Development of a new generation of 4-aminoquinoline antimalarial compounds using predictive pharmacokinetic and toxicology models

Article abstract of DOI:10.1021/jm100057h

Among the known antimalarial drugs, chloroquine (CQ) and other 4-aminoquinolines have shown high potency and good bioavailability. Yet complications associated with drug resistance necessitate the discovery of effective new antimalarial agents. ADMET prediction studies were employed to evaluate a library of new molecules based on the 4-aminoquinolone-related structure of CQ. Extensive in vitro screening and in vivo pharmacokinetic studies in mice helped to identify two lead molecules, 18 and 4, with promising in vitro therapeutic efficacy, improved ADMET properties, low risk for drug-drug interactions, and desirable pharmacokinetic profiles. Both 18 and 4 are highly potent antimalarial compounds, with IC₅₀ values of 5.6 and 17.3 nM, respectively, against the W2 (CQ-resistant) strain of Plasmodium falciparum (for CQ, IC₅₀ = 382 nM). When tested in mice, these compounds were found to have biological half-lives and plasma exposure values similar to or higher than those of CQ; they are therefore desirable candidates to pursue in future clinical trials.

Full text of DOI:10.1021/jm100057h

J. Med. Chem. 2010, 53, 3685–3695 3685
DOI: 10.1021/jm100057h
Development of a New Generation of 4-Aminoquinoline Antimalarial Compounds Using Predictive
Pharmacokinetic and Toxicology Models
Sunetra Ray,† Peter B. Madrid,*,† Paul Catz,† Susanna E. LeValley,† Michael J. Furniss,† Linda L. Rausch,† R. Kiplin Guy,‡
Joseph L. DeRisi,§, Lalitha V. Iyer,† Carol E. Green,† and Jon C. Mirsalis†
†SRI International, 333 Ravenswood Avenue, Menlo Park, California 94025, ‡Department of Chemical Biology and Therapeutics, St. Jude
Children’s Research Hospital, Memphis, Tennessee 38105, §Department of Biochemistry and Biophysics, University of California;San Francisco,
San Francisco, California, and Howard Hughes Medical Institute, University of California;San Francisco, San Francisco, California
Received January 14, 2010
Among the known antimalarial drugs, chloroquine (CQ) and other 4-aminoquinolines have shown high
potency and good bioavailability. Yet complications associated with drug resistance necessitate the
discovery of effective new antimalarial agents. ADMET prediction studies were employed to evaluate a
library of new molecules based on the 4-aminoquinolone-related structure of CQ. Extensive in vitro
screening and in vivo pharmacokinetic studies in mice helped to identify two lead molecules, 18 and 4,
with promising in vitro therapeutic efficacy, improved ADMET properties, low risk for drug-drug
interactions, and desirable pharmacokinetic profiles. Both 18 and 4 are highly potent antimalarial
compounds, with IC50 values of 5.6 and 17.3 nM, respectively, against the W2 (CQ-resistant) strain of
Plasmodium falciparum (for CQ, IC50 = 382 nM). When tested in mice, these compounds were found to
have biological half-lives and plasma exposure values similar to or higher than those of CQ; they are
therefore desirable candidates to pursue in future clinical trials.
Introduction
Therefore, development of suitable antimalarial agents
demands that a number of ADMET considerations be
evaluated, including metabolic stability, potential interactions
with cytochrome P450 (CYP) enzymes, absorption across the
intestinal barrier, and potential for mutagenicity.
Previously we reported on research to develop potent
antimalarial compounds effective against drug-resistant
strains of P. falciparum by varying the chemical substitutions
around the heterocyclic ring and the basic amine side chain of
the popular antimalarial drug chloroquine (CQ).4,5 In the
current study, we screened a panel of these novel antimalarial
compounds for improved leads based on the evaluated
ADMET properties.
Two primary reasons for the clinical failure of drug
candidates are poor pharmacokinetic (PKa) properties
and poor toxicity profiles. Therefore, evaluation of critical
physicochemical and pharmacological properties in the
early stages of drug discovery is essential to accelerate the
conversion of hits into qualified development candidates.1
Both in silico and in vitro studies are often employed to
assess PK parameters, such as absorption, distribution,
metabolism, excretion, and toxicity properties (ADMET),
in parallel with structure-activity relationship (SAR)
studies to select potential candidates early in the drug
discovery process.2
Malaria, a devastating infectious disease caused by the
protozoa Plasmodium falciparum, affects about 200-500
million people worldwide annually, causing mortality in
almost 1% of them.3 Increasing problems associated with
drug resistance warrant the discovery of effective new
antimalarial agents. Although numerous small molecules
demonstrating potent antimalarial activity in vitro are
regularly being discovered, most of these compounds fail
to reach the clinic because of poor PK and toxicity profiles.
Results and Discussion
Chemical Structures and Efficacy. (a) Structures. A series
of 4-aminoquinoline antimalarial compounds were chosen
to represent the structural diversity for the series and to cover
a range of potencies for in vitro antimalarial activity
(Figure 1). The panel includes a small number of CQ an-
alogues with altered substitutions on the quinoline ring,
although the majority of the compounds in the panel contain
substitutions of the alkyl groups attached to the basic
nitrogen position on the aminoalkyl side chain. The SAR
for this series has revealed that structural modifications
at the basic nitrogen position are important for activity
against CQ-resistant strains of P. falciparum and that a
range of chemical substitutions are tolerated at this position.
Therefore, it is critical to determine how the structural
variations at these positions affect the ADMET properties
for compounds within this series.
*To whom correspondences should be addressed. Phone: 1-650-859-
2253. Fax: 1-650-859-3153. E-mail: peter.madrid@sri.com.
a Abbreviations: ADMET, absorption, distribution, metabolism,
excretion, and toxicity; CQ, chloroquine; PK, pharmacokinetics; IC50
,
concentration of inhibitor resulting in 50% inhibition; SAR, structure-
-activity relationship; CYP, cytochrome P450; RBCs, red blood cells;
LC/MS, liquid chromatography/mass spectrometry; MTD, maximum
tolerated dose; AUC, area under the curve; Cmax, maximum plasma
concentration; t1/2, elimination half-life; tmax, time to maximum
concentration.
r
2010 American Chemical Society
Published on Web 04/05/2010
pubs.acs.org/jmc

3686 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
Ray et al.
Figure 1. Structures of 4-aminoquinolines compounds used to profile the ADMET properties for the series.
The calculated physical properties of the compounds in this
panel reveal that all of them are compliant with the Lipinski
“rule of five” criteria6 (Table 1). Individual compounds in the
panel differ significantly in their predicted permeability
(ClogD), solubility, and hydrogen bonding properties,
although all these characteristics are well within the acceptable
“druglike” ranges. Therefore, the results of this informative in
silico analysis do not suggest that any of the compounds should
be eliminated from this purely computational assessment.
(b) In Vitro Efficacy. The synthesized compounds were
tested for their ability to inhibit growth of one drug-sensitive
strain of P. falciparum (3D7) and one drug-resistant strain
(W2). The potencies of the compounds, as indicated by their
IC50 values, are summarized in Table 1. The compounds
were all highly active against the 3D7 strain, with potency
values in the low nanomolar range. Most were equally potent
against the W2 strain at the IC50 level (Table 2).
we hypothesized that the partitioning between human
plasma and RBCs could be an essential property for anti-
malarial activity. The RBC to plasma partition ratio (B/P
ratio) was measured for each of the compounds at 1 and
10 μM. All the compounds partition into RBCs over
plasma with ratios similar to that of CQ at 1 μM (B/P =
3.76; Table 2), but the amount of partitioning in uninfec-
ted RBCs was not significantly correlated with in vitro
activity.
Mutagenicity and Cytotoxicity. Since CQ is known to
intercalate with DNA at a high concentration and to behave
like a weak mutagen,7,8 the compounds were screened for
mutagenicity by the Ames assay.9,10 This assay evaluates the
genotoxicity of a compound by comparing its ability to
induce reverse mutations at selected loci in the employed
bacterial strains. In parallel, the antibacterial activity of
these compounds was assessed by observing substantial
decreases in the number of revertant colonies on the test
plates, clearing or absence of the background bacterial lawn
Since the in vitro screening assay involves targeting the
stage of the parasite inside human red blood cells (RBCs),

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3687
Table 1. Calculated Physical Properties of Antimalarial Compound Panel
calcd sol.a (mM, pH 7)
compd
MW (g/mol)
ClogD (pH 7)
H-bond donor
H-bond acceptor
no. Lipinski violations
CQ
1
319.9
377.5
369.4
353.4
367.9
422
1.7
2.9
2.7
2.4
4.2
5.5
3.7
4.8
4.2
3.6
3.3
4.7
4.5
3.5
5.4
4.7
4.2
4.5
2.1
1.4
1.3
1.3-5.5
29
4.8
1
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
4
2
1.6
2.0
1
4
3
1
3
4
0.10
0.0027
0.17
1
3
5
1
4
6
412
1
4
7
462.8
411.9
368.9
413.9
467.9
392.3
413.9
451.9
415.9
371.9
415.9
369.8
371.9
371.9
319-468
0.042
0.037
0.15
2
4
8
1
5
9
10
1
4
0.30
2
5
11
12
0.0064
0.030
0.24
2
5
1
4
13
14
2
5
0.0024
0.024
0.076
0.029
0.91
1
4
15
16
1
4
1
4
17
18
1
4
2
5
19
20
4.9
6.4
3
5
3
5
overall range
0.0024-29
1-3
3-5
a All calculations were performed using Advanced Chemistry Development Laboratories (ACD/Laboratories) physiochemical property prediction
software; calcd sol. = calculated solubility.
Table 2. Antimalarial Potency of the Test Articles Measured by (a) Efficacy against the Drug-Resistant (W2) and Drug-Sensitive (3D7) Strains of
P. falciparum and (b) RBC Partitioning Efficiency of the Compounds (Concentration in Blood versus Plasma, B/P Ratio)
RBC Partitioning
compd
region of modification
in vitro efficacy, 3D7 IC50, nM
in vitro efficacy, W2 IC50, nM
B/P, 1 μM
B/P, 10 μM
CQ
1
17
382
267
287
290
17.3
34.6
50
3.76
4.80
6.12
4.94
2.58
3.29
3.21
5.69
1.55
3.13
2.85
3.01
2.00
3.13
3.20
2.44
2.74
2.34
5.15
5.73
2.66
4.08
21.31
6.79
5.45
3.67
5.17
3.68
5.36
2.58
2.48
2.52
1.92
1.89
2.48
3.36
2.36
2.23
2.13
3.33
8.77
0.91
heterocycle ring
heterocycle ring
heterocycle ring
115
62
2
3
51
4
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
3°-amine side chain
2°-amine side chain
2°-amine side chain
2°-amine side chain
14.5
20.9
50
5
6
7
45
70
8
2.0
63
3.0
66
9
10
11
12
13
14
15
16
17
18
19
20
5.0
6.0
7.4
1.0
71
9.3
8.0
10.9
3.9
100
80
8.1
13
34
12
5.0
2.5
13
5.6
183.6
45.3
4.7
growth, formation of pinpoint nonrevertant colonies, or
absence of bacterial growth.
(Table 3). For the assessment of metabolic stability, each
of the compounds was incubated with pooled human liver
microsomes in the presence of NADPH, and the disappear-
ance of the parent compound was analyzed by LC/MS/MS.
CQ had a longer half-life than any other compound in
these assays. The ring-substituted compounds with a diethyl
aminoalkyl side chain (1, 2, and 3) had longer half-lives than
the side chain analogues. All the side chains with substituted
tertiary amines had a significantly shorter half-life (<10 min)
than CQ. Since CQ is mostly modified in the side chain,
modifications to the basic side chain may have had a
significant effect on the metabolism of these compounds.
Moreover, the structures containing an aromatic group in
None of the tested compounds showed any mutagenic
activity against either the TA98 frame shift mutation or the
TA100 base substitution strains, indicating that compounds
in this class do not have high mutagenicity. Cytotoxic effects
were observed only at high doses (g50 μM) for all tested
compounds.
Metabolic Stability. The parent compound, CQ, is pri-
marily metabolized by N-dealkylation of the ethyl groups on
its aminoalkyl side chain. This factor prompted us to test for
significant changes in the metabolism of drug candidates, a
majority of which carried modifications in this region

3688 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
Ray et al.
Table 3. Half-Lives (t1/2) of the Test Articles in the Presence of Pooled
Human Microsomes
are often coadministered with antiretroviral drugs. Thus, a
drug-drug interaction with a coadministered drug might
alter the pharmacokinetics of the test compound or vice
versa. This effect is most often the result of drug-induced
induction or inhibition of CYP enzymes.12 Induction of
specific P450 enzymes may alter the metabolic profile of a
drug either by increasing metabolism or by creating an
alternative pathway of metabolism. These changes can have
profound effects on the pharmacology and toxicology of
drugs. Since inhibition of CYP enzymes can have significant
effects on the safety profile of drugs that are coadministered,
the U.S. Food and Drug Administration recommends that
potential drug-drug interactions of any new chemical entity
be investigated prior to regulatory submission. For example,
the classic antimalarial drugs quinine and quinidine are
known to be potent inhibitors of the CYP2D6 enzyme.13
CQ is also known to inhibit CYP2D6 activity significantly.14
CYP2D6 is an important member of the human CYP group
of enzymes. It is estimated that CYP2D6 metabolizes nearly
25% of the drugs commonly used today.15 Studies in the
past have shown that coadministration of CQ can increase
the levels or effects of CYP2D6 substrates, such as selected
β-blockers like metoprolol,16 antiretroviral agents like
indinavir,17 and the antischizophrenic drug thioridazine.18
To evaluate potential drug-drug interactions early in the
antimalarial drug development process, we analyzed the
inhibitory effect of our CQ analogues on a panel of four
common P450 isoforms. In this assay, human liver micro-
somes were incubated with a test article and a cocktail of
specific P450 substrates for each enzyme. The known major
metabolites of the substrates were subsequently quantified
by LC/MS/MS to compute the percentage of inhibition due
to the test compound in comparison to the percentage in no
drug treated controls. As a rule of thumb, enzyme activity
levels of <70% of the level observed for the untreated
controls were considered to be significant inhibition.
As expected, the majority of the compounds inhibited
the CYP2D6 enzyme. As shown in Figure 3 and Table 4,
compounds 5, 6, 7, 10, 12, 14, and 16 showed approximately
40% inhibition of CYP2D6 activity even at 1 μM. The most
potent inhibitor was 8, which showed 77% inhibition at
1 μM. The secondary amine versions of compounds 8, 10,
and 16 (18, 20, and 19, respectively) were found to be much
less powerful inhibitors of CYP2D6. This difference prob-
ably indicates reduced binding to the CYPs and explains why
the secondary amines tend to have longer half-lives than their
tertiary-amine counterparts.
A few of the compounds (specifically 5, 10, 11, and 16) also
somewhat inhibited the CYP3A4 enzyme at the highest
concentration tested (10 μM). Among the 21 compounds
tested, 4 (1, 3, 13, and 20) did not significantly inhibit any of
the CYP isoforms at either of the tested concentrations.
Absorption. Oral administration is the desired route for the
delivery of antimalarial compounds because it maximizes
patient compliance and reduces ancillary care costs. In many
impoverished countries, it is the only practical means of
delivering drugs. For drugs administered orally, the intest-
inal membrane often poses the first barrier to systemic
exposure. Thus, it is useful to evaluate the absorption of a
molecule across the intestinal barrier before it can be con-
sidered as a potential lead candidate.
compd
t1/2 (min), 1 μM
t1/2 (min), 10 μM
CQ
1
749
374.7
116.2
116.3
48.2
13.8
12
NCa
255.6
NCa
3378.5
108.9
13
2
3
4
5
6
17.7
29.3
29.2
14.1
37.1
8.7
7
11.8
8.5
8
9
7.7
7.5
10
11
12
13
14
15
16
17
18
19
20
6.9
6.9
8.2
8.2
6.7
6.4
29.5
8.8
6.3
5.6
38.4
8.6
5.5
94.8
147.6
408.4
276.2
160.8
NCa
a NC = not calculated owing to insufficient decrease in parent
compound.
the side chain, such as 12, 17, 14, and 16, have much shorter
half-lives. In contrast, the three secondary amines lacking the
propyl group on the terminal nitrogen (18, 19, and 20)
showed significantly extended half-lives.
Metabolite Identification. Since metabolic stability is
essential for an effective antimalarial drug and since meta-
bolism pathways can often be blocked by chemical modifica-
tion of the substrate, LC/MS was used to identify the
metabolites of the tertiary amine compounds generated
during incubation with human microsomes. Figure 2 indi-
cates the results obtained for two of these side chain sub-
stituted compounds that showed rapid metabolism. As seen
in studies with CQ,11 the primary metabolites for these
analogues resulted from N-dealkylation reactions of both
the propyl group (N-des-Prop) and the aromatic groups
(N-des-R). The shortened half-lives of the compounds with
benzylic substituted nitrogens on the side chain were
expected because the protons at the benzylic positions are
significantly more acidic than those of an alkyl-substituted
nitrogen; they are therefore more efficient in abstracting
the proton R to the nitrogen, the rate-limiting step in most
N-dealkylation reactions. An important observation was
that the metabolites resulting from the N-dealkylation of
the propyl group were at least as abundant as those obtained
from removal of the benzyl group (similar trends were found
for 11 and 16). This finding explains why we see improved
metabolic stability in compounds 18, 19, and 20 after the
propyl group is removed. Another important finding was
that the methoxy group in 8 and the methylene group in 10,
which give these compounds their strong antimalarial activ-
ity, are metabolically labile (“-methylene-”=23.9% of the
metabolites in the case of 8 and “o-desmethyl”=9.4% of the
metabolites in the case of 10).
Cytochrome P450 Inhibition. Antimalarial drugs are often
administered concurrently with other therapies. For example,
in the African subcontinent, antimalarial treatment is often
confounded by the presence of HIV infection in the same
patient. Under these conditions, antimalarial compounds
The parallel artificial membrane permeation assay (PAMPA)
method uses phospholipid-coated filters to estimate mem-
brane permeability. These assays are highly reproducible and

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3689
Figure 2. Relative percentages of metabolites for 8 (A) and 10 (B) identified by LC/MS after incubation with pooled human microsomes.
and modeling of paracellular absorption. The method is
useful nevertheless because it has the ability to assess the
permeability coefficient (Pe) at a wide range of pH levels, and
pH is often an important factor in drug absorption.
On the basis of a combination of the antimalarial activity
data and the earlier ADMET evaluations, we identified five
major compounds as possible candidates for clinical applica-
tions (1, 3, 4, 18, and 20). We used a double-sink PAMPA
protocol to evaluate their permeability across the gut wall
(Table 5). All the candidates had permeability equal to or
better than that of our parent compound, CQ. Of the five
tested compounds, 3 and 18 had the highest permeability at
pH 7.4. Permeability declined sharply at lower pH values,
indicating that greater absorption would be expected in the
Figure 3. Inhibition of CYP2D6 enzyme activity by the test articles.
Graph shows the amount of the metabolite 10-OH-bufuralol pro-
lower gut than in other parts of the digestive tract.
Hepatotoxicity. Numerous studies have reported hepato-
duced in the presence of each test article (1 μM) as a percentage of
toxicity resulting from use of CQ.19,20 Therefore, we
control (no drug). Values of <70% indicate significant inhibition.
decided to screen our five lead candidates for any potential
amenable for high throughput, although they can be misleading
owing to the lack of metabolizing enzymes and transporters,
hepatotoxicity in our final stage of ADMET screening
(Figures 4). Two independent methods were used to assess

3690 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
Ray et al.
Table 4. Inhibition of Cytochrome P450 Isoforms by the Test Articles at 1 and 10 μMa
CYP inhibition (metabolite produced as a percent of control)
CYP2C19 CYP2D6
CYP3A4a
CYP2B6
CYP2C9
CYP3A4b
compd
1 μM
10 μM
1 μM
10 μM
1 μM
10 μM
1 μM
10 μM
1 μM
10 μM
1 μM
10 μM
CQ
1
112
119
99
134
118
109
121
112
107
117
108
115
117
113
114
100
128
113
109
107
110
112
98
114
118
105
103
102
109
98
128
115
145
126
121
73
113
120
109
93
111
126
115
107
115
114
120
108
136
110
119
107
102
113
115
127
106
123
121
119
122
99
101
81
94
83
61
55
60
23
85
56
98
61
94
64
73
53
79
76
91
96
90
72
57
94
63
13
22
15
8
105
114
107
94
110
125
120
111
108
30
102
102
108
96
106
125
120
113
105
26
2
3
97
95
4
107
104
104
97
101
89
103
76
5
108
98
6
96
103
101
109
108
103
97
92
94
102
99
82
89
7
111
114
104
96
113
119
107
97
104
122
108
106
79
8
117
108
103
108
90
96
85
107
102
94
84
89
9
49
15
10
13
87
17
20
11
4
10
11
12
13
14
15
16
17
18
19
20
85
63
57
39
98
96
96
98
80
87
79
90
69
75
88
108
98
107
93
128
63
110
107
104
101
94
105
101
100
97
102
68
102
92
53
87
106
94
104
85
100
46
82
57
99
89
46
62
99
94
102
98
93
89
99
66
66
84
96
115
102
108
104
100
113
29
56
69
70
90
103
105
102
108
101
106
102
108
103
108
105
97
a Data expressed as percent metabolism of known substrates for each isoform compared to control. The known substrates are as follows: CYP2B6
(bupropion 25 μM), CYP2C9 (diclofenac 10 μM), CYP2C19 (mephenytoin 50 μM), CYP2D6 (bufuralol 10 μM), CYP3A4a (midazolam 4 μM), and
CYP3A4b (testosterone 50 μM). Values of <70% are considered to be significant inhibition.
Table 5. PAMPA Permeability Measurements (Pe, ꢀ10-6 cm/s) across
Three pH Conditionsa
compd
pH 7.4/7.4sink
pH 6.2/7.4sink
pH 5.0/7.4sink
CQ
1
775 ( 195
1858 ( 499
2704 ( 107
1109 ( 176
2400 ( 652
1363 ( 214
51 ( 1
0.5 ( 1
4 ( 1
2 ( 2
0 ( 0
19 ( 6
0 ( 0
125 ( 78
235 ( 8
2 ( 0
979 ( 129
4 ( 1
3
4
18
20
a Values are the mean ( SD across n g 2 samples.
hepatotoxicity: (a) measurement of the percentage of lactate
dehydrogenase (LDH) released to the medium21 and (b) the
MTT assay, which measures the amount of formazan formed
by mitochondrial dehydrogenase activity in viable cells on
3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium (MTT).22
Compounds at 10, 100, and 500 μM were tested on isolated
rat hepatocytes. The results are presented in Figure 4a
(MTT) and Figure 4b (LDH). The cytotoxicity profiles for
all tested compounds were almost identical in both the MTT
assay and the LDH assay. The parent compound, CQ, had
minimal cytotoxicity at lower doses but considerable toxicity
at higher doses (TC50=50 μM for MTT and TC50=77 μM
for LDH). Among the test compounds, 3 was minimally
toxic (TC50=74 μM for MTT and TC50=135 μM for LDH),
and 18 was highly toxic (TC50=7.5 μM for MTT and TC50=
10 μM for LDH). The MTT assay showed high toxicity for
compound 1 (TC50 < 10 μM), but the compound was only
moderately cytotoxic in the LDH assay (TC50 = 40 μM).
Compounds 20 and 4 have almost identical cytotoxicity
profiles (TC50 ≈ 40 μM for MTT and TC50 = 30 μM for
LDH) that are nearly as good as that of CQ. Note, however,
that compound 18 is an extremely potent antimalarial agent
(Table 2) with IC50 < 10 nM, and thus, it might never be
administered in vivo at dose levels as high as those used in
these cytotoxicity assays. At concentrations above 100 μM,
Figure 4. Cytotoxic effects of the compounds on isolated rat
hepatocytes: (a) MTT assay; (b) LDH assay.
all the compounds were significantly cytotoxic. These results
with isolated rat hepatocytes were similar to those obtained
with multiple immortalized cell lines, including HEK-293,
HepG2, BJ, and Raji.
Pharmacokinetic Parameters Evaluated in Mice. On the
basis of the in vitro ADMET studies, compounds 1, 4, 18,
and 20 were selected for PK profiling in mice. As shown in
Figure 5 and Table 6, the plasma concentration profile of a

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3691
single oral dose of the secondary amine 18 (AUClast
2026.66 pmol h/mL) is similar to that of CQ (AUClast
=
=
Toxicity Studies in Mice. Toxicity studies were performed
in male and female CD-1 mice to determine potential adverse
effects and maximum tolerated dose (MTD) for two second-
ary amines, compounds 18 and 20, which had quite different
in vitro hepatotoxicity profiles. Both male and female CD-1
mice (n=(5/sex)/(dose group)) were treated for 7 days with
compound 18 at 50, 100, or 200 (mg/kg)/day or with com-
pound 20 at 100, 200, or 400 (mg/kg)/day. All mice survived
until the end of the study except one male mouse adminis-
tered compound 18 at 200 (mg/kg)/day, which was found
dead on day 2. All treated groups showed sporadic tachy-
cardia. The body weights of all treated mice remained
normal for the duration of the study when compared to the
vehicle-treated controls. No biologically significant changes
in clinical pathology were observed on day 8, and histo-
pathology revealed no findings related to the administration
of the test articles. Reductions in absolute liver weight and
liver-to-body-weight ratios were observed after treatment
with compound 20 at various dose levels. Although these
findings suggest a possible effect of compound 20 on the
liver, the absence of hepatic effects in both serum clinical
chemistry and histopathologic evaluations of the liver sug-
gests that this effect is of minimal toxicological significance.
The MTDs in mice are estimated to be 100 (mg/kg)/day for
compound 18 and 400 (mg/kg)/day for compound 20 admi-
nistered by oral gavage for 7 days.
3
2639.29 pmol h/mL) except that the maximum plasma con-
3
centration (Cmax) for 18 is about 2-fold lower than that of
CQ. In contrast, the plasma exposure of the other second-
ary amine, compound 20, was much lower (AUClast
=
620.49 pmol h/mL and Cmax =65.9 nM). The PK profile
3
of compound 4 looks quite promising. Compared to CQ,
compound 4 offered a longer period of exposure to the drug
(t1/2 = 15.9 h), a much higher exposure level (AUClast =
6792.4 pmol h/mL), and a similar Cmax. Compound 1 had
3
the shortest half-life; its concentration in the plasma dropped
below 11.8 nM within 24 h. The relative bioavailability (Frel)
of each of the candidates was calculated using CQ as the
reference drug. As expected, compound 18 (Frel = 0.7) is
bioequivalent with CQ. Compound 4 (Frel=2.34) appears to
have a better PK profile than CQ and all the other candidates
tested.
Designing a Progression Decision Matrix To Select Leads.
The Medicines for Malaria Venture (MMV, www.mmv.org)
is a public-private partnership dedicated to delivering effec-
tive antimalarial treatment to future generations. To im-
prove the success rate of antimalarial drug development,
MMV now recommends a suitable product profile for all
antimalarial drugs against uncomplicated malaria caused by
P. falciparum in adults and children.23 On the basis of
the MMV guidelines and current ADMET data for these
4-aminoquinolines, we have derived a compound progression
Figure 5. Plasma drug concentration in male CD-1 mice after a
single oral dose at 50 mg/kg.
Table 6. Pharmacokinetic Profile in Mice Following a Single Oral Dose
a
compd
dose po (mg/kg)
t1/2 (h)
tmax (h)
Cmax (nM)
AUClast (pmol h/mL)
3
relative bioavailability (Frel)
CQ
1
50
50
50
50
50
5.5
3.1
0.5
0.5
0.5
2
444.3
261.4
436.6
240.2
65.9
2639.29
725.77
NA
0.25
2.34
0.70
0.21
4
15.9
6.8
6792.42
2026.66
620.49
18
20
4.8
4
a Relative bioavailability (Frel) = AUClast,test/AUClast,CQ
.
Table 7. Efficacy and ADMET Parameters for the Development of a Treatment for Uncomplicated Malaria Caused by Plasmodium falciparum in
Adults and Children
parameter
minimum essential
<2
ideal
18
4
in silico profiling, Lipinski “rule of five”
violations
0
0
0
in vitro potency (IC50, nM, W2 strain)
bacterial mutagenicity, AMES test
<2000
<100
5.6
no increase
17.3
no increase
<3-fold increase
in mutagenicity
>0.5 h
no increase
in mutagenicity
>4 h
in mutagenicity
1.5 h (1 μM),
4.6 h (10 μM)
in mutagenicity
0.80 h (1 μM),
1.8 h (10 μM)
metabolic stability, t1/2 human
microsomes
permeability, PAMPA
pH 7.4/7.4sink
>100 ꢀ 10-6 cm/s
>1000 ꢀ 10-6 cm/s
2400 ꢀ 10-6 cm/s
1109 ꢀ 10-6 cm/s
in vitro P450 inhibition (inhibition of
CYP isoforms at 1 μM)
cytotoxicity, primary rat hepatocytes,
ratio TC50 to IC50 W2
t1/2 (h, rodent)
only CYP2D6
none
none significant
none significant
>20
>100
1340
2313
>2
>80
>300
>8
>200
>1000
6.8
15.9
436
Cmax (nM, rodent)
AUC (nmol h/L, rodent)
240
2026
6792
3

3692 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
Ray et al.
Scheme 1. Synthesis of Side Chain Substituted Secondary
Synthesis of Side Chain Substituted Tertiary Amines. Synth-
esis of 4, 5, 6, and 7 has previously been described (Madrid
et al.24). Synthesis of 9, 10, 17, 8, 12, 15, 14, 11, 13, and 16 has
also been previously published (Madrid et al.4).
Aminesa
Synthesis of Side Chain Substituted Secondary Amines. See
Scheme 1.
General Procedure for Reductive Alkylation. A stock solution
of N1-(7-chloroquinolin-4-yl)propane-1,3-diamine was made in
dry MeOH, and 1.5 mL (0.04 mmol) was aliquoted to the wells
of a 48-well reaction block. An aldehyde was then added to each
well (0.17 mmol, 4 equiv), and care was taken to reduce exposure
to moisture in the air. The reaction block was sealed with a
rubber gasket and shaken for 16 h at 300 rpm. Then sodium
borohydride was added to each well (6.4 mg, 0.17 mmol).
CAUTION: explosion hazard. After bubbling had ceased, the
solution remaining in each well was transferred into another 48-
position reaction block containing cartridges loaded with SCX-
SPE medium. Each well was washed with 1% TFA in MeOH
(1 mL, 2ꢀ) and MeOH (2 mL, 2ꢀ) and eluted with 5% TEA in
MeOH (2 mL, 2ꢀ). The eluted products were collected into glass
test tubes, and solvent was removed on a GeneVac HT-4 (10 mbar,
4.0 h, 35 °C). The partly purified products were then dissolved in
saturated HCl/MeOH (1 mL) and dried again to form the HCl
salts of each compound. Each dry HCl salt was then dissolved in
1 mL of DMSO-d6 and subjected to LC/MS/CLND for analysis
of purity, identity, and yield. A 10 mM stock solution of each
compound was made in DMSO-d6 based on the CLND quanti-
a Reagents and conditions: (a) RCHO (4 equiv), MeOH, 16 h; (b)
NaBH4 (4 equiv), MeOH, 2 h.
decision matrix to guide the development of a new anti-
malarial drug that will conform with the desired product
profile (Table 7) recommended by the MMV. Among all the
candidates tested in our research, compounds 4 and 18 were
well matched with this profile and should be promising leads
for further development.
Conclusion
In conclusion, our ADMET studies allowed us to evaluate
a set of antimalarial lead compounds and to assess the
parameters that will be essential for further lead optimiza-
tion efforts. A combination of in silico, in vitro, and in vivo
ADMET assays was used to establish efficacy and ADMET
parameters for the selection of a preclinical development
candidate. The in silico evaluation confirmed that the
compounds had “druglike” properties but did not give
information of sufficient value to discriminate between
compounds. The in vitro predictive ADMET assays
revealed important differences in permeability and meta-
bolic stability that must be measured during lead optimiza-
tion efforts to maximize compound efficacy. These in vitro
studies indicated that conversion of the side chain of CQ to a
secondary amine created more potent antimalarial candi-
dates with improved PK profiles. Finally, the in vivo PK
studies in mice helped us to identify two potential drug
candidates with potent in vitro activity against CQ-resistant
P. falciparum along with acceptable PK properties. Future
lead optimization efforts are warranted to produce a fully
optimized antimalarial drug candidate ready for preclinical
development studies.
1
tation data, and compounds were analyzed by H NMR and
LC/MS.
2-{[3-(7-Chloroquinolin-4-ylamino)propylamino]methyl}-6-met-
hoxyphenol (20): 68% yield; 100% pure by HPLC. LCMS (ESI)
m/z calcd for C20H22ClN3O2 [M þ H]þ 372.14. Found: 372.49.
1H NMR (400 MHz, MeOD-d4): δ 8.38, (d, J=6.4, 1H), 8.17
(d, J = 9.2, 1H), 7.81 (d, J=2.0, 1H), 7.50 (dd, J=9.2, 2.0, 1H),
6.96 (dd, J=7.6, 1.2, 1H), 6.86 (dd, J=7.6, 1.2, 1H), 6.79 (t, J=
7.6, 1H), 6.67 (d, J=6.4, 1H), 4.19 (s, 2H), 3.83 (s, 3H), 3.56 (t,
J=6.4, 2H), 3.14 (t, J=7.2, 2H), 2.13 (m, 2H).
2-{[3-(7-Chloroquinolin-4-ylamino)propylamino]methyl}-4-met-
hoxyphenol (19): 46% yield; 100% pure by HPLC. LCMS (ESI)
m/z calcd for C20H22ClN3O2 [M þ H]þ 372.14. Found: 372.48.
1H NMR (400 MHz, MeOD-d4): δ 8.43, (m, 2H), 7.91 (d, J=2.0,
1H), 7.71 (dd, J=9.2, 2.0, 1H), 6.92 (m, 2H), 6.81 (m, 2H), 4.20
(s, 2H), 3.71 (s, 3H), 2.21 (m, 2H).
N-Benzo[1,3]dioxol-4-ylmethyl-N0-(7-chloroquinolin-4-yl)pro-
pane-1,3-diamine (18): 82% yield; 100% pure by HPLC. LCMS
(ESI) m/z calcd for C20H20ClN3O2 [M þ H]þ 370.12. Found:
370.47. 1H NMR (400 MHz, DMSO-d6): δ 8.31, (d, J=6.4, 1H),
7.97 (d, J = 9.2, 1H), 7.74 (d, J=2.0, 1H), 7.32 (dd, J=9.2, 2.0,
1H), 6.72-6.82 (m, 3H), 6.50 (d, J=6.4, 1H), 5.86 (s, 2H), 3.83
(s, 2H), 3.42 (t, J=6.4, 2H), 3.28 (t, J=7.2, 2H), 2.84 (t, J=6.8,
2H), 1.97 (m, 2H).
Screening for Antimalarial Activity. Compounds were
screened for growth inhibition activity against P. falciparum
Experimental Section
General. All reagents and starting materials were purchased
from commercial sources and used without further purification.
1H and 13C NMR were recorded on a Varian Utility 400 MHz
spectrometer, in CDCl3, CD3OD, or DMSO-d6 solvent. Che-
mical shifts were reported as parts per million (ppm) downfield
from an internal tetramethylsilane (TMS) standard (δ=0.0 for
1H NMR) or from solvent reference. Coupling constants (J)
were measured in hertz (Hz). Electrospray mass spectra (ES-
MS) were collected on a Waters ZQ 4000 mass spectrometer.
Purity was assessed using by HPLC using an Xterra reversed-
phase C18 column (4.6 mm ꢀ 20 mm, 3 μm) running a binary
gradient with water (with 0.05% TFA) and methanol (with
0.05% TFA) from 0% to 100% methanol over 20 min. All
compounds were confirmed to have g95% purity.
using a fluorescent-activated cell sorting (FACS) assay
described previously.24 Briefly, the test articles were screened
at two concentrations (30 and 200 nM) against both a drug-
sensitive strain, 3D7, and a highly drug-resistant strain, W2. All
screening was done in triplicate, with no drug and nonactive
4-amino-7-chloroquinoline-treated cultures as negative controls
and CQ as a positive control. IC50 values were estimated by
analyzing the resulting log(dose-response) curves by visual
extrapolation.
RBC Partitioning Experiment. Heparinized rat whole blood
was collected and centrifuged at 1000g for 5 min and then at
3000g for 15 min to obtain the reference plasma. Aliquots of
fresh whole blood and reference plasma were spiked with the test
articles at 0-10 μM (spiking solution used PBS with 10%
methanol) and incubated at 37 °C for 30 min, with shaking.
All concentrations were produced in triplicate. Standards were
prepared similarly. After incubation, the whole blood samples
Chemical Synthesis. Synthesis of Ring Substituted 4-Amino-
quinolines. Synthesis of 1, 2, and 3 has previously been described
(Madrid et al.5).

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3693
were centrifuged as before, and the supernatant (plasma) was
collected from the red blood cells into microfuge tubes and
stored on ice. Aliquots (100 μL) of the plasma obtained from
the whole blood and the reference plasma were then diluted with
400 μL of acetonitrile, vortexed, centrifuged at 14 000 rpm for
5 min, and analyzed by LC-MS/MS. The analyte peak areas
obtained from LC-MS analysis were used to calculate the
RBC-to-plasma partitioning ratio (B/P) using the following
equation:
using a 30 min multistep gradient elution and a 3:7 fixed-ratio
flow splitter so that 30% of the postcolumn flow was diverted to
the in-line PDA and MS detectors. The HPLC mobile phase was
0.1% formic acid in water (A) and 0.1% formic acid in aceto-
nitrile (B), and the gradient ranged from 0% to 90% B. The mass
spectrometer was operated in precursor-ion scan (full-scan)
positive-ion electrospray ionization mode and was optimized
for detection of [M þ H]þ molecular ions. Samples were
evaluated for the presence of metabolite masses (both expected
and unexpected) not detected in the controls.
CYP Inhibition Studies. All substrates and cofactors were
purchased from Sigma-Aldrich, St. Louis, MO. The inhibitory
activity of the test articles at 1 and 10 μM was tested on various
CYPs in pooled human liver microsomes (HLMs) (BD-
Gentest). The compounds were incubated at 37 °C for 20 min
with HLMs (0.2 mg/mL) and a substrate cocktail consisting of
25 μM phenacetin (CYP1A2), 25 μM bupropion (CYP2B6),
10 μM diclofenac (CYP2C9), 50 μM mephenytoin (CYP2C19),
10 μM bufuralol (CYP2D6), 50 μM testosterone (CYP3A4),
and 4 μM midazolam (CYP3A4) and cofactors, along with
2.5 mM NADPH/3.3 mM MgCl2 in 100 mM phosphate buffer,
pH 7.4. Positive control incubations containing known inhibi-
tors of the CYPs of interest were also included. Samples were
analyzed by LC-MS/MS. Inhibition of each test article was
reported as a percentage of the activity obtained with the no
drug control.
Parallel Artificial Membrane Permeability Assay. ptAll ma-
terials used for the PAMPA assay were purchased from pION
Inc., Woburn, MA. In PAMPA, a “sandwich” is formed
between a 96-well microtiter plate loaded with magnetic star
disks and a 96-well filter plate (part no. 110212) so that each
composite well is divided into two chambers (donor at the
bottom and acceptor at the top) separated by a 125 μm micro-
filter disk coated with the GIT-0 lipid solution (part no. 110669).
Stock solutions of the test compounds in DMSO were added to
the buffered donor solutions (part no. 110158) at pH levels of
5.0, 6.2, and 7.4 to obtain a final concentration of 50 μM; the
acceptor wells contained the acceptor buffer at pH 7.4 (part no.
11039). The sandwich was formed and allowed to incubate, with
stirring, for 30 min at 40 μM aqueous boundary layer (ABL) in
the Gut-box (part no. 110205); humidity was held constant to
prevent evaporation. The sandwich was then separated, and
both the donor and acceptor compartments were scanned for
absorbance in the UV range (250-500 nm) by a spectrophoto-
meter (SPECTRAmax 190, Molecular Devices, Sunnyvale,
CA). The permeability coefficient Pe (cm/s) values were calcu-
lated using the PAMPA Explorer Command Software (version
3.3) obtained from pION Inc. All assays were carried out in four
separate wells for each pH condition tested, and the results are
reported as the mean ( SD.
ꢀ
ꢁ
B
P
1
CRefPlasma
¼
þ 1
H
CPlasmaWB - 1
where H is the hematocrit or percentage of total blood cells in
whole blood sample (v/v), CRefPlasma is the concentration in
reference plasma, and CPlasmaWB is the concentration in plasma
obtained from whole blood.
AMES Assay for Mutagenicity and Cytotoxicity Screening.
The microbial mutagenicity screening or Ames assay of the test
articles was performed using the plate incorporation method
with Salmonella typhimurium tester strains TA98 and TA100, in
the presence and absence of an Aroclor 1254 induced rat liver
metabolic activation system containing 10% S-9 (MA).
The test articles were assessed for mutagenicity with
Salmonella tester strains TA98 and TA100 (Dr. Bruce Ames,
University of California, Berkeley) in the presence and in
the absence of a rat liver metabolic activation system, S-9
(Molecular Toxicology, Inc., Boone, NC), using two plates
per dose level over a wide range of doses. Frozen tester strains
were inoculated into 50 mL of oxoid nutrient broth no. 2 (CM
67), allowed to sit unshaken for 2-4 h, then gently shaken (100
rpm) for about 11-14 h at 37 °C. To start the assay, 2 mL of
molten top agar, 0.1 mL of tester organisms (about 108
bacteria), 0.5 mL of metabolic activation mixture (10% of
Aroclor 1254 induced rat liver S9, 8 mM MgCl2, 33 mM KCl,
5 mM glucose 6-phosphate, 4 mM NADP, and 0.2 M of 50%
sodium phosphate buffer, pH 7.4), and appropriate amounts of
the test articles (doses of 5, 10, 50, 100, 500, 1000, and 2000 μg/
plate) were added to sterile 13 mm ꢀ 100 mm test tubes and
placed in a 43 °C heating block. The mixture was stirred gently,
poured onto plates containing about 25 mL of minimal glucose
agar, and incubated at ∼37 °C for about 48 h. The revertant
colonies were counted after the incubation period. Appropriate
sterility, solvent (zero dose), and positive control tests were also
performed in parallel.
Metabolic Stability Studies and Metabolite Identification. Test
articles were incubated at 37 °C with pooled human liver
microsomes (BD-Gentest, BD-Biosciences, Woburn, MA) at
two concentrations (1 and 10 μM). The incubation mix consisted
of test articles (1 or 10 μM), 0.2 mg/mL human microsomal
protein, 2.5 mM NADPH, and 3.3 mM MgCl2 in 100 mM
phosphate buffer, pH 7.4. All incubations were performed in
triplicate. Reactions were stopped at the end of incubation by
the addition of cold acetonitrile (1ꢀ volume of reaction). The
rate of loss of parent compound was determined at 0, 15, 30, and
60 min by LC-MS/MS analysis. The amount of compound in
the samples was expressed as a percentage of remaining com-
pound compared to time point zero (100%). The slope (-k) of
the linear regression from ln[test compound] versus time plot
was determined, and t1/2 was calculated from the equation k=
Cytotoxicity Studies with Rat Hepatocytes. All assays were
performed on freshly isolated hepatocytes from 11-week-old
male Sprague-Dawley rats (Charles Rivers Laboratories).
The procedure for hepatocyte isolation has been previously
described.25
MTT Assay. For the MTT assay, freshly isolated rat hepato-
cytes were added to 96-well Biocoat plates (BD Biosciences, San
Jose, CA) at 20 000 cells per well. The hepatocytes were allowed
to attach for approximately 2 h. After attachment, the medium
was removed along with any unattached cells, and medium
containing the test article (10, 100, and 500 μM) was added in
triplicate. The cells were incubated for 20-24 h. The medium
containing the test article was removed, and the cells were rinsed
with warm medium. The MTT assay was performed using the in
vitro toxicology assay kit (TOX-1, Sigma, St. Louis, MO)
according to the manufacturer’s instructions, except for minor
modifications. Briefly, MTT solution was reconstituted with
medium without phenol red and serum and was returned to
the 37 °C incubator for 3-4 h. At the end of the incubation
period, the tetrazolium ring was cleaved by the mitochondrial
ln(2)/t1/2
.
For metabolite identification, samples were prepared in
duplicate by incubating each test article with human micro-
somes at 10 μM for 15 min. Incubations without any test article
generated samples that served as the negative control. Sample
analysis was performed by LC-MS/MS using a Waters 2795
HPLC with a Waters 996 photodiode array detector coupled to
a Waters Micromass Quattro Ultima triple-quadrupole mass
spectrometer. Samples were chromatographed on a Luna C18(2),
250 mm ꢀ 4.6 mm, 5 μm column at a flow rate of 1.0 mL/min

3694 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
Ray et al.
dehydrogenase to form the MTT formazan crystals that are
solubilized in acidified isopropanol. The absorbance of the
purple solution is measured spectrophotometrically at 570 nm.
The background absorbance measured at 690 nm is subtracted
from this value to obtain the final reading. The results are
expressed as a percentage of the absorbance obtained in the
control wells, which had cells treated with medium only (no test
article).
Acknowledgment. We thank Anna Furimsky and Robert
Swezey for their valuable contribution to the project. The work
was supported by NIAID Grants U01-AI-53862 and U01-AI-
75517 and the American Lebanese Syrian Associated Charities
(ALSAC) and St. Jude Children’s Research Hospital.
References
LDH Assay. LDH activity of treated cells was measured by
the LDH assay kit (TOX-7, Sigma) according to the manufac-
turer’s instructions, except for minor modifications. Briefly,
freshly isolated hepatocytes were plated, treated with test arti-
(1) Kassel, D. B. Applications of high-throughput ADME in drug
discovery. Curr. Opin. Chem. Biol. 2004, 8, 339–345.
(2) Shearer, T. W.; Smith, K. S.; Diaz, D.; Asher, C.; Ramirez, J. The
role of in vitro ADME assays in antimalarial drug discovery and
development. Comb. Chem. High Throughput Screening 2005, 8,
89–98.
cles as before, and incubated for 20-24 h. After incubation, the
1
/
cells were rinsed with PBS, lysed with
volume of lysis
10
(3) World Malaria Report; World Health Organization: Geneva, 2008.
(4) Madrid, P. B.; Liou, A. P.; DeRisi, J. L.; Guy, R. K. Incorporation
of an intramolecular hydrogen-bonding motif in the side chain of
4-aminoquinolines enhances activity against drug-resistant P.
falciparum. J. Med. Chem. 2006, 49, 4535–4543.
(5) Madrid, P. B.; Sherrill, J.; Liou, A. P.; Weisman, J. L.; Derisi, J. L.;
Guy, R. K. Synthesis of ring-substituted 4-aminoquinolines and
evaluation of their antimalarial activities. Bioorg. Med. Chem. Lett.
2005, 15, 1015–1018.
(6) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.
Experimental and computational approaches to estimate solubility
and permeability in drug discovery and development settings. Adv.
Drug Delivery Rev. 2001, 46, 3–26.
(7) Schupbach, M. E. Mutagenicity evaluation of the two antimalarial
agents chloroquine and mefloquine, using a bacterial fluctuation
test. Mutat. Res. 1979, 68, 41–49.
solution, incubated at 37 °C for 45 min, and centrifuged to
collect the supernatant. The supernatant was then diluted 5ꢀ by
PBS. Diluted supernatant was then mixed with equal volumes of
the LDH assay mixture and incubated at room temperature for
20 min protected from light. Finally incubation was terminated
by the addition of 1/10 volume of 1 N HCl, and the absorbance
was measured spectrophotometrically at 490 nm. The back-
ground absorbance measured at 690 nm was subtracted from the
reading at 490 nm to obtain the final reading. The results were
expressed as a percentage of the absorbance obtained in the
control wells that had cells treated with medium only (no test
article).
Pharmacokinetic Studies in Male CD1 Mice. All animal
studies were carried out in accordance with SRI International’s
animal care policies. Male CD-1 mice were purchased from
Charles Rivers Laboratories. The plasma pharmacokinetics of
CQ and of compounds 18, 183, and 4 were determined in male
CD-1 mice following administration of a single dose (50 mg/kg,
n=3) by oral gavage . The vehicle used for CQ and compound 4
was 0.9% saline. Compounds 18 and 20 were administered in
5% DMSO and 95% PEG300. For the PK studies, 0.6 mL of
blood was collected from three mice per test article and at 15,
30, and 60 min and 2, 4, 8, 10, 24, and 48 h after dose
administration. All blood samples were collected from the
retro-orbital sinus under isoflurane anesthesia, then placed
in tubes with EDTA. After collection, blood was processed
to plasma by centrifugation at 2000 rpm for 10-20 min,
and samples were analyzed by LC-MS/MS. PK parameters,
including area under the plasma drug concentration versus
time curve (AUC0-t), maximum plasma drug concentration
(Cmax), time to maximum concentration (Tmax), and elimina-
tion half-life (t1/2), were determined using the WinNonlin
Professional software (Pharsight Corporation, Mountain
View, CA).
(8) Obaseiki-Ebor, E. E.; Obasi, E. E. Aspects of chloroquine muta-
genicity. Mutat. Res. 1986, 175, 51–59.
(9) Fluckiger-Isler, S.; Baumeister, M.; Braun, K.; Gervais, V.; Hasler-
Nguyen, N.; Reimann, R.; Van Gompel, J.; Wunderlich, H. G.;
Engelhardt, G. Assessment of the performance of the Ames II
assay: a collaborative study with 19 coded compounds. Mutat. Res.
2004, 558, 181–197.
(10) McCann, J.; Choi, E.; Yamasaki, E.; Ames, B. N. Detection of
carcinogens as mutagens in the Salmonella/microsome test: assay
of 300 chemicals. Proc. Natl. Acad. Sci. U.S.A. 1975, 72, 5135–
5139.
(11) Ansari, A. M.; Craig, J. C. Metabolites of chloroquine: some
observations on desethylchloroquine and N-acetyldesethylchloro-
quine. J. Pharm. Sci. 1994, 83, 1040–1042.
(12) Zhou, S. F. Drugs behave as substrates, inhibitors and inducers
of human cytochrome P450 3A4. Curr. Drug Metab. 2008, 9, 310–
322.
(13) Hutzler, J. M.; Walker, G. S.; Wienkers, L. C. Inhibition of
cytochrome P450 2D6: structure-activity studies using a series of
quinidine and quinine analogues. Chem. Res. Toxicol. 2003, 16,
450–459.
(14) Adedoyin, A.; Frye, R. F.; Mauro, K.; Branch, R. A. Chloroquine
modulation of specific metabolizing enzymes activities: investiga-
tion with selective five drug cocktail. Br. J. Clin. Pharmacol. 1998,
46, 215–219.
Dose-Range-Finding Toxicity Studies in Male and Female CD-
1 Mice. Male and female CD-1 mice (6-8 weeks, Charles River
Laboratories) were given once daily oral doses of of compound
18 at 50, 100, or 200 mg/kg or of compound 20 at 100, 200, and
400 mg/kg for 7 days (n=(5/sex)/(dose group)). An appropriate
amount of test article was suspended in 5% DMSO/95%
PEG300 v/v on the day of dosing. The control group consisted
of vehicle-treated animals. Body weights were checked prior to
dose administration, on day 4, and prior to necropsy on day 8.
Blood for clinical pathology evaluation was collected on day 8
from the retro-orbital sinus and processed for complete hema-
tology and serum chemistry parameters (e.g., glucose, alkaline
phosphatase, alanine aminotransferase, blood urea nitrogen,
aspartate aminotransferase, total protein, albumin, calcium,
phosphorus, cholesterol, triglycerides, total bilirubin, creati-
nine, globulin, sodium). On the final day, tissues were collected
and retained in 10% neutral buffered formalin for histopathol-
ogy. Toxicology parameters (body weight, clinical pathology,
and organ weight data) were computed as the mean ( SD using
the LABCAT In-Life software, version 6.2. All data were
statistically analyzed by one-way ANOVA and a post hoc
Dunnett’s t test.
(15) Beverage, J. N.; Sissung, T. M.; Sion, A. M.; Danesi, R.; Figg, W.
D. CYP2D6 polymorphisms and the impact on tamoxifen therapy.
J. Pharm. Sci. 2007, 96, 2224–2231.
(16) Lancaster, D. L.; Adio, R. A.; Tai, K. K.; Simooya, O. O.; Broad-
head, G. D.; Tucker, G. T.; Lennard, M. S. Inhibition of meto-
prolol metabolism by chloroquine and other antimalarial drugs.
J. Pharm. Pharmacol. 1990, 42, 267–271.
(17) Savarino, A.; Lucia, M. B.; Rastrelli, E.; Rutella, S.; Golotta, C.;
Morra, E.; Tamburrini, E.; Perno, C. F.; Boelaert, J. R.; Sperber,
K.; Cauda, R. Anti-HIV effects of chloroquine: inhibition of viral
particle glycosylation and synergism with protease inhibitors.
J. Acquired Immune Defic. Syndr. 2004, 35, 223–232.
(18) Traficante, L. J.; Hine, B.; Gershon, S.; Sakalis, G. Chloroquine
potentiation of thioridazine effects in rats and drug-resistant
schizophrenic patients: a preliminary report. Commun. Psycho-
pharmacol. 1977, 1, 407–419.
(19) Jamshidzadeh, A.; Niknahad, H.; Kashafi, H. Cytotoxicity of
chloroquine in isolated rat hepatocytes. J. Appl. Toxicol. 2007,
27, 322–326.
(20) Rossmann-Ringdahl, I.; Olsson, R. Porphyria cutanea tarda:
effects and risk factors for hepatotoxicity from high-dose chloro-
quine treatment. Acta Derm.-Venereol. 2007, 87, 401–405.
(21) Du, X. M.; Irino, N.; Furusho, N.; Hayashi, J.; Shoyama, Y.
Pharmacologically active compounds in the Anoectochilus and
Goodyera species. Nat. Med. (Tokyo) 2008, 62, 132–148.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3695
(22) Mosmann, T. Rapid colorimetric assay for cellular growth and
survival: application to proliferation and cytotoxicity assays.
J. Immunol. Methods 1983, 65, 55–63.
(23) Medicines for Malaria Venture (MMV) Product Profiles for Anti-
malarial Drugs. http://www.mmv.org/IMG/pdf/PRODUCT_
PROFILE_with_logo.pdf.
(24) Madrid, P. B.; Wilson, N. T.; DeRisi, J. L.; Guy, R. K. Parallel
synthesis and antimalarial screening of a 4-aminoquinoline library.
J. Comb. Chem. 2004, 6, 437–442.
(25) Green, C. E.; LeValley, S. E.; Tyson, C. A. Comparison of amphe-
tamine metabolism using isolated hepatocytes from five species
including human. J. Pharmacol. Exp. Ther. 1986, 237, 931–936.