Pyridyl benzamides as a novel class of potent inhibitors for the kinetoplastid Trypanosoma brucei

Article abstract of DOI:10.1021/jm500191u

A whole-organism screen of approximately 87000 compounds against Trypanosoma brucei brucei identified a number of promising compounds for medicinal chemistry optimization. One of these classes of compounds we termed the pyridyl benzamides. While the initial hit had an IC₅₀ of 12 μM, it was small enough to be attractive for further optimization, and we utilized three parallel approaches to develop the structure-activity relationships. We determined that the physicochemical properties for this class are generally favorable with particular positions identified that appear to block metabolism when substituted and others that modulate solubility. Our most active compound is 79, which has an IC₅₀ of 0.045 μM against the human pathogenic strain Trypanosoma brucei rhodesiense and is more than 4000 times less active against the mammalian L6 cell line.

Full text of DOI:10.1021/jm500191u

Article
pubs.acs.org/jmc
Pyridyl Benzamides as a Novel Class of Potent Inhibitors for the
Kinetoplastid Trypanosoma brucei
Lori Ferrins,†,□ Michelle Gazdik,∥,⊥,□ Raphael Rahmani,† Swapna Varghese,∥ Melissa L. Sykes,#
̈
Amy J. Jones,# Vicky M. Avery,# Karen L. White,‡ Eileen Ryan,‡ Susan A. Charman,†,‡ Marcel Kaiser,∇,○
Christel A. S. Bergstrom,§,◆ and Jonathan B. Baell*,†,⊥,¶
̈
†Medicinal Chemistry, ‡Centre for Drug Candidate Optimisation, and §Drug Delivery, Disposition and Dynamics, Monash Institute
of Pharmaceutical Sciences, Monash University, Parkville, Victoria 3052, Australia
∥Department of Chemistry, La Trobe Institute for Molecular Science, La Trobe University, Melbourne, Victoria 3086, Australia
⊥The Walter and Eliza Hall Institute, 1G Royal Parade, Parkville, Victoria 3052, Australia
#Eskitis Institute for Drug Discovery, Griffith University, Brisbane Innovation Park, Don Young Road, Nathan, Queensland 4111,
Australia
∇Swiss Tropical and Public Health Institute, Socinstrasse 57, Basel, 4051, Switzerland
○University of Basel, Petersplatz 1, Basel, 4003, Switzerland
◆Department of Pharmacy, Uppsala University, Biomedical Center P.O. Box 580, SE-751 23 Uppsala, Sweden
¶Department of Medical Biology, University of Melbourne, Parkville, Victoria 3010, Australia
S
* Supporting Information
ABSTRACT: A whole-organism screen of approximately
87000 compounds against Trypanosoma brucei brucei identified
a number of promising compounds for medicinal chemistry
optimization. One of these classes of compounds we termed
the pyridyl benzamides. While the initial hit had an IC50 of
12 μM, it was small enough to be attractive for further
optimization, and we utilized three parallel approaches to
develop the structure−activity relationships. We determined
that the physicochemical properties for this class are generally
favorable with particular positions identified that appear to
block metabolism when substituted and others that modulate solubility. Our most active compound is 79, which has an IC50 of
0.045 μM against the human pathogenic strain Trypanosoma brucei rhodesiense and is more than 4000 times less active against the
mammalian L6 cell line.
poor coordination, and sleep disturbances.2 Without treatment
HAT is eventually fatal. There are currently few treatment
INTRODUCTION
■
Human African trypanosomiasis (HAT) currently affects more
options, making this disease a key target for drug discovery.
than 20000 people according to the latest report published by
For treatment of the first stage of HAT, there are currently
the World Health Organization.1 HAT is caused by the
transmission of Trypanosoma brucei spp. by the tsetse fly to
two drugs registered for use, as shown in Figure 1: pentamidine
(1) and suramin (2). Pentamidine is used to treat infection
humans, and, as a result, this disease is largely restricted to
with T.b. gambiense,3 while suramin is prescribed to patients
Africa where the vector, parasite, and animal reservoirs coexist.2
infected with T.b. rhodesiense.4,5 Neither of these drugs can
Infection is with either Trypanosoma brucei gambiense or
cross the BBB, and they both have significant side effects.
Trypanosoma brucei rhodesiense, and the rate of progression of
Pentamidine has been associated with nephrotoxicity and
the disease is subspecies-dependent, with T.b. gambiense being
accountable for 95% of cases and causing a chronic infection,
diabetes mellitus,6 while suramin can lead to renal failure and
exfoliative dermatitis.4 Melarsoprol (3) was the first drug used
to treat stage two HAT.6,7 It is effective against both pathogenic
subspecies, although recently resistance to the drug has been
observed.7 There are also significant side effects associated with
its use including reactive encephalopathy, which can be fatal in
while T.b. rhodesiense presents as an acute infection.2 The
disease has two stages: the first stage is the hemolymphatic
phase where trypanosomes multiply in subcutaneous tissue,
blood, and the lymphatic system, and is accompanied by fever,
headaches, joint pain, and itching.2 Stage two is the
neurological phase and occurs when the parasites cross the
blood brain barrier (BBB) and infect the central nervous system
(CNS). This typically results in behavioral change, confusion,
Received: February 4, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Drugs used for the treatment of human African trypanosomiasis (1−5), fexinidazole (6), now in phase II/III clinical trials, and SCYX-7158
(7) currently in phase I clinical trials.
3−10% of patients.7 Eflornithine (4), effective against T.b.
gambiense,7 was registered in 1990 and is less toxic than its
predecessor. More recently, a combination therapy of
nifurtimox (5) and eflornithine (4) (NECT) was introduced,
but it is ineffective against T.b. rhodesiense and the need for
parenteral administration is a drawback.8 A number of
Figure 2. Structure and physicochemical parameters of screening hit 8.
compounds have recently entered clinical development for
HAT. Pafuramidine, a constrained analogue of pentamidine,
entered clinical development in 2005 but was abandoned when
Compound 8 possesses a very low molecular weight of 212 g
mol−1, a low polar surface area of 42 Å2 suitable for potential
liver toxicity and delayed renal insufficiency were identified in
CNS penetration to treat stage two HAT, an attractively low
an extended phase 1 trial.9 Drugs currently in clinical trials are
lipophilicity (log P of 2.3), and would be neutral at
fexinidazole (6), a nitroheterocycle that is in phase II/III
physiological pH, aiding BBB penetration. The low molecular
clinical trials,10 and the oxaborole, SCYX-7158 (7), which
weight results in a high ligand efficiency and much scope to
entered phase I trials in March 2012. The outcome of these
optimize; as such, this compound was selected for development
trials is pending.11
of structure−activity relationships (SAR). Herein, we report the
Fortunately, with the advent of recent public−private
results of this investigation.
partnerships and related initiatives,1 it is anticipated that
improved treatments will emerge for many of the so-called
RESULTS AND DISCUSSION
■
“neglected” diseases. There are a number of drug discovery
Mining the HTS database returned 15 compounds, which were
avenues currently being utilized in the search for new
broadly related to 8. However, these compounds were not
antitrypanosomal drugs, which we have recently reviewed.12
sufficiently similar to develop meaningful SAR. Three parallel
approaches were then employed to establish a focused SAR
We recently conducted high-throughput screening (HTS)
against T. brucei using a library of 87296 compounds, and
investigation. The first of these involved identification of close
discovered several classes with promising activity.13 One of
analogues in the literature that were associated with previously
these has been progressed to a hit-to-lead medicinal chemistry
reported biological activity. Of particular interest were those
optimization program.14 Some inhibitor series were not
compounds with demonstrated activity against trypanosomes,
discussed in the initial HTS publication because they did not
meet predefined selection criteria. One of these was a simple
pyridyl benzamide (8), which was slightly less active than the
other classes with an IC50 value of 12 μM, close to the
predefined activity cutoff value of <10 μM, and a selectivity
index (SI) of 28 relative to HEK 293 cells. However, this
compound was attractive to the medicinal chemists as an
optimization candidate as its physicochemical properties were
deemed to be favorable for development (Figure 2).
or whereby T. brucei possesses a target that the compound had
reported activity against in other organisms. From this search, it
was determined that relatively little was known about the
biological activity of 8, although a low level of antifungal activity
against a strain of corn-smut has been reported (Ustilago
maydis).15 The closely related 2-methylbenzanilide system was
reported to have more effective and wide ranging antifungal
activity and demonstrated activity against succinate dehydro-
genase (Complex II in mitrochondria) in corn smut.15 This
B
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. Potential synthetic targets identified in the literature, where pyridyl benzamides 9−15 are counterparts to corresponding benzanilides with
associated antifungal activity.
a
enzyme has been previously identified in T. brucei procyclic
trypomastigotes.16 There is some precedence for cross-species
activity of compounds toward fungi and trypanosomes.17,18 We
therefore targeted pyridyl benzamides containing groups
known to impart great potency to fungal succinate dehydrogen-
ase, in case they did so for T. brucei as well.15 These compounds
(9−15 in Figure 3) have lipophilic groups in the 5-position (9−
11) or 6-position (12) of the pyridine ring, or on the amide
nitrogen atom (13), while 14 and 15 contain a 2-iodo and 2-
trifluoromethyl, respectively, on the benzamide ring instead of a
methyl group.
Scheme 1
The second approach that we adopted comprised searching
databases of compound biological activities such as the TCAMS
set19 and PubChem. A number of analogues were identified
with a variety of biological activities. Those of particular interest
were 16−18 (Figure 4), which were identified as active against
a
(i) X = OH, DIPEA, HBTU, DMF, 60 °C, or EDCI, DMAP, DMF,
t
25°C; X = Cl, pyridine, 16 h, 25°C; (ii) ClSO3H, then BuNH2; (iii)
MeI, K2CO3, TBAB, acetone, 16 h, 25 °C; (iv) alkyne, CuI, PPh3,
Pd(PPh3)2Cl2/Pd(PPh3)4, DMF, Et3N, microwave 120 °C, 10−40
min; (v) boronic acid, K2CO3, TBAB, Pd(PPh3)2Cl2, 1,4-dioxane,
H2O, microwave 130 °C, 15 min.
chlorosulfonylation after coupling followed by reaction with
tert-butylamine was successful.
The biological data for 8−18 are shown in Table 1. Retesting
of a fresh sample of 8 synthesized in-house returned a slightly
Table 1. T.b. brucei Inhibitory Activity of 8−18
Figure 4. Compounds 16−18 were identified through searching
PUBCHEM or the TCAMS set.
ID
IC50 (μM) SEM
percentage inhibition at 10 μM
a
8
3.03 0.7
80%
12%
Plasmodium falciparum. These compounds therefore became
synthetic targets, where any observed activity against T. brucei
could provide information regarding a possible mode of action
for these compounds.
The third approach comprised directed synthesis to develop
the SAR through probing different regions of the molecule with
varying substituents.
As shown in Scheme 1, the synthetic versatility of this
scaffold allowed ready access to all target molecules. Core
assembly was readily constructed via coupling of an
appropriately substituted 2-aminopyridine with an appropri-
ately substituted benzoic acid or benzoyl chloride. In the case of
12, the alkoxy group was introduced prior to an amide
formation in a reaction involving displacement of a labile aryl
halogen with NaH-generated alkoxide in DME. Lipophilic
hydrocarbon substituents in the pyridine ring such as those in
9−11 were introduced with Sonogashira (9) or Suzuki (10, 11)
reactions after coupling, whereas alkylation of the amide
nitrogen atom readily gave 13. In the case of 17,
9
10
11
12
13a
13b
14
15
16
17
18
19%
11%
<10%
11%
17%
38%
<10%
<10%
12%
53%
a
Selectivity index relative to HEK293 cells is 27.5.
improved IC50 (3.03 μM) than originally obtained (12 μM). It
can be seen quite clearly that the structural changes in general
were detrimental to activity. Our interpretation of these results
is that T. brucei SAR for this class of compound is divergent
from SAR for any other known biological activities.
C
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 2. T.b. brucei Inhibitory Activity of SAR Probes of 8 Focusing on the Pyridine and Amide Moieties
a
b
Selectivity relative to HEK293 cells. [N] = endocyclic nitrogen.
Much more promising were the results from exploring 8
(28). The similar activities for the 4-chloro (26) and very bulky
phenylacetylene (29) or phenyl (31) imply this could largely be
a favorable transport effect rather than a specific interaction
with a hydrophobic pocket in an intracellular target site. In the
5-position, methoxy was favorable for activity (32), but methyl
(33), fluoro (34), chloro (37), bromo (38), nitrile (36), or an
endocylic nitrogen (35) were not. As it is hard to envisage that
a methoxy would aid transport, this SAR is consistent with the
methoxy interacting favorably with its target, whether it be
intra- or extracellular. In contrast to substitution at the other
using conventional SAR elaboration. Shown in Table 2 are the
data for compounds with a focus on varying substituents
around the pyridine ring.
Here it can be seen that at the 3-position, methoxy (19),
methyl (20), or fluoro (21) were not well tolerated and neither
was an endocyclic nitrogen (22). In the 4-position, a cyano
(23) was tolerated, an endocyclic nitrogen (24) less so, but in
contrast, good activity accompanied substitution with a range of
hydrophobic groups (25−31), particularly with the 4-fluoro
D
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 3. T.b. brucei Inhibitory Activity of a Number of Analogues of the Tolyl Ring
a
ID
R2
IC50 (μM) SEM
percentage inhibition at 10 μM
SI
64
65
66
67
68
79
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
H
9.07 2.2
2.74 1.0
9.2
30
2-Et
2-F
35%
44%
2-Cl
2-OCF3
2-Ph
13%
<10%
2-CH3, 3-F
3-F
0.89 0.04
7.33 2.1
0.63 0.09
5.7 0.2
94
11
2-CH3, 3-Cl
2-CH3, 3-Br
2-CH3, 3-CH3
132
15
3.5 0.2
24
2-CH3, 3-OCH3
10%
21%
b
2-CH3, 3-[N]
2-CH3, 4-F
1.9 0.3
1.1 0.09
1.1 0.06
44
76
60
2-CH3, 4-Cl
2-CH3, 4-Br
2-CH3, 4-OCH3
2-CH3, 4-CH3
2-CH3, 4-CHCH(CH)2CH3
2-CH3, 4-Ph
<10%
0.98 0.07
85
24%
<10%
<10%
<10%
22%
2-CH3, 5-F
2-CH3, 5-Br
2-CH3, 5-CH3
2-CH3, 5-OCH3
<10%
<10%
<10%
45%
b
2-CH3, 5-[N]
2-CH3, 6-CH3
2-CH3, 6-F
b
2-CH3, 6-[N]
12%
a
b
Selectivity relative to HEK293 cells. [N] = endocyclic nitrogen.
positions, no favorable substitution could be found for the 6-
position, even after extensive probing with a variety of diverse
groups (39−51). Perhaps the only exception was 51, where an
amino group appeared to be tolerated with an IC50 of 4.6 μM.
Removal of the pyridyl nitrogen as with the benzanilide (52),
and the substitution of the pyridine ring with a range of five-
membered heterocycles (54−61), resulted in the complete loss
of activity and demonstrated that the pyridine ring was essential
for activity. It was thought that replacement of the pyridinyl
nitrogen with a nitrile (53) could overcome the observed loss
of activity as nitriles are known to replace water-mediated
hydrogen-bond bridges involving heterocyclic nitrogens,20
although here such a change resulted in loss of activity. Finally,
it was determined through inactivity of the N-methyl (62) and
reverse amide (63) analogues that the secondary amide was
essential for activity.
Table 1. Maintaining the 2-methyl group, other positions were
probed, and it can be seen that while a 3-methoxy (75) or
endocyclic nitrogen (76) were not tolerated, a 3-fluoro (70)
was highly favorable for activity with an IC50 of 0.89 μM. The
methyl group was confirmed to be important as its loss from 70
to give des-methyl analogue 71 led to a near 10-fold loss of
activity. Substitution with 3-chloro (72) was favorable with an
IC50 of 0.63 μM, while the 3-methyl (74) and 3-bromo (73)
were significantly less active with respective IC50 values of 3.5
and 5.7 μM. This suggests that a small, electron-withdrawing
and hydrophobic group is favorable for activity at the 3-
position. In the 4-position, a methoxy (80) was not tolerated;
fluorine (77), chlorine (78), bromine (79), and methyl (81)
substitution all gave favorable results, with IC50 values of 1.9,
1.1, 1.1, and 0.98 μM, respectively. In the 5-position,
substitution of an endocyclic nitrogen (88) led to a loss of
activity, and similarly fluorine (84), bromine (85), methyl (86),
and methoxy (87) in this position rendered the compounds
inactive. It appears as though this position nestles tightly within
the target’s active site and no manipulation is possible.
Likewise, substitution at the 6-position was unfavorable for
activity with methyl (89), fluorine (90), and an endocyclic
nitrogen (91) leading to a complete loss of activity.
The SAR for the tolyl ring was probed next, and the results
are shown in Table 3.
Here, it is observed that the methyl group is important, and
its loss (64) led to a 3-fold decrease in activity as compared to
8. Interestingly, extension of the 2-methyl to a 2-ethyl was well
tolerated, with 65 having an IC50 of 2.74 μM, although fluoro
(66), chloro (67), trifluoromethoxy (68), and phenyl (69)
were not tolerated, complementing the results for 14 and 15 in
E
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Finally, we looked for evidence of additive SAR to see if
further increases in potency could be engineered in a rational
way. We saw previously (Table 3) that when R2 was 2-methyl,
3-fluoro (70) or 2-methyl, 4-fluoro (77), respective IC50 values
of 0.89 and 1.9 μM were obtained, which were better than that
for 8 (3.03 μM). Combining the 2-methyl, 3-fluoro, and 4-
fluoro in the one molecule gave a further improvement in
activity, and 92 (Table 4) returned an IC50 of 0.41 μM. We also
Table 5. Biological Activity Profile of Selected Compounds
against a Parasite Panel (IC50, μM)
a
T.b.
brucei
T.b.
rhod.
T.
cruzi
L.don.
d
P. falc.
Cytotox.
b
c
e
f
ID
axe
K1
L6
8
3.03
2.1
10
152
173
168
71
48
>133
168
36
124
88
156
146
224
198
70
25
26
29
31
70
79
81
92
94
0.36
0.61
0.061
0.97
0.61
0.045
0.24
0.10
0.64
2.1
1.7
>96
9.7
91
2.0
66
>124
>131
>103
>133
>121
>113
Table 4. T.b. brucei Inhibitory Activity of Compounds with a
Combination of Favorable Substitution Patterns around the
Pyridyl and Phenyl Rings in Search of Additive SAR
0.89
1.1
198
100
187
176
147
103
48
310
193
283
308
223
0.98
0.41
0.90
93
17
72
a
b
Values are means of three experiments, 50%. T.b. rhodesiense strain
STIB 900, bloodstream form (trypomastigotes). Melarsoprol was used
c
as a control, IC50 0.002 μg/mL. T. cruzi Tulahaen C4 strain,
amastigote stage. Benznidazole was used as a control, EC50 0.38 μg/
d
a
mL. L. donovani MHOM-ET-67/L82 strain, amastigote stage.
ID
R1
R2
IC50 (μM) SEM
SI
e
Miltefosine was used as a control, EC50 0.20 μg/mL. P. falciparum
3D7 strain, erythrocytic stage. Chloroquine was used as a control, EC50
0.11 μg/mL. Rat skeletal myoblast cell L-6 strain. Podophylotoxin was
used as a control, EC50 0.007 μg/mL.
92
93
94
95
96
97
98
99
100
H
3-F, 4-F
3-F
0.41 0.07
0.83 0.01
0.90 0.44
0.47 0.05
0.53 0.26
0.10 0.04
0.19 0.24
2.3 0.2
205
100
93
f
4-CH3
4-CH3
4-Cl
3-F, 4-F
3-F
177
160
851
450
4
4-Cl
3-F, 4-F
3-F
T. brucei over the other parasites, and they were largely
noncytotoxic with respect to the L6 mammalian cell line.
An assessment of drug-likeness was then made, and various
physicochemical and metabolic parameters for selected
compounds are listed in Table 6. It can be seen that all
compounds are of relatively low molecular weight in the range
226−312 gmol−1. Their polar surface areas are low and well
within the range where good CNS penetration is plausible.21
An in silico prediction model for the potential for CNS
exposure was employed to further assess this possibility.22 Of
the 14 compounds screened, seven were predicted to have high
CNS exposure (nominally defined as a brain to plasma ratio
>0.3), and a further two (41 and 93) had values just below the
nominal cutoff for high CNS penetration. On the basis of these
predictions, this class would be expected to have reasonable
CNS exposure, which could potentially lead to the treatment of
the CNS stage of HAT. That said, CNS penetration is a very
complex phenomenon, given that exposure depends on a
number of factors including protein binding, influx and efflux,
and metabolism,23 and an in vivo assessment of these
compounds is still required. Likewise, distribution coefficients
were favorable in the range 2.3−4.7 at pH 7.4. Aqueous
solubility for some compounds was very good; for example,
both 42 and 79 had solubilities of greater than 100 μg/mL at
both pH 2 and 6.5. This system was strongly influenced by the
attached substituents such that 29 was very poorly soluble at
1.6−3.1 μg/mL at pH 2 and <1.6 μg/mL at pH 6.5. Plasma
protein binding also varied significantly but was acceptable.
Microsomal degradation was moderate to quite high across all
compounds tested. Compound 29 was relatively metabolically
stable with a predicted EH value of 0.51. For this compound,
the pyridine ring is substituted in the 4-position, and it is
possible that the effect is to hinder metabolic oxidation of the
adjacent electron-rich 6-position, which is para to the amide
group. This hypothesis is supported by the observation that the
most metabolically stable compound is 32 with a predicted EH
of 0.47, where the 5-position is blocked by a methoxy group.
The metabolism of compounds 34 and 95 was examined in
greater detail in an effort to determine the key sites of
4-F
4-F
3-F, 4-F
3-F
5-OCH3
5-OCH3
4-F
2.6 0.2
32
a
Selectivity relative to HEK293 cells.
demonstrated earlier (Table 2) that favorable R1 substituents
comprised 4-methyl (25), 4-chloro (26), 4-fluoro (28), or 5-
methoxy (32). When R2 was kept constant as 2-methyl, 3-
fluoro and combined with these R1 substituents, compounds
93, 95, 97, and 99 resulted (Table 4). Here, it can be seen that
for 95 this combination was beneficial with an IC50 of 0.47 μM
and even more so for 97 with an IC50 of 100 nM and an
improved selectivity profile as compared to 8, while a neutral
effect was observed for 93. Although both 93 and 95 show an
improved SI as compared to 8 (27.5), a 16-fold improvement in
the SI was obtained when a fluorine was substituted at the 4-
position (98), and this compound was significantly more
potent than the others, with an IC50 of 190 nM. However, the
combination involving the 5-methoxy (99) was detrimental to
activity (IC50 2.3 μM). This detrimental effect of the R1 5-
methoxy was also observed when R2 comprised 2-methyl, 4-
fluoro (100, IC50 2.6 μM).
To investigate activity against the human pathogenic
subspecies T.b. rhodesiense, the original hit (8) and a selection
of analogues were tested against a panel of parasites including
T.b. rhodesiense, Trypanosoma cruzi, Leishmania donovani, and
Plasmodium falciparum, the causative agents of HAT, Chagas
disease, leishmania, and malaria, respectively. As shown in
Table 5, all of the compounds that were potent against T.b.
rhodesiense were also potent against T.b. brucei. However, there
was not a direct correlation, and some compounds, such as 29
and 80, were significantly more potent against the human
pathogenic species T.b. rhodesiense than might be expected
based on their respective activities against T.b. brucei. Indeed,
the activities of these two compounds against T.b. rhodesiense
were very promising with respective IC50 values of 61 and 45
nM. In all cases, the compounds were exquisitely selective for
F
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 6. Key Physicochemical Parameters and in Vitro Metabolic Stability of Selected Compounds
b
c
log D
solubility (μg/mL)
d
PSA
cPPB
(%)
in vitro CLint
(μL/min/mg protein)
predicted CNS
exposure
a
e
ID
MW
(Å2)
pH 7.4
pH 2
pH 6.5
microsome-predicted EH
22
23
25
26
29
32
34
36
41
70
79
92
93
95
237
237
226
247
312
242
230
237
230
230
291
248
244
265
65.8
65.8
42.0
42.0
42.0
51.2
42.0
65.8
42.0
42.0
42.0
42.0
42.0
42.0
3.9
2.8
2.8
3.4
4.7
2.5
2.7
3.9
3.0
2.6
3.2
2.8
3.5
2.3
6.3−12.5
50−100
>100
6.3−12.5
25−50
25−50
12.5−50
<1.6
97.6
64.1
70.1
86.6
98.3
60.1
56.0
97.5
67.0
67.9
88.1
63.9
72.4
71.7
rapid non-NADPH mediated metabolism
high
high
high
high
high
low
82
0.76
0.83
0.84
0.51
0.47
0.63
129
128
27
25−50
1.6−3.1
>100
12.5−25
25−50
6.3−12.5
>100
23
25−50
6.3−12.5
>100
43
low
rapid non-NADPH mediated metabolism
high
f
191
58
0.88
0.69
0.79
0.86
0.90
0.91
low
50−100
>100
25−50
>100
low
low
low
97
50−100
>100
25−50
25−50
12.5−25
154
237
254
f
low
12.5−25
high
d
a
b
c
Calculated using ACD/Laboratories software, version 9. Measured chromatographically. Kinetic solubility determined by nephelometry. Human
plasma protein binding estimated using a chromatographic method. In vitro intrinsic clearance determined in human liver microsomes and
predicted hepatic extraction ratio calculated from in vitro data. Predicted B:P values of 0.28, that is, just beneath the nominal cutoff value for high
e
f
CNS exposure.
Figure 5. Structures of putative metabolites of 34 and 95. The dotted lines indicate proposed sites of hydroxylation.
metabolism. Metabolism of 34 (Figure 5) in NADPH-
supplemented human microsomal incubations revealed two
mono-oxygenated metabolites (M+16 (I), N-hydroxylation of
the amide bond; and M+16 (II), aromatic hydroxylation of the
tolyl ring) and one bis-oxygenation metabolite (M+32,
dihydroxylation of the tolyl moiety). A putative metabolite
with a molecular ion consistent with oxidative defluorination
(M−2) was also detected; however, the MS/MS signal was too
weak to confirm the structure. In the presence of the dual
cofactors, NADPH and UDPGA, limited glucuronidation of M
+16 (II) and M+32 was observed, while in an incubation
devoid of cofactors, no metabolites were detected, suggesting
that NADPH-mediated hydroxylation to M+16 (I) and (II)
represents the primary metabolic pathways.
Compound 95 was shown to exhibit a high rate of
degradation in NADPH-supplemented human liver micro-
somes (Table 6), and mono-oxygenated (M+16, oxygenation of
the aminopyridine moiety) and bis-oxygenated (M+32, bis-
oxygenation of the aminopyridine moiety) species were
identified (Figure 5). In addition, a putative metabolite with a
molecular ion consistent with N-dearylation (M−111) was
detected; however, the MS/MS signal was too weak to confirm
the structure (Figure 5). In the presence of NADPH and
UDPGA, extensive glucuronidation of the M+16 and M+32
metabolites was observed (i.e., there was a significant decrease
in the detection of M+16 and M+32 species in conjunction
with the appearance of their respective glucuronide conjugates),
while in microsomal samples devoid of cofactors only a very
small peak (in terms of peak area response) for the M+16
metabolite was seen, suggesting that this metabolite is likely to
be formed via NADPH and, to a lesser extent, non-NADPH-
dependent mechanisms. Overall, oxygenation of the amino-
pyridine region appears to be the major metabolic pathway of
95.
Comparing the metabolite profiles of 34 and 95, the
introduction of an aromatic fluorine (as in 95) appeared to
reduce the susceptibility of that region to metabolism.
However, both compounds were still highly metabolized in
the regions without fluorine substitution.
Another important parameter in assessing the therapeutic
potential of compounds that inhibit growth of trypanosomes, or
indeed any pathogenic parasite, is whether the compounds
actually kill the organism or temporarily inhibit their growth.
Those that are not cidal are generally viewed with less interest
as potential therapeutics. It was therefore determined whether
G
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
these compounds were trypanocidal or trypanostatic using cidal
assays, and the results are shown in Table 7.
EXPERIMENTAL SECTION
■
General Chemistry Experimental. See the Supporting Informa-
tion.
Determination of Purity. Purity was determined using high
performance liquid chromatography (HPLC) carried out either on a
Waters Auto Purification System 3100 with a Waters column
(XBridgePrep C18, 5 μm, OBD, 19 × 100 mm) or on a Waters
Alliance HT 2795 with a Phenomenex column (Luna, 5 μm, C18, 100
Å, 150 × 10 mm). The purity of all compounds for biological testing
was >95% in all cases, except where specified otherwise.
Table 7. Percentage of T.b. brucei Trypanosomes Killed
Following Exposure to Compounds over a 72 h Period
percentage of cells killed
ID
24 h
STDEV
48 h
STDEV
72 h
STDEV
92
95
92
6.08
8.73
99.95
83.7
0.07
2.40
99.9
98.8
100
100
0
0
64.23
58.1
0.85
pentamidine
puromycin
DMSO
100
PHYSICOCHEMICAL STUDIES
■
100
100
0
0
Solubility Estimates Using Nephelometry. This was
obtained as described previously.1,24 See the Supporting
Information for more information.
Chromatographic Log D Measurement. Partition
coefficients (log D) were estimated as described previously.1,25
See the Supporting Information for more information.
Chromatographic Protein Binding Estimation. This
was obtained as previously described.1,26 See the Supporting
Information for more information.
In Vitro Metabolism in Human Liver Microsomes. Data
were obtained as described previously.1,27,28 See the Supporting
Information for more information.
The percentage of T.b. brucei trypanosomes killed following
24, 48, and 72 h exposure to the MIC of 92 and 95 as
compared to the untreated DMSO control is shown in Table 7.
After 24 h incubation with compounds 92 and 95, the number
of viable trypanosomes was reduced by >90 and 60%,
respectively. Following further 24 h incubation, only a small
population of viable trypanosomes remained (<20%), and by
72 h, >98.5% of the trypanosomes had been killed. The
continual decrease and almost complete elimination of viable
trypanosomes following 72 h exposure to the MIC of 92 and
95 indicates that the compounds are trypanocidal. No viable
trypanosomes were detected following 24 and 48 h incubation
with puromycin and pentamidine, respectively.
Metabolite ID. Each test compound was incubated at 37 °C
with human liver microsomes (XenoTech, lot no. 1210057),
and the reaction was initiated by the addition of an NADPH-
regenerating system as described above. To maximize the
metabolite yield, a high substrate concentration (10 μM) and
high microsomal protein concentration (1 mg/mL) were used,
and incubations were conducted for up to 60 min. Samples
without test compound and without NADPH were included as
controls. In addition, a single sample initiated by the addition of
the dual cofactor system (containing the NADPH regeneration
buffer described and UDPGA (mixture A, BD Gentest) in the
presence of 0.025 mg/mL alamethacin) was included (at the 60
min time point) for the qualitative assessment of the potential
for glucuronide formation. Reactions were quenched by protein
precipitation with an equal volume of ice-cold acetonitrile
(containing 0.15 μg/mL diazepam as internal standard), and
then centrifuged for 3 min at 10000 rpm. The supernatant was
removed and analyzed for parent degradation and metabolite
formation by LC/MS (Waters Micromass Xevo G2 QTOF
coupled to a Waters Acquity UPLC). The mobile phase
consisted of an acetonitrile−water gradient (containing 0.05%
formic acid). A Supelco Ascentis Express RP-Amide (50 × 2.1
mm) column was used. MS analyses were conducted in positive
mode electrospray ionization under MSE acquisition mode,
which allows simultaneous acquisition of MS spectra at low and
high collision energies. The identity of putative metabolites was
confirmed by accurate mass and MS/MS fragmentation where
possible.
CONCLUSIONS
■
We have described the discovery of a class of pyridyl
benzamides that are novel inhibitors of T.b. brucei and T.b.
rhodesiense. For compound 79, the IC50 value against the
human pathogenic subspecies, T.b. rhodesiense, was 45 nM.
Assessment of predictive ADMET parameters revealed that
these very low molecular weight compounds have very
favorable properties for further optimization. In particular,
they have relatively low log D values, good aqueous solubility,
and low plasma protein binding. Most of the analogues were
predicted to have good CNS penetration, a key feature for the
treatment of second stage HAT. These compounds were
determined to be cidal and not static in their biological action.
Ten compounds have been tested against a panel of parasites,
including T.b. rhodesiense, alongside T. cruzi, P. falciparum, and
L. donovani, and were shown to demonstrate selectivity toward
T. brucei in comparison to the other trypanosome species.
These compounds did not display significant activity against L6
cells. Improvement of the metabolic stability of these
compounds will be a key parameter to optimize during the
next phase of development.
While an efficacy study for these compounds has not yet
been undertaken, it is an important proof of principle
experiment. Given the expense associated with using an animal
model for HAT and the time required to perform the
experiment, further improvement in metabolism is required
before these experiments are performed.
The specific biological target or targets of this series of
compounds remain unknown. However, for some of the more
active compounds, with whole organism IC50 values of less than
100 nM, it is reasonable to assume that target affinity may be in
the low nanomolar range. Target identification is warranted to
enable the facilitation of the next stage of rational optimization.
In Silico Prediction of CNS Exposure. The partial least-
squares projection to latent structures (PLS) model developed
for CNS exposure in 201222 was applied to 14 representative
compounds. SMILES strings were submitted to the software
Corina 3.0 (Molecular Networks, Erlangen, Germany) to
produce the three-dimensional structures. The resulting
structures were used to calculate a large number of
physicochemical properties and molecular descriptors using
DragonX (Talete, Italy). These descriptors were then
submitted to the PLS model, and the CNS exposure was
predicted. On the basis of the predicted brain to plasma (B:P)
H
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
value, the compounds were classified as showing high CNS
exposure (B:P > 0.3) or low CNS exposure (B:P < 0.3).
ACKNOWLEDGMENTS
■
This research was supported by the NHMRC (IRIISS grant
number 361646; NHMRC Senior Research Fellowship
1020411 (J.B.B.), NHMRC Project Grant 1025581); Victorian
Stage Government OIS grant; the Australian Government for
the Australian Postgraduate Award (L.F.); Dr. Jason Dang for
his assistance in obtaining some of the analytical data; and
DNDi for allowing access to their parasite screening platform at
STPHI.
BIOLOGICAL ASSAYS
■
All in vitro assays were carried out at least twice independently
in singleton. The IC50 values are the means of two independent
assays and vary by less than 50%.
P. falciparum Assay. This was undertaken as previously
described.1,29−32 See the Supporting Information for more
information.
L. donovani Axenic Amastigotes Assay. This was
undertaken as previously described.1,32−34 See the Supporting
Information for more information.
T. cruzi Assay. Data were obtained as previously
described.1,32,35 See the Supporting Information for more
information.
ABBREVIATIONS USED
■
CNS, central nervous system; HAT, human African trypano-
somiasis; T.b. brucei, Trypanosoma brucei brucei; T.b.
gambiense, Trypanosoma brucei gambiense; T.b. rhodesiense,
Trypanosoma brucei rhodesiense
T. brucei rhodesiense Assay. This was undertaken as
previously described.1,32,36,37 See the Supporting Information
for more information.
Rat Skeletal Myoblast Cytotoxicity Assay. This was
undertaken as previously described.1,32,38,39 See the Supporting
Information for more information.
T.b. brucei Assay. This was undertaken as previously
described.1,40,41 See the Supporting Information for more
information.
HEK293 Cytotoxicity Assay. This was undertaken as
previously described.1 See the Supporting Information for more
information.
Cidal Assay. The number of viable trypanosomes remaining
following 24, 48, and 72 h exposure to the minimum inhibitory
concentration (MIC) of compounds was determined by
directly visualizing and counting the number of parasites
present at each time point. The assay was performed as
described previously in detail by Sykes et al.13 Briefly, 55 μL of
T.b. brucei parasites was added to a black/clear bottomed 384
well microtiter plate and incubated for 24 h at 37 °C/5% CO2.
Compounds were diluted in DMEM medium, and 5 μL of this
dilution was added to assay plates to give the MIC. After 24, 48,
and 72 h incubation at 37 °C/5% CO2, wells were directly
visualized under a microscope, and the number of parasites
remaining was determined by counting a sample of the wells in
a hemocytometer. The cell counts were compared to that of the
positive control, puromycin (MIC1.15 μM), and the negative
control DMSO (0.4%), and the percentage of trypanosomes
killed was subsequently determined. The MIC of each
compound was defined as the minimum concentration at
which there was a plateau of activity in the Alamar blue assay
(>95% activity). The experiment was performed in duplicate.
REFERENCES
■
(1) WHO. Control and surveillance of human African trypanoso-
miasis. http://apps.who.int/iris/bitstream/10665/95732/1/
9789241209847_eng.pdf (accessed, 10th January, 2014).
(2) Brun, R.; Blum, J.; Chappuis, F.; Burri, C. Human African
trypanosomiasis. Lancet 2010, 375, 148−159.
(3) WHO. Drugs. http://www.who.int/trypanosomiasis_african/
drugs/en/index.html (accessed, July 27, 2012).
(4) Voogd, T. E.; Vansterkenburg, E. L.; Wilting, J.; Janssen, L. H.
Recent research on the biological activity of suramin. Pharmacol. Rev.
1993, 45, 177−203.
(5) de Koning, H. P. Transporters in African trypanosomes: role in
drug action and resistance. Int. J. Parasitol. 2001, 31, 512−522.
(6) Nok, A. J. Arsenicals (melarsoprol), pentamidine and suramin in
the treatment of human African trypanosomiasis. Parasitol. Res. 2003,
90, 71−79.
(7) Seebeck, T.; Maser, P. Drug resistance in African Trypanoso-
miasis. In Antimicrobial Drug Resistance; Mayers, D. L., Ed.; Humana
Press: Totowa, NJ, 2009; pp 589−604.
(8) Priotto, G.; Kasparian, S.; Mutombo, W.; Ngouama, D.;
Ghorashian, S.; Arnold, U.; Ghabri, S.; Baudin, E.; Buard, V.;
Kazadi-Kyanza, S.; Ilunga, M.; Mutangala, W.; Pohlig, G.; Schmid,
C.; Karunakara, U.; Torreele, E.; Kande, V. Nifurtimox-eflornithine
combination therapy for second-stage African Trypanosoma brucei
gambiense trypanosomiasis: a multicentre, randomised, phase III, non-
inferiority trial. Lancet 2009, 374, 56−64.
(9) Barrett, M. P. Potential new drugs for human African
trypanosomiasis: some progress at last. Curr. Opin. Infect. Dis. 2010,
23, 603−608.
(10) DNDi. New Oral Drug Candidate for African Sleeping Sickness.
http://www.dndi.org/media-centre/press-releases/354-media-centre/
press-releases/1505-new-oral-drug-candidate-hat.html (accessed, 9th
July).
(11) DNDi. Oxaborole SCYX-7158 (HAT). http://www.dndi.org/
portfolio/oxaborole-scyx-7158.html (accessed, July 27, 2012).
(12) Ferrins, L.; Rahmani, R.; Baell, J. B. Drug discovery and human
African trypanosomiasis: a disease less neglected? Future Med. Chem.
2013, 5, 1801−1841.
(13) Sykes, M. L.; Baell, J. B.; Kaiser, M.; Chatelain, E.; Moawad, S.
R.; Ganame, D.; Ioset, J.-R.; Avery, V. M. Identification of Compounds
with Anti-Proliferative Activity against Trypanosoma brucei brucei Strain
427 by a Whole Cell Viability Based HTS Campaign. PLoS Neglected
Trop. Dis. 2012, 6, e1896.
(14) Ferrins, L.; Rahmani, R.; Sykes, M. L.; Jones, A. J.; Avery, V. M.;
Teston, E.; Almohaywi, B.; Yin, J.; Smith, J.; Hyland, C.; White, K. L.;
Ryan, E.; Campbell, M.; Charman, S. A.; Kaiser, M.; Baell, J. B. 3-
(Oxazolo[4,5-b]pyridin-2-yl)anilides as a novel class of potent
inhibitors for the kinetoplastid Trypanosoma brucei, the causative
agent for human African trypanosomiasis. Eur. J. Med. Chem. 2013, 66,
450−465.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures for compound synthesis and charac-
terization. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Tel.: +61 9903 9044. E-mail: Jonathan.Baell@monash.edu.
■
Author Contributions
□L.F. and M.G. contributed equally.
Notes
The authors declare no competing financial interest.
I
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(15) White, G. A. Substituted benzanilides: Structural variation and
inhibition of complex II activity in mitochondria from a wild-type
strain and a carboxin-selected mutant strain of Ustilago maydis. Pestic.
Biochem. Physiol. 1987, 27, 249−260.
T. equiperdum, T. evansi, T. rhodesiense and T. gambiense. EMBO J.
1985, 4, 1273−1277.
(37) Raz, B.; Iten, M.; Grether-Buhler, Y.; Kaminsky, R.; Brun, R.
̈
̈
The Alamar Blue assay to determine drug sensitivity of African
trypanosomes (T.b. rhodesiense and T.b. gambiense) in vitro. Acta Trop.
1997, 68, 139−147.
(38) Page, B.; Page, M.; Noel, C. A new fluorometric assay for
cytotoxicity measurements in vitro. Int. J. Oncol. 1993, 3, 473−476.
(39) Ahmed, S. A.; Gogal, R. M.; Walsh, J. E. A new rapid and simple
non-radioactive assay to monitor and determine the proliferation of
lymphocytes: an alternative to [3H]thymidine incorporation assay. J.
Immunol. Methods 1994, 170, 211−224.
(40) Sykes, M. L.; Avery, V. M. Development of an Alamar blue
viability assay in 384-well format for high throughput whole cell
screening of Trypanosoma brucei brucei bloodstream form strain 427.
Am. J. Trop. Med. Hyg. 2009, 81, 665−674.
(41) Hirumi, H.; Hirumi, K. Continuous cultivation of Trypanosoma
brucei blood stream forms in a medium containing a low concentration
of serum protein without feeder cell layers. J. Parasitol. 1989, 75, 985−
989.
(16) Turrens, J. F. The role of succinate in the respiratory chain of
Trypanosoma brucei procyclic trypomastigotes. Biochem. J. 1989, 259,
363−368.
(17) Zeiman, E.; Greenblatt, C. L.; Elgavish, S.; Khozin-Goldberg, I.;
Golenser, J. Mode of action of fenarimol against Leishmania spp. J.
Parasitol. 2008, 94, 280−286.
(18) Urbina, J. A.; Payares, G.; Molina, J.; Sanoja, C.; et al. Cure of
short- and long-term experimental Chagas’ disease using D0870.
Science 1996, 273, 969−971.
(19) Gamo, F.-J.; Sanz, L. M.; Vidal, J.; de Cozar, C.; Alvarez, E.;
Lavandera, J.-L.; Vanderwall, D. E.; Green, D. V. S.; 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.
(20) Meanwell, N. A. Synopsis of some recent tactical application of
bioisosteres in drug design. J. Med. Chem. 2011, 54, 2529−2591.
(21) Pajouhesh, H.; Lenz, G. R. Medicinal chemical properties of
successful central nervous system drugs. NeuroRx 2005, 2, 541−53.
(22) Bergstrom, C. S.; Charman, S.; Nicolazzo, J. Computational
̈
prediction of CNS drug exposure based on a novel in vivo dataset.
Pharm. Res. 2012, 29, 3131−3142.
(23) Di, L.; Rong, H.; Feng, B. Demystifying brain penetration in
central nervous system drug discovery. J. Med. Chem. 2012, 56, 2−12.
(24) Bevan, C. D.; Lloyd, R. S. A high-throughput screening method
for the determination of aqueous drug solubility using laser
nephelometry in microtiter plates. Anal. Chem. 2000, 72, 1781−1787.
(25) Lombardo, F.; Shalaeva, M. Y.; Tupper, K. A.; Gao, F. ElogDoct:
A tool for lipophilicity determination in drug discovery. 2. Basic and
neutral compounds. J. Med. Chem. 2001, 44, 2490−2497.
(26) Valko, K.; Nunhuck, S.; Bevan, C.; Abraham, M. H.; Reynolds,
D. P. Fast gradient HPLC method to determine compounds binding
to human serum albumin. Relationships with octanol/water and
immobilized artificial membrane lipophilicity. J. Pharm. Sci. 2003, 92,
2236−2248.
(27) Obach, R. S. Prediction of human clearance of twenty-nine
drugs from hepatic microsomal intrinsic clearance data: An
examination of in vitro half-life approach and nonspecific binding to
microsomes. Drug Metab. Dispos. 1999, 27, 1350−1359.
(28) Davies, B.; Morris, T. Physiological parameters in laboratory
animals and humans. Pharm. Res. 1993, 10, 1093−1095.
(29) Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.; Chulay, J. D.
Quantitative assessment of antimalarial activity in vitro by a
semiautomated microdilution technique. Antimicrob. Agents Chemother.
1979, 16, 710−718.
(30) Matile, H.; Pink, J. R. L. Plasmodium falciparum malaria parasite
cultures and their use in immunology. In Immunological Methods;
Lefkovits, I., Pernis, B., Eds.; Academic Press: San Diego, CA, 1990.
(31) Ponnudurai, T.; Leeuwenberg, A. D.; Meuwissen, J. H.
Chloroquine sensitivity of isolates of Plasmodium falciparum adapted
to in vitro culture. Trop. Geogr. Med. 1981, 33, 50−54.
(32) Huber, W.; Koella, J. C. A comparison of three methods of
estimating EC50 in studies of drug resistance of malaria parasites. Acta
Trop. 1993, 55, 257−261.
(33) Cunningham, M. L.; Fairlamb, A. H. Trypanothione reductase
from Leishmania donovani. Purification, characterisation and inhibition
by trivalent antimonials. Eur. J. Biochem. 1995, 230, 460−468.
(34) Mikus, J.; Steverding, D. A simple colorimetric method to screen
drug cytotoxicity against Leishmania using the dye Alamar Blue.
Parasitol. Int. 2000, 48, 265−269.
(35) Buckner, F. S.; Verlinde, C. L.; La Famme, A. C.; CVan Voorhis,
W. C. Efficient technique for screening drugs for activity against
Trypanosoma cruzi using parasites expressing beta-galactosidase.
Antimicrob. Agents Chemother. 1996, 40, 2592−2597.
(36) Baltz, T.; Baltz, D.; Giroud, C.; Crockett, J. Cultivation in a
semi-defined medium of animal infective forms of Trypanosoma brucei,
J
dx.doi.org/10.1021/jm500191u | J. Med. Chem. XXXX, XXX, XXX−XXX