Urea-based inhibitors of trypanosoma brucei methionyl-trna synthetase: Selectivity and in vivo characterization

Article abstract of DOI:10.1021/jm300303e

Urea-based methionyl-tRNA synthetase inhibitors were designed, synthesized, and evaluated for their potential toward treating human African trypanosomiasis (HAT). With the aid of a homology model and a structure-activity-relationship approach, low nM inhibitors were discovered that show high selectivity toward the parasite enzyme over the closest human homologue. These compounds inhibit parasite growth with EC₅₀ values as low as 0.15 μM while having low toxicity to mammalian cells. Two compounds (2 and 26) showed excellent membrane permeation in the MDR1-MDCKII model and encouraging oral harmacokinetic properties in mice. Compound 2 was confirmed to enter the CNS in mice. Compound 26 had modest suppressive activity against Trpanosoma brucei rhodesiense in the mouse model, suggesting that more potent analogues or compounds with higher exposures need to be developed. The urea-based inhibitors are thus a promising starting point for further optimization toward the discovery of orally available and CNS active drugs to treat HAT.

Full text of DOI:10.1021/jm300303e

Article
pubs.acs.org/jmc
Urea-Based Inhibitors of Trypanosoma brucei Methionyl-tRNA
Synthetase: Selectivity and in Vivo Characterization
Sayaka Shibata,†,‡ J. Robert Gillespie,§ Ranae M. Ranade,§ Cho Yeow Koh,† Jessica E. Kim,†
Joy U. Laydbak,§ Frank H. Zucker,† Wim G. J. Hol,† Christophe L. M. J. Verlinde,†
Frederick S. Buckner,*,§ and Erkang Fan*,†
‡
§
†Department of Biochemistry, Department of Chemistry, and Department of Medicine, University of Washington, Seattle,
Washington 98195, United States
S
* Supporting Information
ABSTRACT: Urea-based methionyl-tRNA synthetase inhibitors were designed,
synthesized, and evaluated for their potential toward treating human African
trypanosomiasis (HAT). With the aid of a homology model and a structure−
activity-relationship approach, low nM inhibitors were discovered that show high
selectivity toward the parasite enzyme over the closest human homologue. These
compounds inhibit parasite growth with EC50 values as low as 0.15 μM while having
low toxicity to mammalian cells. Two compounds (2 and 26) showed excellent
membrane permeation in the MDR1-MDCKII model and encouraging oral
pharmacokinetic properties in mice. Compound 2 was confirmed to enter the CNS in mice. Compound 26 had modest
suppressive activity against Trpanosoma brucei rhodesiense in the mouse model, suggesting that more potent analogues or
compounds with higher exposures need to be developed. The urea-based inhibitors are thus a promising starting point for further
optimization toward the discovery of orally available and CNS active drugs to treat HAT.
demonstrating that MetRS is an attractive protein drug target
for T. brucei.4 However, aminoquinolone-based MetRS
inhibitors were known from antibacterial work to have poor
bioavailability.5,6 In our own hands, this series of compounds
indeed exhibited poor permeability in the MDR1-MDCKII
model that predicts CNS penetration,7−9 making it unlikely
that they would cross the blood−brain barrier for treating late
stage HAT. These pharmacokinetic (PK) liabilities prompted
us to explore other scaffolds to block T. brucei MetRS in our
investigations toward anti-HAT therapeutics. In this paper, we
report that using a urea moiety to replace the aminoquinolone
group resulted in selective MetRS inhibitors that show good
potency in parasite growth inhibition assays and promising
improvements in bioavailability.
INTRODUCTION
■
Approximately 60 million people in sub-Saharan Africa are at
risk for human African trypanosomiasis (HAT) caused by
Trypanosoma brucei. HAT caused by Trypanosoma brucei
gambiense progresses slowly from early stage to late stage
disease, whereas disease caused by Trypanosoma brucei
rhodesiense progresses rapidly.1,2 In late stage HAT, the central
nervous system (CNS) becomes infected and the untreated
disease is uniformly fatal. Depending on the stage of the disease
and the subspecies of the causative agent, HAT is treated either
with suramin, pentamidine, melarsoprol, eflornithine, or a
combination of nifurtimox and eflornithine.1,2 These currently
used drugs are either highly toxic and/or need to be
administered by injection. Thus, there is an urgent need to
develop new therapeutics that are effective, safe, affordable,
orally administered, and easily stored in tropical conditions
(http://www.dndi.org/diseases/hat/target-product-profile.
html).
Methionyl-tRNA synthetase (MetRS), as one of the
aminoacyl-tRNA synthetases (aaRS), plays an essential role in
the core biological process of translating nucleotide-encoded
gene sequences into proteins. The enzymatic reaction of aaRS
generally consists of the following steps: the recognition of a
specific amino acid and ATP, the formation of an aminoacyl-
adenylate, the recognition of a specific tRNA, and the transfer
of the aminoacyl group to the 3′-end of the tRNA.3 We recently
showed by RNAi knockdown that the single MetRS of T. brucei
is essential for parasite survival.4 Moreover, we synthesized a
series of potent aminoquinolone-based inhibitors of parasite
MetRS that inhibited parasite growth in culture, further
RESULTS AND DISCUSSION
■
Design of Urea-Based Inhibitors. The starting point for
the work in this paper is the predicted binding mode of
aminoquinolone-based compound 1 in a homology model of T.
brucei MetRS that we reported earlier.4 We were able to create
a high quality model because of the disclosure in a conference
poster by the Replidyne company of a cocrystal structure of a
related aminoquinolone-based inhibitor bound to Clostridium
difficile MetRS.10 Compound 1 was successfully docked into the
model, filling two binding pockets. The benzyl fragment
occupies the mostly hydrophobic methionine substrate pocket,
and one of the meta-chlorine atoms resides essentially in the
Received: March 2, 2012
Published: June 21, 2012
© 2012 American Chemical Society
6342
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
Figure 1. Docking study of aminoquinolone 1 and the design of urea and guanidine analogues using a T. brucei MetRS homology model.4 (a)
Docked pose of 1 with the two NHs of the aminoquinolone forming hydrogen bonds with Asp287; (b) design of urea 2 and guanidine 3; (c)
overlaid poses of 1 (carbons in green) and 2 (carbons in yellow) after docking.
a
Scheme 1. Synthetic Routes
a
(a) Ar-CHO, AcOH, NaCNBH3, MeOH, rt; (b) 2-acetyldimedone, DIPEA, CH2Cl2, MeOH rt; (c) di-tert-butyl dicarbonate, K2CO3, MeCN, water,
rt; (d) hydrazine, THF, microwave (65 °C, with 200 W maximum power input); (e) 1-isothiocyanatobenzene, CH2Cl2, rt; (f) HgCl2, 7 M NH3 in
MeOH; (g) TFA/CH2Cl2 (1:4 v/v), rt; or H2O/TIPS/TFA (3:3:94, v/v), rt; (h) Ar-NCO, CH2Cl2, rt; (i) Ar-NH2, CDI, DMF, rt; (j) Ar-NH2, p-
nitrophenyl chloroformate, DIPEA, THF, rt.
same position as the sulfur atom of methionine. The
aminoquinolinone ring occupies an adjacent pocket, only
created upon inhibitor binding, and forms hydrogen bonds
through its NHs with the carboxylate of Asp 287 (Figure 1).
The Replidyne data and our docking study indicated the
importance of a planar NH-X-NH in the aminoquinolone ring
system for forming hydrogen bonds with the carboxylate of
Asp287. This aspartate residue is strictly conserved in all MetRS
enzymes based on a BLAST search that involved 250 sequence
alignments and is responsible for substrate binding by forming
a salt bridge to the α-amino group of methionine.11 Thus the
aminoquinolone targets an enzyme active site amino acid
residue that is unlikely to mutate, which is advantageous for
drug discovery. However, the very same aminoquinolone
moiety was suspected to be the potential cause of the
inhibitor’s poor bioavailability.5,6 Therefore, we decided to
move away from aminoquinolones but to keep a planar NH-X-
NH moiety in our next generation of inhibitors. Conceptually
dissecting the hetero ring system of the aminoquinolone led to
a urea 2 or a guanidine 3 (Figure 1b). Literature search revealed
that GlaxoSmithKline (GSK) has previously reported only one
urea-based MetRS inhibitor for bacterial targets with moderate
cellular activity.6 In addition, Ibis Therapeutics reported a series
of similar urea-based compounds for antibacterial chemo-
therapy with moderate activities although the compounds’
target of action was not identified in their publication.12
Therefore, urea or guanidine-based inhibitors against T. brucei
MetRS warrant further systematic investigation using structure-
based approaches. Molecular modeling demonstrated that the
preferred binding conformations of urea or guanidine analogues
superimposed nicely onto the aminoquinolone-based inhibitor
(Figure 1c for urea 2) and maintained the key hydrogen bond
interactions with Asp287 as described before. As a result,
compounds 2 and 3 were chosen as the initial testing
candidates for synthesis, following Scheme 1.
For the aminoquinolone-based inhibitors, we previously
reported an excellent correlation between T. brucei growth
inhibition (EC50) and T. brucei MetRS thermal shift (ΔTm),4 a
proxy for binding affinity where a 2 °C shift usually indicates
significant binding.13−15 By the same thermal shift assay, 2 and
3 exhibited ΔTm = 7.8 0.3 °C and 4.5 0.5 °C, respectively,
indicating that the urea-based 2 binds to the MetRS enzyme
more tightly than the guanidine-based 3. Hence, we selected
6343
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
the urea-based 2 as the template in our exploration for
inhibitors with good permeability properties.
Affinity, Activity, and Selectivity of Urea-Based
Inhibitors. To explore the structure−activity-relationship
(SAR), first the benzyl ring of 2 was modified by adding a
hydrophobic substituent (−Cl, −OMe, and −OEt, see Scheme
1 and Table 1) to fill the mainly hydrophobic pocket. These
hydrophilic and hydrophobic residues, a variety of substituents
(−OH, −NH2, −F, and −Cl) as well as heteroaromatic rings
(thiophene and pyridine) were tested. From the ΔTm values, it
can be seen that a hydroxyl group in meta- and ortho- positions
is tolerated but that an amino group binds less favorably.
Fluorine is preferred over chlorine. Substituted thiophene
compounds 26−28 bind as well as the phenyl-based 2. Among
pyridine substituted compounds 29−31, only the 3-pyridine 29
was nearly as active as 2.
Table 1. Binding of Urea Compounds to T. brucei MetRS by
Choosing a minimum ΔTm threshold of 7.5 °C we selected a
subset of compounds for confirmation of their inhibitory
activity in an aminoacylation enzyme inhibition assay
implemented for the T. brucei MetRS enzyme (Table 2). In
a Thermal Shift Assay
3
this assay, the incorporation of H-methionine into amino-
acylated tRNA is measured. The IC50s ranged from 19 to 110
nM, thereby confirming the value of the ΔTm screen to find
effective inhibitors, and 27 was the most potent.
aromatic
ring
T. brucei MetRS ΔTm
a
compd
R1
R2
(°C)
2
3,5-diCl
phenyl
H
H
H
H
H
H
H
H
H
H
H
H
H
7.8 0.3
7.7 0.1
6.4 0.3
5.5 0.1
4.7 0.1
2.7 1.3
2.5 0.1
1.9 0.2
1.1 0.3
1.1 0.3
1.3 0.1
1.1 0.1
0.5 0.2
8.4 0.3
7.7 0.1
7.7 0.1
7.3 0.1
6.3 0.5
5.7 0.2
5.2 0.3
5.8 0.3
4.7 0.2
4.3 0.2
8.3 0.9
7.9 0.1
7.6 0.3
7.1 0.1
4.5 0.2
3.1 0.1
4
3-Cl,5-OMe phenyl
We also examined these potent T. brucei MetRS inhibitors
for selectivity. In the active site the sequence identity between
T. brucei and human mitochondrial MetRS is 79%. With the
human cytosolic enzyme, the identity in the active site is only
38%. Earlier, Replidyne scientists found that aminoquinolone
inhibitors were 1000-fold better inhibitors of bacterial MetRS
than the human mitochondrial enzyme and that they were
virtually inactive against the human cytosolic enzyme.16
Therefore, we tested our urea-based inhibitors only against
human mitochondrial MetRS. As shown in Table 2, the tested
compounds exhibited IC50s ranging from 4349 to >10000 nM,
corresponding to a selectivity index of 39 or better.
5
2,3,5-triCl
3-Cl
phenyl
phenyl
phenyl
phenyl
phenyl
phenyl
phenyl
phenyl
phenyl
phenyl
phenyl
Phenyl
Phenyl
Phenyl
phenyl
Phenyl
Phenyl
Phenyl
Phenyl
Phenyl
phenyl
3-thiophene
6
7
3-OMe
3-OEt
8
9
4-Cl
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
2-Cl
4-OMe
2-OEt
H
2-OMe
4-OEt
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
2-OH
3-OH
3-F
The best urea-based inhibitors were tested in a T. brucei
brucei growth inhibition assay (Table 2). The EC50s ranged
from 150 to 757 nM, and 27 was most potent. An excellent
linear correlation (R2 = 0.95) between T. brucei EC50 and T.
brucei MetRS IC50 values was observed for the tested
compounds, strongly supporting that these compounds inhibit
T. brucei growth by targeting the MetRS enzyme (Figure 2).
To further establish that the MetRS inhibitors were acting
“on target” in T. brucei, we performed experiments to
demonstrate in vivo inhibition of protein synthesis. T. brucei
cultures were incubated for 4 h (approximately 1 doubling
2-F
4-F
4-Cl
3-Cl
4-OH
2-NH2
3-NH2
H
3
3-Cl,5-OMe 3-thiophene
H
time) with H-histidine while in the presence of various test
3,5-diCl
3,5-diCl
3,5-diCl
3,5-diCl
2-thiophene
3-pyridine
2-pyridine
4-pyridine
H
compounds. Newly synthesized protein was collected by
precipitation with trichloracetic acid and quantified by
scintillation counting. Both the aminoquinolone MetRS
inhibitor 1 and the urea compound 26 resulted in a dose-
H
H
H
a
3
ΔTm values are the average of three independent runs.
dependent decrease of incorporation of H-amino acid into
proteins similar to the protein synthesis inhibitor, cyclo-
heximide (Figure 3). Two control compounds were also tested
that act by mechanisms not directly related to protein synthesis:
difluoromethylornithine (DFMO) that inhibits ornithine
decarboxylase and subsequently polyamine biosynthesis, and a
protein farnesyltransferase inhibitor (PFT inh) with anti-T.
brucei activity.17 Incubating T. brucei with these two compounds
did not result in a dose-dependent decrease in protein synthesis
(Figure 3). These results provide evidence that the MetRS
inhibitors are blocking protein synthesis in cultured T. brucei as
is expected by their mechanism of interfering with the normal
processing of tRNA.
The compounds were also tested for host cell toxicity using a
human lymphoblast cell line (CRL-8155) and a hepatocellular
carcinoma cell line (Hep G2) (Table 2). These compounds
showed toxicities to mammalian cell lines with EC50s ranging
from 9.5 to 43 μM, which translates into a selectivity index
compounds were synthesized using the second route in Scheme
1 due to the need to vary the benzyl group on the left part of
the molecule, which was attached last during synthesis. All three
substitution positions (ortho-, meta-, and para-) exhibited a
preference for the smaller substituents (−Cl, −OMe), with the
meta- position exhibiting the highest ΔTm values among all
other tested mono- and nonsubstituted compounds 6−15.
Subsequently, the SAR around the aryl urea moiety was
examined while keeping 3,5-dichlorobenzyl moiety on the other
side of the molecule fixed (Scheme 1 and Table 1).
Compounds were prepared using the first route shown in
Scheme 1 so that the common part containing the 3,5-
dichlorobenzyl moiety was synthesized first. As the aromatic
ring of the aryl-urea is predicted to reside in a pocket that is
created upon inhibitor binding, the pocket was anticipated to
be somewhat flexible. Because the pocket consists of both
6344
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
Table 2. Enzymatic and Cellular Inhibitory Activity of Urea Compounds
a
enzyme assays
cell assays
mammalian cell lines
b
selectivity index
T. brucei brucei
T. brucei
MetRS
IC50 (nM)
human mito
MetRS
IC50 (nM)
CRL-8155
EC50 (nM)
Hep G2
EC50 (nM)
compd
human IC50/T. brucei IC50
EC50 (nM)
27
17
16
26
4
19
29
6591
5430
347
187
95
150
170
210
220
330
415
350
757
22000
12000
>20000
12000
20000
12000
9500
22000
12000
>20000
13000
20000
20000
14500
17900
39
3720
28
4349
155
186
>175
48
45
8348
2
57
>10000
3056
28
18
64
110
4246
39
43200
a
b
Enzyme inhibition assay was measured in aminoacylation assays. Selectivity Index: human mitochondrial MetRS IC50 over T. brucei MetRS IC50.
the protein and will be the subject of another manuscript (Koh
et al., manuscript under preparation). However, the binding
mode of the inhibitors to the active site is in strong agreement
with the model generated because the Replidyne poster had
allowed us to incorporate conformational changes in the
homology model. For example, superposition (Cα-trace based)
of the T. brucei MetRS:compound 2 crystal structure with the
docking model of the same inhibitor revealed minimum
differences in the binding site (Figure 4). Comparing the
Figure 2. Correlation between enzyme (IC50) and parasite growth
(EC50) inhibition for compounds listed in Table 2. Correlation
coefficient R2 is 0.95 (p < 0.0001).
Figure 4. Superposition of T. brucei MetRS:2 complex (protein in
pink) with the docking model of the same inhibitor (protein in cyan).
The binding mode of 2 in the crystal structure (carbons in magenta) at
the active site is in good agreement with that of the model (carbons in
yellow).
Figure 3. Protein synthesis inhibition in T. brucei cells. Midlog
bloodstream form T. brucei were cultured with 3H-histidine and
various compounds at the indicated concentrations relating to their 48
h EC50. After 4 h, the cultures were harvested, proteins precipitated
with trichloroacetic acid, and radiation quantified by scintillation
counting.
two, the differences in atomic positions of the inhibitor are all
less than 1 Å. The chlorine substituted benzyl ring binds to the
methionine pocket, while the urea linked benzene forms a
planar moiety that fits into the auxiliary pocket, as predicted by
the model. The hydrophobic parts of Tyr250 and Tyr472−
Val473 stack against the urea linked benzene, validating the
importance of planarity in this part of the inhibitors. All other
major interactions are otherwise similar. The hydrogen bonds
between the urea NHs and Asp287, as well as the positioning of
a meta-chlorine in the same site of the methionine sulfur atom,
are all predicted correctly by the model. Therefore, the validity
between 27 and 147. Compound 27, which was the most
potent T. brucei growth inhibitor with an EC50 of 150 nM,
showed the highest selectivity index of 147.
Binding Mode to the Target MetRS. Recently, we have
obtained crystal structures of T. brucei MetRS in complex with
six of the inhibitors discussed here. The binding of inhibitors to
T. brucei MetRS resulted in large degree of structural changes in
6345
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
of using the docking model to guide the design of new
inhibitors is confirmed by the crystal structure.
Membrane and Brain Permeability. After we achieved
high affinity binding to the target, potent activity against T.
brucei cultures, and selectivity versus human model cells, it was
particularly important to analyze the potential permeability of
compounds to the CNS in order to develop drugs for treating
late stage HAT. Hence, compounds 2, 4, 17, 26, and 27 were
tested for permeability across a monolayer of MDR1-MDCKII
cells that mimic the blood−brain barrier (Table 3).7−9
Table 3. Cell Permeability of Urea Compounds in the
MDR1-MDCKII Model
Figure 5. Concentrations of compounds 1 and 2 60 min after IP
injection in mice (n = 3 per compound). The concentrations of 1 in
the brain were at or below the limit of detection <0.003 μM.
MDR1-MDCKII
Papp (nm/s)
a
a
b
compd
−34
29
+34
AQ
1
8
52 20
455 62
317 82
0.36 0.33
0.21 0.21
0.42 0.28
0.53 0.12
0.49 0.16
compounds 2, 26, and 27 (Figure 6). Compound 26 exhibited
the best Cmax at 3.7 0.01 μM and a half-life of 63 15 min
2
356 83
172 55
97 21
26
27
4
191
3
73
29
9
3
150 31
83 31
17
0.62 0.13
a
b
Compound 34 is a P-gp pump inhibitor (see main text). AQ:
absorptive quotient.9
Consistent with previous reports of poor oral absorption of
the aminoquinolone compounds,6 compound 1 was observed
to have poor permeability through the MDR1-MDCKII
monolayer (Table 3). Compounds that enter the central
nervous system are typically associated with an apparent
permeability Papp >150 nm/s.8 This liability makes the
aminoquinolone scaffold a poor candidate for HAT develop-
ment because it is unlikely to be orally absorbed or pass
through the blood−brain barrier. Two compounds in the urea
series, 2 and 26, exhibited a Papp >300 nm/s in the presence of
P-gp inhibitor 34 (N-[4-[2-(6,7-dimethoxy-3,4-dihydro-1H-
isoquinolin-2-yl)ethyl]phenyl]-5-methoxy-9-oxo-10H-acridine-
4-carboxamide, GF120918),18 indicating excellent intrinsic
permeability of the compounds. Permeation of compounds 4,
17, and 27 were relatively low and were limited by P-gp efflux.
Figure 6. Pharmacokinetics characteristics of compounds 2, 26, and
27 in a mouse model. Mice (n = 3 per group) were orally administered
compounds by a single dose of 50 mg/kg.
Compound 26 exhibited a moderate Papp value of 172
nm/s measured in the absence of 34 compared to Papp value of
317 82 nm/s in the presence of 34, with an absorptive
quotient (AQ)9 value of 0.42
0.28, suggesting that
55
with a moderate AUC level of 464 69 μM-min as an early
lead compound. The overall mouse PK data suggested that the
metabolic stability of compounds 2, 26, and 27 is reasonably
good but may need further improvement by identifying and
chemically modifying metabolically vulnerable fragments to
further advance these aryl-urea moieties in the future.
Regardless, the demonstrated oral bioavailability (Figure 6)
and brain permeability (Figure 5) for the urea series of
compounds are dramatic improvements of pharmacological
properties over the aminoquinolone compounds.6
Next, we selected compound 26 to test in the murine model
of acute T. brucei infection (Figure 7). This model uses the
highly pathogenic T. brucei rhodesiense subspecies that causes
rapidly rising parasitemia and death within about 4−5 days of
infection. The mice were given 26 or vehicle starting 1 day after
infection. At 48, 72, and 96 h postinfection, parasitemia was
significantly reduced in the 26-treated group compared to the
vehicle-treated group by 94%, 80%, and 63%, respectively
(Figure 7). Unfortunately, parasitemia continued to rise in the
compound 26 is a moderate substrate for P-gp. The
permeability of compound 2 in the absence of 34 remained
very high (Papp, 356 83 nm/s), with an AQ value of 0.21
0.21, indicating that P-gp efflux did not significantly limit the
permeation of this compound.
To corroborate the in vitro permeability experiments, two
compounds were tested for brain permeability in mice by a
single dose of 5 mg/kg. As predicted from the in vitro model
above, compound 1 demonstrated nearly undetectable brain
levels at 60 min after IP injection. In contrast, compound 2 was
actually concentrated in the brain (1.80 0.09 μM) compared
to plasma (0.12
0.02 μM) (Figure 5). The mechanism of
concentration remains to be characterized, but this feature is
likely to be advantageous for treating second stage
trypanosomiasis.
Oral Bioavailability and in Vivo Efficacy of Urea-Based
Inhibitors. To further analyze the PK properties of the urea
series, oral mouse PK experiments were performed with
6346
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
achieve EC50 values of 150 nM with a 147-fold selectivity over
mammalian cell lines (compound 27). Protein labeling studies
provide evidence that the MetRS inhibitors block protein
synthesis in T. brucei cells. Importantly, the new urea
compounds (2 and 26) have high cell permeability according
to MDR1-MDCKII cell assays and compound 2 demonstrated
high levels of brain penetration in mice. Additionally, the urea
compounds were observed to give good plasma levels in mice
after oral dosing. Compound 26 led to significant suppression
of T. brucei rhodesiense in mice but was not curative, probably
due to its modest potency of ∼0.4 μM against this strain.
Through structure-based drug design, we hope to develop more
potent compounds with the needed pharmacological properties
to advance the urea series of MetRS inhibitors through
preclinical development.
Figure 7. Suppression of T. brucei parasitemia in mice with 26.
Animals (n = 5 per group) were infected with T. brucei rhodesiense at
time 0, and treatment was initiated at 24 h postinfection. Mice received
26 or vehicle every 12 h for 4 days by oral administration. Mice were
sacrificed when they became moribund (vehicle-treated mice were all
sacrificed on day 4 and 26-treated mice were sacrificed on day 5).
Differences in parasitemia were significant at 48, 72, and 96 h (p =
0.030, p = 0.027, p = 0.008, respectively). The Y-axis is plotted on a
log10 scale.
EXPERIMENTAL METHODS
■
Chemistry. All starting materials were purchased from various
chemical vendors and used without further purification unless noted.
Silica was EMD 1.07734 Silica Gel 60 0.063−0.200 MM, 70−230
mesh ASTM, and TLC plates were Silica Gel IB-F. H and 13C NMR
1
spectra were recorded on a Bruker AV-300 or AV-500. Chemical shifts
(δ) are reported in parts per million (ppm) relative to internal
1
1
CD3OD (δ 3.31 H NMR, δ 49.2 13C NMR) or CDCl3 (δ 7.24 H
NMR), the coupling constants (J) are given in Hz. The abbreviations
used are as follows: s, singlet; d, doublet; dd, double doublet; t, triplet;
m, multiplet; p, pentet. ESI-MS were recorded on an Agilent 1100
LC/MSD Trap mass spectrometer. The purity of all final compounds
2−31 (>98%) was confirmed by reverse phase HPLC analysis
monitored at 230 nm.
26 group and the animals needed to be euthanized when they
became moribund at 120 h postinfection. No toxicity
attributable to 26 was observed. Although there was a
significant suppressive effect on parasitemia and an extension
of survival for 1 day, the modest in vivo activity is easily
explained by considering the pharmacodynamics of the
experiment. First, when 26 was tested in vitro against the
rhodesiense STIB900 strain, it has an EC50 of 0.42 μM. On the
basis of the pharmacokinetic studies (Figure 4), the plasma
concentrations of 26 are above the IC50 for less than 6 h of the
12 hour dosing interval, and this is not considering the
possibility that plasma protein binding may further reduce free
compound levels in blood. In previous work, we found that T.
brucei needed to be maintained with MetRS inhibitor at 8 times
the EC50 for 72 h to lead to irreversible killing in vitro.4 On the
basis of these considerations, it is actually satisfying to see a
significant in vivo suppression of parasitemia with 26. It is
evident that more potent compounds (e.g., EC50 of at least 0.05
μM or less) with good pharmacokinetic profiles will be
necessary to produce complete clearance of parasites in mice.
With the aid of the recent crystal structure of urea compounds
bound to the T. brucei MetRS, we hope to make progress in
designing and synthesizing new compounds with the needed
level of potency. For example, future plans to improve the
potency of the urea inhibitors may involve making more
extensive interactions with the methionyl substrate pocket as it
appears that the current inhibitors do not fill up the entire
volume of the pocket as confirmed by the crystal structure.
Studies to address potential metabolic liabilities are also in
progress, which may provide information on metabolism “hot-
spots” that may be modified using structure-based design.
1-(3-(3,5-Dichlorobenzylamino)propyl)-3-phenylurea (2). To
a solution of tert-butyl 3-aminopropylcarbamate (36 μL, 0.21 mmol) in
anhydrous CH2Cl2 (1 mL) at 0 °C was added 1-isocyanatobenzene
(25 μL, 0.21 mmol). After the mixture was stirred overnight at room
temperature, the solvent was removed in vacuo. Silica gel flash
chromatography (30−60% EtOAt, hexane) yielded tert-butyl 3-(3-
phenylureido)propylcarbamate (70 mg, quantitative) as a white solid.
1H NMR (500 MHz, CDCl3) δ 7.13−7.39 (m, 5H), 7.01−7.09 (m,
1H), 5.73−5.81 (m, 1H), 5.02−5.10 (m, 1H), 3.23−3.31 (m, 2H),
3.15−3.21 (m, 2H), 1.56−1.66 (m, 2H), 1.46 (s, 9H). ESI-MS m/z
294.1 (M + H)+.
To a solution of TFA:CH2Cl2 (1:4) was added tert-butyl 3-(3-
phenylureido)propylcarbamate (20 mg, 0.068 mmol). After the
mixture was stirred for 3 h at room temperature, the solvent was
removed in vacuo. The residues were redissolved in MeOH (10 mL).
Sodium methoxide (4 mg, 0.07 mmol) was added to the mixture, and
stirred for 15 min. Acetic acid (250 μL) and 3,5-dichlorobenzaldehyde
(12 mg, 0.068 mmol) were added to the mixture, and stirred for 30
min, and then NaCNBH3 (13 mg, 0.2 mmol) was added. After the
mixture was stirred overnight at room temperature, the solvent was
removed in vacuo. Silica gel flash chromatography (5% (10% NH4OH,
90% MeOH), CH2Cl2) yieled 2 (13 mg, 54%) as a white solid (mp
96−97 °C). 1H NMR (500 MHz, MeOD) δ 7.32−7.45 (m, 5H), 7.26
(t, J = 7.9 Hz, 2H), 7.00 (t, J = 7.3 Hz, 1H), 3.76 (s, 2H), 3.27 (t, J =
6.7 Hz, 2H), 2.66 (t, J = 7.1 Hz, 2H), 1.66−1.83 (m, 2H). 13C NMR
(126 MHz, MeOD) δ 157.08, 143.58, 139.55, 134.67, 128.39, 126.75,
126.64, 122.04, 118.89, 52.01, 45.80, 37.18, 29.61. ESI-MS m/z 353.3
(M + H)+.
N1-(3,5-Dichlorobenzyl)propane-1,3-diamine (32). 3,5-Di-
chlorobenzaldehyde (100 mg, 0.57 mmol) was added to a solution
of propyldiamine (500 μL, 6 mmol) and AcOH (3 mL) in MeOH (10
mL). After the mixture was stirred for 30 min, NaCNBH3 (50 mg, 0.79
mmol) was added. The mixture was stirred overnight and the solvent
was removed in vacuo. The residues were washed with EtOAc and 1 M
NaOH (aq). The organic layer was dried over MgSO4 and
concentrated in vacuo, yielding the desired product 32 (120 mg,
CONCLUSIONS
■
In summary, we used the T brucei MetRS homology model to
guide rational modifications of the aminoquinolone scaffold and
arrived at the new urea-based compounds that have improved
PK properties. The homology model proved to be a high
quality representation according to the X-ray structure of
compound 2 bound to T. brucei MetRS. We were able to
1
90%) as a colorless oil. ESI-MS m/z 234.1 (M + H)+. H NMR (300
6347
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
MHz, CDCl3) δ 7.25−7.15 (m, 3H), 3.74 (s, 2H), 2.80 (t, J = 6.9 Hz,
2H), 2.68 (t, J = 6.9 Hz, 2H), 1.66 (p, J = 7.5 Hz, 2H).
Protein Expression and Purification. Full-length T. brucei
MetRS was expressed and purified as previously described4 and used
throughout the study except for crystallization experiments. Human
mitochondrial MetRS was expressed as reported19 with plasmid
provided by Dr. Spremulli.
Protein Crystallization and Structure Determination. De-
tailed descriptions of protein crystallization and structure determi-
nation will be reported in a separate manuscript (Koh et al.,
manuscript under preparation). The structure of the complex of T.
brucei MetRS and 2 was deposited to the Protein Data Bank (PDB ID
code: 3TUN).
Thermal Shift Assay. The thermal shift assay was performed as
previously described4,15 with the modifications that samples contained
the MetRS enzyme (0.4 mg/mL for T. brucei enzyme), 100 μM of
inhibitor, and 5% DMSO. The assays were repeated three times
independently.
T. brucei Methionyl-tRNA Synthetase Aminoacylation Assay.
Enzyme activity was quantified by the attachment of [3H]methionine
to tRNA in the presence of the T. brucei MetRS enzyme. Reactions
were performed in 96-well filter plates with Durapore membranes
(MSHVN4B10; Millipore) in volumes of 75 μL. The reaction was
performed with 25 mM HEPES (pH 7.9), 10 mM MgCl2, 50 mM KCl,
0.2 mM spermine, 0.1 mg/mL bovine serum albumin, 2.5 mM
dithiothreitol, 1% DMSO, and 1 U/mL pyrophosphatase (I1643;
Sigma). Recombinant enzyme (10 nM) and compound inhibitors
(starting concentration varied depending on potency and included 12
serial two dilutions) were mixed with the buffer and preincubated for
15 min. To start the reaction, 400 μg/mL bulk Escherichia coli tRNA
(R4251; Sigma), 0.1 mM ATP, and 250 nM [3H]methionine (80 Ci/
mmol) were added. The plate was incubated without shaking at room
temperature for 120 min. The reactions were stopped by the addition
of 100 μL of cold 10% trichloroacetic acid. The reaction components
were separated from tRNA by filtration through a vacuum manifold
and washed three times with cold 10% trichloroacetic acid. The filter
plates were dried overnight, scintillation fluid was added, and the
counts on the plates were determined in a scintillation plate counter.
Samples were run in quadruplicate, and percent inhibition was
calculated using two different controls (no enzyme and no test
compound) with the following formula:
tert-Butyl 3,5-dichlorobenzyl(3-(1-(4,4-dimethyl-2,6-
dioxocyclohexylidene)ethyl-amino)propyl)carbamate (33). 2-
Acetyldimedone (Dde-OH) (70 mg, 0.39 mmol) was added to a
solution of 32 (90 mg, 0.39 mmol) and DIPEA (0.13 mL, 0.77 mmol)
in MeOH (0.5 mL) and CH2Cl2 (10 mL). After the mixture was
stirred overnight at room temperature, the solvent was removed in
vacuo. Silica gel flash chromatography (4% MeOH, CH2Cl2) afforded
the desired intermediate. Di-tert-butyl dicarbonate (72 mg, 0.33
mmol) was added to a solution of the intermediate (130 mg, 0.33
mmol) and K2CO3 (45 mg, 0.33 mmol) in 10% water and 90% MeCN
(15 mL). After the reaction mixture was stirred overnight at room
temperature, the solvent was removed in vacuo. The residues were
washed with EtOAt and brine. The organic layer was dried over
anhydrous MgSO4, filtered, and then concentrated in vacuo. Silica gel
flash chromatography (10−50% EtOAt, hexane) afforded the
corresponding product 33 (100 mg) as a colorless oil. 1H NMR
(500 MHz, MeOD) δ 7.34 (s, 1H), 7.24 (s, 2H), 4.44 (s, 2H), 3.46−
3.53 (m, 2H), 3.34−3.44 (m, 2H), 2.53 (s, 3H), 2.37 (s, 4H), 1.88−
1.96 (m, 2H), 1.47 (s, 9H), 1.03 (s, 6H). ESI-MS m/z 498.1 (M +
H)+.
1-[3-(3,5-Dichloro-benzylamino)-propyl]-3-thiophen-3-yl-
urea (26). Anhydrous hydrazine (6.7 μL, 0.20 mmol) was added to a
solution of 33 (35 mg, 0.071 mmol) in anhydrous THF (1 mL). The
mixture was irradiated with microwave (a CEM Discover system, 200
W maximum input power) to maintain at 65 °C for 30 min. The
solvent was removed in vacuo, and the residue was redissolved in
anhydrous CH2Cl2 (0.5 mL). To the mixture at 0 °C was added 3-
isocyanatothiophene (9 mg, 0.071 mmol), and the mixture was stirred
overnight at room temperature. The solvent was removed in vacuo,
and the residues were purified by silica gel flash chromatography
(EtOAt, hexane). The purified intermediate was added to a solution of
3% H2O, 3% TIPS, and 94% TFA (1 mL). The solvent was removed
in vacuo, and silica gel flash chromatography (8% (10% NH4OH, 90%
MeOH), CH2Cl2) yielded the desired product 26 (20 mg, 79%) as a
white solid (mp 103−104 °C). 1H NMR (500 MHz, MeOD) δ 7.31−
7.40 (m, 3H), 7.27 (dd, J = 3.2, 5.2 Hz, 1H), 7.14 (dd, J = 1.4, 3.2 Hz,
1H), 6.94 (dd, J = 1.4, 5.2 Hz, 1H), 3.77 (s, 2H), 3.27 (t, J = 6.6 Hz,
2H), 2.66 (t, J = 7.1 Hz, 2H), 1.70−1.79 (m, 2H). 13C NMR (126
MHz, MeOD) δ 156.96, 143.17, 137.28, 134.69, 126.81, 126.75,
123.69, 121.06, 106.40, 51.91, 45.73, 37.22, 29.49. ESI-MS m/z 359.2
(M + H)+.
%inhibition = 100 × (Mnd − Dtd)/(Mnd − Mne)
where Mnd is the average no drug control, Dtd is each test drug value,
and Mne is the average no enzyme control. IC50 values were then
calculated by nonlinear regression (sigmoidal dose response) in Prism
3.0.
1-[3-(3-Chloro-5-methoxy-benzylamino)-propyl]-3-thio-
phen-3-yl-urea (27). 3-Isocyanatothiophene (36 mg, 0.29 mol) was
added to a solution of tert-butyl 3-aminopropylcarbamate (50 mg, 0.29
mmol) in anhydrous CH2Cl2 at 0 °C. After the reaction was stirred
overnight at room temperature, the solvent was removed in vacuo.
Silica gel flash chromatography (30−60% EtOAt, hexane) afforded
tert-butyl 3-(3-(thiophen-3-yl)ureido)propylcarbamate (60 mg, 70%).
To a solution 3% H2O, 3% TIPS, and 94% TFA (1 mL) was added
tert-butyl 3-(3-(thiophen-3-yl)ureido)propylcarbamate (28 mg, 0.094
mol). After the mixture was stirred for 1 h at room temperature, the
solvent was removed in vacuo. Residues were redissolved in MeOH
(10 mL). Sodium methoxide (5 mg, 0.094 mol) was added to the
mixture and stirred for 15 min. Acetic acid (250 μL) and 3-chloro-5-
methoxybenzaldehyde (16 mg, 0.094 mol) were added to the mixture,
stirred for 30 min, and then NaCNBH3 (9 mg, 0.14 mmol) was added
to the mixture. After the mixture was stirred overnight, the solvent was
removed in vacuo. Silica gel flash chromatography (9% (10% NH4OH,
90% MeOH), CH2Cl2) yielded the desired product 27 (15 mg, 45%)
Human Mitochondrial Methionyl-tRNA Synthetase Amino-
acylation Assay. Assays were performed as described above except
for the following changes. Enzyme activity was quantified by the
attachment of [3H]methionine to tRNA in the presence of the human
mitochondrial MetRS enzyme. The reaction was performed with 50
mM Tris-HCl (pH 8.0), 2.5 mM MgCl2, 2.5 mM KCl, 0.2 mM
spermine, 0.2 mg/mL bovine serum albumin, 2.5 mM dithiothreitol,
and 1 U/mL pyrophosphatase (I1643; Sigma). Recombinant enzyme
(20 nM) and compound inhibitors (starting concentration varied
depending on potency and included 12 serial two dilutions) were
mixed with the buffer and preincubated for 15 min. To start the
reaction, 200 μg/mL bulk E. coli tRNA (R4251; Sigma), 2.5 mM ATP,
and 250 nM [3H]methionine (80 Ci/mmol) were added. The plate
was incubated without shaking at 37 °C for 120 min.
T. brucei Growth Inhibition Assay. T. brucei brucei (bloodstream
form strain 427 from K. Stuart, Seattle BioMed, Seattle, WA) were
cultured in HMI-9 medium containing 10% fetal bovine serum,
penicillin, and streptomycin at 37 °C with 5% CO2 and T. brucei
rhodesiense (bloodstream form STIB900 from Simon Croft, London
School of Hygiene and Tropical Medicine, UK) were grown in HMI-
18 containing 20% fetal bovine serum, penicillin, and streptomycin.20
Drug sensitivity of the T. brucei strain was determined in 96-well
microtiter plates in triplicate with an initial inoculum of 5 × 104
trypomastigotes per well. Compound stock solutions were prepared in
DMSO at 20 mM and added in serial dilutions for a final volume of
1
as a colorless oil. H NMR (500 MHz, MeOD) δ 7.28 (ddd, J = 1.8,
3.2, 5.1 Hz, 1H), 7.16 (dd, J = 1.4, 3.2 Hz, 1H), 7.03 (q, J = 1.6 Hz,
1H), 6.97 (dd, J = 1.4, 5.1 Hz, 1H), 6.94 (q, J = 1.8 Hz, 1H), 6.91 (q, J
= 2.0 Hz, 1H), 3.85 (s, 2H), 3.82 (s, 3H), 3.29 (t, J = 6.6 Hz, 2H), 2.78
(t, J = 7.1 Hz, 2H), 1.81 (p, J = 5.7, 6.6 Hz, 2H). 13C NMR (126 MHz,
MeOD) δ 160.78, 157.14, 140.34, 137.23, 134.65, 123.72, 121.07,
120.64, 113.19, 112.75, 106.47, 54.70, 52.00, 45.46, 36.93, 28.92. ESI-
MS m/z 354.9 (M + H)+.
6348
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
200 μL/well. Parasite growth was quantified at 48 h by the addition of
Alamar Blue (Alamar Biosciences, Sacramento, CA).21 Pentamidine
isethionate (Sigma-Aldrich) was included in each assay as a positive
control. Standard errors within assays were consistently less than 15%.
Protein Synthesis Inhibition Assay. Midlog phase T. brucei cells
(5 × 105) were added to wells of a 96-well plate in a 100 μL volume of
HMI-9 media. Compounds at concentrations ranging from 1 to 16
times the 48 h EC50 were added to the plated cells in triplicate and
allowed to incubate for 15 min, with a no-drug control included for
each plate. [3H]-L-histidine (1 mCi/ml, 47.9 Ci/mmol) (Moravek
Biochemicals, Brea, CA) in a volume of 20 μL of PBS was added to
each well and incubated with the parasites for 4 h. Cycloheximide and
difluoromethylornithine were purchased from Sigma-Aldrich; the PFT
inhibitor corresponds to compound 4g in a previous publication.17
The experiment was stopped by adding 100 μL of 20% trichloroacetic
acid and placed at 4 °C for 15 min. Plates were harvested onto a
PerkinElmer Unifilter-96 GF/B filter using an Inotech Biosystems cell
harvester and washed twice with 5% trichloroacetic acid. The filters
were dried overnight in a vacuum oven and then finished by raising the
temperature to 40 °C for 3 h. Scintillation fluid was added, and the
filters were counted using a PerkinElmer MicroBeta2 scintillation
counter. These methods were adapted from Rock et al.22
Mammalian Cell Growth Inhibition Assays. The human
lymphocytic cell line CRL-1855 (American Type Culture Collection)
were grown in RPMI medium (Lonza, Walkersville, MD) with 10%
fetal bovine serum at 37 °C with 5% CO2. The human hepatocellular
cell line HepG2 (American Type Culture Collection) were grown in
DMEM/F-12 medium (Lonza, Walkersville, MD) with 10% fetal calf
serum. Cells (5 × 103/well) were added to 96-well plates and
incubated with serial dilutions of compounds for 48 h. At that time,
cell viability was quantified by addition of Alamar Blue, and plates were
incubated for an additional 4 h at 5% CO2, 37 °C. Absorbance
readings (OD570−600) were used to calculate viability referenced against
cells grown with no inhibitors.
MDR1-MDCKII Assay. Compounds and the controls, Amprenavir
(Moravek Biochemicals) and Propranolol (Sigma), were tested in
triplicate at a final concentration of 3 μM with and without 2 μM 34
(Elacridar, Toronto Research Chemicals Inc.). Amprenavir gives
moderate−high permeability without 34 (typically Papp >200) but is
reversed by 34 (AQ >0.5), whereas Propranolol gives high
permeability without reversal with 34 (typically Papp >300 and AQ
<0.5). MDR1-MDCKII cells were cultured at 37 °C in 5% CO2 in cell
culture media containing 500 mL of Dulbecco’s Modified Eagle’s
Medium with GlutaMAX (GIBCO) supplemented with 50 mL of fetal
bovine serum (GIBCO) and 5 mL of penicillin−streptomycin
(Sigma). Then 3.3 × 105 MDR1-MDCKII cells were plated in the
hanging insert (apical chamber) of the 12-well Transwell plates
(Corning-Costar) in 0.5 mL of cell culture media. An additional 1.5
mL of cell culture media was added to the bottom (basolateral)
chamber. The assay was conducted three days later at 37 °C in 5%
CO2 in transport media containing 500 mL of Hank’s Balanced Salt
Solution (Sigma) supplemented with 12.5 mL of 1 M glucose (Sigma)
and 12.5 mL of 1 M HEPES buffer (Sigma).
nm/s and TEER values >200 ohm/cm2. The Papp(A − B) or apparent
permeability coefficient for each compound was calculated and
reported in nm/s.
P
= (dQ /dt)/(A × Co)
app
where dQ is the amount of compound that has crossed the membrane,
dt is the incubation time of the assay, A is the area of the cell
monolayer, and Co is the initial concentration in the donor well.
The absorptive quotient (AQ) previously described by Troutman
and Thakker9 identifies compounds that may be P-gp pump substrates
by quantifying the functional activity of the P-gp pump during
absorptive transport. An AQ value ≥0.5 is associated with high affinity
for the P-gp pump.
Measurement of Brain Permeability in Mice. All studies
described in this paper using mice were done under an approved
IACUC protocol at University of Washington. Six−eight week old
Swiss Webster female mice (CFW, Charles River, Wilmington, MA)
were administered compound 1 or 2 by IP injection with a dose of 5
mg/kg. The compounds were dissolved in a solution of 5% DMSO,
3% EtOH, and 7% Tween 80 in saline. Each compound was
administered to three mice, and at 1 h they were sacrificed by cervical
dislocation, blood was collected in heparinized capillary tubes, and
brains removed and frozen at −80 °C. Plasma was separated from
whole blood and frozen for later analysis. Ten μL of plasma were
extracted with 30 μL of acetonitrile containing an extraction standard.
Thawed brain tissue was homogenized in PBS containing an extraction
standard using a Pyrex tissue grinder and then extracted with 4 mL of
acetonitrile. Blank mouse brains spiked with known amount of
compound were used to standardize the assay. The samples were
centrifuged and the supernatants dried under vacuum. Samples were
dissolved again in acetonitrile, and compound concentrations were
quantified on a Waters Quattro Micro in MS/MS mode attached to an
Agilent HPLC system as described.23
Pharmacokinetic Studies in Mice. Pharmacokinetic studies of
compounds 2, 26, and 27 were performed by orally administering to
mice 50 mg/kg of compounds dissolved in 5% DMSO, 3% EtOH, and
7% Tween 80 in saline solution. Three mice were used per group.
Then 40 μL of blood was collected at time points of 0.5, 1, 2, 3, 4, and
6 h post injection via tail snips. The plasma samples were extracted
with acetonitrile and analyzed as described above.23
Efficacy Studies in Mice. Five female 6−8 week old Swiss
Webster (CFW) mice (Charles River, Wilmington, MA) were used
per group. At time 0, each mouse was infected by IP injection with 1 ×
104 bloodstream forms of T. rhodesiense STIB900 (gift of Simon Croft,
London School of Hygiene and Tropical Medicine, UK) in 0.2 mL.
Treatment with compounds was initiated at 24 h postinfection.
Compound 26 was mixed in vehicle of 5% DMSO, 3% EtOH, and 7%
Tween 80 in saline solution and administered at a dose of 50 mg/kg
every 12 h for 4 days by oral gavage (0.2 mL per mouse). A second
group received vehicle alone every 12 h for 4 days (0.2 mL) by oral
gavage. Parasitemia was monitored daily by microscopic examination
of tail blood on wet mounts. Statistical analysis on individual days was
performed by unpaired t tests.
On the day of the assay, the cell culture media was replaced with
transport media with and without 2 μM 34 and preincubated for
approximately 30 min. Transepithelial electrical resistance (TEER)
values were then recorded using the Millicell-ERS (Millipore). The
transport media was then aspirated, and compounds and controls were
added to the inserts at a final concentration of 3 μM in 0.4 mL of
transport media with and without 2 μM 34. Plates were incubated with
shaking (∼60 rpm) for 60 min. Immediately following incubation, two
100 μL aliquots were taken from both chambers and the amount of
drug was measured by mass spectroscopy. Lucifer Yellow (Aldrich), a
fluorescence compound used to test membrane integrity during the
assay, was then added to the inserts at a final concentration of 250 μM
and incubated for 60 min with shaking (∼60 rpm). After this final
incubation, 100 μL aliquots were transferred to a black 96-well plate
(BD Falcon) and florescence was measured for each membrane, and
TEER values were again recorded. Mass spectroscopy analysis was
performed on samples with Lucifer Yellow Papp(A − B) values <20
ASSOCIATED CONTENT
* Supporting Information
Additional synthesis and compound characterization details.
This material is available free of charge via the Internet at
http://pubs.acs.org.
■
S
Accession Codes
PDB ID code 3TUN.
AUTHOR INFORMATION
Corresponding Author
* For F.B.: phone, (206) 616-9214; fax, (206) 685-6045; E-
mail, fbuckner@uw.edu; address, University of Washington,
Division of Allergy and Infectious Diseases, Box 357185,
■
6349
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
compounds by P-glycoprotein in cell culture models of intestinal
epithelium. Pharm. Res. 2003, 20, 1210−1224.
Seattle, WA 98195. For E.F.: phone, (206) 685-7048; fax,
(206)-685-7002; E-mail: erkang@uw.edu; address, University
of Washington, Department of Biochemistry and Biomolecular
Structure Center, Box 357742, Seattle, WA 98195.
(10) Evans, R.; Green, L.; Sun, X.; Guiles, J.; Lorimer, D.; Burgin, A.;
Janjic, N.; Jarvis, T.; Davies, D. Co-crystal structure of REP3123 bound
to Clostridium difficile methionyl tRNA synthetase, poster F1−2114. In
Abstracts of the 47th International Conference on Antimicrobial Agents
and Chemotherapy; American Society for Microbiology: Washington,
DC, 2007.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
(11) Larson, E. T.; Kim, J. E.; Zucker, F. H.; Kelley, A.; Mueller, N.;
Napuli, A. J.; Verlinde, C. L. M. J.; Fan, E.; Buckner, F. S.; Van
Voorhis, W. C.; Merritt, E. A.; Hol, W. G. J. Structure of Leishmania
major methionyl-tRNA synthetase in complex with intermediate
products methionyladenylate and pyrophosphate. Biochimie 2011,
93, 570−582.
(12) Seth, P. P.; Ranken, R.; Robinson, D. E.; Osgood, S. A.; Risen, L.
A.; Rodgers, E. L.; Migawa, M. T.; Jefferson, E. A.; Swayze, E. E. Aryl
urea analogs with broad-spectrum antibacterial activity. Bioorg. Med.
Chem. Lett. 2004, 14, 5569−5572.
(13) Pantoliano, M. W.; Petrella, E. C.; Kwasnoski, J. D.; Lobanov, V.
S.; Myslik, J.; Graf, E.; Carver, T.; Asel, E.; Springer, B. A.; Lane, P.;
Salemme, F. R. High-density miniaturized thermal shift assays as a
general strategy for drug discovery. J. Biomol. Screening 2001, 6, 429−
440.
(14) Lo, M. C.; Aulabaugh, A.; Jin, G. X.; Cowling, R.; Bard, J.;
Malamas, M.; Ellestad, G. Evaluation of fluorescence-based thermal
shift assays for hit identification in drug discovery. Anal. Biochem.
2004, 332, 153−159.
(15) Crowther, G. J.; Napuli, A. J.; Thomas, A. P.; Chung, D. J.;
Kovzun, K. V.; Leibly, D. J.; Castaneda, L. J.; Bhandari, J.; Damman, C.
J.; Hui, R.; Hol, W. G.; Buckner, F. S.; Verlinde, C. L.; Zhang, Z.; Fan,
E.; Van Voorhis, W. C. Buffer optimization of thermal melt assays of
Plasmodium proteins for detection of small-molecule ligands. J. Biomol.
Screening 2009, 14, 700−707.
(16) Green, L. S.; Bullard, J. M.; Ribble, W.; Dean, F.; Ayers, D. F.;
Ochsner, U. A.; Janjic, N.; Jarvis, T. C. Inhibition of Methionyl-tRNA
Synthetase by REP8839 and Effects of Resistance Mutations on
Enzyme Activity. Antimicrob. Agents Chemother. 2009, 53, 86−94.
(17) Nallan, L.; Bauer, K. D.; Bendale, P.; Rivas, K.; Yokoyama, K.;
Horney, C. P.; Pendyala, P. R.; Floyd, D.; Lombardo, L. J.; Williams,
D. K.; Hamilton, A.; Sebti, S.; Windsor, W. T.; Weber, P. C.; Buckner,
F. S.; Chakrabarti, D.; Gelb, M. H.; Van Voorhis, W. C. Protein
farnesyltransferase inhibitors exhibit potent antimalarial activity. J. Med.
Chem. 2005, 48, 3704−3713.
(18) Polli, J. W.; Wring, S. A.; Humphreys, J. E.; Huang, L. Y.;
Morgan, J. B.; Webster, L. O.; Serabjit-Singh, C. S. Rational use of in
vitro P-glycoprotein assays in drug discovery. J. Pharmacol. Exp. Ther.
2001, 299, 620−628.
(19) Spencer, A. C.; Heck, A.; Takeuchi, N.; Watanabe, K.; Spremulli,
L. L. Characterization of the human mitochondrial methionyl-tRNA
synthetase. Biochemistry 2004, 43, 9743−9754.
(20) 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.
(21) 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.
(22) Rock, F. L.; Mao, W.; Yaremchuk, A.; Tukalo, M.; Crepin, T.;
Zhou, H.; Zhang, Y. K.; Hernandez, V.; Akama, T.; Baker, S. J.;
Plattner, J. J.; Shapiro, L.; Martinis, S. A.; Benkovic, S. J.; Cusack, S.;
Alley, M. R. An antifungal agent inhibits an aminoacyl-tRNA
synthetase by trapping tRNA in the editing site. Science 2007, 316,
1759−1761.
(23) Van Voorhis, W. C.; Rivas, K. L.; Bendale, P.; Nallan, L.;
Horney, C.; Barrett, L. K.; Bauer, K. D.; Smart, B. P.; Ankala, S.;
Hucke, O.; Verlinde, C. L.; Chakrabarti, D.; Strickland, C.; Yokoyama,
K.; Buckner, F. S.; Hamilton, A. D.; Williams, D. K.; Lombardo, L. J.;
Floyd, D.; Gelb, M. H. Efficacy, pharmacokinetics, and metabolism of
■
Support for this research was provided by the National
Institutes of Health grants AI067921 and AI084004. We are
grateful to Colin McMartin and Regine Bohaceck for providing
the FLO/QXP software, Robert Jacobs and Cindy Rewerts
from Scynexis for advice and assistance in MDR1-MDCKII
assay, Zhongsheng Zhang and Baoshun Zhang for technical
assistance, Sharon Creason for performing mammalian cell
cytotoxicity assays, Matthew Hulverson for assisting with
pharmacokinetic experiments, Linda Spremulli for providing
expression plasmid of human MetRS, and Wesley Van Voorhis,
Alberto Napuli, and the protein production unit of the Medical
Structural Genomics of Pathogenic Protozoa (MSGPP)
program for protein expression.
ABBREVIATIONS USED
■
HAT, human African trypanosomiasis; MetRS, methionyl-
tRNA synthetase; aaRS, aminoacyl-tRNA synthetase; MDCK,
Madin−Darby canine kidney; BLAST, Basic Local Alignment
Search Tool; AQ, absorptive quotient
REFERENCES
■
(1) Rodgers, J. Human African trypanosomiasis, chemotherapy and
CNS disease. J. Neuroimmunol. 2009, 211, 16−22.
(2) Brun, R.; Blum, J.; Chappuis, F.; Burri, C. Human African
trypanosomiasis. Lancet 2010, 375, 148−159.
(3) Sheppard, K.; Yuan, J.; Hohn, M. J.; Jester, B.; Devine, K. M.;
Soll, D. From one amino acid to another: tRNA-dependent amino acid
biosynthesis. Nucleic Acids Res. 2008, 36, 1813−1825.
(4) Shibata, S.; Gillespie, J. R.; Kelley, A. M.; Napuli, A. J.; Zhang, Z.;
Kovzun, K. V.; Pefley, R. M.; Lam, J.; Zucker, F. H.; Van Voorhis, W.
C.; Merritt, E. A.; Hol, W. G.; Verlinde, C. L.; Fan, E.; Buckner, F. S.
Selective inhibitors of methionyl-tRNA synthetase have potent activity
against Trypanosoma brucei infection in Mice. Antimicrob. Agents
Chemother. 2011, 55, 1982−1989.
(5) Jarvest, R. L.; Berge, J. M.; Berry, V.; Boyd, H. F.; Brown, M. J.;
Elder, J. S.; Forrest, A. K.; Fosberry, A. P.; Gentry, D. R.; Hibbs, M. J.;
Jaworski, D. D.; O’Hanlon, P. J.; Pope, A. J.; Rittenhouse, S.;
Sheppard, R. J.; Slater-Radosti, C.; Worby, A. Nanomolar inhibitors of
Staphylococcus aureus methionyl tRNA synthetase with potent
antibacterial activity against gram-positive pathogens. J. Med. Chem.
2002, 45, 1959−1962.
(6) Jarvest, R. L.; Armstrong, S. A.; Berge, J. M.; Brown, P.; Elder, J.
S.; Brown, M. J.; Copley, R. C. B.; Forrest, A. K.; Hamprecht, D. W.;
O’Hanlon, P. J.; Mitchell, D. J.; Rittenhouse, S.; Witty, D. R. Definition
of the heterocyclic pharmacophore of bacterial methionyl tRNA
synthetase inhibitors: potent antibacterially active non-quinolone
analogues. Bioorg. Med. Chem. Lett. 2004, 14, 3937−3941.
(7) Wang, Q.; Rager, J. D.; Weinstein, K.; Kardos, P. S.; Dobson, G.
L.; Li, J.; Hidalgo, I. J. Evaluation of the MDR-MDCK cell line as a
permeability screen for the blood−brain barrier. Int. J. Pharm. 2005,
288, 349−359.
(8) Mahar Doan, K. M.; Humphreys, J. E.; Webster, L. O.; Wring, S.
A.; Shampine, L. J.; Serabjit-Singh, C. J.; Adkison, K. K.; Polli, J. W.
Passive Permeability and P-Glycoprotein-Mediated Efflux Differentiate
Central Nervous System (CNS) and Non-CNS Marketed Drugs. J.
Pharmacol. Exp. Ther. 2002, 303, 1029−1037.
(9) Troutman, M. D.; Thakker, D. R. Novel experimental parameters
to quantify the modulation of absorptive and secretory transport of
6350
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351

Journal of Medicinal Chemistry
Article
tetrahydroquinoline inhibitors of Plasmodium falciparum protein
farnesyltransferase. Antimicrob. Agents Chemother. 2007, 51, 3659−
3671.
6351
dx.doi.org/10.1021/jm300303e | J. Med. Chem. 2012, 55, 6342−6351