Mimicking the intramolecular hydrogen bond: Synthesis, biological evaluation, andmolecular modeling of benzoxazines and quinazolines as potential antimalarial agents

Article

pubs.acs.org/jmc

Mimicking the Intramolecular Hydrogen Bond: Synthesis, Biological

Evaluation, and Molecular Modeling of Benzoxazines and

Quinazolines as Potential Antimalarial Agents

Sandra Gemma,^†,‡Caterina Camodeca,^†,‡Margherita Brindisi,^†,‡Simone Brogi,^†,‡Gagan Kukreja,^†,‡

Sanil Kunjir,^†,‡Emanuele Gabellieri,^†,‡Leonardo Lucantoni,^‡,§Annette Habluetzel,^‡,§

Donatella Taramelli,^†,‡,∥Nicoletta Basilico,^†,‡,⊥Roberta Gualdani,^#Francesco Tadini-Buoninsegni,^#

Gianluca Bartolommei,^#Maria Rosa Moncelli,^#Rowena E. Martin,^▽Robert L. Summers,^▽

Stefania Lamponi,^†,‡Luisa Savini,^†,‡Isabella Fiorini,^†,‡Massimo Valoti,^○Ettore Novellino,^†,◆

Giuseppe Campiani,*^,†,‡and Stefania Butini^†,‡

^†European Research Centre for Drug Discovery and Development (NatSynDrugs), University of Siena, Via Aldo Moro, 53100 Siena,

Italy

^‡CIRM Centro Interuniversitario di Ricerche sulla Malaria, Universita

^§Scuola di Scienze del Farmaco e dei Prodotti della Salute, Universita

̀

di Torino, Torino, Italy

̀

di Camerino, 62032 Camerino (MC), Italy

^∥Dipartimento di Scienze Farmacologiche e Biomolecolari, and ^⊥Dipartimento di Scienze Biomediche, Chirurgiche e Odontoiatriche,

̀

Universita di Milano, Via Pascal 36, 20133 Milan, Italy

^#Department of Chemistry “Ugo Schiﬀ”, University of Florence, 50019 Sesto Fiorentino, Italy

^▽Research School of Biology, The Australian National University, Canberra ACT 0200, Australia

^○Dipartimento di Neuroscienze, University of Siena, via A. Moro 2, Siena, Italy

^◆Dipartimento di Chimica Farmaceutica e Tossicologica, University of Naples Federico II, Via D. Montesano 49, 80131 Naples, Italy

S

* Supporting Information

ABSTRACT: The intramolecular hydrogen bond formed between a

protonated amine and a neighboring H-bond acceptor group in the side

chain of amodiaquine and isoquine is thought to play an important role in

their antimalarial activities. Here we describe isoquine-based compounds in

which the intramolecular H-bond is mimicked by a methylene linker. The

antimalarial activities of the resulting benzoxazines, their isosteric

tetrahydroquinazoline derivatives, and febrifugine-based 1,3-quinazolin-4-

ones were examined in vitro (against Plasmodium falciparum) and in vivo

(against Plasmodium berghei). Compounds 6b,c caused modest inhibition

of chloroquine transport via the parasite’s “chloroquine resistance

transporter” (PfCRT) in a Xenopus laevis oocyte expression system. In silico predictions and experimental evaluation of

selected drug-like properties were also performed on compounds 6b,c. Compound 6c emerged from this work as the most

promising analogue of the series; it possessed low toxicity and good antimalarial activity when administered orally to P. berghei-

infected mice.

eﬃcacy made it the frontline antimalarial for more than 40

years.

INTRODUCTION

■

Despite recent reductions in the malaria burden, the disease

continues to impose a devastating toll upon aﬄicted countries

in terms of morbidity, mortality, and signiﬁcantly reduced

economic prosperity.^1,2The quinolines served as the mainstay

for the treatment and prevention of malaria for several

centuries, beginning in the seventeenth century with the use

of extracts from bark of the South American Cinchona tree

(which contained a number of quinolines, including quinine).

Chloroquine (CQ, 1, Chart 1) was deployed in the 1940s as a

synthetic substitute for quinine, and its aﬀordability, safety, and

However, the eventual emergence and spread of CQ-

resistant (CQR) strains of the human malaria parasite,

Plasmodium falciparum, has severely limited the usefulness of

this drug and has necessitated a re-examination of chemo-

therapies for malaria. CQ and the related antimalarial drug

amodiaquine (AQ, 2) are both 4-aminoquinolines, yet AQ

remains eﬀective against several CQR strains of P. falciparum

Received: June 13, 2012

A

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Here we report the synthesis, in vitro and in vivo biological

investigation, potential toxicity proﬁles, and molecular model-

ing of this novel series of 4-aminoquinolines.

Chart 1. Reference and Title Compounds

CHEMISTRY

■

In general, the syntheses undertaken here used inexpensive

chemicals and required relatively few steps. The synthesis of

compounds 5a−c is reported in Scheme 1. For the synthesis of

5a,b, the key intermediates 9a,b were prepared following a

described two-step procedure.^18,21The ﬁrst step involved a

Mannich reaction involving the commercially available 3-

hydroxyacetanilide 7 and the suitable amine to provide the

corresponding Mannich products 8a,b. The amide function was

then hydrolyzed with aqueous hydrochloric acid, and the 3-

aminophenols thus obtained were coupled with 4,7-dichlor-

oquinoline.

Reaction of 9a,b with formaldehyde led to the closure of the

benzoxazine ring, producing compounds 5a,b in reasonable

yield after chromatographic puriﬁcation. The benzyl-derivative

5c was synthesized using an alternative synthetic procedure

(Scheme 1). Accordingly, intermediate 10, prepared as

previously described,^22,23was reacted with the Schiﬀ base

formed by treatment of benzylamine with formaldehyde. The

resulting Mannich product was exposed to an excess of

formaldehyde leading to the ring closure.²⁴Compound 5c was

precipitated from the reaction crude, and the resulting solid was

then puriﬁed by crystallization.

a

Structures of title compounds 5a−c, 6a−c, 15a−d, 16, and 21 are

reported in Table 1.

(Chart 1).³⁻⁵AQ is generally well tolerated when used as a

treatment and was reintroduced in 2002 as a combination

therapy with artesunate to combat multidrug resistant

malaria.⁶⁻⁸Toxicity associated with the extended use of AQ

is thought to be primarily due to its extensive bioactivation in

vivo to form a reactive quinone-imine, which covalently binds

to cell components and thereby disrupts cell function (and may

also lead to antiamodiaquine immune responses).^9,10Several

studies have attempted to improve the toxicity proﬁle of

AQ,¹¹⁻¹⁴for example, isoquine (IQ, 3) was designed to prevent

the formation of the toxic quinone-imine intermediate. Of

particular note, previous work has indicated that 4-amino-

quinolines carrying a single aromatic ring in the side chain are

considerably more active if they contain a hydrogen bond

acceptor close to a basic nitrogen.¹⁵⁻¹⁹This motif was present

in early antimalarial compounds, such as 4,¹⁸and is also present

in AQ and IQ. Furthermore, the side chains of quinine,

meﬂoquine, and halofantrine as well as a number of novel 4-

aminoquinolines each contain a hydrogen bond acceptor group

in close proximity to a protonatable amine.¹⁵⁻¹⁷The neutral

form of ferroquine also contains a similar intramolecular

hydrogen bond.²⁰The recurrence of this intramolecular

hydrogen-bonding motif, together with its importance to

antimalarial activity, led us to explore the eﬀect of mimicking

it by way of a methylene linker. Using the IQ structure as a

template, we introduced a methylene tether between the two

heteroatoms involved in the intramolecular hydrogen bonding.

In the resulting series of isoquine analogues 5a−c (Chart 1 and

Table 1), the amino and hydroxyl groups are constrained to

form a benzofused oxazine ring. In addition, alkyl groups such

as ethyl, n-propyl, or benzyl were placed at the benzoxazine N3

in order to explore the eﬀect of lipophilicity and steric

hindrance on antiplasmodial activity. However, the instability of

the benzoxazine system under acidic hydrolytic conditions

necessitated the replacement of the benzoxazine oxygen with a

nitrogen atom to produce a more hydrolytically stable

tetrahydroquinazoline system (6a−c, Chart 1 and Table 1).

The synthesis of 6a−c, 15a−d, and 16 is outlined in Scheme

2. 4-Nitroanthranilic acid 11²⁵was coupled with the suitable

amines to yield intermediates 12a,b,d.

The latter amides were then treated with formic acid to form

the quinazolinone ring of 13a,b or with trimethylorthoformate

and p-toluensulfonic acid to produce 13d. A diﬀerent synthetic

approach was used to prepare intermediate 13c in one step

from 11; microwave irradiation of the anthranilic acid 11 in the

presence of trimethylorthoformate, p-toluenesulfonic acid

(PTSA), and benzylamine led to the formation of the

quinazolinone 13c, albeit with a moderate yield. The nitro

group of 13a,b,d was reduced by treatment with a Fe/CaCl₂

mixture to obtain compounds 14a,b,d, whereas compound 13c

was reduced to 14c using SnCl₂. Finally, anilines 14a−d were

coupled with 4,7-dichloroquinoline to yield compounds 15a−d.

Reduction of the quinazolinone system with LiAlH₄or with

alane generated in situ led to the tetrahydroquinazolines 6a,b

and 6c, respectively. By contrast, reduction of ferrocenyl-

derivative 6d under a variety of reduction conditions invariably

led to decomposition of the starting material. Reaction of 15c

with oxalyl chloride and NaCNBH₃led to selective imine

reduction, giving the dihydroquinazolinone system of com-

pound 16.

During the course of this study, the quinazolinone 21, an

analogue of 15c bearing a protonatable side chain, was also

synthesized (Scheme 3). 4-Cyanobenzaldehyde 17 was

submitted to a reductive amination protocol which exploits

α-picoline-borane as a less toxic alternative to NaCNBH₃.²⁷

Amine 19 was then obtained through reduction of the nitrile

group of 18 using nickel boride generated in situ from NiCl₂

and NaBH₄.²⁸Coupling of 19 with the anthranilic acid

derivative 11 followed by cyclization led to the key intermediate

20. This latter compound was then reduced, and the resulting

aniline was coupled with 4,7-dichloroquinoline to aﬀord 21.

B

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 1. Ionic Forms, in Vitro Antiplasmodial Activity, and Cytotoxicity of Title and Reference Compounds

C

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 1. continued

a

Percentage of ionic form in parentheses. DP = diprotonated form; P = protonated form; N = neutral form (ACD/pK_aDB, version 11.00, software,

b

Advanced Chemistry Development, Inc., Toronto, Canada). IC₅₀values are the mean of at least three determinations. Standard errors were all

within 10% of the mean. CQS strain. CQR strain. Resistance index = W2-IC₅₀/D10-IC₅₀. TC₅₀(toxic concentration) values are the mean of at

least three determinations. Standard errors were all within 5% of the mean.

c

d

e

f

a

Scheme 1. Synthesis of Benzoxazine Derivatives 5a−c

Scheme 2. Synthesis of 6a−c, 15a−d, and 16

a

Reagents and conditions: (a) RNH₂, HCHO 37%, EtOH, reﬂux, 24

h; (b) HCl 20%, EtOH, reﬂux, 6 h; (c) 4,7-dichloroquinoline, EtOH,

reﬂux, 12 h; (d) KOH, HCHO 37%, MeOH, reﬂux,1 h; (e) HCHO

37%, benzylamine, MeOH, reﬂux, 3 h.

a

Reagents and conditions: (a) EDC, HOBt, Et₃N, CH₂Cl₂, 0 °C, 1 h,

then ethylamine (for 12a), n-propylamine (for 12b), or ferrocenyl-

amine²⁶(for 12d), 25 °C, 12 h; (b) HCOOH, neat, 110 °C, 3 h (for

13a,b) or CH(OMe)₃, PTSA, reﬂux 3 h (for 13d); (c) CH(OMe)₃,

benzylamine (for 13c), PTSA, neat, MW, 150 °C, 150 W, 3 × 10 min;

(d) Fe, CaCl₂, 75% EtOH, reﬂux, 2 h (for 14a,b,d) or SnCl₂·2H₂O,

AcOEt, reﬂux, 2 h (for 14c); (e) 4,7-dichloroquinoline, EtOH, reﬂux,

18−48 h; (f) LiAlH₄, dry THF, reﬂux, 4 h (for 6a,b) or AlH₃

(prepared in situ), dry THF, 25 °C, 1 h (for 6c); (g) (i) (COCl)₂,

CHCl₃, reﬂux, 1 h, (ii) NaCNBH₃, dry THF, 25 °C, 24 h.

RESULTS AND DISCUSSION

■

Antiplasmodial Activity and Structure−Activity Rela-

tionships (SARs). The antiplasmodial activities of the

synthesized compounds were measured in vitro against

human red blood cells infected with either the D10 (CQ-

sensitive; CQS) or the W2 (CQR) strain of P. falciparum. The

IC₅₀values (presented in Table 1) were determined using a

parasite lactate dehydrogenase (pLDH) assay.²⁹

conformations of these drugs contain an intramolecular

hydrogen bond between the phenolic oxygen and the

ammonium nitrogen. These data indicate that the six-

membered ring feature of IQ is closely mimicked by the

benzoxazine ring of derivative 5a. Benzoxazine derivatives 5a−c

showed potent antiplasmodial activity against both CQR and

To verify our design approach, we performed a conforma-

tional analysis on the structure of protonated IQ, AQ, and

halofantrine. As shown in Figure S1 of the Supporting

Information, the analysis conﬁrmed that the lower energy

D

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

6b and 6c using a X. laevis oocyte expression system.³¹The

direction of PfCRT^CQR-mediated CQ transport in this system is

from the mildly acidic extracellular medium (pH 6.0) into the

oocyte cytosol (pH 7.2), which corresponds to the eﬄux of CQ

from the acidic DV (pH ∼ 5.5) into the parasite cytosol (pH

7.3). In an initial experiment in which the derivatives were

tested at an extracellular concentration of 100 μM, both 6b and

6c caused modest but statistically signiﬁcant reductions in the

uptake of CQ via PfCRT^CQR(Figure 1A; P < 0.001, ANOVA)

and did not aﬀect the passive diﬀusion of CQ into oocytes

expressing PfCRT^CQS(P > 0.05, ANOVA). It is worth noting

that in this screen, both 6b and 6c were considerably less

potent inhibitors of PfCRT^CQRthan the CQ resistance-reverser

verapamil.

a

Scheme 3. Synthesis of 21

Moreover, an analysis of the concentration-dependent

inhibition of PfCRT^CQRby 6c (Figure 1B) yielded an IC₅₀

value (228 25 μM; n = 4) that is signiﬁcantly higher than that

measured previously for verapamil (30 1 μM³¹) or saquinavir

(13 0.9 μM;³²the most potent PfCRT^CQRinhibitor reported

to date). Taken together, the data indicate that 6b and 6c have

a

Reagents and conditions: (a) pyrrolidine, AcOH, pic-BH₃, neat, 25

°C, 3 h; (b) NiCl₂·6H₂O, NaBH₄, EtOH, 25 °C, 12 h; (c) 11, EDC,

HOBt, Et₃N, CH₂Cl₂, 0 °C, 1 h, then 19, 25 °C, 12 h; (d) HCOOH,

neat, 110 °C, 3 h; (e) Fe, CaCl₂, EtOH 75%, reﬂux, 2 h; (f) 4,7-

dichloroquinoline, pyridinium chloride, EtOH, reﬂux, 18 h.

relatively low aﬃnities for PfCRT^CQR

.

A third series of isoquine analogues (15a−d) bearing a N3-

substituted 1,3-quinazolin-4-one scaﬀold (based on the potent

antimalarial compound febrifugine) were synthesized and their

antiplasmodial activities tested. The ferrocenylmethyl side chain

of compound 15d is more lipophilic than a benzyl group and

has a peculiar electrochemical behavior, making it an attractive

moiety to be exploited for the development of novel

antimalarials.³³The very high IC₅₀values (>1000 nM) obtained

for 15a−d and the partially reduced intermediate 16 are likely

to be due, at least in part, to poor accumulation of these

compounds in the DV. For instance, compounds 15a−d and 16

are predicted to possess only one protonatable N, moreover the

calculated distribution of protonated species indicates that they

are very weak bases. Hence, these quinazolinone intermediates

are not expected to accumulate to high levels in the DV via the

weak-base trapping mechanism.

However, the addition of another protonatable moiety to 15c

(to give compound 21) greatly improved protonatability (53%

in a diprotonated form at pH 5.5; Table 1) and restored

antiplasmodial activity against CQS and CQR parasites at

nanomolar concentrations.

In Silico Prediction and Experimental Evaluation of

Selected Drug-Like Properties. In silico drug-likeness of

compounds 5a−c and 6a−c was calculated using the OSIRIS

molecular property explorer (Actelion Pharmaceuticals Ltd.,

Allschwil, Switzerland Table S1 of the Supporting Information).

LogP and log S values were calculated, along with drug-likeness

and drug-score parameters, for 5a−c and 6a−c as well as for

IQ, CQ, and AQ (Table 2).

The data indicates that 5a−c and 6a−c are likely to be less

soluble in water than the reference antimalarials. The program

identiﬁed structural fragments within 5a−c and 6c that may

increase the risk of mutagenic side eﬀects. These molecular

fragments are often observed in toxic compounds (as listed in

the RTECS database) but are less frequently observed in

approved drugs (the assumption being that approved drugs are

devoid of major toxicity issues).

These ﬁndings led us to undertake in vitro assessments of

selected drug-like and toxicity properties. Speciﬁcally, the (a)

solubility and chemical stability, (b) metabolic stability, and (c)

cytotoxicity and genotoxicity of the compounds were

determined.

CQS parasites. Indeed, the IC₅₀values determined for 5a−c in

the CQS strain were comparable to those measured in these

parasites for IQ, AQ, and CQ. However, only one of the

compounds (5c) was equally active against both CQS and

CQR strains, with a resistance index (RI) of 0.8 (the RI is the

IC₅₀value measured in the CQR strain divided by that

measured in the CQS strain).

Increases in the lipophilicity of the substituents were

associated with modest increases in potency against the CQR

strain (5c > 5b > 5a), but this trend was not observed in the

CQS strain. Likewise, the ﬁnding that the RI values calculated

for compounds 5a−c decrease with increasing lipophilicity (3.1,

2.0, and 0.8, respectively) supports the idea that activity against

CQR parasites is inﬂuenced by the lipophilicity of the

substituent. NMR-based stability studies revealed that com-

pounds 5a−c were stable in a DMSO/D₂O solution for several

days at 25 °C. However, rapid opening of the benzoxazine ring

occurred in the presence of an aqueous or methanolic solution

of DCl. Furthermore, when incubated at 37 °C, compounds

5a−c were stable at pH 7.4 but became unstable after only 1 h

at pH 2. We therefore performed isosteric replacement of the

benzoxazine O with a NH group in order to increase chemical

stability, which yielded quinazolines 6a−c. NMR studies

conﬁrmed that compounds 6a−c were stable for 24 h at 37

°C under both neutral and acidic conditions. The tetrahy-

droquinazoline derivatives 6a,b (which possess ethyl and n-

propyl substituents, respectively, at the tetrahydroquinazoline

N3) exhibited potent activities against the CQS strain but were

less eﬀective against CQR parasites (RI = 5.4 and 2.5,

respectively). It is worth noting that the two compounds that

bear a lipophilic benzyl group (5c and 6c) were equally active

against CQR and CQS strains (RI = 0.8 and 1.3, respectively).

This suggests that they evade, or are not signiﬁcantly aﬀected

by, the mechanisms underlying resistance to CQ.

The primary determinant of CQ resistance in P. falciparum

are mutations in the “chloroquine resistance transporter”

(PfCRT). The resistance-conferring form of PfCRT

(PfCRT^CQR) mediates the eﬄux of CQ out of the digestive

vacuole (DV), away from its site of accumulation and action,

whereas the sensitive form (PfCRT^CQS) lacks this ability.^30,31

We investigated interactions between PfCRT and compounds

E

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

whereas 6b was moderately soluble under the same conditions.

Compound 6b was also chemically stable at pH 7.4.

b. Metabolic Stability. An in silico analysis of the potential

sites of CYP450-mediated metabolism (soft spots) for

compounds 6b and 6c was performed using the P450 site of

metabolism (SOM) prediction workﬂow³⁴(implemented in

Maestro suite 2011³⁵) that predicts metabolism mediated by

isoforms CYP3A4, CYP2C9, and CYP2D6. The results are

presented in Figure S2 of the Supporting Information. The data

indicates that 6b and 6c both contain soft spots at the quinoline

and tetrahydroquinazoline rings and that the benzylic aromatic

portion of 6c contains an additional set of soft spots.

A preliminary assessment of the metabolic stability of 6b and

6c was obtained using a human recombinant CYP3A4 assay

(Table 3). Compound 6b was more resilient to CYP3A4-

modiﬁcation than was 6c. This ﬁnding is consistent with the

prediction (Figure S2 of the Supporting Information) that both

compounds contain CYP3A4 soft spots and that 6c is likely to

be more susceptible to metabolism than 6b because it contains

an additional set of soft spots in its benzylic tail group.

The metabolic stability of compound 6b was studied further

by determining its in vitro intrinsic clearance (Cl_int) in the

presence of liver microsomes. The clearance of 6b was high in

the presence of either mouse or rat liver microsomes (Table 3).

When assessed in the presence of human liver microsomes

(Figure 2), the natural logarithm of the percent of 6b remaining

in the nonmetabolized form decreased linearly over time,

indicating that the CYP-mediated depletion of 6b followed a

monoexponential relationship. The calculated k (elimination

rate constant) and Cl_intvalues were 0.0192 min⁻¹and 48.0

(μL/min/mg), respectively.

In Vitro Toxicity Assays and Ames Test. Compounds

5a,c and 6b,c were tested for mammalian cytotoxicity in

NIH3T3 (mouse embryonic ﬁbroblast) and HepG2 (human

hepatocellular liver carcinoma) cells. The resulting TC₅₀values

(reported in Table 1) indicate that all of the compounds

possess low cytotoxic potential.

Figure 1. (A) [³H]CQ uptake into oocytes expressing Dd2 PfCRT^CQR

(black bars) or D10 PfCRT^CQS(gray bars) in the presence of 100 μM

of 6b or 6c. Both compounds caused modest but statistically

signiﬁcant reductions in CQ uptake via PfCRT^CQR(P < 0.001;

ANOVA). The level of inhibition by verapamil is represented as a solid

line with the SEM from n = 4 represented by gray dashed lines. (B)

The concentration-dependent eﬀect of 6c on the uptake of [³H]CQ

into oocytes expressing Dd2 PfCRT^CQR(closed circles) or D10

PfCRT^CQS(open circles). The IC₅₀value derived from the data for 6c

The mutagenic potential of compounds 5a,b and 6b,c was

investigated in the Salmonella typhimurium strains TA98 and

TA100 using the Ames test.^36,37This assay examines the ability

of compounds to cause DNA mutations. As shown in Figure 3

(TA100 strain) and in Figure S3 of the Supporting Information

(TA98 strain), none of the compounds were mutagenic over

the concentration range tested (5−2940 μM). The tests

produced similar results when repeated in the presence of the

S9 fraction of rat liver (to mimic metabolic activation of the

compounds; Figure S4 of the Supporting Information).

In Vivo Antimalarial Activity against P. berghei. The

potent antiplasmodial activities and favorable toxicity proﬁles of

5a and 6b,c encouraged a preliminary in vivo evaluation of

these compounds in the P. berghei mouse model of malaria. In

vivo testing was performed on groups of six BALB/c mice

inoculated intraperitoneally with 10⁷P. berghei-infected

erythrocytes according to the Peters 4-day test.³⁸The

antimalarial activity (ED₅₀, mg/kg) was determined by

nonlinear regression (four-parameter logistic curve ﬁt).

Analogue 5a was found to be active after intraperitoneal

administration, but its lack of stability precluded oral

administration (data not shown). Compounds 6b,c both

exhibited excellent oral in vivo activity, with compound 6b

slightly more eﬀective than 6c (ED₅₀values of 8.7 and 15.7 mg/

kg, respectively; Figure 4). This ﬁnding correlates well with the

relative potencies of these two compounds in the in vitro assays

(228

25 μM; n = 4), was obtained by least-squares ﬁt of the

equation Y = Y_min+ [(Y_max− Y_min)/(1 + ([inhibitor]/IC₅₀)^C], where Y

is PfCRT^CQRmeditated CQ transport, Y_minand Y_maxare the minimum

and maximum values of Y, and C is a constant. PfCRT^CQR-mediated

CQ transport was calculated by subtracting the uptake measured in

oocytes expressing D10 PfCRT^CQSfrom that in oocytes expressing

Dd2 PfCRT^CQR. Uptake is shown as the mean SEM from ﬁve (A) or

four (B) separate experiments, within which measurements were made

from 10 oocytes per treatment. Note that noninjected oocytes and

oocytes expressing PfCRT^CQStake up [³H]CQ to similar (low) levels

via simple diﬀusion of the neutral species; this represents the

“background” level of CQ accumulation in oocytes (refer to ref 31

for full data and a detailed discussion).

a. Solubility and Chemical Stability. Compounds 6b,c were

tested for water solubility (at pH 3.0 and 7.4) and chemical

stability (at pH 7.4). The results are summarized in Table 3.

Both compounds were more soluble at pH 3.0 than at pH 7.4,

but 6b was the most soluble compound; the solubility of

compound 6c was below the limit of detection at pH 7.4

F

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Table 2. Calculated Molecular Properties of Compounds 5a−c, 6a−c, and the reference antimalarials IQ, AQ, and CQ

a

Values were calculated using the OSIRIS property explorer (Actelion; http://www.organic-chemistry.org/prog/peo/). A Drug-Likeness value

greater than 0 indicates that the molecule contains fragments that are frequently represented in commercially available drugs. The Drug-Score

combines (i) drug-likeness, (ii) clogP, (iii) log S, (iv) molecular weight, and (v) toxicity risk. The following properties are typical of approved drugs

that have low toxicity: clogP < 5, log S > −4, molecular weight < 450, drug likeness > 0, and a high drug score. Further details can be found at:

http://www.rdchemicals.com/property-explorer-tutorial.html (access date 10/03/2012). Color code: green circles, no risk; yellow circle, medium

risk; red circles, high risk.

important membrane transporters that are known to be

inhibited by 4-aminoquinolines (including CQ).^23,39,40

Reducing the risk of drug-induced cardiac arrhythmia is a

major hurdle in the development of new chemotherapies. The

most common issue is acquired long QT syndrome, which can

be caused by drugs that interfere with hERG K⁺channels and

thereby delay cardiac repolarization and increase the risk of

arrhythmia. Hence, the ability to inhibit hERG K⁺channels is

considered to be an indicator of the compound’s potential for

producing pro-arrythmic eﬀects. We investigated interactions

between hERG K⁺channels and compounds 6b and 6c using

patch-clamp techniques.⁴¹The results are summarized in

Figure 5. Compound 6b was a potent inhibitor of the rat

hERG K⁺channel and was comparable in its inhibitory activity

to the antimalarial halofantrine. By contrast, 6c did not aﬀect

the function of the rat hERG K⁺channel, even when tested at

10 μM.

Table 3. Solubility and the Chemical and in Vitro Metabolic

Stability of Compounds 6b,c

property

6b

6c

a

solubility at pH 7.4 (μM)

solubility at pH 3.0 (μM)

chemical stability (pH 7.4) (%)

53

nd

b

231

100

122

nd

c

metabolic stability (hCYP3A4) (%)

43 (high)

18 (low)

d

Cl_int(μL/min/mg)

mouse

rat

219.0

121.0

a

b

The aqueous solubility was below the limit of quantitation. Results

c

refer to the largest UV peak. Metabolic stability was calculated as the

percentage of compound remaining after a 1 h incubation with

d

recombinant hCYP3A4. In vitro intrinsic clearance determined in the

presence of liver microsomes.

The Na⁺/K⁺-ATPase (sodium−potassium pump) is essential

to several physiological processes, including generation of the

resting membrane potential and the Na⁺gradient as well as the

regulation of cell volume.⁴²We tested the ability of compounds

6b and 6c to inhibit the hydrolytic activity of the Na⁺/K⁺-

ATPase. Membrane fragments extracted from the outer

medulla of rabbit kidney were treated with 10 μM of 6b, 6c,

or ouabain (a highly speciﬁc Na⁺/K⁺-ATPase inhibitor) and the

concentration of P_imeasured using a colorimetric assay. As

shown in Figure 6, ouabain abolished Na⁺/K⁺-ATPase activity,

whereas both 6b and 6c only caused 35−40% reductions in

hydrolytic activity. Therefore, both 6b and 6c have very

moderate inhibitory eﬀects on the steady-state enzyme activity

and can be considered to be poor inhibitors of the Na⁺/K⁺-

ATPase.

Figure 2. Time course for the depletion of 6b in the presence of

human liver microsomes. The depletion of 6b is expressed as the

natural logarithm of the percent of 6b remaining in the non-

metabolized form at each time point.

Computational Study of the hERG K⁺Channel. The

molecular basis for the diﬀerent activities of 6b and 6c against

hERG K⁺channels were investigated by molecular docking

simulation. The computational methodology employed GOLD

software (The Cambridge Crystallographic Data Centre,

Cambridge, UK)⁴³and applied the scoring function Gold-

Score⁴⁴using a homology model of the hERG K⁺channel⁴⁵

and indicates that both analogues represent lead structures for

further development.

Ether-a-go-go Related Gene (hERG) K⁺Channel and

Na⁺/K⁺-ATPase Assays. To characterize the safety proﬁle of

the new antimalarials in greater detail, the eﬀects of compounds

6b and 6c on the rat hERG K⁺channel and the rabbit Na⁺/K⁺-

ATPase were assessed. Both proteins are physiologically

(available from Schrodinger;⁴⁶see Experimental Section for

̈

further details). This computational approach was recently used

G

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 3. Ames test performed on the S. typhimurium TA100 strain with compound 5c (A), 5a (B), 6c (C), or 6b (D). Red star = p < 0.01.

Figure 5. (A) Current traces elicited by depolarizing voltage pulse in

20 mV steps from a holding potential of −80 mV in the absence (top)

or presence (bottom) of 100 nM of compound 6b. (B) Current traces

elicited by depolarizing voltage pulse in 20 mV steps from a holding

potential of −80 mV in the absence (top) or presence (bottom) of 100

nM of compound 6c. (C) Concentration-dependent blockade of the

rat hERG K⁺channel by 6b. The IC₅₀value derived from the data is 22

4 nM, and the n value (Hill coeﬃcient) is 0.64 0.07 (n = 3).

to explain and/or to predict the aﬃnity of a range of

compounds⁴⁷⁻⁵⁶(including antimalarials⁵⁷) for hERG K⁺

channels. The results of the docking simulation are depicted

in Figure 7. The reliability of the docking results was conﬁrmed

by superimposition of docked poses of compounds 6b and 6c

with their calculated lowest energy conformations (Figure S5 of

the Supporting Information).

The two panels of Figure 7 display representative poses of

the most populated cluster of docked solutions for compounds

6b (Figure 7A) and 6c (Figure 7B). Docking poses relative to

Figure 4. In vivo antimalarial activities of compounds 6b,c in P.

the reference compounds CQ, AQ, IQ, and halofantrine have

berghei-infected mice. The blue lines represent the 95% conﬁdence

been also taken into consideration and are displayed in Figure

limits of the ﬁtted curve.

S6 of the Supporting Information. The data indicates that

compound 6b strongly interacts with the key residues Y652 and

H

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 6. Hydrolytic activity of the Na⁺/K⁺-ATPase in the presence of

10 μM of Ouabain, 6b, or 6c. Values were normalized with respect to

the control. The error bars represent sem (n = 3) or range/2 (n = 2).

F656 in the α and β subunits, respectively. The lowest energy

arrangement (Figure 7A) allowed aromatic π−π stacking

interactions between these residues and the quinoline and

quinazoline moieties of 6b. Furthermore, the protonated

nitrogens of compound 6b (quinoline N1 and quinazoline

N3) produced a series of H-bonds with T623α and S649γ, two

of the key residues involved in hERG K⁺channel blockade.⁴⁹

By contrast, the best cluster of docked positions for

compound 6c (Figure 7B and Figure S5B of the Supporting

Information) suggests that it has a low-aﬃnity interaction with

the hERG K⁺channel. This is in part due to partial π−π

stacking with residues Y652 and F656 of the δ subunit because

the distances between these residues and the quinoline and

quinazoline moieties of 6c exceeded 5 Å (which precludes VdW

interactions and full π−π stacking). Furthermore, there was no

potential for interactions between the aromatic side chains of

Y652γ and F656γ and the benzyl group of 6c. In addition, for a

large number of low energy poses of the most populated cluster

of 6c, polar contacts with other residues in the pocket were not

evident; this may be due to steric hindrance by the benzyl

group. Taken together, these observations could explain the

low aﬃnity of 6c for the hERG K⁺channel. Indeed, the

GoldScore value for 6c (50.06) was similar to those calculated

for CQ, IQ, and AQ (49.16, 51.93, and 51.09, respectively;

these three drugs inhibit hERG in the μM range). The

representative binding poses of CQ, IQ, and AQ are depicted in

Figure S6 of the Supporting Information (panels A, B, and C,

respectively). The interaction patterns observed for CQ, AQ,

and IQ in the docking simulations, together with their

respective GoldScores, were diﬀerent from that observed for

halofantrine, which was the most active compound analyzed

(see Figure S5 of the Supporting Information). On the other

hand, the GoldScore obtained for 6b (57.42) was comparable

to that of halofantrine (62.78; Figure S6D of the Supporting

Information). This ﬁnding is consistent with the observation

that both compounds are potent inhibitors of the hERG K⁺

channel (6b IC₅₀= 20 nM, halofantrine IC₅₀= 40 nM²³).

Using the best docked poses of 6b and 6c (Figure 7) as the

starting point, 5 ns of molecular dynamics (MD) were

performed with the Desmond package.^59,60The resulting

predictions were similar to those produced from the static

docking calculations, as shown by superposition of the initial

docked structure and the ﬁnal structures (Figures S7 and S8 of

the Supporting Information). The interactions depicted for

compound 6b in Figure 7 were maintained for the duration of

the simulation with the exception of the H-bond with S649β.

Figure 7. (A) Docked pose of 6b (in orange sticks) in the cavity of the

hERG channel (cyan cartoon; α, β, γ, and δ refers to protein subunits).

It strongly interacts with residues Y652 and F656, and it forms H-

bonds with key residues T623α (quinoline N1) and S649γ (N3 of the

quinazoline moiety). H-bonds are represented by black dotted lines.

(B) Docked pose of 6c (in magenta sticks) in the cavity of hERG

channel (cyan cartoon). It interacts weakly by partial π−π stacking

only with residues Y652 and F656 from δ and γ subunits. Molecular

graphics were generated by PyMOL.⁵⁸Nonpolar hydrogens were

omitted for clarity.

This bond was lost in the initial steps of the simulation and was

replaced by a more favorable H-bond formed between the

quinazoline N3 and the backbone oxygen of residue M645β,

which is known to be located in the hERG binding site (Figure

S9A and Movie S1 of the Supporting Information).^52,61

Additional H-bonds were formed between the quinazoline

N1 of 6b and residues S624 (from both the β and γ subunits)

and T623 (from the β subunit) (Figure S9A and Movie S1of

the Supporting Information). Moreover, as illustrated in

Figures S9C and S9E of the Supporting Information, the

distances monitored during MD simulation for the quinazoline

and quinoline moieties remained favorable (<5 Å) for

producing a strong π−π stacking with Y652 from both the α

I

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

compound was obtained following the above-described procedure in

48% yield as a pale-brown oil. ¹H NMR (300 MHz, CDCl₃) δ 8.24 (br,

1H), 7.02−6.84 (m, 3H), 6.05 (br, 2H), 3.90 (s, 2H), 2.58 (t, J = 6.9

Hz, 2H), 2.00 (s, 3H), 1.54−1.46 (m, 2H), 0.89 (t, J = 7.5 Hz, 3H).

MS (ESI) m/z 223 [M + H]⁺.

and β subunits. By contrast, the MD analysis conﬁrmed that

compound 6c only established two (infrequent) H-bonds with

the hydroxyl group of Y652γ and S649β (Figure S9B and

Movie S2 of the Supporting Information) and that full π−π

stackings were not observed with Y652 or F656 (from γ and δ

subunits, respectively) because the distance between these

residues and 6c was >5 Å (Figure S9D,F,G of the Supporting

Information). These simulation results are again consistent with

the observed high and low aﬃnities, respectively, of 6b and 6c

for the hERG channel, and as such, this work conﬁrms that the

coupling of molecular docking and MD simulations could be

used to predict and/or investigate the aﬃnity of diﬀerent

molecules for a given protein target.

N-(4-((Ethylamino)methyl)-3-hydroxyphenyl)-7-chloro-4-amino-

quinoline (9a). Aqueous HCl (20% solution, 3.5 mL) was added to 8a

(0.63 g, 3.02 mmol), and the solution was heated under reﬂux for 6 h.

The solvent was then removed in vacuo, and the resulting residue

coevaporated with ethanol to give the corresponding hydrochloride

salt. To a solution of the latter salt in ethanol (5.0 mL), 4,7-

dichloroquinoline (0.66 g, 3.33 mmol) was added and the reaction was

heated under reﬂux for 12 h. Thereafter, the solvent was removed

under reduced pressure and the residue puriﬁed by silica gel ﬂash

column chromatography (5% methanol in chloroform) to yield the

quinoline hydrochloride salt. To obtain the compound as a free base,

the solid was dissolved in water and a saturated solution of NaHCO₃

was added. The aqueous solution was extracted with dichloromethane

and dried over Na₂SO₄, and the solvent was removed in vacuo to

aﬀord the desired product in 78% yield as a pale-brown solid.

Spectroscopic data are consistent with those reported in the

literature.²¹

CONCLUSIONS

■

This study describes new benzoxazine and quinazoline-based

analogues of the antimalarial drugs IQ and AQ. Compounds

5a−c possessed excellent in vitro antiplasmodial activities but

were unstable in acidic conditions and were predicted to have

suboptimal ADME+T properties. These issues were addressed

in a series of quinazoline counterparts (compounds 6a−c) that

retained potent antiplasmodial activity, were chemically stable,

and possessed signiﬁcant oral in vivo activity against mouse

malaria. Quinazoline 6c emerged as the most promising

analogue of this series. In addition to its potency in vivo

(EC₅₀of 15.7 mg/kg; oral administration), 6c was equally

active against CQS and CQR strains of P. falciparum.

Furthermore, 6c exhibited little or no in vitro cytotoxicity or

mutagenicity and was a poor inhibitor of both the hERG K⁺

channel and the Na⁺,K⁺-ATPase. However, 6b was found to be

more metabolically stable than 6c. Computational studies

provided a structural basis for the diﬀering aﬃnities of 6b and

6c for the hERG K⁺channel, thereby establishing a guide for

the synthesis of optimized antimalarials.

N-(4-((Propylamino)methyl)-3-hydroxyphenyl)-7-chloro-4-ami-

noquinoline (9b). Starting from 8b (0.59 g, 2.66 mmol), the title

compound was obtained following the above-described procedure in

1

74% yield as a pale-brown solid. H NMR (300 MHz, CDCl₃) δ 8.44

(m, 2H), 7.92 (s, 1H), 7.64 (d, J = 9.3 Hz, 1H), 7.46 (d, J = 7.8 Hz,

1H), 7.04 (m, 3H), 4.23 (s, 2H), 4.03 (s, 1H), 3.03 (t, J = 7.5 Hz, 2H),

2.63 (s, 1H), 1.77 (m, 2H), 1.04 (t, J = 7.2 Hz, 3H). MS (ESI) m/z

342 [M + H]⁺.

N-(3-Ethyl-3,4-dihydro-2H-benzo[e][1,3]oxazin-7-yl)-7-chloro-4-

aminoquinoline (5a). To a mixture of 9a (78.0 mg, 0.24 mmol) and

KOH (17.0 mg, 0.31 mmol) in methanol (2.0 mL), formaldehyde

(37% in water, 58 μL, 0.78 mmol) was added. The reaction was heated

under reﬂux for 1 h; afterward, the solvent was evaporated and the

resulting residue dissolved in water. The aqueous layer was extracted

with chloroform, the combined organic layers were dried over Na₂SO₄,

and the solvent was removed in vacuo. The crude was puriﬁed by silica

gel ﬂash chromatography (5% methanol in chloroform) aﬀording 5a in

1

36% yield as a pale-yellow low melting solid. H NMR (300 MHz,

EXPERIMENTAL SECTION

CDCl₃) δ 8.55 (d, J = 5.3 Hz, 1H), 8.03 (d, J = 2.1 Hz, 1H), 7.84 (d, J

= 9.0 Hz, 1H), 7.45 (dd, J = 9.0, 2.1 Hz, 1H), 7.04−6.97 (m, 2H), 6.79

(dd, J = 8.0, 2.3 Hz, 1H), 6.72 (d, J = 2.3 Hz, 1H), 6.53 (br, 1H), 4.91

(s, 2H), 4.02 (s, 2H), 2.84 (q, J = 7.2 Hz, 2H), 1.20 (t, J = 7.2 Hz,

3H). MS (ESI) m/z 340 [M + H]⁺. HRMS (ESI) calcd for

C₁₉H₁₉ClN₃O [M + H]⁺340.1211, found 340.1215.

■

Chemistry. Starting materials and solvents were purchased from

commercial suppliers and used without further puriﬁcation. Reaction

progress was monitored by TLC using silica gel 60 F254 (0.040−0.063

mm) with detection by UV. Silica Gel 60 (0.040−0.063 mm) or

aluminum oxide 90 standardized were used for column chromatog-

1

N-(3-Propyl-3,4-dihydro-2H-benzo[e][1,3]oxazin-7-yl)-7-chloro-4-

aminoquinoline (5b). Starting from 9b (71 mg, 0.21 mmol), the title

compound was obtained following the above-described procedure in

raphy. Yields refer to puriﬁed materials and are not optimized. H

NMR and ¹³C NMR spectra were recorded on a Varian 300 MHz or a

Bruker 400 MHz spectrometer using the residual signal of the

deuterated solvent as internal standard. Splitting patterns are described

as singlet (s), doublet (d), triplet (t), quartet (q), and broad (br);

chemical shifts (δ) are given in ppm and coupling constants (J) in

hertz (Hz). Mass spectra were recorded utilizing electrospray

ionization (ESI). All moisture-sensitive reactions were performed

under argon atmosphere using oven-dried glassware and anhydrous

solvents. All compounds that were tested in the biological assays were

analyzed by combustion analysis (CHN) to conﬁrm the purity >95%.

N-(4-((Ethylamino)methyl)-3-hydroxyphenyl)acetamide (8a). A

solution of 3-hydroxyacetanilide 7 (2.50 g, 16.5 mmol) in ethanol

(15.0 mL) was added to a stirring mixture of formaldehyde (37% in

water, 1.23 mL, 44.6 mmol) and ethylamine (2.0 M solution) in THF

(8.3 mL, 16.6 mmol). The reaction was allowed to heat under reﬂux

for 24 h; afterward, the solvent was removed under reduced pressure

and the crude material was puriﬁed by chromatography using 5%

methanol in chloroform to furnish the title compound as brown oil in

55% yield. Spectroscopic data are consistent with those reported in the

literature.²¹

1

32% yield as a pale-yellow low melting solid. H NMR (300 MHz,

CDCl₃) δ 8.56 (d, J = 5.4 Hz, 1H), 8.03 (s, 1H), 7.84 (d, J = 8.9 Hz,

1H), 7.45 (dd, J = 8.9, 2.2 Hz, 1H), 7.01 (d, J = 5.4 Hz, 1H), 6.99 (d, J

= 8.1 Hz, 1H), 6.79 (dd, J = 8.1, 2.2 Hz, 1H), 6.72 (s, 1H), 6.50 (s,

1H), 4.90 (s, 2H), 4.01 (s, 2H), 2.74 (t, J = 7.8 Hz, 2H), 1.57 (m, 2H),

0.96 (t, J = 7.5 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 155.5, 152.2,

150.0, 147.6, 139.0, 135.4, 129.3, 128.8, 126.3, 121.3, 118.3, 117.3,

115.1, 110.5, 103.0, 83.0, 53.6, 50.2, 21.5, 11.9. MS (ESI) m/z 354 [M

+ H]⁺. HRMS (ESI) calcd for C₂₀H₂₁ClN₃O [M + H]⁺354.1373,

found 354.1368.

N-(3-Benzyl-3,4-dihydro-2H-benzo[e][1,3]oxazin-7-yl)-7-chloro-

4-aminoquinoline (5c). To a solution of formaldehyde (37% in water,

120 μL, 0.81 mmol) in methanol (3 mL), benzylamine (80 μL, 0.81

mmol) was added and the mixture was heated under reﬂux for 3 h to

allow the formation of the Schiﬀ base. Successively a solution of 10^22,23

(200 mg, 0.73 mmol) in methanol was added and the reaction was

heated at 65 °C for 3 h. Thereafter, the mixture was cooled at 25 °C,

and the resulting precipitate was washed with methanol, aﬀording 5c

1

in 79% yield, as a pale-yellow low melting solid. H NMR (300 MHz,

N-(4-((Propylamino)methyl)-3-hydroxyphenyl)acetamide (8b).

Starting from 7 (2.00 g, 13.2 mmol), formaldehyde (37% in water,

0.91 mL, 33.1 mmol), and propylamine (1.1 mL, 13.2 mmol), the title

CD₃OD) δ 8.57 (d, J = 5.3 Hz, 1H), 8.03 (d, J = 2.1 Hz, 1H), 7.86 (d,

J = 9.0 Hz, 1H), 7.45 (dd, J = 9.0, 2.1 Hz, 1H), 7.30−7.38 (m, 5H),

J

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

1

7.04 (d, J = 5.3 Hz, 1H), 6.97 (d, J = 7.8 Hz, 1H), 6.77−6.83 (m, 2H),

6.61 (br, 1H), 4.92 (s, 2H), 3.99 (s, 2H), 3.96 (s, 2H). MS (ESI) m/z

402 [M + H]⁺. HRMS (ESI) calcd for C₂₄H₂₁ClN₃O [M + H]⁺

402.1373, found 402.1371.

aﬀord 13d in 84% yield as a reddish amorphous solid. H NMR (300

MHz, CDCl₃) δ 8.51 (d, J = 2.2 Hz, 1H), 8.47 (d, J = 8.8 Hz, 1H),

8.24 (dd, J = 8.8, 2.2 Hz, 1H), 8.16 (s, 1H), 4.98 (s, 2H), 4.43−4.31

(m, 2H), 4.30−4.11 (m, 7H). MS (ESI) m/z 388 [M − H]⁻.

7-Amino-3-ethylquinazolin-4(3H)-one (14a). To a solution of 13a

(1.05 g, 4.8 mmol) in 75% aqueous ethanol (25 mL), iron (1.74 g,

31.2 mmol) and CaCl₂(705 mg, 4.8 mmol) were added and the

reaction mixture was heated under reﬂux for 2 h. After cooling to 25

°C, the mixture was ﬁltered through a pad of Celite and the solvent

was evaporated. The resulting residue was partitioned between water

and ethyl acetate. The organic extracts were dried over Na₂SO₄and

evaporated under reduced pressure, aﬀording pure title compound as a

pale-yellow solid in quantitative yield; mp (ethyl acetate/n-hexane)

174−176 °C. ¹H NMR (300 MHz, CDCl₃) δ 8.09 (d, J = 8.0 Hz, 1H),

7.96 (s, 1H), 6.82−6.74 (m, 2H), 4.16 (br, 2H), 4.01 (q, J = 7.2 Hz,

2H), 1.39 (t, J = 7.2 Hz, 3H). MS (ESI) m/z 190 [M + H]⁺, 212 [M +

Na]⁺.

N-Ethyl-2-amino-4-nitrobenzamide (12a). To a suspension of 11

(2.0 g, 11.0 mmol) in dry dichloromethane cooled at 0 °C, EDC (2.3

g, 12.1 mmol), HOBt (1.6 g, 12.1 mmol), and Et₃N (1.7 mL, 12.1

mmol) were added and the mixture was stirred at 0 °C for 1 h.

Successively ethylamine (2 M in THF, 5.5 mL, 11.0 mmol) was added

and the reaction was stirred at 25 °C for 12 h. Thereafter, the organic

mixture was washed with saturated aqueous NaHCO₃, dried over

Na₂SO₄, and evaporated to dryness. The residue was puriﬁed by silica

gel chromatography (5% methanol in chloroform) to aﬀord the title

compound in 80% yield as yellow solid; mp (ethyl acetate/n-hexane)

1

130−132 °C. H NMR (300 MHz, CDCl₃) δ 7.49 (s, 1H), 7.42 (s,

2H), 6.07 (br, 1H), 5.80 (br, 2H), 3.53−3.41 (m, 2H), 1.27 (t, J = 7.5

Hz, 3H). MS (ESI) m/z 210 [M + H]⁺.

N-Propyl-2-amino-4-nitrobenzamide (12b). Starting from 11 (2.0

g, 11.0 mmol) and n-propylamine (900 μL, 11.0 mmol), the title

compound was obtained following the procedure reported for 12a in

91% yield as a yellow solid; mp (ethyl acetate/n-hexane) 128−130 °C.

¹H NMR (300 MHz, CDCl₃) δ 7.51 (s, 1H), 7.43 (s, 2H), 6.07 (br,

1H), 3.48−3.33 (m, 2H), 1.73−1.57 (m, 2H), 1.00 (t, J = 7.4 Hz, 3H).

MS (ESI) m/z 222 [M − H]⁻.

7-Amino-3-propylquinazolin-4(3H)-one (14b). Starting from 13b

(1.23 g, 5.4 mmol), the title compound was obtained following the

procedure reported for 14a in 96% yield as a pale-yellow solid; mp

(ethyl acetate/n-hexane) 146−148 °C. ¹H NMR (300 MHz, CDCl₃) δ

8.04 (d, J = 8.0 Hz, 1H), 7.89 (s, 1H), 6.80−6.72 (m, 2H), 4.37 (br,

2H), 4.02−3.70 (m, 2H), 1.92−1.59 (m, 2H), 0.94 (t, J = 7.4 Hz, 3H).

MS (ESI) m/z 226 [M + Na]⁺.

7-Amino-3-benzylquinazolin-4(3H)-one (14c). To a solution of

13c (310 mg, 1.10 mmol)) in ethyl acetate (8.0 mL), SnCl₂·2H₂O

(1.37 g, 6.1 mmol) was added. The mixture was heated under reﬂux

for 2 h and then poured into saturated aqueous NaHCO₃. The

resulting gelatinous emulsion was ﬁltered through a pad of Celite, and

the biphasic ﬁltrate was extracted with ethyl acetate. The combined

organic extracts were dried over Na₂SO₄and concentrated in vacuo,

and the residue was puriﬁed by silica gel ﬂash chromatography (2%

methanol in dichloromethane), aﬀording 14c in 74% yield as a pale-

yellow solid; mp (ethyl acetate/n-hexane) 165−168 °C. ¹H NMR

(300 MHz, CDCl₃) δ 8.11 (d, J = 7.6 Hz, 1H), 8.00 (s, 1H), 7.41−

7.28 (m, 5H), 6.82−6.73 (m, 2H), 5.15 (s, 2H), 4.22 (br, 2H). ¹³C

NMR (75 MHz, CDCl₃) δ 160.9, 152.3, 150.2, 147.2, 143.7, 136.3,

129.2, 128.9, 128.4, 128.2, 116.1, 109.2, 49.5. MS (ESI) m/z 252 [M +

H]⁺, 274 [M + Na]⁺.

7-Amino-3-(ferrocenylmethyl)quinazolin-4(3H)-one (14d). Start-

ing from 13d (350 mg, 0.88 mmol), the title compound was obtained

following the procedure reported for 14a as yellow−orange solid in

91% yield; mp (ethyl acetate/n-hexane) 190 °C (dec). ¹H NMR (300

MHz, CDCl₃) δ 8.09 (d, J = 9.2 Hz, 1H), 7.97 (s, 1H), 6.86−6.69 (m,

2H), 4.89 (s, 2H), 4.40−4.29 (m, 2H), 4.27−4.08 (m, 7H). MS (ESI)

m/z 360 [M + H]⁺, 382 [M + Na]⁺.

7-((7-Chloroquinolin-4-yl)amino)-3-ethylquinazolin-4(3H)-one

(15a). To a solution of 14a (400 mg, 2.12 mmol) in ethanol (10.0

mL), 4,7-dichloroquinoline (419 mg, 2.12 mmol) was added and the

reaction mixture was heated under reﬂux for 48 h. The resulting

precipitate was collected, washed with ethanol, and dried to give the

title compound in quantitative yield as a yellow solid; mp (methanol)

> 300 °C. ¹H NMR (300 MHz, DMSO-d₆) δ 11.30 (s, 1H), 8.88 (d, J

= 9.1 Hz, 1H), 8.64 (d, J = 6.9 Hz, 1H), 8.48 (s, 1H), 8.27 (d, J = 8.6

Hz, 1H), 8.18 (d, J = 2.0 Hz, 1H), 7.91 (dd, J = 9.1, 1.9 Hz, 1H), 7.76

(d, J = 1.9 Hz, 1H), 7.65 (dd, J = 8.6, 2.0 Hz, 1H), 7.13 (d, J = 6.9 Hz,

1H), 4.03 (q, J = 7.1 Hz, 2H), 1.29 (t, J = 7.1 Hz, 3H). ¹³C NMR (75

MHz, DMSO-d₆) δ 160.1, 154.6, 150.0, 149.7, 145.0, 143.4, 140.3,

139.1, 128.7, 128.3, 126.9, 123.8, 122.1, 120.4, 117.4, 102.2, 42.0, 15.2.

MS (ESI) m/z 351 [M + H]⁺.

N-(Ferrocenylmethyl)-2-amino-4-nitrobenzamide (12d). Starting

from 11 (1.1 g, 6.04 mmol) and ferrocenylamine (1.3 g, 6.04 mmol),

the title compound was obtained following the above-described

1

procedure in 89% yield as a red low melting solid. H NMR (300

MHz, CDCl₃) δ 7.51 (s, 1H), 7.42 (s, 2H), 6.24 (s, 1H), 5.83 (s, 2H),

4.31 (d, J = 5.2 Hz, 2H), 4.29−4.22 (m, 2H), 4.22−4.14 (m, 7H). MS

(ESI) m/z 378 [M − H]⁻.

3-Ethyl-7-nitroquinazolin-4(3H)-one (13a). A mixture of 12a (1.13

g, 5.4 mmol) and formic acid (2.2 mL, 57 mmol) was heated under

reﬂux for 3 h. After cooling to 25 °C, the mixture was poured into

water, and the solution was alkalinized with NaOH and extracted with

chloroform. The organic phase was dried over Na₂SO₄and evaporated

under reduced pressure to aﬀord pure 13a in quantitative yield as a

light-yellow solid; mp (ethyl acetate/n-hexane) 148−150 °C. ¹H NMR

(300 MHz, CDCl₃) δ 8.54 (d, J = 2.2 Hz, 1H), 8.47 (d, J = 8.8 Hz,

1H), 8.25 (dd, J = 8.8, 2.2 Hz, 1H), 8.16 (s, 1H), 4.10 (q, J = 7.2 Hz,

2H), 1.45 (t, J = 7.2 Hz, 3H). MS (ESI) m/z 220 [M + H]⁺.

7-Nitro-3-propylquinazolin-4(3H)-one (13b). Starting from 12b

(1.33 g, 6.0 mmol) and formic acid (2.4 mL, 62 mmol), the title

compound was obtained following the procedure reported for 13a in

quantitative yield as a light-yellow solid; mp (ethyl acetate/n-hexane)

122−124 °C. ¹H NMR (300 MHz, CDCl₃) δ 8.55 (d, J = 2.2 Hz, 1H),

8.47 (d, J = 8.8 Hz, 1H), 8.26 (dd, J = 8.8, 2.2 Hz, 1H), 8.13 (s, 1H),

4.04−3.96 (m, 2H), 1.94−1.77 (m, 2H), 1.02 (t, J = 7.4 Hz, 3H). MS

(ESI) m/z 234 [M + H]⁺.

3-Benzyl-7-nitroquinazolin-4(3H)-one (13c). A mixture of 4-

nitroanthranilic acid 11 (1.1 g, 6.04 mmol), trimethyl orthoformate

(792 μL, 7.25 mmol), benzylamine (790 μL, 7.25 mmol), and PTSA

(57 mg, 0.3 mmol) was heated by MW irradiation: 150 °C, 150 W, 3 ×

10 min. Thereafter water was added, followed by alkalinization with

NaOH and extraction with ethyl acetate. The organic layer was dried

over Na₂SO₄, and the solvent was evaporated under reduced pressure.

The crude was puriﬁed by silica gel chromatography (25% ethyl

acetate in n-hexane) to aﬀord the title compound in 53% yield as a

1

pale-yellow solid; mp (methanol) 163−164 °C. H NMR (300 MHz,

CDCl₃) δ 8.55 (d, J = 2.2 Hz, 1H), 8.49 (d, J = 8.8 Hz, 1H), 8.27 (dd,

J = 8.8, 2.2 Hz, 1H), 8.21 (s, 1H), 7.37 (s, 5H), 5.22 (s, 2H). ¹³C

NMR (75 MHz, CDCl₃) δ 160.1, 151.7, 148.8, 148.4, 135.2, 129.4

(2C), 129.1, 128.9, 128.4 (2C), 126.5, 123.4, 121.2, 50.2.

7-((7-Chloroquinolin-4-yl)amino)-3-propylquinazolin-4(3H)-one

(15b). Starting from 14b (400 mg, 1.97 mmol) and 4,7-dichloroquino-

line (390 mg, 1.97 mmol), the title compound was obtained following

the procedure described for 15a as a yellow solid in quantitative yield;

mp (methanol) > 300 °C. ¹H NMR (300 MHz, CD₃OD) δ 8.62 (d, J

= 9.1 Hz, 1H), 8.50 (d, J = 7.0 Hz, 1H), 8.44−8.38 (m, 2H), 8.01 (d, J

= 2.1 Hz, 1H), 7.85 (dd, J = 9.1, 2.1 Hz, 1H), 7.79 (d, J = 2.1 Hz, 1H),

7.68 (dd, J = 8.6, 2.1 Hz, 1H), 7.21 (d, J = 7.0 Hz, 1H), 4.17−3.97 (m,

2H), 1.96−1.73 (m, 2H), 1.01 (t, J = 7.4 Hz, 3H). MS (ESI) m/z 365

[M + H]⁺.

3-Ferrocenylmethyl-7-nitro-quinazolin-4(3H)-one (13d). A mix-

ture of 12d (515 mg, 1.36 mmol), trimethyl orthoformate (178 μL,

1.63 mmol), and PTSA (13 mg, 0.07 mmol) was heated under reﬂux

for 3 h. Thereafter water was added followed by extraction with

dichloromethane. The organic layer was dried over Na₂SO₄and

evaporated under reduced pressure, and the crude was puriﬁed by

silica gel ﬂash chromatography (25% ethyl acetate in n-hexane) to

K

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

3-Benzyl-7-((7-chloroquinolin-4-yl)amino)quinazolin-4(3H)-one

(15c). Starting from 14c (60 mg, 0.24 mmol) and 4,7-dichloroquino-

line (47 mg, 0.24 mmol), the title compound was obtained following

the procedure described for 15a as a yellow solid in quantitative yield;

3-Benzyl-7-((7-chloroquinolin-4-yl)amino)-2,3-dihydroquinazo-

lin-4(1H)-one (16). To a suspension of 15c (100 mg, 0.24 mmol) in

chloroform, oxalyl chloride (63 μL, 0.73 mmol) was added dropwise at

0 °C. After reﬂuxing for 1 h, the solvent was evaporated and the

residue dissolved in dry THF, and NaCNBH₃(76 mg, 1.21 mmol) was

added at 0 °C. The reaction mixture was stirred at 25 °C for 24 h, then

cooled at 0 °C and treated with 5 N NaOH. The aqueous layer was

extracted with ethyl acetate, and the combined organic layers were

washed with 5 N NaOH, dried over Na₂SO₄, and evaporated under

reduced pressure. The solid residue was washed with methanol,

collected, and dried to aﬀord 16 as a pale-yellow solid in 65% yield;

mp (methanol) 152−155 °C. ¹H NMR (300 MHz, DMSO-d₆) δ 10.41

(s, 1H), 8.65 (d, J = 9.0 Hz, 1H), 8.47 (d, J = 6.6 Hz, 1H), 8.25 (s,

1H), 7.87−7.77 (m, 2H), 7.34−7.29 (m, 5H), 7.12−6.89 (m, 2H),

6.89−6.67 (m, 2H), 4.64 (s, 2H), 4.57 (s, 2H). ¹³C NMR (75 MHz,

DMSO) δ 163.1, 153.5, 152.7, 150.7, 143.2, 138.2, 138.1, 130.4, 129.2,

128.3, 127.9, 127.3, 126.7, 121.9, 118.8, 114.2, 113.9, 109.6, 102.7,

59.1, 48.1. MS (ESI) m/z 415 [M + H]⁺.

4-(Pyrrolidin-1-ylmethyl)benzonitrile (18). To a mixture of 17

(1.00 g, 7.63 mmol), pyrrolidine (630 μL, 7.63 mmol), and acetic acid

(300 μL), pic-BH₃(0.816 g, 7.63 mmol) was added over 5 min and

the reaction mixture was stirred at 25 °C for 3 h. Thereafter, 10%

acetic acid was added and the solution was stirred at 25 °C for 30 min.

A saturated solution of Na₂CO₃was added, the aqueous layer was

extracted with ethyl acetate, and the combined organic layers were

washed with brine, dried over Na₂SO₄, and evaporated under reduced

pressure. The crude product was puriﬁed by silica gel ﬂash

chromatography (10% water in acetonitrile) to aﬀord 18 as a pale-

yellow oil in 60% yield. ¹H NMR (300 MHz, CDCl₃) δ 7.59 (d, J = 8.2

Hz, 2H), 7.45 (d, J = 8.2 Hz, 2H), 3.66 (s, 2H), 2.61−2.40 (m, 4H),

1.86−1.73 (m, 4H). MS (ESI) m/z 187 [M + H]⁺.

1

mp (methanol) >300 °C. H NMR (300 MHz, DMSO-d₆) δ 11.38

(br, 1H), 8.90 (d, J = 9.1 Hz, 1H), 8.64−8.60 (m, 2H), 8.26 (d, J = 8.5

Hz, 1H), 8.19 (s, 1H), 7.88 (d, J = 9.1 Hz, 1H), 7.78 (s, 1H), 7.66 (d, J

= 8.5 Hz, 1H), 7.38−7.28 (m, 5H), 7.13 (d, J = 7.0 Hz, 1H), 5.23 (s,

2H). ¹³C NMR (75 MHz, DMSO-d₆) δ 160.2, 154.8, 149.9, 149.7,

144.7, 143.6, 140.0, 139.2, 137.5, 129.3, 128.8, 128.4, 128.3, 127.1,

124.1, 122.3, 120.5, 120.1, 117.3, 102.2, 49.6. MS (ESI) m/z 413 [M +

H]⁺.

7-((7-Chloroquinolin-4-yl)amino)-3-ferrocenylquinazolin-4(3H)-

one (15d). A solution of 14d (65 mg, 0.18 mmol) and 4,7-

dichloroquinoline (36 mg, 0.18 mmol) in ethanol (3.0 mL), was

reﬂuxed for 18 h. After cooling to 25 °C, the solvent was removed in

vacuo and the crude puriﬁed by silica gel ﬂash chromatography (5%

methanol in chloroform), aﬀording 15d as a yellow amorphous solid in

1

50% yield. H NMR (300 MHz, CDCl₃) δ 8.69 (d, J = 5.2 Hz, 1H),

8.29 (d, J = 8.7 Hz, 1H), 8.08 (d, J = 2.1 Hz, 1H), 8.05 (s, 1H), 7.92

(d, J = 9.0 Hz, 1H), 7.48 (dd, J = 9.0, 2.1 Hz, 1H), 7.45 (d, J = 2.2 Hz,

1H), 7.38−7.27 (m, 2H), 6.93 (br, 1H), 4.94 (s, 2H), 4.47−4.30 (m,

2H), 4.30−4.15 (m, 7H). MS (ESI) m/z 521 [M + H]⁺.

N-(3-Ethyl-1,2,3,4-tetrahydroquinazolin-7-yl)-7-chloro-4-amino-

quinoline (6a). To a suspension of LiAlH₄(108 mg, 2.86 mmol) in

dry THF (3.0 mL), a solution of 15a (250 mg, 0.71 mmol) in THF

(5.0 mL) was added dropwise and the reaction was heated under reﬂux

for 4 h. The cooled solution was diluted with diethyl ether and washed

with brine, and the organic phase was dried over Na₂SO₄and

concentrated under reduced pressure. The crude was puriﬁed by silica

gel ﬂash chromatography (5% methanol in chloroform) aﬀording 6a in

1

N-((4-(Pyrrolidin-1-ylmethyl)phenyl)methyl)amine (19). To a

solution of 18 (0.65 g, 3.50 mmol) in ethanol (15.0 mL), NiCl₂·6H₂O

(0.83 g, 3.50 mmol) was added. Afterward, NaBH₄(1.036 g, 28 mmol)

was added very cautiously while stirring the solution vigorously. The

reaction was stirred at 25 °C for 12 h and then was ﬁltered through a

pad of Celite. The solvent was evaporated, and the crude dissolved in

water and extracted with chloroform. The combined extracts were

dried over Na₂SO₄and evaporated in vacuo to aﬀord pure 19 in 86%

70% yield; mp (ethyl acetate/n-hexane) 177−179 °C. H NMR (300

MHz, DMSO-d₆) δ 8.86 (s, 1H), 8.41−8.38 (m, 2H), 7.84 (d, J = 2.2

Hz, 1H), 7.51 (dd, J = 9.0, 2.2 Hz, 1H), 6.88−6.82 (m, 2H), 6.50−

6.45 (m, 2H), 6.00 (s, 1H), 3.92 (s, 2H), 3.70 (s, 2H), 2.57−2.42 (m,

2H), 1.07 (t, J = 7.2 Hz, 3H). ¹³C NMR (75 MHz, DMSO-d₆) δ 152.3,

149.7, 149.4, 144.8, 138.9, 134.8, 128.7, 127.7, 125.6, 125.0, 118.6,

116.0, 112.2, 108.8, 102.3, 62.3, 53.0, 46.8, 13.1. MS (ESI) m/z 339

[M + H]⁺.

1

yield as a pale-yellow oil. H NMR (300 MHz, CDCl₃) δ 7.37−7.18

N-(3-Propyl-1,2,3,4-tetrahydroquinazolin-7-yl)-7-chloro-4-ami-

noquinoline (6b). Starting from 15b (250 mg, 0.68 mmol), the title

compound was prepared following the procedure reported for 6a in

50% yield as a pale-yellow solid; mp (ethyl acetate/n-hexane) 182−

184 °C. ¹H NMR (300 MHz, CDCl₃) δ 8.52 (d, J = 5.3 Hz, 1H), 8.01

(d, J = 2.0 Hz, 1H), 7.82 (d, J = 9.0 Hz, 1H), 7.43 (dd, J = 9.0, 2.0 Hz,

1H), 7.06−6.83 (m, 2H) 6.60 (dd, J = 8.0, 2.0 Hz, 1H), 6.48 (s, 2H),

4.08 (s, 2H), 3.84 (s, 2H), 2.64−2.43 (m, 2H), 1.74−1.47 (m, 2H),

0.97 (t, J = 7.4 Hz, 3H). MS (ESI) m/z 353 [M + H]⁺. HRMS (ESI)

calcd for C₂₀H₂₂ClN₄[M + H]⁺353.1533, found 353.1411.

N-(3-Benzyl-1,2,3,4-tetrahydroquinazolin-7-yl)-7-chloro-4-ami-

noquinoline (6c). To a suspension of LiAlH₄(110 mg, 2.91 mmol) in

dry THF (3.0 mL) cooled at 0 °C, AlCl₃(130 mg, 0.97 mmol) in dry

THF (2.0 mL) was added dropwise. The mixture was stirred at 0 °C

for 1 h, and then a solution of 15c (200 mg, 0.48 mmol) in dry THF

(3.0 mL) was added and the reaction was stirred at 25 °C for 1 h. After

cooling at 0 °C, the reaction mixture was diluted with ethyl acetate and

quenched with water. The aqueous layer was extracted with ethyl

acetate, the organic phase was dried over Na₂SO₄and evaporated in

vacuo. The crude was puriﬁed by silica gel ﬂash chromatography (50%

ethyl acetate in n-hexane) to aﬀord 6c as a pale-yellow solid in 20%

yield; mp (ethyl acetate/n-hexane) 165−166 °C. ¹H NMR (300 MHz,

CDCl₃) δ 8.50 (d, J = 5.3 Hz, 1H), 7.99 (d, J = 2.1 Hz, 1H), 7.84 (d, J

= 9.0 Hz, 1H), 7.47−7.28 (m, 6H), 6.95 (d, J = 5.3 Hz, 1H), 6.90 (d, J

= 8.0 Hz, 1H), 6.77 (br, 1H), 6.59 (d, J = 8.0, 1H), 6.48 (s, 1H), 4.08

(s, 2H), 3.86 (s, 2H), 3.76 (s, 2H). ¹³C NMR (75 MHz, CD₃OD) δ

151.1, 150.4, 149.0, 144.4, 139.0, 137.6, 135.5, 129.5, 128.3, 127.4,

126.5, 125.3, 123.5, 118.3, 116.0, 113.1, 109.5, 101.6, 62.2, 57.0, 53.0.

MS (ESI) m/z 401 [M + H]⁺. HRMS (ESI) calcd for C₂₄H₂₂ClN₄[M

+ H]⁺401.1533, found 401.1553.

(m, 4H), 3.86−3.76 (m, 2H), 3.61 (s, 2H), 2.52 (s, 4H), 1.88−1.74

(m, 4H). MS (ESI) m/z 191 [M + H]⁺.

7-Nitro-3-(4-(pyrrolidin-1-ylmethyl)benzyl)quinazolin-4(3H)-one

(20). To a suspension of 11 (1.03 g, 5.65 mmol) in dry

dichloromethane (20.0 mL) cooled at 0 °C, EDC (1.19 g, 6.22

mmol), HOBt (0.84, 6.22 mmol), and Et₃N (865 μL, 6.22 mmol)

were added and the mixture was stirred at 0 °C for 1 h. Successively 19

(1.07 g, 5.65 mmol) was added and the reaction was stirred at rt for 12

h. Thereafter, the organic mixture was washed with saturated aqueous

NaHCO₃, dried over Na₂SO₄, and evaporated to dryness. The residue

was puriﬁed by silica gel ﬂash chromatography (5% methanol in

chloroform) to aﬀord N-(4-(pyrrolidin-1-ylmethyl)benzyl)-2-amino-4-

1

nitrobenzamide intermediate in 70% yield as yellow−orange oil. H

NMR (300 MHz, CDCl₃) δ 7.50−7.37 (m, 2H), 7.37−7.20 (m, 5H),

6.66 (br, 1H), 5.86 (br, 2H), 4.55 (d, J = 5.6 Hz, 2H), 3.60 (s, 2H),

2.50 (s, 4H), 1.76 (s, 4H). MS (ESI) m/z 355 [M + H]⁺. The above

compound (1.29 g, 3.64 mmol) and formic acid (1.5 mL, 38 mmol)

were heated under reﬂux for 3 h. After cooling to 25 °C, the mixture

was poured into water, and the solution was alkalinized with diluted

NaOH and extracted with chloroform. The solvent was dried over

Na₂SO₄and evaporated under reduced pressure to aﬀord pure 20 in

91% yield as pale-yellow amorphous solid. ¹H NMR (300 MHz,

CDCl₃) δ 8.53 (d, J = 2.1 Hz, 1H), 8.47 (d, J = 8.8 Hz, 1H), 8.25 (dd,

J = 8.8, 2.1 Hz, 1H), 8.20 (s, 1H), 7.39−7.27 (m, 4H), 5.19 (s, 2H),

3.59 (s, 2H), 2.47 (s, 4H), 1.75 (s, 4H). MS (ESI) m/z 365 [M + H]⁺.

7-((7-Chloroquinolin-4-yl)amino)-3-(4-(pyrrolidin-1-ylmethyl)-

benzyl)quinazolin-4(3H)-one (21). To a solution of 20 (100 mg, 0.27

mmol) in 75% aqueous ethanol (3.0 mL), iron (99 mg, 1.78 mmol)

and CaCl₂(40 mg, 0.27 mmol) were added and the reaction was

heated under reﬂux for 2 h. After cooling to 25 °C, the mixture was

L

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

ﬁltered through a pad of Celite and the solvent was evaporated, and

the resulting residue was suspended in water and the aqueous solution

extracted with ethyl acetate. The crude was puriﬁed by silica gel ﬂash

chromatography (5% methanol in chloroform), aﬀording 7-amino-3-

(4-(pyrrolidin-1-ylmethyl)benzyl)quinazolin-4(3H)-one as a pale-

yellow solid in 60% yield; mp (ethyl acetate/n-hexane) 193−195 °C.

¹H NMR (300 MHz, CDCl₃) δ 8.10 (d, J = 9.1 Hz, 1H), 7.99 (s, 1H),

7.46 (d, J = 7.9 Hz, 2H), 7.34 (d, J = 7.9 Hz, 2H), 6.85−6.74 (m, 2H),

5.14 (s, 2H), 4.24 (s, 2H), 3.84 (s, 2H), 2.80 (br, 4H), 1.92 (br, 4H).

MS (ESI) m/z 335 [M + H]⁺. To a solution of the above compound

(130 mg, 0.39 mmol) in ethanol (3.0 mL), 4,7-dichloroquinoline (77

mg, 0.39 mmol) and pyridinium chloride (49 mg, 0.43 mmol) were

added. The reaction was heated under reﬂux for 18 h. Thereafter, the

solvent was removed and the resulting residue treated with 1 N

NaHCO₃and extracted with chloroform. The crude was then puriﬁed

by alumina column chromatography (chloroform), aﬀording 21 as

white solid in 78% yield; mp (n-hexane) 205−207 °C. ¹H NMR (300

MHz, CDCl₃) δ 8.70 (d, J = 5.2 Hz, 1H), 8.30 (d, J = 8.7 Hz, 1H),

8.08 (s, 2H), 7.93 (d, J = 9.0 Hz, 1H), 7.55−7.43 (m, 2H), 7.39−7.27

(m, 5H), 6.98 (s, 1H), 5.17 (s, 2H), 3.60 (s, 2H), 2.49 (br, 4H), 1.77

(br, 4H). ¹³C NMR (75 MHz, CDCl₃) δ 160.6, 152.2, 150.2, 150.1,

147.6, 146.4, 145.4, 135.9, 134.5, 129.8, 129.5, 129.1, 128.2, 127.1,

121.9, 119.7, 119.6, 117.5, 115.5, 106.3, 60.4, 54.4, 49.5, 23.6. MS

(ESI) m/z 496 [M + H]⁺.

24 h at 25 °C under orbital shaking to achieve “pseudo-

thermodynamic equilibrium” and to presaturate the membrane ﬁlter.

Product suspensions/solutions were then ﬁltered using centrifugation,

diluted 1:2 with the same buﬀer solution, and analyzed by UPLC/UV/

TOF-MS, using UV detection at 254 nm for quantitation. Solubility

was calculated by comparing the sample and standard UV areas: S =

(A_smp× FD × C_st)/A_st, where S was the solubility of the compound

(μM), A_smpwas the UV area of the sample solution, FD was the

dilution factor (2), C_stwas the standard concentration (250 μM), and

A_stwas the UV area of the standard solution.

Metabolic Stability Assay. Compounds in 10 mM DMSO solution

were added to an incubation mixture in a 96-well microplate

containing 20 pmol/mL of hCYP3A4 (0.1−0.2 mg/mL protein).

The mixture was split in two aliquots: one receiving a NADPH

regenerating system, the other an equal amount of phosphate buﬀer.

The ﬁnal substrate concentration was 1 μM along with 0.25% of

organic solvent. Incubation proceeded for 1 h at 37 °C and was

stopped by addition of acetonitrile to precipitate proteins. Metabolic

stability was given as the percent remaining following incubation with

cofactor (NADPH) with reference to the incubation mixture without

NADPH: % remaining = Area NADPH × 100/Area ctrl, where Area

ctrl was the MS peak area of the sample solution without NADPH and

Area NADPH was the MS area of the sample solution with NADPH.

The % CV obtained was typically within 10%.

In Vitro Intrinsic Clearance (Human, Rat and Mouse Liver

Microsomes). Test compound (6b) was incubated separately at 1 μM

concentration in 100 mM phosphate buﬀer (pH 7.4) with 0.2 mg/mL

rat and mouse hepatic microsomal protein or 0.4 mg/mL for human

liver microsomes. (Pooled human liver; BD Biosciences Woburn, MA,

USA). The enzymatic reaction was started by addition of a NADPH

regenerating system (ﬁnal concentrations: 1 mM β-nicotinamide

adenine dinucleotide phosphate reduced (NADPH) + 10 mM glucose-

6-phosphate (G6P) + 0.4 U/mL glucose-6-phosphate dehydrogenase

(G6PDH) as previously reported.⁶³Reactions were stopped at regular

time intervals (0−45 min) by adding an equal volume of acetonitrile.

All incubations were performed in duplicate. Verapamil as positive

control for the assay was incubated in parallel under the same

conditions. Samples were analyzed by HPLC-MS. Quantitative data

were automatically produced using the OpenLynx software. Calcu-

lations Substrate depletion data (peak area at diﬀerent time points)

were ﬁtted to a monoexponential decay model under the assumption

that the concentration of 1 μM was far below the K_mof the test

compound, the intrinsic clearance (Cl_int) was calculated by eq 1

Measurements of CQ Transport in X. laevis Oocytes

Expressing PfCRT. Expression of wild-type and CQ resistance-

conferring forms of PfCRT (from the strains D10 and Dd2,

respectively) at the plasma membrane of X. laevis oocytes was

achieved as described previously.³¹Brieﬂy, oocytes were injected with

cRNA encoding PfCRT (20 ng per oocyte) and the uptake of [³H]CQ

(0.3 μM; 20 Ci/mmol; ARC) was measured 3−5 days postinjection as

detailed elsewhere.³¹Uptake measurements were made over 1.5−2 h

at 27.5 °C in medium that contained 96 mM NaCl, 2 mM KCl, 1 mM

MgCl₂, 1.8 mM CaCl₂, 10 mM MES, 10 mM Tris-base (pH 6.0), and

15 μM unlabeled CQ. Statistical comparisons were made with the

Student’s t-test for unpaired samples or with paired ANOVA in

conjunction with Tukey’s multiple comparisons test.

Parasite Growth and Drug Susceptibility Assay. P. falciparum

D10 and W2 Strains. The CQS (D10) and the CQR (W2) strains of

P. falciparum were sustained in vitro as described by Trager and

Jensen.⁶²Parasites were maintained at 5% hematocrit (human type A-

positive red blood cells) in RPMI 1640 (EuroClone, Celbio) medium

with the addition of 1% AlbuMax (Invitrogen, Milan, Italy), 0.01%

hypoxanthine, 20 mM Hepes, and 2 mM glutamine. All cultures were

maintained at 37 °C in a standard gas mixture consisting of 1% O₂, 5%

CO₂, and 94% N₂. Compounds were dissolved in DMSO and then

diluted with medium to achieve the required concentrations (ﬁnal

DMSO concentration <1%, which is nontoxic to the parasite). Drugs

were placed in 96-well ﬂat-bottom microplates (COSTAR) and serial

dilutions made. Asynchronous cultures with parasitemia of 1−1.5%

and 1% ﬁnal hematocrit were aliquoted into the plates and incubated

for 72 h at 37 °C. Parasite growth was determined spectrophoto-

metrically (OD₆₅₀) by measuring the activity of the parasite lactate

dehydrogenase (pLDH), according to a modiﬁed version of Makler’s

method in control and drug-treated cultures. Antiplasmodial activity is

expressed as the 50% inhibitory concentrations (IC₅₀), and each IC₅₀

Cl_int= k(min −1) × [V]/[P]

(1)

where k is the rate costant for the depletion of substrate, V is the

volume of incubation in μL, and P is the amount of microsomal

proteins in the incubation in mg according to Baranczewski et al.⁶⁴

Compounds were deﬁned as low, medium, or highly metabolized

based on the in vitro CL_intvalues: < 3.4 low, 3.4−92.4 medium, > 92.4

high.

Toxicity Evaluation. Materials. Dulbecco’s Modiﬁed Eagle’s

Medium (DMEM) and Eagle’s Minimum Essential Medium

(EMEM), trypsin solution, and all the solvents used for cell culture

were purchased from Lonza (Switzerland). Mouse immortalized

ﬁbroblasts NIH3T3 were purchased from American Type Culture

Collection (USA) and human caucasian hepatocyte carcinoma cells

HepG2 from The European Collection of Cell Cultures (UK). The

mutagenicity assay was supplied by Biologik srl (Trieste, Italy).

Cell Cultures and Cytotoxicity Assay. The NIH3T3 and HepG2

cell lines were used in the cytotoxicity experiments. NIH3T3 cells were

maintained in DMEM and HepG2 cells in EMEM at 37 °C in a

humidiﬁed atmosphere containing 5% CO₂. The culture media were

supplemented with 10% fetal calf serum (FCS), 1% ^L-glutamine−

penicillin−streptomycin solution, and 1% MEM Non-Essential Amino

Acid Solution. Once at conﬂuence, cells were washed with PBS 0.1 M,

taken up with trypsin−EDTA solution, and then centrifuged at 1000

rpm for 5 min. The pellet was resuspended in medium solution

(dilution 1:15). Cell viability after 24 h of incubation with the diﬀerent

compounds was evaluated by Neutral Red Uptake (Sigma-Aldrich,

value is the mean

standard deviation of at least three separate

experiments performed in duplicate.

Solubility Assay. Standard and sample solutions were prepared

from a 10 mM DMSO stock solution using an automated dilution

procedure. For each compound, three solutions were prepared: one to

be used as standard and the other two as test solutions. Standard: 250

μM standard solution in acetonitrile/buﬀer, with a ﬁnal DMSO

content of 2.5% (v/v). Test sample for pH 3.0: 250 μM sample

solution in acetic acid 50 mM, pH 3, with a ﬁnal DMSO content of

2.5% (v/v). Test sample for pH 7.4: a 250 μM sample solution in

ammonium acetate buﬀer 50 mM, pH 7.4, with a ﬁnal DMSO content

of 2.5% (v/v). The 250 μM product suspensions/solutions in the

aqueous buﬀers were prepared directly in Millipore MultiScreen-96

ﬁlter plates (0.4 μm PTCE membrane) and sealed. Plates were left for

M

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Switzerland) by the procedure previously reported.⁶⁵The data

processing included the Student’s t test with p < 0.05 taken as

signiﬁcance level.

Instruments, Union City, CA) and OriginPro 8.1 (OriginLab

Corporation, Northampton, MA).

Na⁺,K⁺-ATPase assay. Isolation of Na⁺,K⁺-ATPase. Membrane

fragments containing Na⁺,K⁺-ATPase were obtained by extraction

from the outer medulla of rabbit kidneys using procedure C of

Jørgensen.⁶⁷Protein concentration was determined by the Lowry

method⁶⁸using bovine serum albumin as a standard. The total protein

content of membrane fragments was usually comprised between 1.5

and 2.5 mg/mL and speciﬁc ATPase activity was in the range of

1700−2100 μmol P_i/(h·mg protein) at 37 °C.

Mutagenicity Assay: Ames Test. The TA100 and TA98 strains of

Salmonella typhimurium were utilized for mutagenicity assay.

Approximately 10⁷bacteria were exposed to six concentrations of

each test compound in the presence or in the absence of rat liver S9

fraction, as well as a positive and a negative control, for 90 min in

medium containing suﬃcient histidine to support approximately two

cell divisions. After 90 min, the exposure cultures were diluted in pH

indicator medium lacking histidine and aliquoted into 48 wells of a

384-well plate. Within two days, cells which had undergone the

reversion to His grew into colonies. Metabolism by the bacterial

colonies reduced the pH of the medium, changing the color of that

well. This color change can be detected visually or by microplate

reader. The number of wells containing revertant colonies were

counted for each dose and compared to a zero dose control. Each dose

was tested six times.

In Vivo Antiplasmodial Activity Studies. The in vivo evaluation

of therapeutic activity was performed on compounds 5a and 6b,c

according to the standard Peters 4-day test, using the rodent malaria

parasite Plasmodium berghei ANKA (CQ-sensitive) and, as hosts,

female, 5-week-old BALB/c mice, weighing 19 2 g each. For each

compound, groups of each mice each were intraperitoneally inoculated

with 2 × 10⁷P. berghei-infected erythrocytes. Freshly prepared aqueous

stock solutions (100 mg/mL) of compounds 6b,c were serially diluted

1.5-fold to obtain 6−7 escalating doses from 4.4 to 50 mg/kg. Each

experimental group was administered daily with one dose through oral

gavages of 200 μL each. Compound 5a was administered intra-

peritoneally at 50 mg/kg to a single group of six mice. Treatments

started 4 h after the infection and were repeated until day 4. Twenty-

four hours after the last treatment, % parasitemia was counted on

Giemsa smears taken from tail blood, and the mice were sacriﬁced by

CO₂asphyxiation. The compounds’ ED₅₀mg/kg was determined by

nonlinear regression (four-parameter logistic curve ﬁt), using

SigmaPlot v.10.

hERG Assay. mRNA in Vitro Transcription. The hERG clone was

propagated in the E. coli strains TOP10. In vitro transcription was

performed with the “mMessage mMachine” kit (Ambion, Paisley, UK)

with T7 RNA polymerase using a EcoRI restriction enzyme

recognition site for the linearization. To record macroscopic current,

cRNA encoding for the channel was injected in Xenopus laevis oocytes

(0.2 mg/mL), and K⁺currents were recorded with the patch-clamp

technique 2−3 days after injection.

Oocyte Preparation and Injection. Oocytes were dissected from

anesthetized Xenopus leavis (African clawed frogs) by surgical harvest

of oocytes. Oocytes were treated with collagenase (2 mg/mL) for 1 h,

rinsed with a Ca²⁺-free OR-2 solution (NaCl 82.5 mM, KCl 2.5 mM,

MgCl₂1 mM, Hepes-NaOH 5 mM, pH 7.6) and stored in SOS

solution (NaCl 96 mM, KCl 2 mM, CaCl₂1.8 mM, MgCl₂1 mM,

Hepes-NaOH 5 mM, pH 7.6, fresh penicillin 100 units/mL,

streptomycin 100 μg/mL, gentamycin 50 μg/mL).

Measurements of ATPase Activity. ATPase activity was determined

by measuring production of P_iby the malachite green colorimetric

method.⁶⁹The reaction mixture included 25 mM MOPS, pH 7.0, 100

mM NaCl, 20 mM KCl, and 3 mM MgCl₂. Membrane fragments

containing the Na⁺,K⁺-ATPase were added to yield approximately 1

μg/mL protein concentration. The temperature was maintained at 37

°C, and the reaction was started by addition of 1 mM ATP. Samples

were then taken at serial times for P_idetermination. Background

activity was determined in the absence of NaCl.

Computational Details. Preparation of Molecules. Three-

dimensional structure building was carried out on Cooler Master

Centurion 5 (Intel Core2 Quad CPU Q6600 @ 2.40 GHz) with

Ubuntu 10.04 LTS (long-term support) operating system running

Maestro 9.2 (Schrodinger, LLC, New York, NY, 2011).³⁵Molecular

̈

energy minimizations were performed in MacroModel⁷⁰using the

Optimized Potentials for Liquid Simulations-all atom (OPLS-AA)

force ﬁeld 2005.^71,72The solvent eﬀects are simulated using the

analytical Generalized-Born/Surface-Area (GB/SA) model,⁷³and no

cutoﬀ for nonbonded interactions was selected. Polak-Ribiere

conjugate gradient (PRCG)⁷⁴method with 5000 maximum iterations

and 0.001 gradient convergence threshold was employed. The

conformational searches were carried out by application of the

MCMM (Monte Carlo Multiple Minimum) torsional sampling

method, performing automatic setup with 21 kJ/mol (5.02 kcal/

mol) in the energy window for saving structure and a 0.5 Å cutoﬀ

distance for redundant conformers. All compounds reported in this

paper were treated by LigPrep application (version 2.5, Schrodinger,

̈

LLC, New York, NY, 2011), implemented in Maestro suite 2011,

generating the most probable ionization state of any possible

enantiomers and tautomers at vacuolar pH values (5.5

0.5) for

compound 3 and at cellular pH (7 2) for all compounds. The

intramolecular H-bonds for selected compounds were retrieved by

measurements application tool implemented in Maestro graphical

interface.

P450 SOM Workﬂow. Sites of metabolism for all molecules

reported in this study were predicted by P450 site of metabolism

(P450 SOM)⁷⁵workﬂow implemented in Schrodinger suite 2011

̈

(Schrodinger, LLC, New York, NY, 2011). For each molecule, P450

̈

SOM performed a calculation of intrinsic reactivity coupled to induced

ﬁt docking (IFD)⁷⁶for the selected isoform of cytochrome. The

reactivity rules have been parametrized in P450 SOM to predict

atomic reactivity proﬁles for promiscuous P450 enzymes that are

thought to be mostly independent of structural restrictions on the

binding poses. The reactivity was predicted with a linear free energy

approach based on the Hammett and Taft scheme where the reactivity

of a given atom is the sum of a baseline reactivity rate and a series of

perturbations determined by the connectivity.⁷⁵The induced-ﬁt

docking approach is a variation on the normal protocol.⁷⁷The initial

sampling was enhanced by generating multiple starting conformations

so that a wider range of poses is found in the initial docking stage. The

initial docking includes van der Waals scaling of the receptor and

alanine mutation of the most ﬂexible residues. In the Prime reﬁnement

stage, any residue with an atom within 5 Å of any ligand pose was

selected for side chain prediction. The subsequent minimization

included the ligand, side chains, and backbones of the ﬂexible residues.

The ligand was then redocked into each of the low-energy protein

conformations, determined by a 40 kcal/mol cutoﬀ. There was no ﬁnal

scoring stage because all poses were considered in determining which

atoms are suﬃciently accessible to the reactive heme iron. Any atom

Electrophysiological Experiments. hERG currents were recorded

from macropatches 2−3 days after injection.⁶⁶The pipet and the bath

solutions had the following composition: KOH 98 mM, MgSO₄1

mM, CaCl₂1.8 mM, HEPES 5 mM, Na-pyruvate 2.5 mM (pH 7.5

with methanesulfonic acid).⁶⁶All drugs were prepared as 10 mM stock

solutions in water and stored at −20 °C. On the day of experiments,

aliquots of the stock solutions were diluted to the desired

concentrations with the bath solution. Patch pipettes were pulled

from glass capillary tubing (Warner G85150T-4) on a programmable

Flaming/Brown type puller (Sutter P-87) and ﬁre polished on a

microforge using the resistive heat from a platinum wire. Patch

pipettes were applied to the surface of freshly devitellinized oocytes to

form electrical seals (>1 GΩ), allowing a voltage-clamp on patches of

plasma membrane. Data were acquired using a Multiclamp 700B

patch-clamp ampliﬁer and a Digidata 1440A acquisition system with

pClamp 10 software (Molecular Devices, Sunnyvale, CA). All

experiments were performed at room temperature, 22−24 °C. Data

analysis was carried out using Clampﬁt 10.2 software (Axon

N

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

within the cutoﬀ distance of 5 Å from the heme iron was considered as

a potential site of metabolism.⁷⁵

In Silico ADME+T Predictions. Biosafety of developed new

molecules was also explored. Properties of new compounds were

studied by applying in silico tool OSIRIS property explorer (http://

www.organic-chemistry.org/prog/peo/ access date 12/03/2012)

released by Actelion Pharmaceutical Ltd.

sequently, one individual trajectory for each complex of 5 ns was

calculated. The trajectory ﬁles were analyzed by Simulation Event

Analysis and Simulation Quality Analysis implemented in the

Desmond package. The same applications were used to generate all

plots concerning MD simulation presented in this study. The distances

between quinazoline and quinoline moieties and key residues on

hERG channel were measured by taking into account the centroids of

each moiety. The movies were generated by Desmond trajectory

viewer and converted to appropriate size by Virtualdub application

licensed under the GNU General Public License (GPL) (http://www.

virtualdub.org/).

Molecular Docking Studies. Homology model of hERG channel

generated by Farid and co-workers⁴⁵was obtained by Schrodinger

̈

Web site. The model was chosen because the open conformations

appear suitable for drug binding. Moreover, the authors used this

model of hERG channel as a docking target for binding mode

prediction of well-characterized hERG blockers, including terfenadine,

sertindole, ibutilide, and cloﬁlium. Their studies yielded converging

binding modes for all mentioned ligands, which added conﬁdence to

the poses obtained.

ASSOCIATED CONTENT

* Supporting Information

■

S

Low energy conformers of isoquine (IQ), amodiaquine (AQ)

and halofantrine, site of metabolism prediction for compound

6b,c, Ames test performed on S. typhimurium TA98 strains for

compounds 5c,a and 6c,b, Ames test performed on S.

typhimurium TA98 and TA100 strains in the presence of the

liver fraction S9 for compounds 5a, 6b, and 6c, superposition

between docked poses of 6b and 6c and their low energy

confomers, superposition between initial docked structure of 6b

and after 5 ns of MD simulation, superposition between initial

docked structure of 6c and after 5 ns of MD simulation, plots

derived from MD simulation, Elemental analysis. Molecular

dynamics simulation movies (AVI). This material is available

free of charge via the Internet at http://pubs.acs.org.

a. Protein Preparation. The three-dimensional structure of the

hERG channel was imported into Schrodinger Maestro molecular

̈

modeling environment,³⁵it was submitted to protein preparation

wizard implemented in Maestro suite 2011 (Protein Preparation

Wizard workﬂow 2011; http://www.schrodinger.com/supportdocs/

18/16). This protocol allowed us to obtain a reasonable starting

structure of protein for molecular docking calculations by a series of

computational steps. In particular, we performed three steps to (1) add

hydrogens, (2) optimize the orientation of hydroxyl groups, Asn, and

Gln, and the protonation state of His, and (3) perform a constrained

reﬁnement with the impref utility, setting the max RMSD of 0.30. The

impref utility consists of a cycles of energy minimization based on the

impact molecular mechanics engine and on the OPLS_2005 force

ﬁeld.^71,72

b. Molecular Docking. Molecular Docking was carried out using

GOLD 5.0.1 (Genetic Optimization for Ligand Docking) software

from Cambridge Crystallographic Data Center, UK, that uses the

Genetic algorithm (GA)⁴³running under Ubuntu 10.04 LTS OS. This

method allows a partial ﬂexibility of protein and full ﬂexibility of

ligand. For each of the 100 independent GA runs, a maximum number

of 125000 GA operations were performed. The search eﬃciency values

were set on 200% in order to increase the ﬂexibility of the ligands

docked. The active site radius of 8 Å was chosen by XYZ coordinates

from the center of the terfenadine docked by IFD into hERG

homology model.⁴⁵Default cutoﬀ values of 2.5 Å (dH-X) for

hydrogen bonds and 4.0 Å for van der Waals distance were employed.

When the top three solutions attained RMSD values within 1.5 Å, GA

docking was terminated. The ﬁtness function GoldScore⁴⁴was

evaluated.

Molecular Dynamics. All molecular dynamics (MD) simulations

were carried out by Desmond 3.0 package^59,60using Maestro

Molecular Modeling environment as graphical interface.³⁵As a ﬁrst

step, the complexes hERG−6b and hERG−6c derived from docking

calculations (Figure 7A and B, respectively) were imported in Maestro

and in the system builder step were placed into POPC membrane and

solvated into an orthorhombic box ﬁlled with water, simulated by

TIP3P model.⁷⁸OPLS_2005 force ﬁeld^71,72was applied for MD

calculations. Na⁺and Cl⁻ions were added to provide a ﬁnal salt

concentration of 0.15 M in order to simulate physiological

concentration of monovalent ions.⁵⁹Constant temperature (300 K),

pressure (1.01325 bar), and surface tension (4000 barÅ) were

employed with NPγT (constant number of particles, pressure, surface

tension, and temperature) as ensemble class. RESPA integrator⁷⁹was

used in order to integrate the equations of motion, with an inner time

step of 2.0 fs for bonded interactions and nonbonded interactions

within the short-range cutoﬀ. Nose−Hoover thermostats⁸⁰were used

to keep the constant simulation temperature, and the Martyna−

Tobias−Klein method⁸¹was applied to control the pressure. Long-

range electrostatic interactions were calculated by particle-mesh Ewald

method (PME).⁸²The cutoﬀ for van der Waals and short-range

electrostatic interactions was set at 9.0 Å. The equilibration of the

system was performed with the default protocol provided in Desmond,

which consists of a series of restrained minimizations and molecular

dynamics simulations utilized to slowly relax the system. Con-

AUTHOR INFORMATION

Corresponding Author

*Phone: +390577234172. Fax: +390577234333. E-mail:

campiani@unisi.it.

■

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

Authors thank MIUR for ﬁnancial support, Prof. Annarosa

Arcangeli for providing hERG cDNA subcloned into the

pcDNA3.1 vector, and Siena Biotech SpA for evaluation of

selected drug-like properties. This work was also supported by

the Australian National Health and Medical Research Council

(NHMRC) (grant 1007035). R.E.M. was supported by an

NHMRC Australian Biomedical Fellowship (fellowship

520320).

ABBREVIATIONS USED

■

WHO, World Health Organization; CQ, chloroquine; CQR,

chloroquine-resistant; AQ, amodiaquine; IQ, isoquine; DV,

digestive vacuole; CQS, chloroquine-sensitive; pLDH, parasite

lactate dehydrogenase; PfCRT, Plasmodium falciparum chlor-

oquine-resistance transporter; IFD, induced-ﬁt docking; hERG,

ether-a-go-go related gene; DMEM, Dulbecco’s Modiﬁed

Eagle’s Medium; EMEM, Eagle’s Minimum Essential Meidum;

FCS, fetal calf serum

REFERENCES

■

(1) WHO World Malaria Report 2011; World Health Organization:

Geneva, 2011; http://www.who.int/malaria/world_malaria_report_

2011/en/.

(2) Murray, C. J.; Rosenfeld, L. C.; Lim, S. S.; Andrews, K. G.;

Foreman, K. J.; Haring, D.; Fullman, N.; Naghavi, M.; Lozano, R.;

Lopez, A. D. Global malaria mortality between 1980 and 2010: a

systematic analysis. Lancet 2012, 379, 413−431.

O

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(3) Watkins, W. M.; Sixsmith, D. G.; Spencer, H. C.; Boriga, D. A.;

Kariuki, D. M.; Kipingor, T.; Koech, D. K. Effectiveness of

amodiaquine as treatment for chloroquine-resistant Plasmodium

falciparum infections in Kenya. Lancet 1984, 1, 357−359.

(4) Olliaro, P.; Nevill, C.; LeBras, J.; Ringwald, P.; Mussano, P.;

Garner, P.; Brasseur, P. Systematic review of amodiaquine treatment in

uncomplicated malaria. Lancet 1996, 348, 1196−1201.

(5) Anvikar, A. R.; Sharma, B.; Sharma, S. K.; Ghosh, S. K.; Bhatt, R.

M.; Kumar, A.; Mohanty, S. S.; Pillai, C. R.; Dash, A. P.; Valecha, N. In

vitro assessment of drug resistance in Plasmodium falciparum in five

States of India. Indian J. Med. Res. 2012, 135, 494−499.

(6) Neftel, K. A.; Woodtly, W.; Schmid, M.; Frick, P. G.; Fehr, J.

Amodiaquine induced agranulocytosis and liver damage. Br. Med. J.

(Clin. Res. Ed.) 1986, 292, 721−723.

(21) O’Neill, P. M.; Mukhtar, A.; Stocks, P. A.; Randle, L. E.;

Hindley, S.; Ward, S. A.; Storr, R. C.; Bickley, J. F.; O’Neil, I. A.;

Maggs, J. L.; Hughes, R. H.; Winstanley, P. A.; Bray, P. G.; Park, B. K.

Isoquine and related amodiaquine analogues: a new generation of

improved 4-aminoquinoline antimalarials. J. Med. Chem. 2003, 46,

4933−4945.

(22) Lawrence, R. M.; Dennis, K. C.; O’Neill, P. M.; Hahn, D. U.;

Roeder, M.; Struppe, C. Development of a Scalable Synthetic Route to

GSK369796 (N-tert-Butyl Isoquine), a Novel 4-Aminoquinoline

Antimalarial Drug. Org. Process Res. Dev. 2008, 12, 294−297.

(23) O’Neill, P. M.; Park, B. K.; Shone, A. E.; Maggs, J. L.; Roberts,

P.; Stocks, P. A.; Biagini, G. A.; Bray, P. G.; Gibbons, P.; Berry, N.;

Winstanley, P. A.; Mukhtar, A.; Bonar-Law, R.; Hindley, S.; Bambal, R.

B.; Davis, C. B.; Bates, M.; Hart, T. K.; Gresham, S. L.; Lawrence, R.

M.; Brigandi, R. A.; Gomez-delas-Heras, F. M.; Gargallo, D. V.; Ward,

S. A. Candidate selection and preclinical evaluation of N-tert-butyl

isoquine (GSK369796), an affordable and effective 4-aminoquinoline

antimalarial for the 21st century. J. Med. Chem. 2009, 52, 1408−1415.

(24) Omura, Y.; Taruno, Y.; Irisa, Y.; Morimoto, M.; Saimoto, H.;

Shigemasa, Y. Regioselective Mannich reaction of phenolic com-

pounds and its application to the synthesis of new chitosan derivatives.

Tetrahedron Lett. 2001, 42, 7273−7275.

(25) Zhao, H.; Fu, H.; Qiao, R. Copper-catalyzed direct amination of

ortho-functionalized haloarenes with sodium azide as the amino

source. J. Org. Chem. 2010, 750, 3311−3316.

(26) Biot, C.; Glorian, G.; Maciejewski, L. A.; Brocard, J. S. Synthesis

and antimalarial activity in vitro and in vivo of a new ferrocene−

chloroquine analogue. J. Med. Chem. 1997, 40, 3715−3718.

(27) Sato, S.; Sakamoto, T.; Miyazawa, E.; Kikugawa, Y. One-pot

reductive amination of aldehydes and ketones with alpha-picoline-

borane in methanol, in water, and in neat conditions. Tetrahedron

2004, 60, 7899−7906.

(7) Hatton, C. S.; Peto, T. E.; Bunch, C.; Pasvol, G.; Russell, S. J.;

Singer, C. R.; Edwards, G.; Winstanley, P. Frequency of severe

neutropenia associated with amodiaquine prophylaxis against malaria.

Lancet 1986, 1, 411−414.

(8) Thomas, F.; Erhart, A.; D’Alessandro, U. Can amodiaquine be

used safely during pregnancy? Lancet Infect. Dis. 2004, 4, 235−239.

(9) Maggs, J. L.; Tingle, M. D.; Kitteringham, N. R.; Park, B. K.

Drug−protein conjugates. XIV. Mechanisms of formation of protein-

arylating intermediates from amodiaquine, a myelotoxin and

hepatotoxin in man. Biochem. Pharmacol. 1988, 37, 303−311.

(10) Harrison, A. C.; Kitteringham, N. R.; Clarke, J. B.; Park, B. K.

The mechanism of bioactivation and antigen formation of

amodiaquine in the rat. Biochem. Pharmacol. 1992, 43, 1421−1430.

(11) Werbel, L. M.; Cook, P. D.; Elslager, E. F.; Hung, J. H.; Johnson,

J. L.; Kesten, S. J.; Mcnamara, D. J.; Ortwine, D. F.; Worth, D. F.

Antimalarial Drugs 0.60. Synthesis, Antimalarial Activity, and

Quantitative Structure−Activity Relationships of Tebuquine and a

Series of Related 5-[(7-Chloro-4-quinolinyl)amino]-3-[(alkylamino)-

methyl][1,1′-biphenyl]-2-ols and N-omega-Oxides1,2. J. Med. Chem.

1986, 29, 924−939.

(12) 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.

(13) O’Neill, P. M.; Harrison, A. C.; Storr, R. C.; Hawley, S. R.;

Ward, S. A.; Park, B. K. The Effect of Fluorine Substitution on the

Metabolism and Antimalarial Activity of Amodiaquine. J. Med. Chem.

1994, 37, 1362−1370.

(14) Raynes, K. J.; Stocks, P. A.; O’Neill, P. M.; Park, B. K.; Ward, S.

A. New 4-aminoquinoline mannich base antimalarials. 1. Effect of an

alkyl substituent in the 5 ′-position of the 4 ′-hydroxyanilino side

chain. J. Med. Chem. 1999, 42, 2747−2751.

(15) 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.

(16) 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.

(17) 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.

(28) Khurana, J. M.; Kukreja, G. Rapid reduction of nitriles to

primary amines with nickel boride at ambient temperature. Synth.

Commun. 2002, 32, 1265−1269.

(29) Gemma, S.; Kunjir, S.; Coccone, S. S.; Brindisi, M.; Moretti, V.;

Brogi, S.; Novellino, E.; Basilico, N.; Parapini, S.; Taramelli, D.;

Campiani, G.; Butini, S. Synthesis and antiplasmodial activity of

bicyclic dioxanes as simplified dihydroplakortin analogues. J. Med.

Chem. 2011, 54, 5949−5953.

(30) Summers, R. L.; Martin, R. E. Functional characteristics of the

malaria parasite’s “chloroquine resistance transporter”: implications for

chemotherapy. Virulence 2010, 1, 304−308.

(31) Martin, R. E.; Marchetti, R. V.; Cowan, A. I.; Howitt, S. M.;

Broer, S.; Kirk, K. Chloroquine transport via the malaria parasite’s

chloroquine resistance transporter. Science 2009, 325, 1680−1682.

(32) Martin, R. E.; Butterworth, A. S.; Gardiner, D. L.; Kirk, K.;

McCarthy, J. S.; Skinner-Adams, T. S. Saquinavir inhibits the malaria

parasite’s chloroquine resistance transporter. Antimicrob. Agents

Chemother. 2012, 56, 2283−2289.

(33) Dive, D.; Biot, C. Ferrocene conjugates of chloroquine and

other antimalarials: the development of ferroquine, a new antimalarial.

ChemMedChem 2008, 3, 383−391.

(34) P450. Site of Metabolism Prediction, version 1.1; Schrodinger,

̈

LLC: New York, 2011.

(35) Maestro, version 9.2; Schrodinger, LLC: New York, 2011.

̈

(36) Gartiser, S.; Hafner, C.; Kronenberger-Schafer, K.; Happel, O.;

Trautwein, C.; Kummerer, K. Approach for detecting mutagenicity of

biodegraded and ozonated pharmaceuticals, metabolites and trans-

formation products from a drinking water perspective. Environ. Sci.

Pollut. Res. Int. 2012, 19, 3597−3609.

(18) Burckhalter, J. H.; Tendick, F. H.; et al. Aminoalkylphenols as

antimalarials (heterocyclicamino)-alpha-amino-o-cresols; the synthesis

of camoquin. J. Am. Chem. Soc. 1948, 70, 1363−1373.

(19) O’Neill, P. M.; Ward, S. A.; Berry, N. G.; Jeyadevan, J. P.;

Biagini, G. A.; Asadollaly, E.; Park, B. K.; Bray, P. G. A medicinal

chemistry perspective on 4-aminoquinoline antimalarial drugs. Curr.

Top. Med. Chem. 2006, 6, 479−507.

(20) Biot, C.; Taramelli, D.; Forfar-Bares, I.; Maciejewski, L. A.;

Boyce, M.; Nowogrocki, G.; Brocard, J. S.; Basilico, N.; Olliaro, P.;

Egan, T. J. Insights into the mechanism of action of ferroquine.

Relationship between physicochemical properties and antiplasmodial

activity. Mol. Pharmaceutics 2005, 2, 185−193.

(37) Nunes, L. G.; Gontijo, D. C.; Souza, C. J.; Fietto, L. G.;

Carvalho, A. F.; Leite, J. P. The mutagenic, DNA-damaging and

antioxidative properties of bark and leaf extracts from Coutarea

hexandra (Jacq.) K. Schum. Environ. Toxicol. Pharmacol. 2012, 33,

297−303.

(38) Peters, W. Drug resistance in Plasmodium berghei Vincke and

Lips, 1948. I. Chloroquine resistance. Exp. Parasitol. 1965, 17, 80−89.

P

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(39) Bhattacharyya, D.; Sen, P. C. The effect of binding of

chlorpromazine and chloroquine to ion transporting ATPases. Mol.

Cell. Biochem. 1999, 198, 179−185.

(40) Mazumder, B.; Mukherjee, S.; NagDas, S. K.; Sen, P. C. The

interaction of chloroquine with transport ATPase and acetylcholine

esterase in microsomal membranes of rat in vitro and in vivo. Biochem.

Int. 1988, 16, 35−44.

(41) Kongsamut, S.; Kang, J.; Chen, X. L.; Roehr, J.; Rampe, D. A

comparison of the receptor binding and HERG channel affinities for a

series of antipsychotic drugs. Eur. J. Pharmacol. 2002, 450, 37−41.

(42) Kaplan, J. H. Biochemistry of Na,K-ATPase. Annu. Rev. Biochem.

2002, 71, 511−535.

(43) Jones, G.; Willett, P.; Glen, R. C. Molecular recognition of

receptor sites using a genetic algorithm with a description of

desolvation. J. Mol. Biol. 1995, 245, 43−53.

(44) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.

Development and validation of a genetic algorithm for flexible docking.

J. Mol. Biol. 1997, 267, 727−748.

(45) Farid, R.; Day, T.; Friesner, R. A.; Pearlstein, R. A. New insights

about HERG blockade obtained from protein modeling, potential

energy mapping, and docking studies. Bioorg. Med. Chem. 2006, 14,

3160−3173.

(46) Induced Fit: A novel method for fast and accurate prediction of

ligand induced conformational changes in receptor active sites ;

(58) The PyMOL Molecular Graphics System, version 1.5.0.1;

Schrodinger LLC: New York, 2012.

(59) Desmond Molecular Dynamics System, version 3.0; D. E. Shaw

Research, New York, 2011; Maestro-Desmond Interoperability Tools ,

version 3.0; Schrodinger LLC: New York, 2011.

(60) Bowers, K. J.; Chow, E.; Xu, H.; Dror, R. O.; Eastwood, M. P.;

Gregersen, B. A.; Klepeis, J. L.; Kolossvary, I.; Moraes, M. A.;

Sacerdoti, F. D.; Salmon, J. K.; Shan, Y.; Shaw, D. E. Scalable

algorithms for molecular dynamics simulations on commodity clusters.

In SC ’06: Proceedings of the 2006 ACM/IEEE Conference on

Supercomputing, Tampa, FL, November 11-17, 2006.

(61) Garg, V.; Stary-Weinzinger, A.; Sachse, F.; Sanguinetti, M. C.

Molecular determinants for activation of human ether-a-go-go-related

gene 1 potassium channels by 3-nitro-N-(4-phenoxyphenyl) benza-

mide. Mol. Pharmacol. 2011, 80, 630−637.

(62) Trager, W.; Jensen, J. B. Human malaria parasites in continuous

culture. Science 1976, 193, 673−675.

(63) D’Elia, P.; De Matteis, F.; Dragoni, S.; Shah, A.; Sgaragli, G.;

Valoti, M. DP7, a novel dihydropyridine multidrug resistance reverter,

shows only weak inhibitory activity on human CYP3A enzyme(s). Eur.

J. Pharmacol. 2009, 614, 7−13.

(64) Baranczewski, P.; Stanczak, A.; Sundberg, K.; Svensson, R.;

Wallin, A.; Jansson, J.; Garberg, P.; Postlind, H. Introduction to in

vitro estimation of metabolic stability and drug interactions of new

chemical entities in drug discovery and development. Pharmacol. Rep.

2006, 58, 453−472.

Schrodinger, LLC: New York, 2006; www.schrodinger.com/

̈

productpage/14/6/75/.

(47) Boukharta, L.; Keranen, H.; Stary-Weinzinger, A.; Wallin, G.; de

Groot, B. L.; Aqvist, J. Computer simulations of structure−activity

relationships for HERG channel blockers. Biochemistry 2011, 50,

6146−6156.

(48) Du-Cuny, L.; Chen, L.; Zhang, S. A critical assessment of

combined ligand- and structure-based approaches to HERG channel

blocker modeling. J. Chem. Inf. Model. 2011, 51, 2948−2960.

(49) Durdagi, S.; Duff, H. J.; Noskov, S. Y. Combined receptor and

ligand-based approach to the universal pharmacophore model

development for studies of drug blockade to the hERG1 pore domain.

J. Chem. Inf. Model. 2011, 51, 463−474.

(65) Rossi, C.; Foletti, A.; Magnani, A.; Lamponi, S. New

perspectives in cell communication: bioelectromagnetic interactions.

Semin. Cancer Biol. 2011, 21, 207−214.

(66) Jiang, M.; Dun, W.; Fan, J. S.; Tseng, G. N. Use-dependent

“agonist” effect of azimilide on the HERG channel. J. Pharmacol. Exp.

Ther. 1999, 291, 1324−1336.

(67) Jorgensen, P. L. Isolation of (Na⁺plus K⁺)-ATPase. Methods

Enzymol. 1974, 32, 277−290.

(68) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J.

Protein measurement with the Folin phenol reagent. J. Biol. Chem.

1951, 193, 265−275.

(50) Imai, Y. N.; Ryu, S.; Oiki, S. Docking model of drug binding to

the human ether-a-go-go potassium channel guided by tandem dimer

mutant patch-clamp data: a synergic approach. J. Med. Chem. 2009, 52,

1630−1638.

(69) Lanzetta, P. A.; Alvarez, L. J.; Reinach, P. S.; Candia, O. A. An

improved assay for nanomole amounts of inorganic phosphate. Anal.

Biochem. 1979, 100, 95−97.

(70) MacroModel, version 9.9; Schrodinger, LLC, New York, 2011.

̈

(51) Lees-Miller, J. P.; Subbotina, J. O.; Guo, J.; Yarov-Yarovoy, V.;

Noskov, S. Y.; Duff, H. J. Interactions of H562 in the S5 helix with

T618 and S621 in the pore helix are important determinants of

hERG1 potassium channel structure and function. Biophys. J. 2009, 96,

3600−3610.

(52) Masetti, M.; Cavalli, A.; Recanatini, M. Modeling the hERG

potassium channel in a phospholipid bilayer: molecular dynamics and

drug docking studies. J. Comput. Chem. 2008, 29, 795−808.

(53) Mitcheson, J. S.; Chen, J.; Lin, M.; Culberson, C.; Sanguinetti,

M. C. A structural basis for drug-induced long QT syndrome. Proc.

Natl. Acad. Sci. U. S. A. 2000, 97, 12329−12333.

(54) Pearlstein, R. A.; Vaz, R. J.; Kang, J.; Chen, X. L.;

Preobrazhenskaya, M.; Shchekotikhin, A. E.; Korolev, A. M.;

Lysenkova, L. N.; Miroshnikova, O. V.; Hendrix, J.; Rampe, D.

Characterization of HERG potassium channel inhibition using

CoMSiA 3D QSAR and homology modeling approaches. Bioorg.

Med. Chem. Lett. 2003, 13, 1829−1835.

(55) Recanatini, M.; Cavalli, A.; Masetti, M. Modeling HERG and its

interactions with drugs: recent advances in light of current potassium

channel simulations. ChemMedChem 2008, 3, 523−535.

(56) Sintra Grilo, L.; Carrupt, P. A.; Abriel, H.; Daina, A. Block of the

hERG channel by bupivacaine: electrophysiological and modeling

insights towards stereochemical optimization. Eur. J. Med. Chem. 2011,

46, 3486−3498.

(71) Jorgensen, W. L.; Maxwell, D. S.; TiradoRives, J. Development

and testing of the OPLS all atom force field on conformational

energetics and properties of organic liquids. J. Am. Chem. Soc. 1996,

118, 11225−11236.

(72) Kaminski, G. A.; Friesner, R. A.; Tirado-Rives, J.; Jorgensen, W.

L. Evaluation and reparametrization of the OPLS-AA force field for

proteins via comparison with accurate quantum chemical calculations

on peptides. J. Phys. Chem. B 2001, 105, 6474−6487.

(73) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T.

Semianalytical Treatment of Solvation for Molecular Mechanics and

Dynamics. J. Am. Chem. Soc. 1990, 112, 6127−6129.

(74) Polak, E.; Ribiere, G. Note sur la convergence de directions

conjuguee

37.

(75) P450 Site of Metabolism Prediction, User Manual, version 1.1;

́ ́

s. Rev. Fr. Inform. Rech. Operation., 3e Annee 1969, 16, 35−

Schrodinger Press LLC: New York, 2011.

̈

(76) Sherman, W.; Day, T.; Jacobson, M. P.; Friesner, R. A.; Farid, R.

Novel procedure for modeling ligand/receptor induced fit effects. J.

Med. Chem. 2006, 49, 534−553.

(77) Schrodinger Suite 2011 Induced Fit Docking Protocol: Glide,

̈

version 5.7; Schrodinger, LLC: New York, 2011; Prime, version 3.0;

̈

Schrodinger, LLC: New York, 2011.

̈

(78) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R.

W.; Klein, M. L. Comparison of simple potential function for

simulating liquid water. J. Chem. Phys. 1983, 79, 926−935.

(79) Humphreys, D. D.; Friesner, R. A.; Berne, B. J. A Multiple-

Time-Step Molecular-Dynamics Algorithm for Macromolecules. J.

Phys. Chem. 1994, 98, 6885−6892.

(57) Sanchez-Chapula, J. A.; Navarro-Polanco, R. A.; Culberson, C.;

Chen, J.; Sanguinetti, M. C. Molecular determinants of voltage-

dependent human ether-a-go-go related gene (HERG) K⁺channel

block. J. Biol. Chem. 2002, 277, 23587−23595.

Q

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(80) Hoover, W. G. Canonical DynamicsEquilibrium Phase-Space

Distributions. Phys. Rev. A 1985, 31, 1695−1697.

(81) Martyna, G. J.; Tobias, D. J.; Klein, M. L. Constant-Pressure

Molecular-Dynamics Algorithms. J. Chem. Phys. 1994, 101, 4177−

4189.

(82) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.;

Pedersen, L. G. A Smooth Particle Mesh Ewald Method. J. Chem. Phys.

1995, 103, 8577−8593.

R

dx.doi.org/10.1021/jm300831b | J. Med. Chem. XXXX, XXX, XXX−XXX

Article Doi

Mimicking the intramolecular hydrogen bond: Synthesis, biological evaluation, andmolecular modeling of benzoxazines and quinazolines as potential antimalarial agents

DOI: 10.1021/jm300831b

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm300831b

Full text of DOI:10.1021/jm300831b

Products guided by the article

R&D Labs maybe for 1414781-90-0

Relevant to this article

Hot Product