Structural specificity of chloroquine-hematin binding related to inhibition of hematin polymerization and parasite growth

Article abstract of DOI:10.1021/jm9902180

Considerable data now support the hypothesis that chloroquine (CQ)- hematin binding in the parasite food vacuole leads to inhibition of hematin polymerization and parasite death by hematin poisoning. To better understand the structural specificity of CQ-hematin binding, 13 CQ analogues were chosen and their hematin binding affinity, inhibition of hematin polymerization, and inhibition of parasite growth were measured. As determined by isothermal titration calorimetry (ITC), the stoichiometry data and exothermic binding enthalpies indicated that, like CQ, these analogues bind to two or more hematin μ-oxo dimers in a cofacial π-π sandwich-type complex. Association constants (K(a)'s) ranged from 0.46 to 2.9 x 10⁵ M^-1 compared to 4.0 x 10⁵ M^-1 for CQ. Remarkably, we were not able to measure any significant interaction between hematin μ-oxo dimer and 11, the 6-chloro analogue of CQ. This result indicates that the 7-chloro substituent in CQ is a critical structural determinant in its binding affinity to hematin μ-oxo dimer. Molecular modeling experiments reinforce the view that the enthalpically favorable π-π interaction observed in the CQ-hematin μ-oxo dimer complex derives from a favorable alignment of the out-of-plane π-electron density in CQ and hematin μ-oxo dimer at the points of intermolecular contact. For 4- aminoquinolines related to CQ, our data suggest that electron-withdrawing functional groups at the 7-position of the quinoline ring are required for activity against both hematin polymerization and parasite growth and that chlorine substitution at position 7 is optimal. Our results also confirm that the CQ diaminoalkyl side chain, especially the aliphatic tertiary nitrogen atom, is an important structural determinant in CQ drug resistance. For CQ analogues 1-13, the lack of correlation between K(a) and hematin polymerization IC₅₀ values suggests that other properties of the CQ-hematin μ-oxo dimer complex, rather than its association constant alone, play a role in the inhibition of hematin polymerization. However, there was a modest correlation between inhibition of hematin polymerization and inhibition of parasite growth when hematin polymerization IC₅₀ values were normalized for hematin μ-oxo dimer binding affinities, adding further evidence that antimalarial 4-aminoquinolines act by this mechanism.

Full text of DOI:10.1021/jm9902180

4630
J . Med. Chem. 1999, 42, 4630-4639
Str u ctu r a l Sp ecificity of Ch lor oqu in e-Hem a tin Bin d in g Rela ted to In h ibition of
Hem a tin P olym er iza tion a n d P a r a site Gr ow th
Sudha Rani Vippagunta,† Arnulf Dorn,§ Hugues Matile,§ Apurba K. Bhattacharjee,‡ J ean M. Karle,‡
William Y. Ellis,‡ Robert G. Ridley,§ and J onathan L. Vennerstrom*,†
College of Pharmacy, University of Nebraska Medical Center, 986025 Nebraska Medical Center, Omaha, Nebraska 68198-6025,
Pharma Division, Preclinical Research, F. Hoffmann-La Roche Ltd., CH-4070 Basel, Switzerland, and Division of
Experimental Therapeutics, Walter Reed Army Institute of Research, Washington, D.C. 20307-5100
Received May 3, 1999
Considerable data now support the hypothesis that chloroquine (CQ)-hematin binding in the
parasite food vacuole leads to inhibition of hematin polymerization and parasite death by
hematin poisoning. To better understand the structural specificity of CQ-hematin binding,
13 CQ analogues were chosen and their hematin binding affinity, inhibition of hematin
polymerization, and inhibition of parasite growth were measured. As determined by isothermal
titration calorimetry (ITC), the stoichiometry data and exothermic binding enthalpies indicated
that, like CQ, these analogues bind to two or more hematin µ-oxo dimers in a cofacial π-π
sandwich-type complex. Association constants (Ka’s) ranged from 0.46 to 2.9 × 105 M-1 compared
to 4.0 × 105 M-1 for CQ. Remarkably, we were not able to measure any significant interaction
between hematin µ-oxo dimer and 11, the 6-chloro analogue of CQ. This result indicates that
the 7-chloro substituent in CQ is a critical structural determinant in its binding affinity to
hematin µ-oxo dimer. Molecular modeling experiments reinforce the view that the enthalpically
favorable π-π interaction observed in the CQ-hematin µ-oxo dimer complex derives from a
favorable alignment of the out-of-plane π-electron density in CQ and hematin µ-oxo dimer at
the points of intermolecular contact. For 4-aminoquinolines related to CQ, our data suggest
that electron-withdrawing functional groups at the 7-position of the quinoline ring are required
for activity against both hematin polymerization and parasite growth and that chlorine
substitution at position 7 is optimal. Our results also confirm that the CQ diaminoalkyl side
chain, especially the aliphatic tertiary nitrogen atom, is an important structural determinant
in CQ drug resistance. For CQ analogues 1-13, the lack of correlation between Ka and hematin
polymerization IC50 values suggests that other properties of the CQ-hematin µ-oxo dimer
complex, rather than its association constant alone, play a role in the inhibition of hematin
polymerization. However, there was a modest correlation between inhibition of hematin
polymerization and inhibition of parasite growth when hematin polymerization IC50 values
were normalized for hematin µ-oxo dimer binding affinities, adding further evidence that
antimalarial 4-aminoquinolines act by this mechanism.
In tr od u ction
correlation between inhibition of hematin polymeriza-
tion and hematin binding affinity for CQ and seven
other antimalarial quinolines.3f Therefore, it is probable
that the interaction of CQ with hematin is an important
event in the manifestation of the antimalarial activity
of this drug.3e,f,5a
With widespread chloroquine (CQ)-resistant Plasmo-
dium falciparum malaria,1 there is a need to design new
antimalarials to take the place of this very important,
but unfortunately now less useful, drug.2 To identify
potential alternatives to CQ, a molecular understanding
of its mechanism of action would be most helpful.
Considerable data now support the hypothesis that CQ
and other antimalarial quinolines inhibit parasite growth
by binding to hematin3 leading to inhibition of hematin
polymerization3c,f,4 and death of the parasite by hematin
poisoning.5 Hematin polymerization, a unique parasite
mechanism to detoxify hematin, is a nonenzymatic
process4d in which the hematin released from parasite
digestion of hemoglobin is converted to hemozoin or
malaria pigment,6 an insoluble ionic polymer in which
adjacent hematin units are linked via a carboxylate-
iron(III) coordinate bond. We recently found a positive
NMR,7 Mo¨ssbauer spectroscopy,8 UV-Vis,5f,9 and
equilibrium dialysis3a experiments provide substantial
evidence for complex formation between CQ and hema-
tin. Recent isothermal titration calorimetry (ITC)
experiments3f suggest that CQ binds to hematin in the
form of two hematin µ-oxo dimers in a cofacial π-π
sandwich-type complex supporting an original hypoth-
esis by Moreau et al.7 We hypothesized3f that this
† University of Nebraska Medical Center.
§ F. Hoffmann-La Roche Ltd.
‡ Walter Reed Army Institute of Research.
10.1021/jm9902180 CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/19/1999

Structural Specificity of CQ-Hematin Binding
Ch a r t 1. Side Chain-Modified CQ Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22 4631
(Scheme 1). Condensation of equimolar amounts of
3-nitroaniline with diethyl ethoxymethylenemalonate
yielded the corresponding arylaminomethylene ester 14
which was cyclized to the 7-nitroquinoline ester 15 with
high regioselectivity. Alkaline hydrolysis of the ester
yielded the corresponding acid 16, the silver salt of
which was decarboxylated in refluxing diphenyl ether
to produce 4-hydroxy-7-nitroquinoline (17). Chlorination
with phosphorus oxychloride produced 7-nitro-4-chlo-
roquinoline (18) which was treated with 2-amino-5-
(diethylamino)pentane at 165 °C under nitrogen to
afford 9. Catalytic hydrogenation of 9 (H2, Pd/C, 50 atm)
yielded 10, isolated as its trihydrochloride salt. The 7H
(7)13 and 7-bromo (8)14 analogues of CQ were synthe-
sized according to the reaction sequence described
above.
In ter a ction w ith Hem a tin . As we have noted
before,3f it would be preferable to carry out both drug-
hematin binding and hematin polymerization experi-
ments at pH 4.8-5.4, the pH of the parasite food
vacuole.15 However, due to the poor solubility of hematin
at low pH, we settled on pH 6.53f as a compromise,
because at this pH hematin is fully soluble and hematin
polymerization still proceeds reasonably.
Ch a r t 2. Quinoline Ring-Modified CQ Analogues
The thermodynamic parameters of hematin binding
for CQ analogues 1-13 are shown in Table 1. In solution
hematin exists almost exclusively in the form of its µ-oxo
dimer,16 and this is reflected in the observed binding
stoichiometries. With the exception of 6, all of the CQ
analogues bound to at least two hematin µ-oxo dimers.
The lower binding stoichiometry for 6 probably derives
from the presence of the ethanol cosolvent required to
solubilize this compound as organic solvents impede
hematin dimerization and aggregation.17 Conversely,
the binding stoichiometry for 7 suggests that this
compound binds to an aggregate of five hematin µ-oxo
dimers. As a whole, the stoichiometry data support a
model proposed by Moreau et al.7 in which CQ binds to
hematin µ-oxo dimers in a cofacial π-π sandwich-type
complex.
binding produces a forward shift in the equilibrium
between hematin monomer and hematin µ-oxo dimer10
thereby decreasing the concentration of hematin mono-
mer available for incorporation into hemozoin, a mo-
lecular mechanism to account for CQ-mediated inhibi-
tion of hematin polymerization.
To advance our understanding of the structural
specificity of CQ-hematin binding, and to test the
hypothesis that hematin can serve as a potential target
for the design of new antimalarial agents,11 we selected
a set of 13 CQ analogues for further study (Charts 1
and 2). For these compounds we measured hematin
binding affinity, inhibition of hematin polymerization,
and inhibition of parasite growth against both CQ-
sensitive and CQ-resistant strains of P. falciparum. In
the course of this work, we also anticipated that useful
structural insights related to CQ drug resistance might
also be gained.
CQ analogues 1-13 had hematin µ-oxo dimer as-
sociation constants (Ka’s) ranging from 0.46 to 2.9 × 105
M-1 (Table 3). Suprisingly, not one of the analogues
bound more avidly to hematin µ-oxo dimer than did CQ.
The 8-9-fold decrease (P ) 0.0071) in binding affinities
for analogues 1 and 2 suggests that the alkyl side chain
of CQ may position the quinoline heterocycle in an
optimal stacking orientation for favorable π-π orbital
overlap with hematin µ-oxo dimer. For CQ analogues
3-5, more subtle structural variation in the alkyl side
chain of CQ reduced binding affinities only 1.5 to 3-fold
(0.029 < P < 0.249). However, replacement of the
tertiary nitrogen atom of CQ with a methine carbon
atom in 6 decreased binding affinity by 5-fold (P )
0.011). As addition of ethanol cosolvent was required
for the titration of hematin with 6 because of its poor
solubility, these data must be considered with caution
because polar protic solvents such as ethanol signifi-
cantly disrupt π-π ligand-receptor interactions.18
Resu lts
Ch em istr y. 4-Amino-7-chloroquinoline (1) and 4-(eth-
ylamino)-7-chloroquinoline (2) were prepared by reac-
tion of 4,7-dichloroquinoline with ammonia and ethyl-
amine, respectively. The 7-NO2 analogue of CQ (9) was
synthesized according to the procedure for the synthesis
of 4-aminoquinolines described by Price and Roberts12
For 7, the 7H CQ analogue, binding affinity decreased
5-fold (P ) 0.011). For 8-10, replacement of the 7-chloro
substituent with bromo, nitro, and amino functional
groups also lowered binding affinities, but only 1.5-3-
fold (0.019 < P < 0.339) compared to CQ. Remarkably,

4632 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22
Vippagunta et al.
Sch em e 1a
a
Key: (a) 90 °C; (b) biphenyl ether, 255 °C; (c) 10% aq NaOH then HCl; (d) NH4OH, AgNO3; (e) biphenyl ether, 255 °C; (f) POCl3, 110
°C; (g) 2-amino-5-(diethylamino)pentane, 120 °C; (h) H2, Pd/C, EtOH then HCl(g).
Ta ble 1. Thermodynamic Characterization of CQ
Analogue-Hematin µ-Oxo Dimer Binding
analogues 12 and 13, methyl substitution at the 2 and
3 positions of the quinoline heterocycle decreased bind-
∆G
∆H
T∆S
ing affinity 3- and 4-fold (P ) 0.012), respectively. This
suggests that the methyl groups sterically prevent the
approach of the quinoline heterocycle to hematin µ-oxo
dimer, and thus for 12 and 13, the efficiency of the
enthalpically favorable π-π stacking interaction may
decrease.
compda
nb
1/n (kcal/mol) (kcal/mol) (kcal/mol)
1
2c
3
0.23 ( 0.02
0.20 ( 0.01
0.21 ( 0.01
0.21 ( 0.01
0.15 ( 0.01
0.43 ( 0.03
4.4 -6.7 ( 0.1 -9.1 ( 1.0 -2.4 ( 1.2
5.0 -6.6 ( 0.2 -9.4 ( 0.5 -2.8 ( 0.6
4.8 -7.7 ( 0.2 -8.7 ( 0.3 -1.0 ( 0.5
4.8 -7.4 ( 0.2 -9.7 ( 0.3 -2.3 ( 0.5
6.7 -7.3 ( 0.2 -9.7 ( 0.4 -2.4 ( 0.4
2.3 -6.9 ( 0.3 -6.3 ( 2.0
-7.0 ( 0.1 -13 ( 2
3.6 -7.2 ( 0.1 -6.1 ( 0.2
5.6 -7.7 ( 0.2 -12 ( 1
7.7 -7.7 ( 0.3 -10 ( 2
4.0 -7.3 ( 0.1 -10 ( 1
5.9 -7.0 ( 0.1 -6.0 ( 0.6
4.2 -7.9 ( 0.4 -10 ( 1
4
5c
6d
7
0.6 ( 0.1
-6.0 ( 2.4
1.1 ( 0.3
As for CQ, drug-hematin µ-oxo dimer binding was
primarily enthalpy-driven for each of the CQ analogues
excepting 11 (Table 1). We attribute this exothermic
binding enthalpy largely to π-π interactions between
the compounds and hematin µ-oxo dimer, although
release of perturbed water molecules from the am-
phiphilic surfaces19 of hematin µ-oxo dimer and these
CQ analogues could also contribute to this favorable
exothermic enthalpy and drive this CQ analogue-
hematin µ-oxo dimer association. For these CQ ana-
logues, a slope of 315.5° (compensating temperature)
from an enthalpy-entropy compensation plot20 (data not
shown) is diagnostic of the participation of hydrophobic
or structural water in the interaction between these
compounds and hematin µ-oxo dimer. The observed
linear enthalpy-entropy compensation also indicates
similar complex geometry. This correlates with results
from molecular modeling experiments (vide infra).
0.10 ( 0.01 10
8
9
10
12
13
CQe
0.28 ( 0.04
0.18 ( 0.01
0.13 ( 0.04
0.25 ( 0.03
0.17 ( 0.03
0.24 ( 0.03
-4.3 ( 1.2
-2.3 ( 1.8
-2.7 ( 0.3
1.0 ( 0.7
-2.1 ( 1.2
a
Analogue 11 was insoluble in phosphate buffer, pH 6.5, but it
was sufficiently soluble in 20 mM MES buffer, pH 5.6.26 Under
these conditions, no measurable binding enthalpy was detectable
by ITC for analogue 11. b Binding stoichiometry. c Analogues 2 and
5 were first dissolved in 3-4 drops of 200 mM acetate buffer, pH
4.0 and then made up to volume with 250 mM phosphate buffer,
d
pH 6.5 Analogue 6 was not soluble in 250 mM phosphate buffer,
so it was first dissolved in EtOH (0.5 mL) and then diluted with
H2O (9.5 mL) to a concentration of 2 mM. e Data from Dorn et al.8
Ta ble 2. Atomic Electrostatic Charge Densities of
7-Chloro-4-aminoquinoline (1) and 6-Chloro-4-aminoquinoline
Calculated at the 6-31G** Level
For 6, 8, and 13, hematin µ-oxo dimer binding was
also entropy-driven. Except that 8 is more hydrophobic
than CQ, there is no obvious structurally based expla-
nation for the albeit modest favorable binding entropy
(9-15%) for these three CQ analogues. Binding entro-
pies, whether favored or disfavored, largely reflect net
changes in the hydrogen-bonded network of water19,21,22
in the hydrophobic hydration shell of the CQ analogue-
hematin µ-oxo dimer complexes vs free CQ analogue and
free hematin µ-oxo dimer assuming a similar unfavor-
able entropy due to the bimolecular ordering of the
complex. It is interesting to note that although 7 and 9
had significantly higher enthalpies (∆H) of hematin
µ-oxo dimer binding than did CQ, the entropic penalty
(T∆S) for this binding also increased such that the free
energy (∆G) of binding changed little due to enthalpy-
entropy compensation.23 This suggests that even though
the π-π interactions between hematin µ-oxo dimer and
7 and 9 were probably more favorable as reflected in
their higher binding enthalpies, there may be an upper
limit to 4-aminoquinoline-hematin µ-oxo dimer binding
affinity and CQ may be near this barrier.
atom
7-chloro
6-chloro
N1
C2
C3
C4
C4a
C5
C6
C7
C8
-0.75
+0.48
-0.70
+0.64
-0.31
-0.13
-0.23
+0.20
-0.38
+0.66
-0.90
-0.14
-0.79
+0.45
-0.78
+0.90
-0.66
-0.10
-0.09
+0.09
-0.55
+0.92
-1.09
-0.13
C8a
NH2
Cl
we were not able to measure any significant interaction
between hematin µ-oxo dimer and 11, the 6-chloro
analogue of CQ. This result indicates that the 7-chloro
substituent in CQ is a critical structural determinant
in its binding affinity to hematin µ-oxo dimer. For CQ

Structural Specificity of CQ-Hematin Binding
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22 4633
F igu r e 2. Model of a 1:2 CQ-hematin µ-oxo dimer π-π
sandwich-type complex.
enthalpically favorable π-π interaction observed in the
CQ-hematin µ-oxo dimer complex derives from a favor-
able alignment of the out-of-plane π-electron density in
CQ and hematin µ-oxo dimer at the points of intermo-
lecular contact.24
These results support a model originally proposed by
Moreau et al.7 in which CQ forms a π-π sandwhich-
type complex with two hematin µ-oxo dimers (Figure
2). As is evident in this model, the electrostatic potential
surface of the hematin µ-oxo dimer indicates an electron-
deficient region around the iron atom on both sides of
the dimer. We conclude that enthalpy-driven CQ-
hematin µ-oxo dimer binding derives largely from at-
tractive electrostatic24 interactions between the quino-
line and porphyrin rings of CQ and hematin µ-oxo
dimer, respectively.
In h ibition of Hem a tin P olym er iza tion . Inhibition
of hematin polymerization and inhibition of plasmodial
growth for CQ analogues 1-13 are shown in Table 3.
CQ analogue 4, the N-desethyl metabolite of CQ, 8, the
7-bromo analogue of CQ, and 12 and 13, the two methyl-
substituted CQ analogues, were as effective as CQ in
inhibition of hematin polymerization in vitro. Remark-
ably, CQ analogue 1, which contains an unsubstituted
4-amino functional group, also shared this distinction.
The other CQ analogues 2, 3, 5, and 6 with modified
alkyl side chains were somewhat less potent. Replace-
ment of the 7-chloro substituent of CQ with a 7-nitro
(9) decreased potency by 6-fold (P ) 0.0001). However,
replacement of the chlorine with a hydrogen (7) or
amino group (10) or shifting the chlorine from the 7 to
the 6 position (11) abolished hematin polymerization
inhibitory activity altogether, results that illustrate the
predominating influence of the quinoline heterocycle
structure on inhibition of hematin polymerization. This
is reinforced by noting that 1, 6, and CQ, with nearly
identical electrostatic isopotential surfaces (Figure 1),
are potent inhibitors of hematin polymerization, whereas
7 and 11, with substantially different electrostatic
F igu r e 1. Three-dimensional isopotential contours of molec-
ular electrostatic potentials at -5 kcal/mol for CQ and CQ
analogues 1, 6, 7, and 11.
Molecu la r Mod elin g. Results of our molecular mod-
eling studies are also consistent with the experimental
observations of the present investigation. Structures for
CQ and CQ analogues 1, 6, 7, and 11 were generated
using ab initio calculations at the 3-21G* level to more
fully understand the structural aspects of CQ-hematin
µ-oxo dimer binding and to shed light on the underlying
basis for the complete lack of binding interaction
between hematin µ-oxo dimer and 11, the 6-chloro
analogue of CQ. Three-dimensional electrostatic poten-
tial profiles (Figure 1) of CQ and four CQ analogues
distinguish 11 from 1, 6, 7, and CQ in terms of their
hematin µ-oxo dimer binding affinities, substantiating
the view that π-π interactions predominate in CQ-
hematin µ-oxo dimer binding.
The contiguous three-dimensional electrostatic po-
tential surfaces of CQ, 1, and 6 were almost identical;
for 7 this electrostatic potential surface was somewhat
smaller which illustrates the effect of the 7-chloro
substituent on the electron distribution in the π orbitals
of the quinoline heterocycle. Conversely for 11, the
6-chloro CQ analogue, there was a substantial gap in
the same electrostatic potential surface at the N1-C8
edge of the quinoline heterocycle, a region which ap-
pears to be critical for binding with the hematin µ-oxo
dimer. Subsequent ab initio calculations at the 6-31G**
level for 7-chloro-4-aminoquinoline (1) and 6-chloro-4-
aminoquinoline, structurally simplified surrogates for
CQ and 11, respectively, allowed us to more accurately
determine atomic electrostatic charge densities for these
molecules (Table 2). As illustrated by these data, shift-
ing the chlorine atom from the 7 to the 6 position
significantly altered the π-electron density distribution
at positions C4, C4a, C8, and C8a. The striking loss of
binding affinity for 11 reinforces the view that the

4634 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22
Vippagunta et al.
Ta ble 3. Hematin µ-Oxo Dimer Binding Affinities (Ka) and
IC50 Values for Inhibition of Hematin Polymerization and
Parasite Growth for CQ Analogues 1-13
P. falciparum
hematin
polymerization
IC50 (nM)a
compd
Ka (105 M-1
)
IC50 (µM)
NF54
K1
1
2
3
4
5
6
7
8
9
10
11
12
13
CQ
0.52 ( 0.11
0.46 ( 0.14
2.7 ( 0.9
1.8 ( 0.5
1.5 ( 0.4
0.79 ( 0.31
0.85 ( 0.07
1.2 ( 0.2
2.9 ( 0.8
2.9 ( 1.1
NA
16 ( 2
70 ( 1
18 ( 1
15 ( 1
60 ( 1
30 ( 7
>2500
18 ( 2
100 ( 10
>2500
>2500
17 ( 1
15 ( 1
15 ( 5
1700
820
41
45
24
54
350
93
1000
870
93
1300
420
200
8000
1400
68
4300
4900
1100
59
15
2600
170
150
20
F igu r e 3. Correlation of normalized hematin polymerization
IC50 values and inhibition of parasite growth. Data points for
7, 10, and 11 were not included as these CQ analogues were
inactive against hematin polymerization in vitro.
0.91 ( 0.20
1.5 ( 0.1
4.0 ( 1.7b
13
350
a
b
Average of n ) 2. Data from Dorn et al., 1998.
that reported by De et al.13 for 7 and 11. Although this
decline in antimalarial potency was less marked for 8,
the 7-bromo analogue of CQ, these results confirm that
a 7-chloro substituent confers optimal antimalarial
potency for 4-aminoquinolines active against P. falci-
parum in vitro.13,27 In contrast to a recent study28 in
which the degree of drug resistance to antimalarial
quinolines correlated with relatively general physico-
chemical properties such as lipophilicity and molar
refractivity, and not on closely defined structural fea-
tures, the antimalarial data for these 13 CQ analogues
illustrate that small structural changes can produce
dramatic effects in relative drug resistance.
Cor r ela tion s. First, it is instructive to note that the
three CQ analogues (7, 10, 11) that did not inhibit
hematin polymerization also had the weakest activity
against the two P. falciparum clones (average IC50
values between 2500 and 5000 nM). For CQ and the
remaining CQ analogues, no correlations exist between
either hematin µ-oxo dimer binding affinities and he-
matin polymerization IC50 values (r ) 0.34, P ) 0.27)
or hematin polymerization and P. falciparum IC50
values (r ) 0.065, P ) 0.85). This result demonstrates
that a compound with excellent activity against hematin
polymerization in vitro does not necessarily possess
antimalarial properties, and, as we have previously
observed3f for primaquine, hematin µ-oxo dimer binding
affinity does not always predict inhibition of hematin
polymerization. Yet, if we normalized the inhibition of
polymerization IC50 data by dividing by Ka, a significant
but modest correlation (r ) 0.61, P ) 0.046) with the P.
falciparum IC50 data was evident (Figure 3). This
correlation improved (r ) 0.93, P < 0.0001) when the
data for CQ analogue 1 was omitted. Additional refine-
ments in this correlation analysis would have to con-
sider the relative accumulation of these CQ analogues
in the parasite food vacuole as elegantly demonstrated
by Hawley et al.29 for CQ, amodiaquine, and several
amodiaquine analogues.
isopotential surfaces (Figure 1), are completely inactive.
For 4-aminoquinolines related to CQ, these data suggest
that electron-withdrawing functional groups at the 7
position of the quinoline ring are required for activity
against hematin polymerization and that chlorine sub-
stitution at position 7 is optimal.
An t im a la r ia l Act ivit y. Each CQ analogue was
screened against the CQ-sensitive NF54 strain and the
CQ-resistant K1 strain of P. falciparum in vitro (Table
3). The 4-amino (1) and 4-ethylamino (2) analogues of
CQ had very little activity against either clone, but for
these two analogues, the differences in potency between
the two clones was insignificant. This result is consis-
tent with observations25 that the structure of the CQ
alkyl side chain is an important determinant in CQ drug
resistance. The excellent activity of 3 (Ro 48-0346), the
Roche short-chain CQ analogue, against both P. falci-
parum strains is consistent with previous IC50 data13,25a
for this compound. Analogue 4, the N-desethyl metabo-
lite of CQ, had very little activity against the K1 strain,
data consistent with previous observations.26 CQ ana-
logue 5, in which the length of the alkyl side chain was
extended by one carbon, was only slightly less potent
than CQ against the two P. falciparum clones. For CQ
analogue 6, in which the tertiary nitrogen atom of CQ
was replaced with a methine carbon atom, potency
against the NF54 strain decreased 4-fold, whereas
potency against the K1 strain increased 1.7-fold, impli-
cating the aliphatic tertiary nitrogen atom of CQ as a
structural feature associated with parasite drug resis-
tance.
With the exception of 13, the 3-methyl analogue of
CQ (sontoquine), and 9, the 7-nitro analogue of CQ, the
remaining CQ analogues that differed from CQ in
quinoline ring substitution were substantially less
potent than CQ against the NF54 strain. In this series,
the 7-amino analogue of CQ (10) was notable for its very
weak activity against the NF54 strain. Conversely, 9
and 13 were more potent than CQ against the K1 strain
of the parasite, and each had a much lower resistance
index than did CQ. Replacement of the chlorine atom
at position 7 with a hydrogen (7) or moving the chlorine
to position 6 (11) decreased the antimalarial activity by
an order of magnitude compared to CQ, but the resis-
tance index was largely unchanged, data consistent with
Discu ssion
This work provides an experimental test of some
theoretical predictions to emerge from a recent molec-
ular modeling study30 probing the structural specificity
of CQ-hematin binding. In the two-point attachment
model proposed for CQ-hematin binding by O’Neill et

Structural Specificity of CQ-Hematin Binding
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22 4635
al.,30 hydrogen bonding and ionic interactions between
the two carboxylates of hematin and the charged
tertiary nitrogen atom of CQ were thought to be
important. However, we have demonstrated31 that
hydrogen bonding and ionic interactions probably con-
tribute little to CQ-hematin µ-oxo dimer binding as this
binding is pH-independent and is nearly unaffected by
high salt concentrations. That replacement of the ter-
tiary nitrogen atom of CQ with a methine carbon atom
(6) decreases hematin µ-oxo dimer binding affinity only
5-fold further supports this interpretation.
CQ. Notwithstanding any difference in food vacuole
accumulation between 7 and CQ, 7 has no effect on
hematin polymerization, suggesting that it exerts its
antimalarial activity, feeble as it is, by another mech-
anism. Nonetheless, if hematin binding contributes to
the specific accumulation of 4-aminoquinolines in the
parasite food vacuole,3a,e,39,40 it is conceivable that the
diminished accumulation and much weaker antima-
larial potency of 7 vs CQ is partly a function of its 5-fold
weaker hematin binding affinity. Similar conclusions
were drawn by Hawley et al.41 from data where parasite
accumulation of 4-aminoquinolines correlated with in-
hibition of parasite growth.
Koh et al.32 demonstrated that CQ can adopt a
conformation with an inter-nitrogen separation (Nquin
-
Ndiethyl) of 8.38 Å,33 closely matching the distance of 8.20
Å between the central iron atom and the carboxylates
of hematin30 as calculated from the X-ray structure of
hemin. However hematin µ-oxo dimer binding data for
3, the short-chain analogue of CQ, and for 5, the CQ
analogue with an extra carbon atom in the side chain,
show that hematin µ-oxo dimer binding affinity is
largely independent of this inter-nitrogen separation.
In addition, NMR data7,34 suggest that no iron coordina-
tion occurs between the hematin iron atom and the
three nitrogen atoms of chloroquine, and if hematin
µ-oxo dimer is considered, iron coordination trans to the
Fe-O-Fe bonds would be minimal as the iron atoms
are displaced some 0.5 Å out of the porphyrin planes
toward the bridging oxygen atom.35
For CQ analogues 1-13, the lack of correlation
between Ka and hematin polymerization IC50 values
suggests that other properties of the CQ-hematin µ-oxo
dimer complex, rather than its association constant
alone, play a role in the inhibition of hematin polym-
erization. For example, CQ-hematin µ-oxo dimer com-
plexation decreases hematin solubility and enhances
hematin µ-oxo dimer aggregation,3a,42 both of which may
also contribute to the CQ-mediated inhibition of hema-
tin polymerization. Moreover, Sullivan et al.3d suggest
that inhibition of hematin polymerization by CQ may
be a function of its binding interaction with both
hematin and hemozoin. Finally, one might also consider
that CQ analogues 7, 10, and 11 had almost no activity
either as hematin polymerization inhibitors or as anti-
malarials, limiting the utility of a correlation analysis.
Instead, our experimental and molecular modeling
data both suggest that enthalpy-driven CQ-hematin
µ-oxo dimer binding derives largely from an energeti-
cally favorable π-π molecular recognition24,36,37 interac-
tion between the quinoline heterocycle of CQ and the
metalloporphyrin ring of hematin µ-oxo dimer. For this
binding interaction, Egan et al.3b suggest that the
hydrophobicity of the aromatic ring of the drug may play
a predominant role. This interpretation is supported by
our enthalpy-entropy compensation temperature of
315.5.20 Yet, replacement of the 7-chloro substituent in
CQ with a 7-amino or 7-nitro substituent decreased
binding affinity only marginally, despite the marked
differences in polarity and electronic nature of the amino
and nitro functional groups. We also found that 8, the
more hydrophobic 7-bromo analogue of CQ, bound 3-fold
less strongly to hematin µ-oxo dimer than did CQ. Most
significantly 11, the 6-chloro analogue of CQ, displayed
no measurable affinity for hematin µ-oxo dimer. Hence,
our data suggest that this π-π binding interaction,
which is apparently optimal for 4-aminoquinolines with
a 7-chloro substituent, derives from an energetically
favorable alignment of the out-of-plane π-electron den-
sity in CQ and hematin µ-oxo dimer at the points of
intermolecular contact.24 As shown by the minimal
decrease in hematin binding affinity for 1 and 2 vs CQ,
this π-π interaction is largely independent of the
structure of the 4-alkylamino side chain.
For this series of CQ analogues, inhibition of hematin
polymerization correlated with observed antimalarial
potencies against the CQ-sensitive NF54 strain of P.
falciparum when hematin polymerization IC50 values
were normalized for hematin µ-oxo dimer binding af-
finities. For a compound to inhibit hematin polymeri-
zation, that compound must accumulate to sufficiently
high concentration in the parasite food vacuole.43
Whether CQ analogues 1-13 meet this criterion is not
known, as the structural modifications in the CQ
analogues may have affected the physicochemical prop-
erties relevant to their cellular accumulation in the
parasite.30 If differences in the CQ analogue food vacuole
accumulation were known and considered, the correla-
tion between inhibition of parasite growth and inhibi-
tion of hematin polymerization would likely improve.29
Our in vitro antimalarial data clearly support the
well-known observation that the 7-chloro substituent is
associated with optimal antimalarial activity, although
we did find that 9, the 7-nitro CQ derivative, was
equipotent to CQ and had a much lower resistance
index. The low in vitro antimalarial activity of 7, the
7H analogue of CQ, supports existing data13,27,44 for this
compound. In contrast, our data show that 11, the
6-chloro analogue of CQ, was 13-fold less potent than
CQ in vitro, although it is as active as CQ in P.
gallinaceum-infected chicks.45 Similarly 8, the 7-bromo
analogue of CQ, was 4-7-fold less potent than CQ in
vitro, although it is as effective as CQ in P. berghei-
infected mice.46 Our data also indicate that replacement
of the tertiary aliphatic nitrogen atom of CQ with a
methine carbon atom (6) does not completely abolish
antimalarial activity as predicted by Thompson and
Werbel.45 Although there was no correlation between
IC50 values against P. falciparum in vitro and compara-
In a study with 7, the 7H analogue of CQ, Veignie
and Moreau27 showed that the weak base properties of
CQ are insufficient to account for its accumulation in
the parasite food vacuole, the organelle in which hema-
tin polymerization takes place. They found that 7 (pKa’s
of 8.8 and 10.2) apparently accumulates 20-50-fold less
than does CQ (pKa’s of 8.1 and 10.2)38 in the parasite,
despite the very similar weak base properties of 7 and

4636 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22
Vippagunta et al.
separated, and the aqueous layer was washed with fresh ether
and basified with 2 N NaOH to afford 1 (15.6 g, 88%) as an
off-white solid. Crystallization from benzene yielded 1 as white
needles: mp 147-148 °C (lit.56 mp 148.5-149.5 °C); 1H NMR
(CDCl3) δ 4.86 (br s, 2H), 6.6 (d, J ) 5.1 Hz, 1H), 7.4 (dd, J )
9.0 Hz, 2.1 Hz, 1H), 7.72 (d, J ) 9.0 Hz, 1H), 8.0 (d, J ) 2.1
Hz, 1H), 8.53 (d, J ) 5.1 Hz, 1H); 13C NMR (CDCl3) δ 103.94,
117.1, 121.71, 125.65, 128.74, 135.23, 149.48, 149.64, 151.74.
7-Ch lor o-4-(eth yla m in o)qu in olin e (2). A mixture of 4,7-
dichloroquinoline (5.94 g, 30 mmol) and ethylamine (22.0 g)
were heated with stirring in a bomb at an oil bath temperature
of 90-95 °C for 64 h. After cooling, excess ethylamine was
allowed to evaporate; 1 N KOH (200 mL) was added and an
off-white solid (5.97 g, 96%) was obtained after filtration and
washing with water and cold ether. Recrystallization from
ether afforded 2 as a white crystalline solid: mp 172-173 °C;
1H NMR (CDCl3) δ 1.4 (t, J ) 7.3 Hz, 3H), 3.3-3.4 (m, 2H),
5.1 (br s, 1H), 6.4 (d, J ) 5.1 Hz, 1H), 7.32 (dd, J ) 9.0, 1.5
Hz, 1H), 7.7 (d, J ) 9.0 Hz, 1H), 7.95 (d, J ) 1.5 Hz, 1H), 8.53
(d, J ) 5.5 Hz, 1H); 13C NMR (CDCl3) δ 14.15, 37.83, 98.99,
117.07, 120.99, 125.13, 128.65, 134.72, 149.04, 149.65, 151.97.
Eth yl R-Car beth oxy-â-(3-n itr oan ilin o)acr ylate (14). 3-Ni-
troaniline (50.0 g, 0.362 mol) and diethyl ethoxymethylene-
malonate (78.3 g, 0.362 mol) were heated in a three-necked
flask for 1 h at 90 °C. On cooling and recrystallization from
95% EtOH, 14 (107.5 g, 96%) was obtained as a yellow solid:
mp 81-82 °C (lit.57 mp 81-82 °C); 1H NMR (CDCl3) δ 1.3-1.4
(m, 6H), 4.26-4.33 (q, J ) 7.2 Hz, 4H), 7.4 (d, J ) 8.1 Hz,
1H), 7.6 (tt, J ) 7.9, 2.6 Hz, 1H), 8.0 (br d), J ) 9.4 Hz, 2H),
8.5 (dd, J ) 13.2, 2.7 Hz, 1H).
3-(E t h oxyca r b on yl)-4-h yd r oxy-7-n it r oq u in olin e (15).
14 (107.5 g, 0.35 mol) dissolved in biphenyl ether (150 mL)
was refluxed at 255 °C for 0.5 h during which time the
liberated ethanol was collected by distillation. The reaction
mixture was cooled and filtered and the solid reaction residue
was washed thoroughly with acetone to give a light yellow
solid. Recrystallization from DMSO afforded 15 (28.1 g, 31%)
as a pale yellow crystalline solid: mp 290-296 °C dec; 1H NMR
(DMSO-d6) δ 1.3 (t, J ) 7.1 Hz, 3H), 4.2 (q, J ) 7.2 Hz, 2H),
8.1 (dd, J ) 9.0, 2.3 Hz, 1H), 8.4 (d, J ) 8.8 Hz, 1H), 8.5 (d, J
) 2.2 Hz, 1H), 8.8 (s, 1H).
tive quantitative antimalarial activities against P. galli-
naceum-infected chicks for CQ and CQ analogues 4, 7,
8, 12, and 13 (data not shown), this vitro antimalarial
data and that of De et al.13 reinforce the assertion by
Thompson and Werbel45 that chemical modification of
the quinoline heterocycle of CQ tends to have a more
pronounced impact on vivo antimalarial activity than
do modifications of the side chain.
In summary, as determined by isothermal titration
calorimetry (ITC), the stoichiometry data and exother-
mic binding enthalpies indicate that CQ and these 13
CQ analogues bind to two or more hematin µ-oxo dimers
in a cofacial π-π sandwich-type complex. Remarkably,
we were not able to measure any significant interaction
between hematin µ-oxo dimer and 11, the 6-chloro
analogue of CQ. This result indicates that the 7-chloro
substituent in CQ is a critical structural determinant
in its binding affinity to hematin µ-oxo dimer. For
4-aminoquinolines related to CQ, our data suggest that
electron-withdrawing functional groups at the 7-position
of the quinoline ring are required for activity against
both hematin polymerization and parasite growth and
that chlorine substitution at position 7 is optimal. For
this set of CQ analogues, inhibition of hematin polym-
erization did correlate with inhibition of parasite growth
when hematin polymerization IC50 values were normal-
ized for hematin µ-oxo dimer binding affinities, adding
further evidence that antimalarial 4-aminoquinolines
act by this mechanism. This work also confirms that
the CQ alkyl side chain, especially the aliphatic tertiary
nitrogen atom, is an important structural determinant
in CQ drug resistance. Some of these CQ analogues may
be useful probes to elucidate the structural basis of CQ
drug resistance in future work with proteins encoded
by CQ-resistance associated genes.47,48 Moreover, these
data also support an assertion by Ridley and Hudson49
that a new 4-aminoquinoline active against drug-
resistant malaria could be identified.
3-Ca r boxy-4-h yd r oxy-7-n itr oqu in olin e (16). 15 (20.0 g,
0.076 mol) was stirred in 10% aqueous NaOH (150 mL) for
0.5 h at 100 °C. The reaction mixture was then cooled and
neutralized with HCl to obtain 16 (11.4 g, 64%) as a reddish
1
brown solid: mp 274-275 °C dec (lit.57 mp 287 °C); H NMR
Exp er im en ta l Section
(DMSO-d6) δ 8.3 (dd, J ) 8.94, 2.2 Hz, 1H), 8.5 (d, J ) 8.9 Hz,
Ch em istr y. CQ analogues 4 (7-chloro-4-(4-(ethylamino)-1-
methylbutyl)aminoquinoline dioxalate),50 5 (7-chloro-4-(5-(di-
ethylamino)-1-methylamyl)aminoquinoline),51 6 (7-chloro-4-(5-
ethyl-1-methylheptyl)aminoquinoline hydrochloride), 8 (7-
bromo-4-(4-(diethylamino)-1-methylbutyl)aminoquinoline),14 11
(6-chloro-4-(4-(diethylamino)-1-methylbutyl)aminoquino-
line),51 12 (7-chloro-2-methyl-4-(4-(diethylamino)-1-methylbu-
tyl)aminoquinoline),52,53 and 13 (7-chloro-3-methyl-4-(4-(dieth-
ylamino)-1-methylbutyl)aminoquinoline)54 were obtained from
the Division of Experimental Therapeutics inventory at the
Walter Reed Army Institute of Research. CQ analogue 3 (7-
chloro-4-(4-(diethylamino)-1-methylethyl)aminoquinoline)55 was
obtained from the Hoffmann-La Roche inventory. Other
chemicals and reagents were obtained from Aldrich or Sigma
Chemical Co. Melting points were determined on a Mel-Temp
apparatus and are uncorrected. NMR spectra were recorded
using a Varian XL-300 spectrometer. Elemental analysis were
performed by M-H-W Laboratories, Phoenix, AZ. Thin-layer
chromatography was carried out on precoated Analtech silica
gel coated plates.
1H), 8.7 (d, J ) 2.3 Hz, 1H), 9.1 (s, 1H).
4-Hyd r oxy-7-n itr oqu in olin e (17). 16 was decarboxylated
by way of its silver salt as previously reported.57,58 To 16 (11.4
g, 49 mmol) suspended in boiling water was added concen-
trated NH4OH (0.82 g, 49 mmol) to obtain a clear solution. To
this clear solution was added saturated aqueous silver nitrate
(8.23 g, 49 mmol) to obtain a light gray solid. The reaction
mixture was refluxed for 0.5 h and then allowed to cool to
afford the silver salt of 16 (16.3 g, 99%). To the dried silver
salt was added biphenyl ether (100 mL) and the reaction
mixture was heated at 255 °C for 1 h. On cooling, the reaction
solid was filtered and washed thoroughly with hexane. This
crude reaction product was dissolved in ethanol (300 mL) and
refluxed for 6 h to obtain after cooling 17 (3.14 g, 34%) as a
light yellow solid: mp 314-317 °C dec (lit.57 mp 317-320 °C);
1H NMR (DMSO-d6) δ 6.2 (d, J ) 7.4 Hz, 1H), 8.1 (dd, J ) 8.9,
2.2 Hz, 1H), 8.12 (d, J ) 7.5 Hz, 1H), 8.3 (d, J ) 8.7 Hz, 1H),
8.5 (d, J ) 2.2 Hz, 1H).
4-Ch lor o-7-n itr oqu in olin e (18). 17 (3.1 g, 0.016 mol) was
treated with phosphorus oxychloride (7.0 mL) and the mixture
was refluxed at 110 °C for 45 min under N2. Ice water was
added slowly to the cooled reaction mixture until evolution of
HCl ceased. Basification with 10% aqueous NaOH afforded a
light yellow precipitate (3.34 g, 98%): mp 156-160 °C (lit.57
4-Am in o-7-ch lor oqu in olin e (1). 4,7-Dichloroquinoline (19.8
g, 0.10 mol) was dissolved in phenol (100 g) with stirring and
heating at 165 °C. Ammonia was bubbled through this mixture
for 15 min. The temperature was then raised to 200 °C and
the bubbling of ammonia was continued for 2.5 h. After cooling,
the reaction mixture was treated with 0.2 M acetate buffer,
pH 4.0 (400 mL), and ether (250 mL). The layers were
1
mp 171-173 °C); H NMR (DMSO-d6) δ 8.04 (d, J ) 4.7 Hz,
1H), 8.48 (d, J ) 1.7 Hz, 2H), 8.9 (br d, 1H), 9.1 (d, J ) 4.7
Hz, 1H). Anal. (C9H5ClN2O2) C, H, N, Cl.

Structural Specificity of CQ-Hematin Binding
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22 4637
7-Nitr o-4-(4-(d ieth yla m in o)-1-m eth ylbu tyl)a m in oqu in -
olin e (9). A mixture of 18 (0.5 g, 2.4 mmol) and 2-amino-5-
(diethylamino)pentane (1.9 g, 12 mmol) was heated at 120 °C
for 7 h under N2; 20% aqueous NaOH (20 mL) and CHCl3 (50
mL) were added to the crude reaction product. The CHCl3 layer
was washed thoroughly with 20% aqueous NaOH (4 × 30 mL)
and water (100 mL) and dried over anhydrous sodium bicar-
bonate. Removal of the CHCl3 in vacuo afforded a gummy
residue which was dissolved in 10% acetic acid and filtered to
remove unreacted 4-chloro-7-nitroquinoline. The filtrate was
neutralized with 10% aqueous NaOH and extracted with
CHCl3 (3 × 10 mL), and the combined CHCl3 extracts were
dried over K2CO3. The residue was chromatographed on
alumina (Brockmann, Grade 1) eluting with dry EtOAc.
Removal of the EtOAc in vacuo afforded 9 (220 mg, 28%) as a
al.4d Compound inhibition of parasite growth was determined
by a semiautomated microdilution assay using asynchronous
stock cultures.25a,62
Molecu la r Mod elin g. Gaussian 94, revision A.1 (Gaussian,
Inc.), on a SGI Power Indigo R8000 workstation and Spartan
5.0 (Wavefunction, Inc.) on a SGI Indigo Extreme R4000
workstation were used for the molecular modeling experi-
ments. The structure of hemin was obtained from the Cam-
bridge Crystallographic Data Bank and uploaded into Spartan
5.0. Hematin µ-oxo dimer was built with constraints of 168°
for the Fe-O-Fe angle and 1.76 Å for the Fe-O distance, data
derived from the X-ray structure of ferritetraphenylporphyrin
µ-oxo dimer.63 The minimum energy conformer of hematin
µ-oxo dimer was established by rotating one of the hematin
moieties along the Fe-O-Fe axis through 360° in 5° steps and
calculating the relative energies using the Merck molecular
force field (MMFF94).64 The minimum energy conformer was
then subject to PM3 calculation to determine its electrostatic
potential surface.65,66
The energy-minimized structures of CQ and CQ analogues
1, 6, 7, and 11 were subject to a conformational search at the
AM1 single-point level to obtain the low-energy conformers.
Using Gaussian 94, the lowest and most abundant (population
density > 75%) conformers were optimized at the ab initio
quantum chemical level with the 3-21G* split-valence basis
set. Molecular electrostatic potentials and three-dimensional
electronic profiles were calculated on the 3-21G*-optimized
geometries for the five structures. A similar sequence was
followed for the energy minimization of 1 and 6-chloro-4-
aminoquinoline followed by optimization at the ab initio
quantum chemical level with the 6-31G** basis set. From the
energy-minimized structure of hematin µ-oxo dimer and CQ,
a model of a 1:2 CQ-hematin µ-oxo dimer complex3f,7 was built
in which the electron-deficient electrostatic potential surfaces
of the hematin µ-oxo dimer interact favorably with the
electron-rich electrostatic potential surface of the quinoline
heterocycle of CQ.
1
yellow solid: mp 92-96 °C; H NMR δ (CDCl3) 1.02 (t, J )
7.0 Hz, 6H), 1.34 (d, J ) 6.4 Hz, 3H), 1.6-1.8 (m, 4H), 2.4-
2.58 (m, 6H), 3.7-3.75 (m, 1H), 5.8 (d, J ) 5.9 Hz, 1H), 6.5 (d,
J ) 5.5 Hz, 1H), 7.9 (d, J ) 9.1 Hz, 1H), 8.13 (dd, J ) 9.2, 2.3
Hz, 1H), 8.64 (d, J ) 5.2 Hz, 1H), 8.8 (d, J ) 2.3 Hz, 1H); 13
C
NMR (CDCl3) δ 11.2, 19.8, 23.7, 34.3, 46.7, 48.4, 52.3, 100.9,
117.1, 121.9, 122.5, 125.7, 147.6, 147.9, 148.9, 152.9. Anal.
(C18H26N4O2) C, H, N.
7-Am in o-4-(4-(dieth ylam in o)-1-m eth ylbu tyl)am in oqu in -
olin e (10). To 9 (85 mg, 0.25 mmol) dissolved in EtOH (10
mL) was added 10% Pd/C (8.4 mg) and the mixture was
hydrogenated at 50 atm H2 for 3 h. The reaction solution was
filtered through Celite and the solvent was removed in vacuo
to produce a reddish brown oily residue. The crude reaction
product was dissolved in dry ether (4 mL), cooled to 0 °C, and
exposed to a stream of HCl gas for 5 min to afford a yellow
solid. Excess HCl gas was removed with a stream of N2 and
the ether was decanted to yield the trihydrochloride salt of 10
1
(60 mg, 58%): mp 58-64 °C; H NMR (CDCl3) δ 1.04 (t, J )
7.2 Hz, 6H), 1.3 (d, J ) 6.4 Hz, 3H), 1.6-1.8 (m, 4H), 2.4-2.7
(m, 6H), 3.7-3.75 (m, 1H), 4.2-4.4 (br s, 1H), 5.1 (d, J ) 5.9
Hz, 1H), 6.2 (d, J ) 6.1 Hz, 1H), 6.8 (dd, J ) 9.0, 2.3 Hz, 1H),
7.2 (d, J ) 2.1 Hz, 1H), 7.7 (d, J ) 9.0 Hz, 1H), 8.32 (d, J )
6.0 Hz, 1H); 13C NMR (CDCl3) δ 11.3, 20.3, 23.7, 34.6, 46.7,
48.1, 52.5, 97.0, 110.3, 111.9, 115.7, 120.6, 147.1, 149.0, 150.1,
151.1. Anal. (C18H31N4Cl3) C, H, N.
Ack n ow led gm en t . The financial support for this
work was obtained in part from NIH Grant AI39670-
01. We thank Dr. Luis A. Marky for many helpful
discussions about the ITC experiments.
Isoth er m a l Titr a tion Ca lor im etr y (ITC). All ITC binding
experiments were conducted in 250 mM phosphate buffer, pH
6.5 at 37 °C, using a MCS titration calorimeter from Microcal
Inc. (Northampton, MA). An aliquot of a 1 mM stock solution
of hematin in 0.01 N NaOH was diluted up to the required
concentrations (0.1-0.3 mM) in the phosphate buffer. Hematin
was titrated with 0.5-5 mM solutions of CQ analogues 1-13
using a 100-µL syringe rotating at 400 rpm. Titration experi-
ments consisted of 15-30 injections of 2-6 µL each, and three
to four experiments were performed for each analogue. Heats
of dilution and mixing were obtained by 15-30 injections of
each analogue into the buffer medium. After having first
subtracted heats of dilution and mixing from each injection
heat pulse, the binding isotherms were fitted for a model of a
single set of identical binding sites by nonlinear least-squares
fitting of the titration data using Origin software. Association
constants (Ka’s), enthalpy change (∆H), and stoichiometry (n)
were obtained from binding isotherms using the Marquardt
algorithm.59
In h ibition of Hem a tin P olym er iza tion . As previously
described3f,4d purified hemozoin from the food vacuole of the
malarial parasite P. falciparum60 was used to initiate hematin
polymerization at pH 6.5. Compounds were added to the
reaction mixture as DMSO solutions up to a maximum DMSO
concentration of 10%. The dpm values obtained from the assay
were expressed as percent inhibition relative to hemozoin
formation in a drug-free control. The values of triplicate assays
were plotted semilogarithmically (CA-Cricket Graph III 1.5.2)
and the IC50 values (µM) calculated graphically (SD (standard
deviation).
Refer en ces
(1) Bray, P. G.; Ward, S. A. A Comparison of the Phenomenology
and Genetics of Multidrug Resistance in Cancer Cells and
Quinoline Resistance in Plasmodium falciparum. Pharmacol.
Ther. 1998, 77, 1-28.
(2) (a) Ridley, R. G. Haemoglobin Degradation and Haem Polym-
erization as Antimalarial Drug Targets. J . Pharm. Pharmacol.
1997, 49S, 43-48. (b) Ridley, R. G.; Dorn, A.; Vippagunta, S.
R.; Vennerstrom, J . L. Haematin (haem) Polymerization and its
Inhibition by Quinoline Antimalarials. Ann. Trop. Med. Para-
sitol. 1997, 91, 559-566.
(3) (a) Chou, A. C.; Chevli, R.; Fitch, C. D. Ferriprotoporphyrin IX
Fulfills the Criteria for Identification as the Chloroquine Recep-
tor of Malaria Parasites. Biochemistry 1980, 19, 1543-1549. (b)
Egan, T. J .; Mavuso, W. W.; Ross, D. C.; Marques, H. M.
Thermodynamic Factors Controlling the Interaction of Quinoline
Antimalarial Drugs With Ferriprotoporphyrin IX. J . Inorg.
Biochem. 1997, 68, 137-145. (c) Sullivan, D. J ., J r.; Gluzman,
I. Y.; Russell, D. G.; Goldberg, D. E. On the Molecular Mecha-
nism of Chloroquine’s Antimalarial Action. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 11865-11870. (d) Sullivan, D. J ., J r.; Matile,
H.; Ridley, R. G.; Goldberg, D. E. A Common Mechanism for
Blockade of Heme Polymerization by Antimalarial Quinolines.
J . Biol. Chem. 1998, 273, 31103-31107. (e) Bray, P. G.; Mung-
thin, M.; Ridley, R. G.; Ward, S. A. Access to Hematin: The Basis
of Chloroquine Resistance. Mol. Pharmacol. 1998, 54, 170-179.
(f) Dorn, A.; Vippagunta, S. R.; Matile, H.; J aquet, C.; Venner-
strom, J . L.; Ridley, R. G. An Assessment of Drug-Hematin
Binding as a Mechanism for Inhibition of Hematin Polymeri-
sation by Quinoline Antimalarials. Biochem. Pharmacol. 1998,
55, 727-736.
(4) (a) Slater, A. F. G.; Cerami, A. Inhibition By Chloroquine of a
Novel Haem Polymerase Enzyme Activity in Malaria Tropho-
zoites. Nature 1992, 355, 167-169. (b) Orjih, A. U.; Fitch, C. D.
Hemozoin Production by Plasmodium falciparum: Variation
With Strain and Exposure to Chloroquine. Biochim. Biophys.
In h ibition of P a r a site Gr ow th . P. falciparum strains
NF54 and K1 were cultured according to the method of Trager
and J ensen61 with minor variations as described by Dorn et

4638 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22
Vippagunta et al.
Acta 1993, 1157, 270-274. (c) Egan, T. J .; Ross, D. C.; Adams,
P. A. Quinoline Anti-Malarial Drugs Inhibit Spontaneous For-
mation of â-Hematin (Malaria Pigment). FEBS Lett. 1994, 352,
54-57. (d) Dorn, A.; Stoffel, R.; Matile, H.; Bubendorf, A.; Ridley,
R. G. Malarial Haemozoin/â-Hematin Supports Haem Polym-
erization in the Absence of Protein. Nature 1995, 374, 269-271.
(5) (a) Chou, A. C.; Fitch, C. D. Hemolysis of Mouse Erythrocytes
by Ferriprotoporphyrin IX and Chloroquine. J . Clin. Invest.
1980, 66, 856-858. (b) Orjih, A. U.; Banyal, H. S.; Chevli, R.;
Fitch, C. D. Hemin Lyses Malaria Parasites. Science 1981, 214,
667-669. (c) Ginsburg, H.; Demel, R. A. The Effect of Ferripro-
toporphyrin IX and Chloroquine on Phospholipid Monolayers
and the Possible Implications to Antimalarial Activity. Biochim.
Biophys. Acta 1983, 732, 316-319. (d) Hebbel, R. P.; Eaton, J .
W. Pathobiology of Heme Interaction with the Erythrocyte
Membrane. Semin. Hematol. 1989, 26, 136-149. (e) Ginsburg,
H.; Famin, O.; Zhang, J .; Krugliak, M. Inhibition of Glutathione-
Dependent Degradation of Heme by Chloroquine and Amodi-
aquine as a Possible Basis for Their Antimalarial Mode of Action.
Biochem. Pharmacol. 1998, 56, 1305-1313. (f) Sugioka, Y.;
Suzuki, M. The Chemical Basis for the Ferriprotoporphyrin IX-
Chloroquine Complex Induced Lipid Peroxidation. Biochim.
Biophys. Acta 1991, 1074, 19-24.
(6) (a) Fitch, C. D.; Kanjananggulpan, P. The State of Ferriproto-
porphyrin IX in Malaria Pigment. J . Biol. Chem. 1987, 262,
15552-15555. (b) Slater, A. F. G.; Swiggard, W. J .; Orton, B.
R.; Flitter, W. D.; Goldberg, D. E.; Cerami, A.; Henderson, G. B.
An Iron-Carboxylate Bond Links the Heme Units of Malaria
Pigment. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 325-329. (c)
Bohle, S. D.; Dinnebier, R. E.; Madsen, S.; Stephens, P. W.
Characterization of the Products of the Heme Detoxification
Pathway in Malarial Late Trophozoites by X-ray Diffraction. J .
Biol. Chem. 1997, 272, 713-716.
(7) Moreau, S.; Perly, B.; Biguet, J . Interactions de la Chloroquine
Avec la Ferriprotoporphyrine IX. EÄ tude par Re´sonance Magne´-
tique Nucle´aire. Biochimie 1982, 64, 1015-1025.
(8) Adams, P. A.; Berman, P. A. M.; Egan, T. J .; Marsh, P. J .; Silver,
J . The Iron Environment in Heme and Heme-Antimalarial
Complexes of Pharmacological Interest. J . Inorg. Biochem. 1996,
63, 69-77.
(9) (a) Cohen, S. N.; Phifer, K. O.; Yielding, K. L. Complex Formation
Between Chloroquine and Ferrihaemic Acid In Vitro, and its
Effect on the Antimalarial Action of Chloroquine. Nature 1964,
202, 805-806. (b) Blauer, G.; Ginsburg, H. Complexes of
Antimalarial Drugs with Ferriprotoporphyrin IX, Biochem. Int.
1982, 5, 519-523. (c) J earnpipatkul, A.; Govitrapong, P.;
Yuthavong, Y.; Wilairat, P.; Panijpan, B. Binding of Antimalarial
Drugs to Hemozoin from Plasmodium berghei. Experientia 1980,
36, 1063-1064. (d) J earnpipatkul, A.; Panijpan, B. Molecular
complexes of quinoline antimalarials with iron-porphyrin com-
ponents of protease-digested methemoglobin. Chem. Biol. Inter-
act. 1980, 33, 83-90.
(10) de Almeida Ribeiro, M. C.; Augusto, O.; da Costa Ferreira, A.
M. Inhibitory Effect of Chloroquine on the Peroxidase Activity
of Ferriprotoporphyrin IX. J . Chem. Soc., Dalton Trans. 1995,
3759-3766.
(11) O′Neill, P. M.; Bray, P. G.; Hawley, S. R.; Ward, S. A.; Park, B.
K. 4-Aminoquinolines-Past, Present, and Future: A Chemical
Perspective. Pharmacol. Ther. 1998, 77, 29-58.
(12) Price, C. C.; Roberts, R. M. The Synthesis of 4-Hydroxyquino-
lines. I. Through Ethoxymethylenemalonic Ester. J . Am. Chem.
Soc. 1946, 68, 1204-1208.
(13) De, D.; Krogstad, F. M.; Cogswell, F. B.; Krogstad, D. J .
Aminoquinolines that Circumvent Resistance in Plasmodium
falciparum In Vitro. Am. J . Trop. Med. Hyg. 1996, 55, 579-
583.
(14) Surrey, A. R.; Hammer, H. F. Some 7-Substituted 4-Amino-
quinoline Derivatives. J . Am. Chem. Soc. 1946, 68, 113-116.
(15) (a) Yayon, A.; Cabantchik, Z. I.; Ginsburg, H. Identification of
the Acidic Compartment of Plasmodium falciparum-Infected
Erythrocytes as the Target of the Antimalarial Drug Chloro-
quine. EMBO J . 1984, 3, 2695-2700. (b) Krogstad, D. J .;
Schlesinger, P. H.; Gluzman, I. Y. Antimalarials Increase Vesicle
pH in Plasmodium falciparum. J . Cell. Biol. 1985, 101, 2302-
2309.
(16) (a) Brown, S. B.; Dean, T. C.; J ones, P. Aggregation of Ferri-
haems. Dimerization and Protolytic Equilibria of Protoferrihaem
and Deuterioferrihaem in Aqueous Solution. Biochem. J . 1970,
117, 733-739. (b) Silver, J .; Lukas, B. Mo¨ssbauer Studies on
Protoporphyrin IX Iron(III) Solutions. Inorg. Chim. Acta 1983,
78, 219-224. (c) Bednarski, T. M.; J ordan, J . Electron-Transfer
Characteristics of the Prosthetic Group of Hemoproteins. J . Am.
Chem. Soc. 1967, 89, 1552-1558. (d) Buchler, J . W. Hemoglobin-
An Inspiration for Research in Coordination Chemistry. Angew.
Chem., Int. Ed. Engl. 1978, 17, 407-423. At low concentrations
of free hematin monomer and at low pH, conditions likely to
prevail in the parasite food vacuole, the hematin monomer:µ-
oxo dimer equilibrium shifts toward the monomer, although the
Kobs estimated at 2.8 × 105 M-1 at pH 4.8 (ref 16a) still strongly
favors the µ-oxo dimer.
(17) White, W. I. Aggregation of Porphyrins and Metalloporphyrins.
In The Porphyrins, Vol. V; Dolphin, D., Ed.; Physical Chemistry,
Part C; Academic Press: New York, 1978; pp 303-339.
(18) Diederich, F.; Smithrud, D. B.; Sanford, E. M.; Wyman, T. B.;
Ferguson, S. B.; Carcanague, D. R.; Chao, I.; Houk, K. N. Solvent
Effects in Molecular Recognition. Acta Chem. Scand. 1992, 46,
205-215.
(19) Lemieux, R. U. How Water Provides the Impetus for Molecular
Recognition in Aqueous Solution. Acc. Chem. Res. 1996, 29, 373-
380.
(20) Lumry, R.; Rajender, S. Enthalpy-Entropy Compensation Phe-
nomena in Water Solutions of Proteins and Small Molecules: A
Ubiquitous Property of Water. Biopolymers 1970, 9, 1125-1227.
(21) Blokzijl, W.; Engberts, J . B. F. N. Hydrophobe Effektes
Ansichten und Tatsachen. Angew. Chem. 1993, 105, 1610-1648.
(22) Muller, N. Search for a Realistic View of Hydrophobic Effects.
Acc. Chem. Res. 1990, 23, 24-28.
(23) Gilli, P.; Ferretti, V.; Gilli, G.; Borea, P. A. Enthalpy-Entropy
Compensation in Drug-Receptor Binding. J . Phys. Chem. 1994,
98, 1515-1518.
(24) (a) Hunter, C. A.; Sanders, J . K. M. The Nature of π-π
Interactions. J . Am. Chem. Soc. 1990, 112, 5525-5534. (b)
Hunter, C. A. Arene-Arene Interactions: Electrostatic or Charge
Transfer? Angew. Chem., Int. Ed. Engl. 1993, 32, 1584-1586.
(25) (a) Ridley, R. G.; Hofheinz, W.; Matile, H.; J aquet, C.; Dorn, A.;
Masciadri, R.; J olidon, S.; Richter, W. F.; Guenzi, A.; Girometta,
M. A.; Urwyler, H.; Huber, W.; Thaithong, C.; Peters, W.
4-Aminoquinoline Analogues of Chloroquine With Shortened
Side Chains Retain Acitivity Against Chloroquine-Resistant
Plasmodium falciparum. Antimicrob. Agents Chemother. 1996,
40, 1846-1854. (b) De, D.; Krogstad, F. M.; Byers, L. D.;
Krogstad, D. J . Structure-Activity Relationships for Antiplas-
modial Activity among 7-Substituted 4-Aminoquinolines. J . Med.
Chem. 1998, 41, 4918-4926.
(26) (a) Geary, T. G.; Divo, A. A.; J ensen, J . B. Activity of Quinoline
Containing Antimalarials Against Chloroquine-Sensitive and
Resistant Strains of Plasmodium falciparum In Vitro. Trans.
R. Soc. Trop. Med. Hyg. 1987, 81, 499-503. (b) Aderounmu, A.
F. In Vitro Assessment of the Antimalarial Activity of Chloro-
quine and Its Major Metabolites. Ann. Trop. Med. Parasitol.
1984, 78, 581-585.
(27) Veignei, E.; Moreau, S. The Mode of Action of Chloroquine. Non-
Weak Base Properties of 4-Aminoquinolines and Antimalarial
Effects on Strains of Plasmodium. Ann. Trop. Med. Parasitol.
1991, 85, 229-237.
(28) Bray, P. G.; Hawley, S. R.; Mungthin, M.; Ward, S. A. Physico-
chemical Properties Correlated with Drug Resistance and the
Reversal of Drug Resistance in Plasmodium falciparum. Mol.
Pharmacol. 1996, 50, 1559-1566.
(29) Hawley, S. R.; Bray, P. G.; Mungthin, M.; Atkinson, J . D.;
O’Neill, P. M.; Ward, S. A. Relationship Between Antimalarial
Drug Activity, Accumulation, and Inhibition of Heme Polymer-
ization in Plasmodium falciparum In Vitro. Antimicrob. Agents
Chemother. 1998, 42, 682-686.
(30) O’Neill, P. M.; Willock, D. J .; Hawley, S. R.; Bray, P. G.; Storr,
R. C.; Ward, S. A.; Park, B. K. Synthesis, Antimalarial Activity,
and Molecular Modeling of Tebuquine Analogues. J . Med. Chem.
1997, 40, 437-448.
(31) Vippagunta, S. R.; Dorn, A.; Ridley, R. G.; Vennerstrom, J . L.
Unpublished results.
(32) Koh, H. L.; Go, M. L.; Ngiam, T. L.; Mark, J . W. Conformational
and Structural Features Determining In Vitro Antimalarial
Activity in Some Indolo[3,2-c]Quinolines, Anilinoquinolines and
Tetrahydroindolo[3,2-d]Benzazepines. Eur. J . Med. Chem. 1994,
29, 107-113.
(33) The Nquin-Ndiethyl distances are 8.615 Å (3-21G*) and 8.508 Å
from X-ray data (Karle, J . M.; Karle, I. L. Redetermination of
the Molecular Structure of the Antimalarial Chloroquine Bis-
(dihydrogenphosphate) Dihydrate. Acta Crystallogr. 1988, C44,
1605-1608).
(34) Constantinides, I.; Satterlee, J . D. UV-Visible and Carbon NMR
Studies of Chloroquine Binding to Urohemin I Chloride and
Uroporphyrin I in Aqueous Solutions. J . Am. Chem. Soc. 1988,
110, 4391-4395.
(35) Caughey, W. S.; Barlow, C. H.; O’Keeffe, D. H.; O’Toole, M. C.
Spectroscopic Studies of Cis and Trans Effects in Hemes and
Hemins. Ann. N.Y. Acad. Sci. 1973, 206, 296-309.
(36) Heaton, N. J .; Bello, P.; Herrado´n, B.; del Campo, A.; J ime´nez-
Barbero, J . NMR Study of Intramolecular Interactions between
Aromatic Groups: van der Waals, Charge-Transfer, or Quadru-
polar Interactions? J . Am. Chem. Soc. 1998, 120, 12371-12384.
(37) Schneider, H.; Wang, M. Ligand-Porphyrin Complexes: Quan-
titative Evaluation of Stacking and Ionic Contributions. J . Org.
Chem. 1994, 59, 7464-7472.

Structural Specificity of CQ-Hematin Binding
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 22 4639
(38) Ferrari, V.; Cutler, D. J . Temperature Dependence of the Acid
Dissociation Constants of Chloroquine. J . Pharm. Sci. 1987, 76,
554-556.
(51) Drake, N. L.; Creech, H. J .; Garman, J . A.; Haywood, S. T.; Peck,
R. M.; VanHook, J . O.; Walton, E. Synthetic Antimalarials. The
Preparation of Certain 4-Aminoquinolines. J . Am. Chem. Soc.
1946, 68, 1208-1213.
(52) Price, C. C.; Leonard, N. J .; Reitsema, R. H. Synthesis of
4-Hydroxyquinolines. V. A Direct Synthesis from â-Anilinoacry-
lates. J . Am. Chem. Soc. 1946, 68, 1256-1259.
(53) Steck, E. A.; Hallock, L. L.; Holland, A. J .; Fletcher, L. T.
Quinolines. V. Some Polysubstituted 4-(4′-Diethylamino-1′-
methylbutylamino)quinolines. J . Am. Chem. Soc. 1948, 70,
1012-1015.
(54) Steck, E. A.; Hallock, L. L.; Holland, A. J . Quinolines. III. The
Synthesis of 5- and 7-Chloro and Bromo-3-methyl-4-dialkyl-
aminoalkylaminoquinolines. J . Am. Chem. Soc. 1946, 68, 380-
383.
(55) Hofheinz, W.; J aquet, C.; J olidon, S. Aminochinolin-Derivate mit
einer Wirksamkeit gegen Malariaerreger. European Patent
Application 94116281.0, J une 1995.
(56) Baker, R. H.; Dodson, R. M.; Riegel, B. The Dealkylation of 4-(4-
Diethylamino-1-methylbutylamino)-7-chloroquinoline, SN 7618.
J . Am. Chem. Soc. 1947, 69, 704.
(39) Macomber, D. B.; Sprinz, H.; Tousimis, A. J . Morphological
Effects of Chloroquine on Plasmodium berghei in Mice. Nature
1967, 214, 937.
(40) Fitch, C. D.; Chevli, R.; Gonzalez, Y. Chloroquine-Resistant
Plasmodium falciparum: Effect of Substrate on Chloroquine and
Amodiaquine Accumulation. Antimicrob. Agents Chemother.
1974, 6, 757-762.
(41) Hawley, S. R.; Bray, P. G.; O’Neill, P. M.; Kevin Park, B.; Ward,
S. A. The Role of Drug Accumulation in 4-Aminoquinoline
Antimalarial Potency. Biochem. Pharmacol. 1996, 52, 723-733.
(42) (a) Vippagunta, S. R.; Dorn, A.; Bubendorf, A.; Ridley, R. G.;
Vennerstrom, J . L. Deferoxamine: Stimulation of Hematin
Polymerization and Antagonism of its Inhibition by Chloroquine.
Biochem. Pharmacol. 1999, 58, 817-824. (b) Blauer, G.; Gins-
burg, H. Complexes of Antimalarial Drugs With Ferriprotopor-
phyrin IX. Biochem. Int. 1982, 5, 519-523. (c) Grinberg, L. N.;
Soprunov, F. F. Metabolism of Erythrocyte Infected with Malaria
Parasite and the Action of Antimalarial Drugs. Biomed. Biochim.
Acta 1983, 11/ 12, S317-S321.
(43) Saliba, K. J .; Folb, P. I.; Smith, P. J . Role for the Plasmodium
falciparum Digestive Vacuole in Chloroquine Resistance. Bio-
chem. Pharmacol. 1998, 56, 313-320.
(44) Bass, G. E.; Hudson, D. R.; Parker, J . E.; Purcell, W. P.
Mechanism of Antimalarial Activity of Chloroquine Analogues
From Quantitative Structure Activity Studies. Free Energy
Related Model. J . Med. Chem. 1971, 14, 275-282.
(45) Thompson, P. E.; Werbel, L. M. 4-Aminoquinolines. Antimalarial
Agents: Chemistry and Pharmacology; Academic Press: New
York, 1972; pp 150-193.
(46) Coatney, G. R.; Cooper, W. C.; Eddy, N. B.; Greenberg, J . Survey
of Antimalarial Agents; Public Health Monograph No. 9; U.S.
Government Printing Office: Washington, DC, 1953.
(47) Su, X.; Kirkman, L A.; Fujioka, H.; Wellems, T. E. Complex
Polymorphisms in an ∼330 Kda Protein are Linked to Chloro-
quine-Resistant P. falciparum in Southeast Asia and Africa. Cell
1997, 91, 593-603.
(48) Wu¨nsch, S.; Sanchez, C. P.; Gekle, M.; Grobe-Wortmann, L.;
Wiesner, J . Differential Stimulation of the Na(/H( Exchanger
Determines Chloroquine Uptake in Plasmodium falciparum. J .
Cell. Biol. 1998, 140, 335-345.
(49) Ridley, R. G.; Hudson, A. T. Quinoline Antimalarials. Exp. Opin.
Ther. Patents 1998, 8, 121-135.
(57) Ellis, J .; Gellert, E.; Robson, J . Synthesis of Some New Iodo-
quinolines. Aust. J . Chem. 1973, 26, 907-911.
(58) Baker, R. H.; Lappin, G. R.; Albisetti, C. J ., J r.; Riegel, B. The
Decarboxylation of Nitro Substituted 3-Carboxy-4-quinolinols by
Pyrolysis of Their Silver Salts. J . Am. Chem. Soc. 1946, 68, 1267.
(59) Bevington, P. R. Data Reduction and Error Analysis in the
Physical Sciences; McGraw-Hill: New York, 1969; p 236.
(60) Ashong, J . O.; Blench, I. P.; Warhurst, D. C. The Composition
of Haemozoin from Plasmodium falciparum. Trans. R. Soc. Trop.
Med. Hyg. 1989, 83, 167-172.
(61) Trager, W.; J ensen, J . B. Human Malaria Parasites in Continu-
ous Culture. Science 1976, 193, 673-675.
(62) Desjardins, R. E.; Canfield, C. J .; Haynes, J . D.; Chulay, J . D.
Quantitative Assessment of Antimalarial Activity In Vitro by a
Semiautomated Microdilution Technique. Antimicrob. Agents
Chemother. 1979, 16, 710-718.
(63) Fleischer, E. B.; Srivastava, T. S. The structure and properties
of µ-Oxo bis(tetraphenylporphineiron (III)). J . Am. Chem. Soc.
1969, 91, 2403-2405.
(64) Halgren, T. A. Merck Molecular Force Field. 1. Basis, Form,
Scope, Parametrization, and Performance of MMFF94. J . Com-
put. Chem. 1996, 17, 490-641.
(65) Politzer, P.; Truhlar, D. G. Chemical Applications of Atomic and
Molecular Electrostatic Potentials; Plenum: New York, 1981.
(66) Naray-Szabo, G.; Ferenczy, G. G. Molecular Electrostatics. Chem.
Rev. 1995, 95, 829-847.
(50) Carmack, M.; Bullitt, J r., O. H.; Handrick, G. R.; Kissinger, L.
W.; Von, I. The Synthesis of 4-(4-Amino-1-methylbutylamino)-
7-chloroquinoline and Some 4-(4-Monoalkylamino-1-methylbu-
tylamino)-7-chloroquinolines. J . Am. Chem. Soc. 1946, 68, 1220-
1225.
J M9902180