Antimalarial pyrido[1,2-a]benzimidazoles

Article abstract of DOI:10.1021/jm200227r

A novel class of antimalarial pyrido[1,2-a]benzimidazoles were synthesized and evaluated for antiplasmodial activity and cytotoxicity following hits identified from screening commercially available compound collections. The most active of these, TDR86919 (4c), showed improved in vitro activity vs the drug-resistant K1 strain of Plasmodium falciparum relative to chloroquine (IC₅₀ = 0.047 μM v 0.17 μM); potency was retained against a range of drug-sensitive and drug-resistant strains, with negligible cytotoxicity against the mammalian (L-6) cell line (selectivity index of >600). 4c and several close analogues (as HCl or mesylate salts) showed significant efficacy in P. berghei infected mice following both intraperitoneal (ip) and oral (po) administration, with >90% inhibition of parasitemia, accompanied by an increase in the mean survival time (MSD). The pyrido[1,2-a]benzimidazoles appeared to be relatively slow acting in vivo compared to chloroquine, and metabolic stability of the alkylamino side chain was identified as a key issue in influencing in vivo activity.

Full text of DOI:10.1021/jm200227r

ARTICLE
pubs.acs.org/jmc
Antimalarial Pyrido[1,2-a]benzimidazoles
Albert J. Ndakala,† Richard K. Gessner,† Patricia W. Gitari,† Natasha October,† Karen L. White,§
Alan Hudson,|| Foluke Fakorede,^ David M. Shackleford,§ Marcel Kaiser,# Clive Yeates,¥
Susan A. Charman,§ and Kelly Chibale*,†,‡
†Department of Chemistry, and ‡Institute of Infectious Disease and Molecular Medicine, University of Cape Town, Rondebosch 7701,
South Africa
§Centre for Drug Candidate Optimisation, Monash University (Parkville Campus), 381 Royal Parade, Parkville, Victoria 3052, Australia
Pharmacons, Kent, U.K.
^Tropical Diseases Research, World Health Organization, 20 Avenue Appia, CH-1211, Geneva 27, Switzerland
#Swiss Tropical and Public Health Institute, Postfach, Socinstrasse 57, 4002 Basel, Switzerland
¥InPharma Consultancy, Herts, U.K.
S Supporting Information
b
ABSTRACT: A novel class of antimalarial pyrido[1,2-a]benzimid-
azoles were synthesized and evaluated for antiplasmodial activity
and cytotoxicity following hits identified from screening commer-
cially available compound collections. The most active of these,
TDR86919 (4c), showed improved in vitro activity vs the drug-
resistant K1 strain of Plasmodium falciparum relative to chloro-
quine (IC50 = 0.047 μM v 0.17 μM); potency was retained against
a range of drug-sensitive and drug-resistant strains, with negligible
cytotoxicity against the mammalian (L-6) cell line (selectivity
index of >600). 4c and several close analogues (as HCl or mesylate
salts) showed significant efficacy in P. berghei infected mice following both intraperitoneal (ip) and oral (po) administration, with
>90% inhibition of parasitemia, accompanied by an increase in the mean survival time (MSD). The pyrido[1,2-a]benz-
imidazoles appeared to be relatively slow acting in vivo compared to chloroquine, and metabolic stability of the alkylamino side
chain was identified as a key issue in influencing in vivo activity.
’ INTRODUCTION
(1994), artemether/lumefantrine (1999), atovaquone/proguanil
(1999), chlorproguanil/dapsone (2003), but all come with some
issue limiting use.3 There is thus a clear need to develop new
more affordable and effective antimalarials.4
As part of a TDR collaboration with the Belgian company
Tibotec, in 2000 a small library of 1440 diverse nonpropriety
compounds donated by SPECS were screened against a panel of
protozoa in vitro. This led to the identification of a pyrido[1,2-a]-
benzimidazole coded TDR15087 (1, Figure 1), with moderate in
Malaria continues to take an enormous toll on human health,
particularly in tropical regions. The drugs used to treat this
disease are far from ideal, and many of these were introduced
decades ago. The utility of these medicaments in resource-poor
settings is limited by a number of factors, such as high cost, poor
compliance, drug resistance, low efficacy, and poor safety.1
Malaria results from infection with four different species of the
genus Plasmodium, all transmitted by mosquitoes, namely,
falciparum, vivax, ovale, and malariae. The most important of
these in terms of virulence and mortality is P. falciparum,
although P. vivax also has a huge impact on populations with
regard to morbidity.2 The enormous public health problem
posed by malaria across the developing world is reflected by
the grim estimates that each year the disease causes between 1.7
and 2.5 million deaths. Over the past few decades the mainstays
of antimalarial chemotherapy, chloroquine and pyrimethamine/
sulfadoxine, have been significantly compromised in many
regions by the spread of drug-resistant parasites. To counter
this, a range of newer drugs and combinations have gradually
been introduced into use, e.g., mefloquine (1984), artemisinins
vitro activity toward P. falciparum GHA and W2 strains (IC50
=
0.17ꢀ0.37 μM), significant because there is no published prior
art relating to the antimalarial activity of pyrido[1,2-a]benz-
imidazoles, although derivatives related to 1 have previously been
investigated for antifungal,5 antibacterial,6 and antitumor6ꢀ11
activity.
Other pyrido[1,2-a]benzimidazoles were then sought to
further explore the in vitro antimalarial structureꢀactivity rela-
tionships (SARs). Initially this involved the selection of 535
Received: February 28, 2011
Published: June 06, 2011
r
2011 American Chemical Society
4581
dx.doi.org/10.1021/jm200227r J. Med. Chem. 2011, 54, 4581–4589
|

Journal of Medicinal Chemistry
ARTICLE
commercially available analogues, primarily on the basis of
providing diversity around the pyridobenzimidazole “core”,
which were evaluated in a medium throughput screen (MTS)
using a multidrug resistant strain of P. falciparum (K1),
with cytotoxicity assessed against the murine L-6 cell-line; actives
from the MTS were re-evaluated for IC50 against both
P. falciparum and L-6. From this exercise, 49 compounds
were identified with P. falciparum IC50 of <0.1 μg/mL; notable
were the N-benzylpiperazinyl derivatives TDR35885 (2) and
TDR44047 (3) (Figure 1), which showed good selectivity and
greater activity in vitro compared to 1 (Table 2).
However, 1ꢀ3 all proved to be inactive in the standard
P. berghei mouse model at doses up to 4 ꢁ 100 mg/kg ip,
most likely because of a combination of poor solubility
and metabolic stability. To find compounds active in vivo in
the P. berghei mouse model, we have investigated the anti-
malarial SAR of pyridobenzimidazoles further. Here we de-
scribe part of these investigations focused on 3-aryl deriva-
tives, with alkylamino side chains (Figure 2). Metabolic
stability emerged as a significant factor influencing activity
in vivo, and in vitro microsome stability studies proved to be
useful in guiding the selection of compounds for animal
assessment.
’ RESULTS AND DISCUSSION
Chemistry. A relatively straightforward synthetic approach
(Scheme 1) was followed for the synthesis of target pyrido-
[1,2-a]benzimidazoles, adapting published procedures.6ꢀ12
The final stage involving nucleophilic substitution with the
appropriate amine was initially carried out with external heating,
requiring relatively long reaction times of up to 18 h; microwave
irradiation was subsequently found to reduce the reaction time to
approximately 20 min. As the pyridobenzimidazoles generally
show low aqueous solubility, hydrochloride salts were made of
compounds selected for in vivo evaluation to aid formulation; for a
few compounds mesylate salts were also made for comparison to
provide confidence that the salt form was not significantly affecting
activity. Chloroquine, administered as the free base, shows sig-
nificantly lower activity in the P. berghei mouse model compared to
the diphosphate salt, the standard used in our studies.
Hydrochloride salts were initially prepared by bubbling excess
HCl(g) through a mixture of the free base in dichloromethane or
more conveniently with HCl(aq) in methanol; the compounds in the
free base form were generally insoluble in methanol, and solubilization
occurred on addition of the aqueous acid. The latter method allowed
for easy isolation of the salts by removal of the solvent followed by
trituration with dichloromethane to remove any impurities. Mesylate
salts were prepared by addition of methanesulfonic acid to a solution
of the appropriate freebase in dichloromethane.
Biology. In Vitro Activity. Initial SAR studies (Figure 2,
Table 1) involved exploration around the aminoalkyl side chain
and variation in substitution at position R1 (including CF3 and
positions 30 and 40 substituted aromatic groups). Compounds
4aꢀ10f were evaluated for antimalarial activity against the
multidrug-resistant P. falciparum K1 strain and cytotoxicity
against the mammalian L-6 cell line (Table 1).
Many of the compounds showed antimalarial activity compar-
able to that of chloroquine (IC50 = 0.17ꢀ0.20 μM), with
compound 4c being the most active (IC50 = 0.047 μM, ∼3.5
times more potent than chloroquine) and 6a the least active
(IC50 = 4.48 μM, ∼22 times less potent than chloroquine and
∼95 times less potent than 4c). In general lipophilicity appears to
be an important factor influencing in vitro activity and selectivity,
although other structural features in the alkylamino side chain
also seem to have a significant influence:
Figure 1. Pyrido[1,2-a]benzimidazole screening hits with potent activ-
ity toward P. falciparum.
1. N-Alkylated derivatives (R3 = N(Me)2, N(Et)2, or N(H)-
(Et)) showed greater potency (∼2ꢀ20 times) than their
N-dealkylated (R3 = NH2) counterparts and also had more
favorable selectivity to L-6.
2. Phenyl or monosubstituted aryl groups at R1 in place of a CF3
group showed superior potency of up to 10-fold in all cases,
and activity was generally further enhanced when the R1 aryl
group was substituted (CF3, Cl, or F) at the 40 position.
Figure 2. Target pyrido[1,2-a]benzimidazoles for SAR exploration.
Scheme 1. General Synthetic Approach to 4-Cyanopyrido[1,2-a]benzimidazolesa
a Reagents and conditions: (i) NH4OAc, 140ꢀ150 ꢀC, 30ꢀ60 min; (ii) POCl3, reflux, 2 h; (iii) DMF or THF, Et3N, 80ꢀ90 ꢀC, 18 h or microwave
(150 W), 20 min; R1, R2, R3, n as defined in Table 1.
4582
dx.doi.org/10.1021/jm200227r |J. Med. Chem. 2011, 54, 4581–4589

Journal of Medicinal Chemistry
ARTICLE
Table 1. In Vitro Antimalarial Activity Evaluation of Pyrido[1,2a]benzimidazole Derivatives
IC50 (μM)a
cytotox (L6)
R1
R2
R3
n
log DpH7.4
P. fal (K1)
SI
b
compd
chloroquine
0.17ꢀ0.20
0.006ꢀ0.011
artemisinin
podophyllotoxin
0.009ꢀ0.014
5.11
11.7
18.9
40.1
>210
7.14
>80
4a
5a
6a
7a
4b
5b
6b
7b
8b
9b
10b
4c
5c
6c
7c
8c
9c
10c
4d
5d
6d
4e
5e
6e
7e
8e
9e
10e
4f
4-CF3Ph
3-CF3Ph
Ph
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Et
H
H
H
H
H
NH2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
2
1.80
1.80
0.83
1.37
2.45
2.45
1.48
2.01
1.41
1.50
2.05
2.94
2.94
1.89
2.51
1.90
1.99
2.54
1.46
1.46
0.49
2.49
2.49
1.52
2.06
1.45
1.54
2.09
2.97
2.97
2.00
2.54
1.93
2.02
2.57
3.52
2.01
3.78
2.58
3.72
3.79
1.49
1.00
4.48
3.09
0.13
0.29
1.49
3.43
0.28
0.84
0.31
0.047
0.42
1.11
1.90
0.25
0.55
0.06
0.92
0.63
2.27
0.44
0.38
1.52
3.16
0.27
0.53
0.21
0.30
0.19
1.53
2.08
0.71
0.46
0.17
0.94
0.12
0.054
0.053
0.49
1.22
3.4
NH2
11.7
4.2
NH2
CF3
NH2
13
4-CF3Ph
3-CF3Ph
Ph
N(Me)2
N(Me)2
N(Me)2
N(Me)2
N(Me)2
N(Me)2
N(Me)2
N(Et)2
N(Et)2
N(Et)2
N(Et)2
N(Et)2
N(Et)2
N(Et)2
NH2
>1600
24.6
>50
13
CF3
44.7
19.6
5.88
19.5
28.8
102
4-FPh
3-FPh
4-ClPh
4-CF3Ph
3-CF3Ph
Ph
70
7
62.9
612
243
108
63.7
226
225
683
6.8
120
CF3
121
4-FPh
3-FPh
4-ClPh
4-CF3Ph
3-CF3Ph
Ph
56.5
124
41
6.26
4.33
11.9
>200
180
NH2
6.9
NH2
5.2
4-CF3Ph
3-CF3Ph
Ph
N(Me)2
N(Me)2
N(Me)2
N(Me)2
N(Me)2
N(Me)2
N(Me)2
N(Et)2
N(Et)2
N(Et)2
N(Et)2
N(Et)2
N(Et)2
N(Et)2
N(H)(Et)
N(H)(Et)
piperidine
pyrrolidine
morpholine
morpholine
455
474
71
108
CF3
166
52.5
43
4-FPh
3-FPh
4-ClPh
4-CF3Ph
3-CF3Ph
Ph
11.6
15.4
>210
>190
24.3
>230
87.8
86.6
55.7
48.8
21.5
2.46
193
29
>1000
>633
128
>150
42.2
122
121
287
22.9
20.5
3581
2751
68
5f
6f
7f
CF3
8f
4-FPh
3-FPh
4-ClPh
4-CF3Ph
4-CF3Ph
4-CF3Ph
4-CF3Ph
4-CF3Ph
4-CF3Ph
9f
10f
4g
4h
4i
4j
146
4k
33.3
4l
98.3
80.6
a Mean from n g 2 independent experiments. Individual values varied by less than a factor of 2. b The log DpH7.4 values were calculated using the ACD
LogD Suite of software (version 9.0, Advanced Chemistry Development Inc., Toronto, Canada).
4583
dx.doi.org/10.1021/jm200227r |J. Med. Chem. 2011, 54, 4581–4589

Journal of Medicinal Chemistry
ARTICLE
Table 2. In Vitro Activity against Drug-Sensitive and Drug-Resistant Strains of P. falciparum
P. falciparum IC50 (μM)
W2 mef
HB3
NF54
FC27
FCR3
MAD20
K1
chloroquine
0.118
0.008
0.011
29.5
0.041
0.003
0.012
1.146
ND
0.006
0.005
0.013
0.019
ND
0.004
0.008
<0.040
<0.060
ND
0.085
0.005
0.005
0.012
ND
0.010
0.008
0.013
<0.060
ND
0.151ꢀ0.173
0.005
artemesinin
mefloquine
0.008
pyrimethamine
9.9
1
ND
0.17ꢀ0.36
0.05ꢀ0.078
0.023ꢀ0.095
0.047ꢀ0.058
0.116
2
ND
ND
0.052
0.053
0.11
ND
ND
ND
3
ND
ND
ND
ND
ND
4c
4h
0.084
0.123
0.060
0.090
0.175
0.11
0.137
0.154
0.104
0.106
0.11
Table 3. In Vivo Antimalarial Activity of Pyrido[1,2a]benzimidazole Aminoalkyl Derivatives and Corresponding Salts
P. berghei in vivo
% reduction parasitemia
IC50 (μM)
cytotox (L6)
(MSD)a
compd
X
R
P. fal (K1)
ip
po
chloroquineb
99.6% (20)
99.9% (17)
control
4b
(5ꢀ7)
40-CF3
40-CF3
40-CF3
40-CF3
40-CF3
40-CF3
40-CF3
40-CF3
30-CF3
30-CF3
40-CF3
40-CF3
30-CF3
30-CF3
ꢀNH(CH2)2N(Me)2
2HCl salt of 4b
0.13
0.1
>210
147.2
>175
28.8
89.6
123.6
>200
21.1
180
4ba
4bb
4c
0%c
mesylate salt of 4b
ꢀNH(CH2)2N(Et)2
2HCl salt of 4c
0.14
0.05
0.08
0.09
0.44
0.26
0.38
0.49
0.30
0.16
0.19
0.25
94.1% (11)
96.8% (13.7)
4ca
4cb
4e
91.9% (10)
95.2% (11)
96.2% (16)
97.4% (11)
mesylate salt of 4c
ꢀNH(CH2)3N(Me)2
2HCl salt of 4e
4ea
5e
30.7%c
0%c
ꢀNH(CH2)3N(Me)2
2HCl salt of 5e
5ea
4f
19.2
>190
2.23
24.3
11.8
ꢀNH(CH2)3N(Et)2
2HCl salt of 4f
4fa
5f
72.6% (9)
0%c
93.4% (11)
ꢀNH(CH2)3N(Et)2
2HCl salt of 5f
5fa
a MSD = mean survival time (in days). 50 (mg/kg)/day ꢁ 4 ip or 100 (mg/kg)/day ꢁ 4 po (formulated in 10% aq DMSO). b Chloroquine diphosphate
at 10 (mg/kg)/day ꢁ 4 ip or po. c Mice euthanized on day 4, 24 h after last treatment, because of inactivity.
3. Variation in the alkyl chain length had practically no
significant effect on the in vitro antiplasmodial activity,
although this was only varied by one methylene unit.
4. The morpholino derivative 4k was ∼10ꢁ less active than
other cycloalkylamino derivatives 4i and 4j.
5. In the one example in which there was additional alkyl
substitution on the N attached to the ring (4g), activity was
reduced compared to the des-ethyl derivative (4h).
more or less invariant across a broad range of drug sensitive
and drug resistant strains of P. falciparum (Table 2).
In Vivo Studies. Six compounds with the most favorable
activity and selectivity from the in vitro screen (4b, 4c, 4e, 5e,
4f, and 5f) were chosen for preliminary in vivo evaluation in P.
berghei infected mice. The corresponding hydrochloride salts
(4baꢀ5fa), as well as two mesylate salts (4bb and 4cb), were
used for the in vivo studies. At a repeat dose of 50 (mg/kg)/day ip
for 4 days four of these (4bb, 4ca, 4cb, and 4fa) showed significant
efficacy. Subsequently they were also found to be active following
oral (po) administration with 100 (mg/kg)/day ꢁ 4, giving >90%
6. 4h, the N-des-ethyl metabolite of compound 4c, retained
its activity to an extent, albeit with reduced selectivity vs
L-6.The activity of compounds 4c and 4h was found to be
4584
dx.doi.org/10.1021/jm200227r |J. Med. Chem. 2011, 54, 4581–4589

Journal of Medicinal Chemistry
ARTICLE
Table 4. Oral Efficacy of Selected Compounds (Bis-hydrochloride Salts) in the P. berghei Mouse Modela
a Test compounds were formulated in hydroxypropyl methylcellulose (HPMC) and administered po (3 mice per group).
inhibition of P. berghei parasitemia and a significant increase in the
mean survival time (MSD) of the mice from 10 to 16 days, relative
to controls (5ꢀ7 days, Table 3). The lack of in vivo activity with
the hydrochloride salt of 4b (4ba) compared to the mesylate salt
(4bb) when given ip may be due to a poorer solubility or
dissolution rate adversely affecting distribution.
In single dose studies in the P. berghei mouse model (Table 4),
the hydrochloride salts of four of the most active compounds 4ca,
4ha, 4ja, 4ia showed significant activity at 25 mg/kg, which was
not significantly improved on going to a higher dose of 50 mg/kg,
presumably because of saturation of systemic exposure; 4ia was
ineffectual at 50 mg/kg, possibly because of a problem with
formulation arising from poor solubility (cloudiness of the
solution was observed during administration of 4ia at both
doses). In any case the rate of reduction in parasitemia after
treatment with the pyridobenzimidazoles was significantly slower
than with chloroquine, with maximum reduction generally ob-
served on day 5 compared to day 2 with chloroquine.
In Vitro Metabolic Stability. Studies were conducted using
human, mouse, and rat liver microsomes (Table 5). Compounds
4ca and 4jaꢀ4ka exhibited moderate to high rates of metabolic
degradation, while degradation of 4ha was in the low to moderate
range. Generally for the alkylated derivatives 4ca and 4ha,
N-dealkylated metabolites were observed, and for the cycloalky-
lated derivatives (4jaꢀ4la), there was evidence for deamination,
N-dealkylation, or ring cleavage.
administration of the bis-hydrochloride salts (i.e., 4ca and 4ha) at
dose levels of 5 mg/kg intravenously and 20 mg/kg orally to male
SpragueꢀDawley rats (Table 6).
For compound 4c, the apparent half-life rangedbetween6 and 8
h; volume of distribution was high, and plasma clearance was
moderate. The rate of absorption was relatively slow after oral
administration, and the apparent oral bioavailability was ∼22%.
The des-ethyl derivative 4h was identified as a metabolite of 4c in
plasma samples following both routes of administration, in agree-
ment with the in vitro microsome studies. The in vivo conversion
of 4c to 4h was estimated to be approximately 70%, suggesting that
N-dealkylation is likely to be a major in vivo clearance pathway for
4c. Direct urinary excretion of 4c was negligible following both
intravenous and oral administration. When 4h was administered
intravenously as the bis-hydrochloride salt (4ha) at a dose of 5
mg/kg, it exhibited a long apparent half-life, high volume of
distribution, and low plasma clearance; direct urinary excretion
of 4h was negligible. Compound 4h exhibited high protein binding
in human and mouse plasma with fraction bound values being in
the ranges 98.3ꢀ99.9% and 95.3ꢀ96.9%, respectively. In compar-
ison 4c had 96.7ꢀ99.2% (human) and 97.9% (mouse) protein
binding. On the basis of the high conversion of 4c to 4h observed
in vivo, the long half-life, and low blood clearance of 4h and its
intrinsic activity, it is likely that the N-dealkylation product 4h
contributes significantly to the efficacy of 4c.
Single-dose and multidose oral efficacy studies of 4ha against
P. berghei ANKA GFP in mice indicated that the activity plateaus
(reaches a limit) with dose concentrations higher than 25 mg/kg
In Vivo Pharmacokinetic Studies. The in vivo pharmacoki-
netic properties of both 4c and 4h were assessed following
4585
dx.doi.org/10.1021/jm200227r |J. Med. Chem. 2011, 54, 4581–4589

Journal of Medicinal Chemistry
ARTICLE
Table 5. In Vitro Microsomal Stability
a EH (hepatic extraction ratio) = fraction of dose entering liver, which is metabolized during one pass through the liver. b P-28: monodeethylation. P-56:
bis-deethylation. P + 14: carbonyl addition. P-53: deamination of the pyrrolidine ring. P-97, P-68, P-127, P-113: N-dealkylation of the respective R side
chain. P + 16: oxygenation. P-70: morpholine dealkylation. P-26: morpholine ring cleavage.
Table 6. Pharmacokinetic Parameters for 4c and 4h after iv and Oral Administration of Their Bis-hydrochloride Salts (4ca and
4ha) to Male Rats
4c
4h (after administration of 4ca)
4h
parameter
iva
orala
iva
orala
iva
2.6
orala
nominal dose (mg/kg)b
apparent t1/2 (h)
plasma CLtotal
VZ (L/kg)
4.3
19.0
6.4
19.1
cncd
7.3
cncd
cncd
∼16.3
5.4
30
18.9
0.39
7.5
% dose in urinec
0.05
0.38
248
290
>22
0.08
0.54
640
666
0.05
0.92
1200
1032
1.13
0.21
0.95
960
Cmax (μM)
Tmax (min)
AUC0ꢀtlast (μM min)
304
1000
2076
>28e
3
bioavailability (%)
a Compounds were formulated as suspensions in 0.5% hydroxypropyl methylcellulose. Values are the mean from two animals. b As the
bishydrochloride salt. c % of dose present in pooled urine (collected over 0ꢀ24 or 0ꢀ48 h). d Given the flat nature of the profile, the terminal
elimination half-life could not be determined. e The terminal elimination half-life and oral bioavailability could not be accurately determined. On the basis
of the dose-normalized AUC0ꢀtlast, the exposure after oral administration was approximately 25ꢀ30% of that after iv administration; however, this will
likely be an underestimation of the actual oral bioavailability.
in vivo (Tables 7 and 8). A similar plateau in efficacy was also
observed following subcutaneous administration.
As a preliminary indication of systemic exposure, in vivo studies
in mice following oral and subcutaneous administration of 4ha at
4586
dx.doi.org/10.1021/jm200227r |J. Med. Chem. 2011, 54, 4581–4589

Journal of Medicinal Chemistry
ARTICLE
Table 7. Single-Dose Oral Efficacy of 4ha against P. berghei
ANKA GFP in Mice
escalating dose levels (6ꢀ100 mg/kg) were conducted. Exposure
was found to become saturated at dose levels above 25 mg/kg
(orally) or 50 mg/kg (subcutaneously), consistent with the
observed plateau in efficacy (Figure 3). It is likely that solubility
or dissolution-limited absorption is a contributory factor in limit-
ing systemic exposure and efficacy with increasing dose.
% reduction parasitemia
compd
4ha
dose (mg/kg)a day 2 day 3 day 4 day 5 MSD (day)
1 ꢁ 25
1 ꢁ 50
1 ꢁ 10
23.51 55.45 62.25 68.92
30.78 50.24 71.22 79.79
99.9
13.3
12.7
9
4ha
’ CONCLUSIONS
chloroquine
control
a Compounds were formulated in HPMC.
A novel series of pyrido[1,2-a]benzimidazole derivatives have
been identified that combine good in vitro activity against P.
falciparum with oral efficacy in a P. berghei mouse model. The
pyridobenzimidazoles appear to be slower acting in vivo relative
to chloroquine, pointing to a different mode of action, which has
not been established. The most significant feature of the series is
that the pharmacokinetic profile of the lead compounds needs
considerable improvement if a pyridobenzimidazole is to be
identified as worthy of further progression. Although 4h shows
good stability in rat and mouse microsomes and has a long half-
life in rats, pharmacokinetic studies indicate oral absorption
becomes saturated at relatively low doses, most likely because
of poor dissolution or solubility. Further work is needed to
identify compounds that have the potential for improved phar-
macokinetics, most likely achievable through a combination of
improved solubility and metabolic stability.
5
Table 8. Multidose Oral Efficacy of 4ha against P. berghei
ANKA GFP in Mice
% reduction
4ha (mg/kg)a
cured/infected
parasitemia at day 4
MSD (day)
4 ꢁ 3
0/3
0/3
0/3
0/3
0/3
0
4b
4 ꢁ 6
38.35
80.73
95.72
95.56
7
4 ꢁ 12.5
4 ꢁ 25
4 ꢁ 50
control
14
14.3
14
6.5
a Compounds were formulated in HPMC. b Mice euthanized on day 4,
24 h after last treatment, because of inactivity.
’ EXPERIMENTAL SECTION
Chemistry. All commercially available chemicals were purchased
from either Sigma-Aldrich or Merck. All solvents were dried by appro-
priate techniques. Unless otherwise stated, all solvents used were
anhydrous. Reactions were monitored by TLC using Merck silica gel
plates (60 F-254), and were visualized by ultraviolet light. Silica gel
chromatography was performed using Merck Kieselgel 60: 70ꢀ230
mesh for gravity columns. Melting points were determined on a Reich-
ert-Jung Thermovar hotstage microscope and are uncorrected. Infrared
spectra were recorded on a Thermo Nicolete FTIR instrument in the
4000ꢀ500 cmꢀ1 range using KBr disks. Microanalyses were determined
using a Fisons EA 1108 CHNO-S instrument. Mass spectra were
recorded at the School of Chemistry, University of the Witwatersrand,
South Africa. NMR spectra were recorded on either a Varian Mercury
300 (1H, 300.13 MHz; 13C, 75.5 MHz) or a Varian Unity 400 (1H,
400.13 MHz; 13C, 100.6 MHz) spectrometer. Chemical shifts (δ) are
given in ppm downfield from TMS as the internal standard. Coupling
constants, J, are recorded in hertz (Hz). LC purity traces were obtained
using the Kinetex C18 (2.1 mm ꢁ 150 mm, 2.6 mm fused-core particles)
column, 1 mL injection volume, flow of 0.4 mL/min, gradient 0ꢀ100%
B in 9 min (hold 3 min) (mobile phase A of 10 mM ammonium formate,
pH 3, in 10% MeCN and mobile phase B of 10 mM ammonium formate,
pH 3, in 90% MeCN) with a diode array detector operating at a
wavelength range from 190 to 400 nm.
Purity was determined by combustion analysis and/or HPLC, and all
compounds were confirmed to have >95% purity.
General Procedure for the Synthesis of 1-Oxo-3-alkyl/
aryl-5H-pyrido[1,2-a]benzimidazole-4-carbonitriles (A). A
mixture of 2-benzimidazole acetonitrile (1.0 g, 6.36 mmol), NH4OAc
(0.98 g, 12.72 mmol), and ethyl (4-alkanoyl/aroyl)acetate (7.63 mmol)
was heated to reflux at 150 ꢀC for 1 h and allowed to cool to 100 ꢀC.
MeCN (10 mL) was added. The mixture was stirred for 15 min, allowed
to cool to room temperature, and then cooled on ice. The cold mixture
was filtered and the residue washed with cold MeCN (4 ꢁ 10 mL), dried
in vacuo, and used without further purification.
1-Oxo-3-[4-(trifluoromethyl)phenyl]-5H-pyrido[1,2-a]benz-
imidazole-4-carbonitrile. Silvery tan powder, mp 341ꢀ342 ꢀC
Figure 3. Systemic exposure of 4h following single dose oral and
subcutaneous administration of the bis-hydrochloride salt (4ha) to mice.
4587
dx.doi.org/10.1021/jm200227r |J. Med. Chem. 2011, 54, 4581–4589

Journal of Medicinal Chemistry
ARTICLE
(ethanol); IR (KBr) 3250ꢀ2500 bm, 2203 m (CN), 1664 s (CO), 1548
s, 1509 s, 1458 m, 1324 s, 1169 m, 1107 s, 1075 m, 1064 s cmꢀ1; 1H
NMR (400 MHz, DMF-d7) δ 8.64 (1H, d, J = 7.3 Hz, ArH), 7.93 (4H, s,
ArH), 7.64 (1H, d, J = 7.3 Hz, ArH), 7.57 (1H, dd, J = 7.3, 7.3 Hz, ArH),
7.42 (1H, dd, J = 8.8, 7.3 Hz, ArH), 6.11 (1H, s, dCHꢀ); 13C NMR
(100 MHz, DMF-d7) δ 157.9, 150.9, 147.2, 140.9, 131.6, 130.0, 129.7,
128.4, 127.6, 126.4, 125.1, 122.5, 122.3, 115.9, 115.7, 111.0, 104.6, 67.9;
LRMS (EI) m/z 354.2 (M + H).
General Procedure for the Synthesis of 1-(Alkylamino/
piperido/morpholino/pyrrolidino)ethyl/propylamino)-3-al-
kyl/arylpyrido[1,2-a]benzimidazolecarbonitrile Bis-hydro-
chloride Salts. HCl in methanol (1.25 M, 0.57 mL, 1.32 mmol) was
added to a stirred mixture of the pyrido[1,2-a]benzimidazole-carboni-
trile derivative (0.354 mmol) in methanol (20 mL). After the mixture
was stirred at room temperature for 2.5 h, the solvent was removed in
vacuo and the residue washed with minimum amounts of ice-cold
methanol followed by DCM (4 ꢁ 3 mL), dried in vacuo, and used
without further purification.
General Procedure for the Synthesis of 1-Chloro-3-alkyl/
arylpyrido[1,2-a]benzimidazole-4-carbonitriles (B). A mixture
1-(2-Diethylaminoethylamino)-3-[4-(trifluoromethyl)phenyl]-
pyrido[1,2-a]benzimidazole-4-carbonitrile Bis-hydrochlor-
ide Salt (4ca). Pale yellow, mp 161ꢀ162 ꢀC; H NMR (300 MHz,
of
1-oxo-3-alkyl/aryl-5H-pyrido[1,2-a]benzimidazole-4-carbonitrile
(2.83 mmol) and POCl3 (8.69 g, 5.28 mL, 56.68 mmol) was heated to
reflux at 130 ꢀC for 2 h. Excess POCl3 was removed under reduced
pressure and ice-cold water (20 mL) added to the residue, stirring to yield
a precipitate. The mixture was neutralized with saturated NaHCO3 and
filtered. The resultant solid was washed with ice-cold water (4 ꢁ 15 mL),
dried in vacuo, and used without further purification.
1
DMSO) δ 11.07 (1H, broad s, NH), 8.88 (1H, d, J = 8.4 Hz, ArH), 8.00
(4H, q, J = 8.4 Hz, ArH), 7.87 (1H, d, J = 7.5 Hz, ArH), 7.68 (1H, d, J =
7.7 Hz, ArH), 7.48 (1H, ddd, J = 1.2, 7.6, 8.4, ArH), 6.74 (1H,
s, dCHꢀ), 4.14 (2H, t, J = 6.6 Hz), 3.47 (2H, td, J = 5.1, 6.0 Hz), 3.23
(2H, quintet, J = 4.8, 7.2 Hz), 1.25 (6H, t, J = 7.2 Hz). 13C NMR (100
MHz, DMSO): δ 151.2, 148.4, 147.6, 130.0, 127.1, 125.5, 121.7, 116.2,
93.4, 48.9, 46.4, 8.1 (CH3). LRMS (APCI): m/z 452 (M+ + 1 ꢀ HCl).
1-(2-Ethylaminoethylamino)-3-[4-(trifluoromethyl)phenyl]-
pyrido[1,2-a]benzimidazole-4-carbonitrile Bis-hydrochloride
Salt (4ha). Pale yellow solid, mp 181ꢀ183 ꢀC; purity 97.8% by LC
(tR = 5.94 min); 1H NMR (300 MHz, DMSO) δ 9.36 (2H, broad s),
8.85 (1H, d, J = 8.4 Hz, ArH), 8.00 (5H, m, ArH), 7.88 (1H, d, J = 8.1 Hz,
ArH), 7.68 (1H, t, J = 7.5 Hz, ArH), 7.48 (1H, t, J = 7.5 Hz, ArH), 6.70
(1H, broad s, dCHꢀ), 4.08 (2H, m, ꢀNHCH2Cꢀ), 3.31 (2H, m,
ꢀNHCH2CH3), 3.02 (2H, m, ꢀNHCH2Cꢀ), 1.25 (3H, m, ꢀCH3);
13C NMR (100 MHz, DMSO) δ 151.0, 148.5, 147.6, 140.3, 129.7, 127.1,
126.8, 125.4, 121.4, 116.5, 116.0, 92.8, 44.3, 42.0, 10.7.
1-Chloro-3-[4-(trifluoromethyl)phenyl]pyrido[1,2-a]benz-
imidazole-4-carbonitrile. Yellow solid, mp 247ꢀ248 ꢀC(ethanol);IR
(KBr) 3094 w, 3065 w, 2217 m (CN), 1617 w, 1591 m, 1487 s, 1444 s, 1339 s,
1172 s, 1129 s, 1067 s cmꢀ1; 1H NMR (300 MHz, CDCl3) δ8.59 (1H, d, J=
8.8 Hz, ArH), 8.10 (1H, d, J = 7.8 Hz, ArH), 7.85 (4H, s, ArH), 7.67 (1H, dd,
J = 8.8, 6.8 Hz, ArH), 7.49 (1H, dd, J = 8.8, 5.9 Hz, ArH), 7.05 (1H,
s, dCHꢀ); 13C NMR (100 MHz, CDCl3) δ 147.2, 145.3, 138.6, 134.5,
132.7, 132.3, 132.0, 129.7, 129.1, 127.3, 126.2, 125.0, 123.0, 122.3, 120.8, 115.4,
114.2, 111.8, 98.4; LRMS (EI) m/z 372.1 (M + H), 374.0 (M + 2 + H);
General Procedure for the Synthesis of 1-[(Alkylamino/
piperido/morpholino/pyrrolidino)ethyl/propylamino]-3-alkyl/
arylpyrido[1,2-a]benzimidazolecarbonitriles (C). Method 1.
The appropriate amine (2.69 mmol) was added to a stirred mixture
of the 1-chloro-3-alkyl/arylpyrido[1,2-a]benzimidazole-4-carboni-
trile (1.345 mmol) and triethylamine (0.27 g, 0.37 mL, 2.69 mmol)
in THF or DMF (10 mL). The mixture was heated at 80ꢀ90 ꢀC for 18 h,
filtered hot, and allowed to cool. The solvent was removed in vacuo, and
the residue was washed with minimum amounts of ice-cold ethanol. The
resulting solid was recrystallized from acetone or ethanol.
’ ASSOCIATED CONTENT
S
Supporting Information. Additional details of the char-
b
acterization of selected compounds and the procedures used for
the in vitro and in vivo antimalarial studies and cytotoxicity
assays. This material is available free of charge via the Internet at
http://pubs.acs.org.
Method 2. Microwave irradiation (150 W) was substituted for
external heating, reducing the reaction time to approximately 20 min.
Workup followed the same protocol as method 1.
1-(2-Diethylaminoethylamino)-3-[4-(trifluoromethyl)phenyl]-
pyrido[1,2-a]benzimidazole-4-carbonitrile (4c). Yellow pow-
der, mp 219ꢀ220 ꢀC (ethanol); purity 98% by LC (tR = 5.82 min); IR
(KBr) 3333 bm, 2971 m, 2841 w, 2210 s (CN), 1624 s, 1595 s, 1552 s,
’ AUTHOR INFORMATION
Corresponding Author
*Phone: +27 21 650 2553. Fax: +27 21 650 5195. E-mail:
Kelly.Chibale@uct.ac.za.
1458 m, 1371 m, 1371 m, 1324 s, 1165 m, 1129 s, 1067 s, 1014 m cmꢀ1
;
1H NMR (300 MHz, CDCl3) δ 8.13 (1H, d, J = 7.8 Hz, ArH), 8.05 (1H,
d, J = 7.8 Hz, ArH), 7.83 (2H, d, J = 7.8 Hz, ArH), 7.77 (2H, d, J = 8.8 Hz,
ArH), 7.59 (1H, t, J = 7.8 Hz, ArH), 7.44 (1H, br s, NH), 7.36 (1H, dd,
J = 7.8, 6.8 Hz, ArH), 5.90 (1H, s, dCHꢀ), 3.49 (2H, t, J = 5.9 Hz,
CH2CH2N(Et)2), 2.97 (2H, t, J = 5.9 Hz, CH2CH2N(Et)2), 2.73 (4H, q,
’ ACKNOWLEDGMENT
We thank the WHO Special Programme for Research and
Training in Tropical Diseases for financial support for this research
(Project A50868). The University of Cape Town, South African
Medical Research Council (MRC), and South African Research
Chairs Initiative (SARChI) of the Department of Science and
Technology (DST) administered through the South African National
Research Foundation are gratefully acknowledged for support (K.C.).
We also acknowledge the generous gifts of the early samples of
TDR15087 and related analogues from SPECS, with the valuable
roles of Herman Verheij in the initial selection of the 1440 compound
library, Louis Maes in the screening of this library at Tibotec, and Reto
Brun in overseeing further screening at Swiss TPH.
J = 7.8 Hz, N(CH2CH3)2), 1.16 (6H, t, J = 7.8 Hz, N(CH2CH3)2); 13
C
NMR (75 MHz, CDCl3) δ 150.5, 149.0, 147.9, 145.7, 141.0, 129.0,
128.0, 126.0, 125.7, 121.0, 120.1, 116.7, 112.8, 89.3, 50.1, 46.0, 39.9,
11.7; LRMS (EI) m/z 452.2 (M + H).
1-(2-Ethylaminoethylamino)-3-[4-(trifluoromethyl)phenyl]-
pyrido[1,2-a]benzimidazole-4-carbonitrile (4h). Yellow fluffy
powder, mp 222ꢀ224 ꢀC (dec). Purity >99% by LC (tR = 5.94 min). 1H
NMR (400 MHz, CDCl3) δ: 8.57 (1H, d, J = 8.4 Hz, ArH), 7.94 (4H, m,
ArH), 7.79 (1H, d, J = 8.0 Hz, ArH), 7.52 (1H, m, ArH), 7.33 (1H, m,
ArH), 6.18 (1H, s, =CH-), 3.11 (2H, t, J = 6.2 Hz, -NHCH2Cꢀ), 2.83
(2H, q, J = 7.2 Hz, -NHCH2CH3), 2.48 (2H, m, -NHCH2Cꢀ), 1.14
(3H, t, J = 7.2 Hz, ꢀCH3). 13C NMR (100 MHz, CDCl3) δ: 150.5,
149.1, 147.9, 145.8, 141.1, 129.0, 128.1, 126.1, 125.8, 121.0, 120.3, 116.7,
112.9, 89.3, 50.1, 46.1, 40.0, 11.8. m/z (EI, positive ion) 423.5 (M+), 417,
352, 333, 274, 237, 210, 162, 88.
’ ABBREVIATIONS USED
ip, intraperitoneal; po, oral administration;HPLC, high pressure
liquid chromatography;HPMC, hydroxypropyl methylcellulose;
4588
dx.doi.org/10.1021/jm200227r |J. Med. Chem. 2011, 54, 4581–4589

Journal of Medicinal Chemistry
ARTICLE
MSD, meansurvival time; MTS, mediumthroughputscreen; SAR,
structureꢀactivity relationship; TDR, tropical diseases research;
TLC, thin layer chromatography;TMS, tetramethylsilane; WHO,
World Health Organization
’ REFERENCES
(1) Nwaka, S.; Ridley, R. G. Virtual drug discovery and development
for neglected diseases through publicꢀprivate partnerships. Nat. Rev.
Drug Discovery 2003, 2, 919–928.
(2) Ridley, R. G.; Hudson, A. Quinoline antimalarials. Expert Opin.
Ther. Pat. 1998, 8, 121–136.
(3) Pink, R.; Hudson, A.; Mouries, M. A.; Bendig, M. Opportunities
and challenges in antiparasitic drug discovery. Nat. Rev Drug Discovery
2005, 4, 727–740.
(4) Ridley, R. G. Medical need, scientific opportunity and the drive
for antimalarial drugs. Nature 2002, 415, 686–693.
(5) Takemura, M.; Takashi, H.; Kawakami, K.; Takeshita, H.;
Kimura, Y.; Watanabe, J.; Sugimoto, Y.; Kitamura, A.; Nakajima, R.;
Kanai, K.; Fujisawa, T. Imidazo[1,2-a]pyridine derivative. Eur. Pat. Appl.
1479681, 2004.
(6) Rida, S. M.; Soliman, F. S. G.; Badawey, E.-S. A. M.; El-Ghazzawi,
E.; Kader, O.; Kappe, T. Benzimidazole condensed ring systems. 1.
Syntheses and biological investigations of some substituted pyrido[1,2-
a]benzimidazoles. J. Heterocycl. Chem. 1988, 25, 1087–1093.
(7) Badawey, E.-S. A. M.; Kappe, C. Benzimidazole condensed ring
system. IX. Potential antineoplastics. New synthesis of some pyrido[1,2-
a]benzimidazoles and related derivatives. Eur. J. Med. Chem. 1995,
30, 327–332.
(8) Badawey, E.-S. A. M.; Kappe, C. Synthesis of some new imidazo-
[1,2-a]pyrimidine-5(1H)-ones as potential antineoplastic agents. J. Het-
erocycl. Chem. 1995, 32, 1003–1006.
(9) Russell, R. K.; van Niewalt, C. E. The synthesis of
cycloalkylpyrido[1,2-a]benzimidazole carbonitrile analogs. J. Heterocycl.
Chem. 1995, 32, 299–306.
(10) Badawey, E.-S. A. M.; Kappe, T. Benzimidazole condensed ring
systems. XI. Synthesis of some substituted cycloalkyl pyrido[1,2-
a]benzimidazoles with anticipated antineoplastic activity. Eur. J. Med.
Chem. 1999, 34, 663–637.
(11) El-Hawash, S. A. M.; Badawey, E.-S. A. M.; Kappe, T. Benzi-
midazole condensed ring systems. XII. Synthesis and anticancer evalua-
tion of certain pyrido[1,2-a]benzimidazole derivatives. Pharmazie 1999,
54, 341–346.
(12) Rida, S. M.; Soliman, F. S. G.; Badawey, E.-S. A. M.; El Sayed,
A. M.; Kappe, T. Benzimidazole condensed ring systems. 2. New
synthesis of substituted 1-oxo-1H,5H-pyrido[1,2-a] benzimidazole-4-
carbonitriles and related derivatives. J. Heterocycl. Chem. 1988, 25,
1725–1728.
4589
dx.doi.org/10.1021/jm200227r |J. Med. Chem. 2011, 54, 4581–4589