindole scaffold with the appropriate
pinacol boronate ester. The crude material was reduced in
situ using cobalt(II) chloride hexahydrate and sodium borohydride
to provide the corresponding amine 12 (Scheme ).
Scheme 2
Synthesis of 4-(2-(5-(Aminomethyl)-2-methoxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
Acid 12
Reagents and conditions: (i)
4-methoxy-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzonitrile,
Pd(PPh3)4, Na2CO3, 1,4-dioxane,
110 °C, 18 h; (ii) CH3OH, K2CO3, 55 °C, 18 h; (iii) 2-isopropyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoic
acid, Pd(dppf)Cl2.CH2Cl2, Na2CO3, 1,4-dioxane, 110 °C, 0.5 h, MW; and (iv)
CoCl2·6H2O, NaBH4, CH3OH.
Synthesis of 4-(2-(5-(Aminomethyl)-2-methoxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
Acid 12
Reagents and conditions: (i)
4-methoxy-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzonitrile,
Pd(PPh3)4, Na2CO3, 1,4-dioxane,
110 °C, 18 h; (ii) CH3OH, K2CO3, 55 °C, 18 h; (iii) 2-isopropyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoic
acid, Pd(dppf)Cl2.CH2Cl2, Na2CO3, 1,4-dioxane, 110 °C, 0.5 h, MW; and (iv)
CoCl2·6H2O, NaBH4, CH3OH.Similarly, analogue 15 was
synthesized via Suzuki coupling of 4 with 2-methoxyphenyl boronic acid,
followed by tosyl deprotection and Suzuki coupling (Scheme ).
Scheme 3
Synthesis of 2-Isopropyl-4-(2-(2-methoxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)benzoic Acid 15, 4-(2-(5-((Diethylamino)methyl)-2-hydroxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
Acid 19, 4-(2-(3-((Diethylamino)methyl)phenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
Acid 23 and [4-(2-(3-((Diethylamino)methyl)-2-methoxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
Acid] 27
Reagents and conditions: (i)
(2-methoxyphenyl)boronic acid, Pd(PPh3)4, Na2CO3, 1,4-dioxane, 110 °C, 12 h; (ii) CH3OH, K2CO3, 55 °C, 18 h; (iii) 2-isopropyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoic
acid, Pd(dppf)Cl2·CH2Cl2, Na2CO3, 1,4-dioxane, 110 °C, 0.5 h, MW; (iv)
4-(methoxymethoxy)-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzaldehyde
or 3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzaldehyde, Pd(PPh3)4, Na2CO3, 1,4-dioxane,
110 °C, 18 h; (v) Et2NH, NaBH(AcO)3, 1,4-dioxane,
rt, 18 h; and (vi) HCl, MeCN/H2O 3:1.
Synthesis of 2-Isopropyl-4-(2-(2-methoxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)benzoic Acid 15, 4-(2-(5-((Diethylamino)methyl)-2-hydroxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
Acid 19, 4-(2-(3-((Diethylamino)methyl)phenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
Acid 23 and [4-(2-(3-((Diethylamino)methyl)-2-methoxyphenyl)-1H-pyrrolo[2,3-b]pyridin-4-yl)-2-isopropylbenzoic
Acid] 27
Reagents and conditions: (i)
(2-methoxyphenyl)boronic acid, Pd(PPh3)4, Na2CO3, 1,4-dioxane, 110 °C, 12 h; (ii) CH3OH, K2CO3, 55 °C, 18 h; (iii) 2-isopropyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzoic
acid, Pd(dppf)Cl2·CH2Cl2, Na2CO3, 1,4-dioxane, 110 °C, 0.5 h, MW; (iv)
4-(methoxymethoxy)-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzaldehyde
or 3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzaldehyde, Pd(PPh3)4, Na2CO3, 1,4-dioxane,
110 °C, 18 h; (v) Et2NH, NaBH(AcO)3, 1,4-dioxane,
rt, 18 h; and (vi) HCl, MeCN/H2O 3:1.To investigate the precise role of the methoxy group of ring A
on activity, we prepared analogues 19 and 23 in which the OMe is replaced by a hydroxy group (19) and OMe is removed (23) (Scheme ). The synthetic route toward 19 starts with Suzuki coupling of 4 with methoxymethyl
ether (MOM)-protected boronate ester (see the Supporting Information for boronate ester synthesis) to yield 16. Reductive amination of 16 followed by removal
of the tosyl group gave 18. Finally, 18 was
coupled with boronate ester under Suzuki conditions, and the MOM protecting
group was removed under acidic conditions to provide the desired analogue 19 in excellent yields. Analogue 23 was synthesized
in four steps using the same methods.To investigate whether
the substitution pattern on ring A is important
for antimalarial activity, we produced analogue 27 in
which the methoxy group is moved from para to the ortho position relative
to the methylene N-diethyl functionality (Scheme ).To investigate the effect of the
isopropyl substituent (ring B)
on efficacy, we prepared analogues 28–29, as shown
in Scheme . 4-Bromo-7-azaindole 7a was coupled with various boronate esters under Suzuki conditions
to yield the desired 4-aryl-7-azaindole analogues 28–29.
Scheme 4
Synthesis of 4-Aryl-7-azaindole Analogues, 28–32
Reagents and conditions: (i)
boronate esters, Pd(dppf)Cl2.CH2Cl2, Na2CO3, 1-4-dioxane, 110 °C, 0.5 h,
MW.
Synthesis of 4-Aryl-7-azaindole Analogues, 28–32
Reagents and conditions: (i)
boronate esters, Pd(dppf)Cl2.CH2Cl2, Na2CO3, 1-4-dioxane, 110 °C, 0.5 h,
MW.To investigate the role of the carboxylic
acid substituent of ring
B on antimalarial activity, several analogues 30–32 were synthesized, as depicted in Scheme . Synthesis of analogues 30–32 was achieved by Suzuki coupling 7a with various
boronate esters (R2 = 1H-tetrazole or H) to yield target compounds 30–31 in good yields.An isomer, in which the isopropyl and carboxylic
acids were switched, 32 was synthesized to investigate
the role of these functional
groups on antimalarial activity (Scheme ).
Results and Discussion
In this work, we have maintained the core 7-azaindole molecular
scaffold of TCMDC-135051 and focused on the SARs of the substitutes
on ring A and -B to assess the requirement for this functionality.
All synthesized analogues were assessed in a time-resolved fluorescence
energy transfer (TR-FRET) PfCLK3 in vitro kinase assay against the full recombinant protein kinase (Tables and 2). Analogues which gave low nanomolar activity were then further
assessed in live parasite viability (parasiticidal) assays using laboratory
strain 3D7 (chloroquine-sensitive) P. falciparum (Tables and 2). The synthesized analogues were evaluated for
log D7.4 (distribution co-efficient) and
metabolic stability, and the results are shown in Tables and 2.
Table 1
Physicochemical Properties and Activity
Data of TCMDC-135051 Ring A Analogues
PfCLK3
3D7
analogue
R1
R2
R3
IC50 (nM)a
pIC50
EC50 (nM)b
pEC50
log D7.4c
clint (mL/min/g liver)d
1
NEt2
OMe
H
40
7.4 (±0.221)
180
6.7 (±0.126)
0.93
1.12
8a
NMe2
OMe
H
29
7.5 (±0.224)
457
6.3 (±0.129)
0.85
2.53
8b
N-pyrrolidinyl
OMe
H
38
7.4 (±0.113)
382
6.4 (±0.081)
2.43
1.94
8c
N-morpholinyl
OMe
H
9
8.0 (±0.191)
1339
5.9 (±0.118)
1.20
1.60
12
NH2
OMe
H
76
7.1 (±0.142)
2801
5.6 (±0.104)
0.61
2.92
15
H
OMe
H
79
7.1 (±0.132)
1456
5.8 (±0.152)
2.45
2.54
19
NEt2
OH
H
22
7.7 (±0.115)
3529
5.5 (±0.133)
0.59
1.94
23
NEt2
H
H
25
7.6 (±0.089)
309
6.5 (±0.114)
0.80
0.85
27
NEt2
H
OMe
17
7.7 (±0.116)
3167
5.6 (±0.109)
0.74
1.65
IC50 (the concentration
of an inhibitor where the response is reduced by half).
EC50 (the concentration
of a drug that gives half-maximal response). IC50 and EC50 values are means ± SEM of three independent experiments
run in triplicates (n = 3).
log D7.4 (distribution
co-efficient) was estimated using HPLC chromatography.
In vitro intrinsic
clearance in mouse liver microsomes.
Table 2
Physicochemical Properties and Activity
Data of TCMDC-135051 Ring B Analogues
PfCLK3
3D7
analogue
R1
R2
IC50 (nM)a
pIC50
ED50 (nM)b
pEC50
log D7.4c
Clint (mL/min/g liver)d
28
CH3
CO2H
24
7.6 (±0.10)
1185
5.6 (±0.097)
0.66
1.33
29
H
CO2H
34
7.5
(±0.089)
3272
5.5 (±0.124)
0.65
11.01
30
CH(CH3)2
1H-tetrazole
19
7.7 (±0.089)
270
6.6 (±0.158)
0.93
2.32
31
CH(CH3)2
H
1300
6.0 (±0.091)
NDe
NDe
4.45
9.54
9
CH(CH3)2
CO2CH2CH3
390
6.4 (±0.087)
NDe
NDe
2.45
16.06
32
CO2H
CH(CH3)2
1385
4.9 (±0.093)
NDe
NDe
NDe
NDe
IC50 (the concentration
of an inhibitor where the response is reduced by half).
EC50 (the concentration
of a drug that gives half-maximal response). IC50 and EC50 values are means ± SEM of three independent experiments
run in triplicates (n = 3).
log D7.4 (distribution
co-efficient) was estimated using HPLC chromatography.
In vitro intrinsic
clearance in mouse liver microsomes.
ND, not determined.
IC50 (the concentration
of an inhibitor where the response is reduced by half).EC50 (the concentration
of a drug that gives half-maximal response). IC50 and EC50 values are means ± SEM of three independent experiments
run in triplicates (n = 3).log D7.4 (distribution
co-efficient) was estimated using HPLC chromatography.In vitro intrinsic
clearance in mouse liver microsomes.IC50 (the concentration
of an inhibitor where the response is reduced by half).EC50 (the concentration
of a drug that gives half-maximal response). IC50 and EC50 values are means ± SEM of three independent experiments
run in triplicates (n = 3).log D7.4 (distribution
co-efficient) was estimated using HPLC chromatography.In vitro intrinsic
clearance in mouse liver microsomes.ND, not determined.SAR1 corresponding to analogues 8a–c, 12, and 15 were designed to examine the effect
of the N-diethyl group of ring A on antimalarial activity. In analogue 8a, the N-diethyl group was replaced with an N-dimethyl group
to investigate the effect of alkyl group size and molecular lipophilicity
(log D7.4 = 0.85). For in vitro kinase activity of 8a, the half maximal inhibitory
activity (IC50) IC50 = 29 nM (pIC50 = 7.5 ± 0.224) remains the same in recombinant PfCLK3 kinase, whereas a 2-fold decrease in the half maximal parasite
growth inhibition effect (EC50) EC50 = 457 nM
(pEC50 = 6.3 ± 0.129) was observed (Table ). To further investigate the
steric requirements of this functionality and polarity, pyrrolidine
(8b) and morpholino groups (8c) were introduced.
Analogue 8b shows comparable in vitro kinase activity, IC50 = 38 nM (pIC50 = 7.4
± 0.113), and a 2-fold decrease in parasite growth inhibition 8b, EC50 = 382 nM (pEC50 = 6.4 ±
0.081), was observed. However, the more polar 8c shows
a slight improvement in vitro kinase activity of
IC50 = 9 nM (pIC50 = 8.0 ± 0.191), a 7-fold
decrease in parasite growth inhibition was observed 8c, EC50 = 1339 nM (pEC50 = 5.9 ± 0.118)
(Table ). To further
investigate the polarity of this moiety, we replaced the N-diethyl
functionality with a more polar primary amine (log D7.4 = 0.61). Analogue 12 incorporating a
primary amine gave an IC50 = 76 nM (pIC50 =
7.1 ± 0.142), thus was 2-fold less potent in vitro and showed a dramatic loss of efficacy, EC50 = 2801 nM
(pEC50 = 5.6 ± 0.104), in parasites, indicating the
need to decrease the polarity of the amine group for optimal parasite
growth inhibition (Table ). Analogue 15, with the alkyl amine group removed
(log D7.4 = 2.45) was less potent in vitro, IC50 = 79.0 nM (pIC50 =
7.1 ± 0.132) against recombinant PfCLK3 protein
and in parasites, EC50 = 1456 nM (pEC50 = 5.8
± 0.152). There was a 7-fold drop in the parasite growth inhibition,
suggesting that the alkyl amine on ring A is important for antimalarial
activity (Table ).To explore the importance of the methoxy group (OMe) on ring A,
two analogues, hydroxyl 19 and 23, with
methoxy removed, were prepared. We first replaced the OMe group with
a hydroxyl to investigate the role of polarity on activity (log D7.4 = 0.59). Next, we replaced the OMe group
with hydrogen (23) to investigate the importance of this
functionality for activity. For both compounds 19 and 23, in vitro kinase potency was comparable,
IC50 = 22 nM (pIC50 = 7.7 ± 0.115) and
IC50 = 25 nM (pIC50 = 7.6 ± 0.089), respectively
(Table ). When tested
in parasites, 19 shows significant loss of activity,
EC50 = 3529 nM (pEC50 = 5.5 ± 0.133). Interestingly,
replacing the OMe group with a hydrogen, 23 also shows
a 1.5-fold decrease in potency, EC50 = 309 nM (pEC50 = 6.5 ± 0.114) against parasites. These data demonstrated
that the OMe group on ring A is required for in vitro kinase and parasite growth inhibition. Overall, analogues (8a–c, 12, 15, 19, and 23) showed good metabolic stability in mouse liver
microsomes (Table ).Next, we turned our attention to study the SAR of various
functional
groups on ring B. First, we examined the role of binding through van
der Waals interactions in vitro and lipophilicity
in parasites, imparted by the ring B isopropyl group, on antimalarial
activity. In analogue 28 (log D7.4 = 0.66), the isopropyl group was replaced with methyl,
whereas in analogue 29 (log D7.4 = 0.65), the isopropyl was removed and replaced with hydrogen. Analogues 28 and 29 had a modest increase on in
vitro kinase activity, IC50 = 24 nM (pIC50 = 7.6 ± 0.10) and IC50 = 34 nM (pIC50 = 7.5 ± 0.089), respectively. However, a dramatic loss of potency
was observed for both compounds 28 and 29 with EC50 = 1185 nM (pEC50 = 5.6 ± 0.097)
and EC50 = 3272 nM (pEC50 = 5.5 ± 0.124),
respectively, representing a 6-fold and 16-fold decrease in parasite
growth inhibition (Table ). Therefore, this group is not required for binding in vitro; however, the effect of lipophilicity appears to
be very important for parasite growth inhibition. Analogues 28 and 29 were then evaluated for intrinsic clearance
in mouse microsome. Analogue 28 (Clint = 1.33
mL/min/g liver) showed a similar clearance value to that of 1 (TCMDC-051) in mouse liver microsomes, whereas 29 had a higher clearance under the same conditions (Clint = 11.01 mL/min/g liver) indicating the importance of the isopropyl
group.At this point, we decided to increase the lipophilicity
and replace
the isopropyl group to the larger tert-butyl group.
However, due to a restriction of the chemistry of boronate esters
(instability of tert-butyl anion required for SNAr reaction), we could not synthesize the proposed molecule.
From these data, the alkyl substituent on ring B appears not to be
particularly important in the molecular recognition event with PfCLK3, however is key for overall molecular lipophilicity
that contributes to parasite growth inhibition.To investigate
the requirements (i.e., ionic and H-bonding) of
the carboxylic acid (ring B) for antimalarial activity, we employed
a series of structural changes. The presence of carboxylic acid functionality
in a drug molecule has several potential drawbacks, including limited
permeability across biological membranes, metabolic instability, and
potential idiosyncratic toxicities. A common isostere of a carboxylic
acid, that overcomes many of these physicochemical limitations, is
a tetrazole. We therefore designed tetrazole analogue 30. This change was orchestrated to increase the lipophilicity of the
molecule, retain H-bonding capability, and investigate the role of
potential ionic interactions with the enzyme. The tetrazole analogue 30 shows improved in vitro kinase potency
with an IC50 = 19 nM (pIC50 = 7.7 ± 0.089),
with comparable parasite growth inhibition, EC50 = 270
nM (pEC50 = 6.6 ± 0.158) (Table ). The potency of analogue 30 was tested against resistant mutant PfCLK3 (PfCLK3_G449P) 3D7 parasites.[13] The parasiticidal activity of TCMDC-135051 1 against
asexual 3D7 parasites was 180 nM (pEC50 = 6.7), and 1806
nM (pEC50 = 5.74) against mutant G449P. A 15-fold shift
in sensitivity to the inhibitor. The tetrazole 30, with
an EC50 = 270 nM (pEC50 = 6.6) in wild type
parasites showed a significant reduction is potency against the mutant
G449P parasites, EC50 = 3494 nM (pEC50 = 5.4).
This represents a 13-fold shift in potency.Analogue 31 where the carboxylic acid group was replaced
with a hydrogen shows significant loss of in vitro kinase activity, IC50 = 1300 nM (pIC50 = 6.0
± 0.091). Analogue 31 has high lipophilicity (log D7.4 = 4.45) and in vitro high
intrinsic clearance (Clint = 9.54 mL/min/g liver) demonstrating
the importance of the carboxylic acid group (Table ).Analogue 9 has improved
lipophilicity (log D7.4 = 2.45), yet the in vitro kinase potency was low, IC50 = 390 nM
(pIC50 = 6.4 ± 0.087), indicating that the presence
of a functional
group capable of donating a hydrogen bond is important for activity
(Table ). This analogue
was not tested in parasites because of the low potency against recombinant PfCLK3 kinase.The final part of our SAR assessment
was dedicated to exploring
the possibility of orientation of key substituents. Varying the position
of the OMe group from para to the ortho position 27 resulted
in a 2-fold increase in in vitro kinase activity,
IC50 = 17 nM (pIC50 = 7.7 ± 0.116). However,
when 27 was tested in parasites, the activity was lowered
by 15-fold, EC50 = 3167 nM (pEC50 = 5.6 ±
0.109) (Table ).We next investigated the position of the isopropyl and carboxylic
acid substituents of ring B 32 on antimalarial activity.
The change of orientation of the substituent is detrimental for the in vitro kinase activity, IC50 = 1385 nM (pIC50 = 4.9 ± 0.093), and therefore was not tested in parasites.
These data suggested that the positioning of the isopropyl and carboxylic
acid group was essential for binding to its cellular target.No X-ray crystal structure has been reported for PfCLK3 and so a homology model of PfCLK3 was created
using SWISS-MODEL to provide evidence of our proposed binding mechanism
(i.e., hinge binder), which we hypothesized based on other literature
examples of 7-azaindole scaffolds binding at the hinge region of the
kinase domain. The structure of the human PRPF4B kinase domain was
selected as the template as the closest homologue [PDB 6CNH (sequence identity
53.2%)].[30] Model accuracy was determined
to be reasonable using SWISS-MODEL with QMEAN (qualitative model energy
analysis) score = −2.26 and GMQE (global model quality estimation)
= 0.77. Overlay with Human Jnk1alpha kinase with 4-phenyl-7-azaindole
IKK2 inhibitor bound (PDB 4AWI) facilitated identification of the proposed binding
pocket.[31] Based on this model, we propose
that the 7-aza-indole scaffold interacts with the hinge region in
the flipped conformation, that is, H-bonding to the peptide backbone
of the hinge region (Figure ). The benzoic acid on ring B appears to occupy the ribose
pocket and interacts with Lys-394. In the model, the ring B isopropyl
group occupies a hydrophobic back pocket. Ring A projects into the
solvent-exposed space. The diethyl-amine and methoxy substituents
are solvent exposed and may contribute to the orientation of the 7-azaindole
in the flipped binding conformation.
Figure 3
Putative
binding mode of TCMDC-135051 1 in a PfCLK3
homology model.
We previously demonstrated
that TCMDC-135051 showed selective inhibition
of PfCLK3 when compared against the closely related
human kinases (PRPF4B and CLK2), as well as the closest parasite kinase
(PfCLK1), and other parasite kinases (PfPKG and PfCDPK1).[13] To
further assess kinase selectivity, and the potential for off-target
toxicity, TCMDC-135051 was screened against 140 human kinases at 1
μM concentration. Only nine kinases were found to have less
than 20% activity at this concentration (see the Supporting Information).
TCMDC-135051 Inhibits the Growth of Clinical
Field Isolates
An important property for next-generation
antimalarials is effectiveness
against parasites that are resistant to currently used antimalarials.
To determine whether the parent molecule TCMDC-135051 1 might offer such activity, we tested the efficacy of this compound
against parasites that were collected from malaria patients in The
Gambia. These parasite strains were sequenced and those that contained
genetic markers of resistance were selected (Table ). Parasites were then tested for resistance
to pyrimethamine, a frontline antimalarial where resistance has been
documented in parasites originating from The Gambia (Figure A).[32]
Table 3
Genotypes of Field
Isolated Parasite
Strains: Whole Genome Sanger Sequencing Identified Mutations in Three
Parasite Genes (CRT, PfMDR1, and PfDHFR) Associated with Drug Resistancea
PfCRT
PfMDR1
PfDHFR
sample ID
K76T
N86Y
Y184F
N1042D
N51I
C59R
S108N
I164L
GB0006
X
X
X
N
X
X
N
L
GB0004
K
N
Y
N
N
R
N
L
GB0026
K
N
X
N
N
R
N
L
GB002
T
N
X
N
I
C
N
L
GB0021
K
N
F
N
X
X
N
L
GB0048
X
N
F
N
I
C
N
L
GB0071
X
N
Y
N
N
C
N
L
GB0087
T
N
Y
N
I
C
N
L
GB0020
T
Y
Y
N
I
C
N
L
Shown are the amino acid changes
associated with each of these three genes and whether the parasite
strain contained the mutation.
Figure 4
PRR of clinical isolates
comparing pyrimethamine and TCMDC-135051
at EC50 and 10 times the EC50 of each drug.
This represent the average of triplicate.
Putative
binding mode of TCMDC-135051 1 in a PfCLK3
homology model.Shown are the amino acid changes
associated with each of these three genes and whether the parasite
strain contained the mutation.This analysis identified nine parasite isolates from patients with
varying degrees of resistance to pyrimethamine (Figure A,B) and which contained mutations in one
or more of the three genes; PfCRT (P. falciparum chloroquine-resistant transporter gene), PfMDR1 (P. falciparum multidrug-resistant
gene 1), and PfDHFR (P. falciparum dihydrofolate reductase gene). The mutations identified in these
genes (Table ) have
previously been associated with resistance to commonly used antimalarials,
including chloroquine and pyrimethamine.[33] An EC50 concentration of pyrimethamine was then determined
using the laboratory strain parasite 3D7 (EC50 = 18 nM).
Using this concentration in parasite reduction rate (PRR) assays,
all isolates showed a reduction in parasitemia by at least 70% except
for three isolates (GB0020, GB002, and GB0026), which showed resistance
to pyrimethamine treatment. These three isolates uniquely contained
triple mutations in the PfDHFR gene and a single
mutation in PfCRT (Table ). For these isolates, resistance was evident
after 24 h of treatment with pyrimethamine where parasitemia was only
reduced by 45, 64, and 36%, respectively (Figure A). Resistance to pyrimethamine of these
three isolates was still evident after 72 h of drug treatment (Figure A) and in parasites
exposed to 10× EC50 (180 nM) of pyrimethamine (Figure B).PRR of clinical isolates
comparing pyrimethamine and TCMDC-135051
at EC50 and 10 times the EC50 of each drug.
This represent the average of triplicate.In contrast, all isolates treated with the EC50 of TCMDC-135051,
as determined in 3D7 parasites (EC50 = 180 nM), demonstrated
total susceptibility, including isolates GB0020, GB002, and GB0026
(Figure C). This was
true both at the EC50 concentration and 10× EC50 for TCMDC-135051 (Figure D). Hence in the field isolates tested, TCMDC-135051
showed equivalent activity against parasites that carried genetic
markers of resistance and those that showed actual resistance to a
frontline antimalarial. This indicates the possibility that the parent
molecule, and potentially analogues of this molecule, would be active
against naturally circulating malaria parasites harboring mutations
that promoted resistance to current antimalarial drugs. The PRR assay
was setup using clinical isolates previously cryopreserved in liquid
nitrogen (LN2). These were thawed and placed in culture for at least
one cycle (∼48 h ± 2). The life cycle stage of the parasites
was monitored using a blood film, and greater than 95% of the parasites
are expected to egress and form new invading rings. These new rings,
mainly 0–3 h old postinvasion, were used as time point 0 hours
(t = 0) for the assay. The parasites are then cultured
with drugs for 24, 48, and 72 h. For 24 h treatment, the drug was
removed by washing twice with wash media and for longer treatments
(48 and 72 h time points), drug was replenished with fresh drug every
24 h. After the treatment period, parasites were grown in fresh media
for an additional 48 h in freshly stained erythrocytes to allow invasion
from any viable parasites.
Conclusions
In
summary, we report the synthesis of hit PfCLK3
inhibitor TCMDC-135051 1 (PfCLK3 IC50 = 40 nM, 3D7 EC50 = 180 nM) and a series of related
7-azaindole-based analogues. Of the 14 analogues, 11 had low nanomolar
activity and were further assessed in live parasite viability assays
using the 3D7 (chloroquine-sensitive) strain of P.
falciparum. Tetrazole analogue 30 was
identified with improved in vitro kinase activity
(PfCLK3, IC50 = 19 nM) and comparable
activity in parasites 3D7 (EC50 = 271 nM). SAR was established
for both ring A, highlighting the importance of H-bonding functionality
in the 4-aryl position, and for the alkyl amino group on ring B. Together
these data provide a good starting point for the hit to lead development
of novel PfCLK3 inhibitors based on TCMDC-135051
(1).
Experimental Section
General
Information
Chemicals and solvents were purchased
from standard suppliers and used without additional purification.
All glassware were dried with a flame under flushing argon gas or
stored in the oven and allowed to cool under an inert atmosphere prior
to use. Anhydrous solvents (THF, DCM, and Et2O) were obtained
by passage through solvent filtration systems (Pure Solv), and solvents
were transferred by the syringe. PET ether refers to petroleum (bp
40–60 °C, reagent grade, Fisher Scientific). All reactions
were carried out under a blanket of nitrogen in an inert or dry atmosphere.
Thin-layer chromatography (TLC) was performed using aluminium plates
precoated with silica gel (0.25 mm, 60 A° pore-size) impregnated
with a fluorescent indicator (254 nm). Visualization on TLC was achieved
by the use of UV light (254 nm). Flash column chromatography was undertaken
on silica gel (400–630 mesh). Proton nuclear magnetic resonance
spectra (1H NMR) were recorded on an AVANCE III 400 Bruker
(400 MHz). Proton chemical shifts are expressed in parts per million
(ppm, δ scale) and are referenced to residual protium in the
NMR solvent (CDCl3, δ 7.26; CD3OD, δ
3.31 and DMSO-d6, δ 2.50). The following
abbreviations were used to describe peak patterns when appropriate:
br = broad, s = singlet, d = doublet, t = triplet, q = quadruplet,
sept = septet, and m = multiplet. Coupling constants, J, were reported in hertz unit (Hz). Carbon 13 nuclear magnetic resonance
spectroscopy (13C NMR) was recorded on an AVANCE III 400
Bruker (101 MHz) and was fully decoupled by broad-band decoupling.
Chemical shifts were reported in ppm referenced to the centerline
of a triplet at 77.0, 49.0, and 39.5 ppm of CDCl3, CD3OD, and DMSO-d6. Low-resolution
mass spectrometry was performed on a Thermo Scientific LCQ Fleet quadrupole
mass spectrometer using electrospray ionization in the positive mode
(ESI+), employing a 150 mm × 4 mm C18 column (Dr.
Maisch Reprosil Gold). High-resolution mass spectrometry (HRMS) was
performed on a Bruker microTOF-Q II (ESI+). Preparative
high-performance liquid chromatography (HPLC) was carried out on a
Dionex HPLC system equipped with Dionex P680 pumps and a Dionex UVD170U
UV–vis detector (monitoring at 214 and 280 nm), using a Phenomenex,
Gemini, C18, 5 μm, 250 × 21.2 mm column. Gradients were
performed using solvents consisting of A (H2O + 0.1% TFA)
and B (CH3CN + 0.1% TFA), and fractions were lyophilized
on a Christ Alpha 2-4 LO plus freeze dryer. Final molecules were analyzed
on a Shimadzu reverse-phase HPLC (RP-HPLC) system equipped with Shimadzu
LC-20AT pumps, a SIL-20A autosampler, and a SPD-20A UV–vis
detector (monitoring at 214 nm) using a Phenomenex, Aeris, 5 μm,
peptide XB-C18, 150 × 4.6 mm column at a flow rate of 1 mL/min.
RP-HPLC gradients were run using a solvent system consisting of solution
A (5% CH3CN in H2O + 0.1% TFA) and B (5% H2O in CH3CN + 0.1% TFA). A gradient from 0 to 100%
solution B over 20 min was run. Purity of all final compounds is >95%,
as determined by RP-HPLC.
Experimental Procedures and Characterization
Data
4-Bromo-1-tosyl-1H-pyrrolo[2,3-b]pyridine (3)
To a solution of sodium hydride
(1.83 g, 76.1 mmol, 3 equiv) and tetrabutylammonium bromide (0.25
g, 0.76 mmol, 0.03 equiv) in dichloromethane (80 mL) at 0 °C
was added 4-bromo-1H-pyrrolo[2,3-b]pyridine, 2 (5 g, 25.4 mmol, 1 equiv). The mixture
was then left to stir at 0 °C for 15 min. Toluene sulfonylchloride
(5.81 g, 30.5 mmol, 1.2 equiv) in dichloromethane (20 mL) was slowly
added over 5 min. The mixture was then left to warm up to room temperature
and stirred for 1 h. The reaction was quenched by addition of water
and extracted with dichloromethane. The organic layer was washed with
brine and dried over magnesium sulfate. The residue was then purified
by flash column chromatography (10% ethyl acetate–PET ether)
to give 3 as a colorless solid (8.83 g, 99%); 1H NMR (400 MHz, CDCl3): δ 8.22 (d, J = 5.2 Hz, 1H), 8.06 (d, J = 8.4 Hz, 2H), 7.78 (d, J = 4.0 Hz, 1H), 7.35 (d, J = 5.3 Hz, 1H),
7.28 (d, J = 8.0 Hz, 2H), 6.64 (d, J = 4.0 Hz, 1H), 2.37 (s, 3H); 13C NMR (101 MHz, CDCl3): δ 146.8, 145.5, 145.0, 135.1, 129.7, 128.2, 127.0,
125.7, 124.4, 122.1, 104.9, 21.7; HRMS m/z: calcd for C14H11BrN2NaO2S [M + Na]+, 372.9617; found, 372.9608
(Δ = 2.3 ppm).
n-Butyllithium
(2.5 M; 6.3 mL, 15.6 mmol, 1.1 equiv) was added dropwise to diisopropylamine
(2.4 mL, 17.2 mmol, 1.2 equiv) in diethyl ether (30 mL) at −78
°C over a period of 5 min. The resulting solution was stirred
at −78 °C for 60 min and then slowly added via cannula to a solution of 4-bromo-1-tosyl-1H-pyrrolo[2,3-b]pyridine, 3 (5.0 g, 14.2 mmol, 1 equiv) and
tetramethylethylenediamine (2.3 mL, 15.7 mmol, 1.1 equiv) in diethyl
ether (170 mL) over a period of 10 min at −78 °C. The
resulting solution was then stirred at −78 °C for 90 min.
Iodine (5.4 g, 21.4 mmol, 1.5 equiv) was added in one portion, and
the reaction mixture was stirred at −78 °C for 60 min.
The reaction was quenched with saturated ammonium chloride solution,
and the organic layer was washed with aqueous sodium thiosulfate and
brine before drying over magnesium sulfate. The residue was then purified
by column chromatography (20% ethyl acetate–PET ether) to give 4 as a colorless solid (5.59 g, 85%; 1H NMR (400
MHz, CDCl3): δ 8.11 (d, J = 5.2
Hz, 1H), 8.01 (d, J = 8.5 Hz, 2H), 7.23 (d, J = 5.2 Hz, 1H), 7.22–7.19 (m, 2H), 6.96 (s, 1H),
2.30 (s, 3H); 13C NMR (101 MHz, CDCl3): δ
149.1, 145.7, 144.7, 135.4, 129.8, 128.3, 125.3, 123.6, 122.4, 119.4,
21.7; HRMS m/z: calcd for C14H10BrIN2NaO2S [M + Na]+, 498.8583; found, 498.8602 (Δ = −3.8 ppm).
Method A: General Method of Suzuki Cross-Coupling
To
a solution of desired aryl bromide (1.0 equiv) and tetrakis(triphenylphosphine)palladium(0)
(0.05 equiv) in 1,4-dioxane was added boronic acid/ester (1.1 equiv)
under a nitrogen atmosphere. Aqueous sodium carbonate (2 M, 7.0 equiv)
was then added, and the reaction mixture was left to stir at 110 °C
for 18 h. Solvent was removed under vacuum, and the crude was dissolved
in ethyl acetate and poured into water and extracted with ethyl acetate.
The organic layer was washed with brine before drying over magnesium
sulfate and purified by flash column chromatography as indicated.
Method B: Reductive Amination of Aldehydes
To a solution
of aryl aldehyde (1.0 equiv) in 1,4-dioxane was added amine (1.5 equiv),
and the solution was allowed to stir for 2 min before the addition
of sodium triacetoxyborohydride (2.5 equiv). The reaction mixture
was stirred at room temperature for 18 h before quenching with ammonium
hydroxide. The reaction mixture was extracted with ethyl acetate and
washed with brine. The organic layer was dried over magnesium sulfate,
and the residue was purified by flash column chromatography as indicated.
Method C: Deprotection of Azaindole
To a solution of
protected 7-azaindole (1 equiv) in methanol was added potassium carbonate
(3.5 equiv) and refluxed for 18 h. The reaction was poured into a
mixture of EtOAc (10 mL) and H2O in a separatory funnel.
Solvent was then removed under vacuum, and the residue was then purified
by flash column chromatography as indicated.
Method D: Suzuki Cross-Coupling
with Boronate Ester
To a 10 mL microwave vial containing
the required bromo-7-azaindole
(1 equiv) in 1,4-dioxane was added boronic acid/ester (1.1 equiv),
Pd(dppf)Cl2·DCM complex (0.05 equiv) under a nitrogen
atmosphere. The solution was purged with nitrogen for 5 min, and the
reaction microwaved at 110 °C for 0.5 h. The reaction was allowed
to cool to room temperature, and the mixture was filtered through
Celite eluting with methanol. The filtrate was evaporated, and the
resulting residue was purified by preparative HPLC: 10–95%
acetonitrile in water + 0.1% TFA to give the desired products.
Method
E: Synthesis of Boronate Ester
Boronate esters
required for Suzuki coupling were prepared according to the procedure
reported in the literature.[34] To a solution
of aryl bromide (1 equiv), bis(pinacolato)diboron (1.5 equiv) and
potassium acetate (3 equiv) in 1,4-dioxane (20 mL), PdCI2(dppf)·CH2Cl2 complex (0.1 equiv) were
added under nitrogen and stirred at 100 °C for 3 h. The reaction
mixture was quenched with saturated NaHCO3 and extracted
with ethyl acetate. The organic phase was dried (Na2SO4), filtered, and concentrated to dryness. The crude product
was purified by chromatography (20% ethyl acetate–PET ether)
to give the desired boronate esters.
P. falciparum cultures were maintained
in RPMI-1640 media (Invitrogen) supplemented with 0.2% sodium bicarbonate,
0.5% AlbuMAX II, 2.0 mM l-glutamine (Sigma), and 10 mg/L
gentamycin. For continuous culture, the parasites were kept at 4%
hematocrit in human erythrocytes from 0 + blood donors and between
0.5 and 3% parasitemia maintained in an incubator at 37 °C, 5%
carbon dioxide (CO2), 5% oxygen (O2), and 90%
nitrogen (N2). To obtain highly synchronous ring-stage
parasites for the drug assays, cultures were double synchronized using
Percoll and sorbitol synchronization, as previously described.[35,36] First, highly segmented schizonts were enriched by centrifugation
on a 70% Percoll (GE Healthcare) cushion gradient. The Schizont pellet
was collected and washed before fresh erythrocytes were added to a
final hematocrit of 4%. The schizonts were incubated for about 1–2
h shaking continuously to allow egress and reinvasion of new erythrocytes.
Residual schizonts were then removed by treating the pellet with sorbitol
to generate highly synchronous 1–2 h old ring-stage parasites.
Determining the EC50 of Compound Inhibitors and Drugs Ex Vivo
To determine the EC50 of the
molecules in parasites (P. falciparum 3D7) ex vivo, the molecules were diluted 1 in 3
from a starting concentration of 100 μM for 12 dilution points.
In total, 50 mL of freshly diluted drugs, at twice the required final
concentrations, were aliquoted into black 96-well plates. To the drug
plates, 50 μL of parasites prepared at 8% hematocrit at parasitemia
(0.3–0.5%) were added and mixed by pipetting up and down several
times giving a final culture volume of 100 μL at the required
drug concentration (top concentration of 100 μM) and 4% hematocrit.
To the “no drug” control, growth media was added, and
uninfected erythrocytes were included on the plate as blank. The outer
wells were filled with media to reduce evaporation from the experimental
wells, and the plates were incubated for 50 h (±2 h) to allow
the parasites sufficient time to reinvade before they are collected
and frozen. To quantify growth inhibition, the plates were thawed
at room temperature for at least 1 h, and 100 μL of lysis buffer
(20 mM Tris-HCl; 5 mM EDTA; 0.004% saponin and triton X-100) in PBS
containing Sybr Green I (1 μL in 5 mL) was added to each well
and mixed by pipetting up and down several times and incubated for
1 h in the dark with shaking. Using a Fluoroskan/ClarioStar plate
reader at excitation of 485 nm and emission of 538 nm, plate absorbances
were acquired. The data were normalized against the controls, and
graphs were generated using GraphPad prism 8 to determine the EC50 values using the nonlinear regression log (inhibitor) versus
response (three parameter) curve.
TR-FRET to Determine the
IC50 of the Inhibitors with
Full-Length PfCLK3 Recombinant Protein
The
TR-FRET assays, a high-throughput inhibition assay, as described previously,[13] was used to determine the potency of the small
molecules generated against full-length PfCLK3 recombinant
protein in a kinase buffer (containing 50 mM HEPES, 10 mM MgCl2, 2 mM DTT, 0.01% Tween 20, and 1 mM EGTA), with the ULight-labeled peptide substrate MBP peptide (sequence: CFFKNIVTPRTPPPSQGK).
First, in a 10 μL reaction volume, 5 μL of twice the required
enzyme concentration (50 nM) and 2.5 μL of four times the required
substrate concentration mix containing cold ATP, and the serially
diluted drugs were mixed in a black 384-plate well plate and incubated
at 37 °C for 1 h. The reaction was stopped after incubation by
adding the stopping/detection solution (containing 30 mM EDTA in 1×
Lance detection buffer and 3 nM Europium-labeled antiphospho-specific
antibody) and incubated for another hour at RT before phosphorylation
signals were measured using the ClarioStar.For each test compound,
percent inhibition (response) which was calculated using the formula: was plotted against
the log molar concentration
of compound to calculate the IC50 (potency) of each inhibitor
molecule and plotted using GraphPad prism software. All experiments
were carried out in triplicates, and the data presented were the standard
error of the mean (SEM) of three independent experiments run in triplicates.PfCLK3 phosphorylation of substrate results in
the Europium-labeled antiphospho-specific antibody recognizing the
phosphorylated site on the substrate. The Europium donor fluorophore
is excited at 320 or 340 nm, and energy is transferred to the ULight acceptor dye on the substrate, which finally results
in the emission of light at 665 nm. The level of ULight peptide phosphorylation correlates with the intensity of the emission.
For normalization, a no kinase and a no inhibitor reaction wells were
included, and all experiments were conducted in triplicates. Drug
dilutions, protein concentrations, and incubation times were the same
for easy comparison of results.
Microsomal Stability
Compounds were incubated at 37
°C at a concentration of 1 μM with CD1 mouse liver microsomes
(GIBCO, Thermo Fisher Scientific) in a suspension of 50 mM potassium
phosphate buffer (pH 7.4) with a final protein concentration of 0.5
mg/mL. The reaction was started by the addition of excess NADPH and
then quenched at several time points starting from time 0, then at
3, 6, 9, 15, and 30 min addition of acetonitrile to an aliquot of
the sample. An internal standard was added to each sample before centrifugation
to remove any precipitates before monitoring loss of parent compound
by HPLC analysis using Shimadzu LC-20A (Shimadzu, UK). Prism (Graphpad,
USA) was used to fit an exponential decay for substrate depletion
and subsequently rate constant (k) from the peak
area of the parent compound to the internal standard at each time
point. The rate of intrinsic clearance (CLint) was then
calculated according to the methods of Obach using the equation[37]where V is the incubation
volume (volume/mg protein), and microsomal protein yield is assumed
to be 52.5 mg protein/g liver with verapamil used as a positive control.
Distribution Coefficient (log D7.4)
Distribution coefficient (log D7.4)
was estimated by correlation of the compounds’ chromatographic
retention properties to those of 10 standard compounds with known
distribution coefficients ranging from −0.5 to 5.5 at pH 7.4.
A fast gradient HPLC methodology was used based on the method developed
by Valkó et al.[38]
Kinase Screen
Method
Each enzyme is assayed in its
linear range with 0.3 μM substrate in 50 mM Tris pH 7.5, 0.1
mM EGTA, 0.01 mM DTT, and relevant Mg/ATP (5, 20, or 50 μM)
for 30 min at room temp. Assays are stopped by the addition of 3%
orthophosphoric acid and harvested onto a p81 filter paper using the
PerkinElmer unifilter harvester. Once dried, they are read on a PerkinElmer
Topcount NXT scintillation counter for 30 s/well.[39]
Authors: Matt Addie; Peter Ballard; David Buttar; Claire Crafter; Gordon Currie; Barry R Davies; Judit Debreczeni; Hannah Dry; Philippa Dudley; Ryan Greenwood; Paul D Johnson; Jason G Kettle; Clare Lane; Gillian Lamont; Andrew Leach; Richard W A Luke; Jeff Morris; Donald Ogilvie; Ken Page; Martin Pass; Stuart Pearson; Linette Ruston Journal: J Med Chem Date: 2013-02-26 Impact factor: 7.446
Authors: John Liddle; Paul Bamborough; Michael D Barker; Sebastien Campos; Chun-Wa Chung; Rick P C Cousins; Paul Faulder; Michelle L Heathcote; Heather Hobbs; Duncan S Holmes; Chris Ioannou; Cesar Ramirez-Molina; Mary A Morse; Ruth Osborn; Jeremy J Payne; John M Pritchard; William L Rumsey; Daniel T Tape; Giorgia Vicentini; Caroline Whitworth; Rick A Williamson Journal: Bioorg Med Chem Lett Date: 2012-06-26 Impact factor: 2.823
Authors: Nicholas D Adams; Jerry L Adams; Joelle L Burgess; Amita M Chaudhari; Robert A Copeland; Carla A Donatelli; David H Drewry; Kelly E Fisher; Toshihiro Hamajima; Mary Ann Hardwicke; William F Huffman; Kristin K Koretke-Brown; Zhihong V Lai; Octerloney B McDonald; Hiroko Nakamura; Ken A Newlander; Catherine A Oleykowski; Cynthia A Parrish; Denis R Patrick; Ramona Plant; Martha A Sarpong; Kosuke Sasaki; Stanley J Schmidt; Domingos J Silva; David Sutton; Jun Tang; Christine S Thompson; Peter J Tummino; Jamin C Wang; Hong Xiang; Jingsong Yang; Dashyant Dhanak Journal: J Med Chem Date: 2010-05-27 Impact factor: 7.446
Authors: Judith Straimer; Nina F Gnädig; Benoit Witkowski; Chanaki Amaratunga; Valentine Duru; Arba Pramundita Ramadani; Mélanie Dacheux; Nimol Khim; Lei Zhang; Stephen Lam; Philip D Gregory; Fyodor D Urnov; Odile Mercereau-Puijalon; Françoise Benoit-Vical; Rick M Fairhurst; Didier Ménard; David A Fidock Journal: Science Date: 2014-12-11 Impact factor: 47.728
Authors: Kareem A Galal; Anna Truong; Frank Kwarcinski; Chandi de Silva; Krisha Avalani; Tammy M Havener; Michael E Chirgwin; Eric Merten; Han Wee Ong; Caleb Willis; Ahmad Abdelwaly; Mohamed A Helal; Emily R Derbyshire; Reena Zutshi; David H Drewry Journal: J Med Chem Date: 2022-09-27 Impact factor: 8.039