Reinvestigating Old Pharmacophores: Are 4-Aminoquinolines and Tetraoxanes Potential Two-Stage Antimalarials?

Article abstract of DOI:10.1021/acs.jmedchem.5b01374

The syntheses and antiplasmodial activities of various substituted aminoquinolines coupled to an adamantane carrier are described. The compounds exhibited pronounced in vitro and in vivo activity against Plasmodium berghei in the Thompson test. Tethering a fluorine atom to the aminoquinoline C(3) position afforded fluoroaminoquinolines that act as intrahepatocytic parasite inhibitors, with compound 25 having an IC₅₀ = 0.31 μM and reducing the liver load in mice by up to 92% at 80 mg/kg dose. Screening our peroxides as inhibitors of liver stage infection revealed that the tetraoxane pharmacophore itself is also an excellent liver stage P. berghei inhibitor (78: IC₅₀ = 0.33 μM). Up to 91% reduction of the parasite liver load in mice was achieved at 100 mg/kg. Examination of tetraoxane 78 against the transgenic 3D7 strain expressing luciferase under a gametocyte-specific promoter revealed its activity against stage IV-V Plasmodium falciparum gametocytes (IC₅₀ = 1.16 ± 0.37 μM). To the best of our knowledge, compounds 25 and 78 are the first examples of either an 4-aminoquinoline or a tetraoxane liver stage inhibitors.

Full text of DOI:10.1021/acs.jmedchem.5b01374

Subscriber access provided by - Access paid by the | University Library, UC Santa Cruz
Article
Re-investigating old pharmacophores: Are 4-aminoquinolines
and tetraoxanes potential two-stage antimalarials?
Bogdan A. Šolaja, Natasa Terzic Jovanovic, Jelena Konstantinovic, Miklos Tot, Jovana
Burojevic, Olgica Djurkovic Djakovic, Jelena Srbljanovic, Tijana Stajner, Tatjana Z
Verbic, Mario Zlatovi#, Marta Monteiro Maia Machado, Inês S. Albuquerque, Miguel
Prudencio, Richard J. Sciotti, Stevan Pecic, Sarah D’Alessandro, and Donatella Taramelli
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01374 • Publication Date (Web): 07 Dec 2015
Downloaded from http://pubs.acs.org on December 10, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 65
1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Reꢀinvestigating old pharmacophores: Are 4ꢀaminoquinolines and tetraoxanes potential
twoꢀstage antimalarials?
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
#
¶
¶
¶
Natasa Terzić, Jelena Konstantinović, Mikloš Tot, Jovana Burojević, Olgica Djurkovićꢀ
∇
∇
∇
¶
¶
Djaković, Jelena Srbljanović, Tijana Štajner, Tatjana Verbić, Mario Zlatović, Marta
Machado, Inês S. Albuquerque, Miguel Prudêncio, Richard J. Sciotti, Stevan Pecic,∀
§
§
§,*
†
‡
‡
¶,
Sarah D’Alessandro, Donatella Taramelli, Bogdan A. Šolaja *
#
Institute of Chemistry, Technology, and Metallurgy, Belgrade, Serbia
¶
Faculty of Chemistry, University of Belgrade, Studentski trg 16, P.O. Box 51, 11158,
Belgrade, Serbia
∇Institute for Medical Research, University of Belgrade, Dr.Subotića 4, 11129 Belgrade,
Serbia
§
Instituto de Medicina Molecular, Faculdade de Medicina, Universidade de Lisboa, 1649ꢀ028
Lisboa, Portugal
†
Experimental Therapeutics Branch, Walter Reed Army Institute of Research, Silver Spring,
MD20910, USA
∀
Division of Experimental Therapeutics Department of Medicine, Columbia University, New
York, NY 10032, USA
‡
Dipartimento di Scienze Farmacologiche e Biomolecolari, Università degli Studi di Milano,
20133 Milan, Italy
ACS Paragon Plus Environment

Page 3 of 65
Journal of Medicinal Chemistry
2
1
2
3
4
5
6
7
8
9
Abstract
The syntheses and antiplasmodial activities of various substituted aminoquinolines coupled to
an adamantane carrier are described. The compounds exhibited pronounced in vitro, and in
vivo activity against P. berghei in the Thompson test. Tethering a fluorine atom to the
aminoquinoline C(3) position afforded fluoroaminoquinolines that act as intraꢀhepatocytic
parasite inhibitors, with compound 25 having an IC50 = 0.31 µM and reducing the liver load in
mice by up to 92% at 80 mg/kg dose. Screening our peroxides as inhibitors of liver stage
infection revealed that the tetraoxane pharmacophore itself is also an excellent liver stage P.
berghei inhibitor (78: IC50= 0.33 µM). Up to 91% reduction of the parasite liver load in mice
was achieved at 100 mg/kg. Examination of tetraoxane 78 against the transgenic 3D7 strain
expressing luciferase under a gametocyteꢀspecific promoter revealed its activity against stage
IVꢀV P. falciparum gametocytes (IC50 = 1.16 ± 0.37 µM). To the best of our knowledge,
compounds 25 and 78 are the first examples of either an 4ꢀaminoquinoline or a tetraoxane
liver stage inhibitors.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 65
3
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Introduction
Malaria is a widespread parasitic disease affecting mainly the population of Subꢀ
Saharan Africa, SouthꢀEast Asia and South America. With its heavy burden of ca. 207 million
cases in 2013 and a death toll of ca. 627,000, malaria is one of the major public health
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
problems worldwide (WHO Malaria report 2014). In addition, with as much as ca. 10% of
the global population being transiently ill and temporarily unable to work, malaria poses an
immense economic burden to malariaꢀendemic regions. Fortunately, over the last 15 years the
death toll and the number of cases decreased steadily by the combined action of the
widespread use of insecticideꢀtreated bed nets, the implementation of artemisininꢀbased
2
combination therapies (ACTs) and appropriate diagnosis.
In particular, ACTs, in which a fast parasitemia clearing peroxide molecule is
combined with a slowꢀacting antimalarial pharmacophore, an aminoquinoline or alike, proved
to be a very effective therapeutic approach. Thanks to the public and private contributions and
the efforts of Medicine for Malaria Venture (MMV), new combinations (i.e. DHAꢀ
piperaquine; pyronaridineꢀartesunate) or formulations (i.e., artemetherꢀlumefantrine
3
dispersible for children) have been approved and distributed.
Nevertheless, drug resistance already pushed aside many good and lowꢀtoxic antiꢀ
malarial drugs, most relevantly chloroquine. A few years ago, the first signs of resistance of
Plasmodium falciparum (Pf) to artemisinin in Western Cambodia were reported as unusual
4
, 5
prolonged parasite clearance phenotype following treatment with an ACT. Artemisinin
6
resistance may presently be spreading throughout SouthꢀEast Asia. Multiple point mutations
in the propeller domains of the Pf kelch gene on chromosome 13 (K13), have been associated
with the resistance phenotype in vivo and represent a molecular marker for tracking the
7
spread of artemisinin resistance.
ACS Paragon Plus Environment

Page 5 of 65
Journal of Medicinal Chemistry
4
1
2
3
4
5
6
7
8
9
In this context, to counter the problems outlined above, new, cheap and effective antiꢀ
plasmodial drugs are needed. The life cycle of Plasmodium parasites includes several stages,
in both its mammalian and its invertebrate hosts. Malaria infection in mammals starts with the
bite of an infected female Anopheles mosquito, followed by migration of Plasmodium
sporozoites to the liver, where they infect and develop inside hepatocytes. Hepatic infection is
asymptomatic and malaria symptoms do not arise until hepatic merozoites are released into
the blood stream and infect erythrocytes. Therefore, Plasmodium liver stages appear to be
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
promising targets for new antimalarial drugs aiming at prophylactic intervention. In the
blood, the asexual cyclic replication of the parasites causes the symptoms of malaria and leads
to the differentiation of the parasite’s sexual forms, the gametocytes, which are then taken up
by a mosquito, where the sexual phase of the parasite’s life cycle takes place. The discovery
and development of antimalarials that act through several, if not all, stages of the parasite life
cycle would greatly enhance their effectiveness, thus allowing much cheaper and considerably
more effective regimens. This is also the most recent indication of MMV for new
3
antimalarials.
The activity of several chemotypes that target multiple stages of the parasite’s life
9
10
cycle was reported in the last few years: (+)ꢀSJ733, imidazopyrazines, quinoloneꢀ3ꢀ
11
12
13
14
diarylethers, flavones, Nꢀcinnamoylated chloroquine analogues, quinolinꢀ4(1H)ꢀimines,
1
5
16
17
3
,5ꢀbis(benzylidene)ꢀ4ꢀpiperidones, imidazopyrazines, primaquine–chloroquine hybrid,
1
8
19
20
ketotifen, norketotifen, decoquinate, thienopyrimidinones. All these multiple stage
inhibitors are active against both liver and blood Plasmodium asexual stages. Remarkably,
10,16
11
17
imidazopyrazines,
quinoloneꢀ3ꢀdiarylethers, primaquine–chloroquine hybrid, and
1
9
decoquinate are active against blood sexual stages too. In vivo liverꢀstage (LS) activity was
10,16
11
confirmed for imidazopyrazines,
quinoloneꢀ3ꢀdiarylethers, primaquine–chloroquine
1
7
18
19
21
hybrid, ketotifen, norketotifen, decoquinate, and fluoroquinoline DDD107498.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 65
5
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Peroxideꢀcontaining hybrid compounds, in which the LS drug primaquine was bound
to an artemisinin or tetraoxane moiety, were synthesized as potential new antimalarial drugs,
active against both the liver and the erythrocytic asexual stages of the parasite. In fact, both
peroxide classes possess excellent intraꢀerythrocytic activity. Several primaquineꢀartemisinin
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
23
(
PQꢀART) hybrids, tetraoxane–pyrimidine nitrile hybrids, and tetraoxaneꢀ, and
2
4
artemisininꢀprimaquine hybrids were reported. The PQꢀART and tetraoxaneꢀprimaquine
amalgams showed very interesting LS activity in vitro. However, the observed LS activity
could be primarily ascribed to the partner employed in the hybrids molecules (based on the
23, 25, 26
relative LS inactivity of the respective partner peroxide entities –tetraoxanes and ART
).
22
The in vivo LS activity was confirmed for primaquineꢀART hybrids.
Recently, it was observed that reducing drug (chloroquine) pressure results in reꢀ
2
7
emergence of chloroquineꢀsensitive P. falciparum malaria. Although it is very unlikely that
any aminoquinoline would be used in monotherapy regimens, new aminoquinolines were
2
8, 29,
developed as promising antimalarials, and as possible partners in combination treatments.
3
0, 31, 32, 33, 34, 35, 36
7ꢀchloroꢀ4ꢀaminoquinolines (7ꢀACQ) are considered remarkably active
2
6
against asexual blood stages, but not against LS parasites. It is therefore not surprising that
1
7
only one successful effort towards designing a LSꢀactive 7ꢀACQ has been reported. Lödige
et al. showed that a PQꢀCQ hybrid is efficient against asexual blood stages of CQꢀsusceptible
and CQꢀresistant P. falciparum strains 3D7, Dd2, and K1, and, importantly, against in vitro
17
LS of P. berghei. In addition, they showed that the PQꢀCQ hybrid molecule is able to
appreciably decrease the parasite liver load when dosed at 3× 90 mg/kg.
Our own efforts in the field of 7ꢀACQ antimalarials resulted in promising compounds
37
38
with adamantane and thiopheneꢀbased carriers (Figure 1).
ACS Paragon Plus Environment

Page 7 of 65
Journal of Medicinal Chemistry
6
1
2
3
4
5
6
7
8
9
H
N
H
N
(
)n
NH
(
)m
( )m
HN
N
S
R
Cl
N
Cl
1
: m=1, n=1; 2: m=2, n=1
IC (D6)=3-4 nM; IC (W2)=8-12 nM
5: R = F, m = 1; 6: R = F, m = 2
IC (D6)=8-14 nM; IC (W2)=19-26 nM
37
38
50
50
50
50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
: m=1, n=2; 4: m=2, n=2
7: R = CN, m =1; 8: R = CN, m =2
37
38
IC (D6)=9-11 nM; IC (W2)=16-34 nM
IC (D6)=3-15 nM; IC (W2)=8-24 nM
50
50
50
50
R1
H
N
R2
H
N
HN ( )
( )n
m
( )m
NH
N
X
H(F)
S
Liver stage
Blood stage
(
H)Cl
N
This publication
Blood stage
Cl(H)
Figure 1. Recently developed aminoquinoline antimalarials.
Here, we report on a new generation of aminoquinolines with adamantane carriers and
benzothiophene terminal amine substitutions. We show that compounds of the new series
have improved blood stage antiplasmodial activity in comparison to the initial series. In
addition, the first LS activity of an 4ꢀaminoquinoline is reported for the first time, followed by
firstꢀinꢀclass LS activity of a tetraoxane antimalarial.
Results
Chemistry
Here, we present the synthesis of three types of aminoquinoline antiplasmodials: a)
adamantane derivatives, b) benzothiophene derivatives, and c) 3ꢀfluoroꢀ4ꢀaminoquinolines,
which include chloroquine and AQꢀ13 fluoro derivatives (Scheme 1, Scheme 2). The key
reaction for the preparation of adamantane and benzothiophene P. falciparum inhibitors is the
coupling of prepared fragments by reductive, or BuchwaldꢀHartwig amination as outlined in
Scheme 1. Full details on the syntheses of these compounds and their respective precursors
are given as part of Supporting information.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 65
7
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
R1
R1
H
NH2
( )m
N
HN
N
AdCHO/AdCH CHO
2
HN
N
( )m ( )n
reductive amination
R = Me, Et;
m = 1, 2, 3;
n = 1, 2.
1
Cl
Cl
9, 23, 27, 29, 33, 36, 40, 42
Cl
R1 R2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R3
n
H
N
( )
HN
( )
m
N
H
H N
(H)Cl
N
2
( )m
R1 R2
(
)n
R3 R = H, Me;
nucleophilic substitution /
Buchwald-Hartwing
amination
1
R = H, Me, Et;
2
R = H, Me;
3
(
H)Cl
N
m = 0, 1, 2;
n = 0, 1.
1
0-12, 22, 24, 28, 30, 31,
3
4, 35, 37, 41, 43, 44
R1
NH2
CHO
HN
( )m
N
X
S
HN
( )m
X
reductive
amination
S
R = H, Me, i-Pr;
1
(
H)Cl
N
X = H, F, CN;
m = 1, 2, 3, 5.
(
H)Cl
N
46 - 67
Scheme 1. Amination affords antiplasmodials 9ꢀ12, 22ꢀ24, 27ꢀ31, 33ꢀ37, 40ꢀ44, 46ꢀ67.
The synthesis of the 3ꢀfluoroꢀsubstituted aminoquinoline core was initiated with the
introduction of fluorine at the C(3) position via corresponding diazonium salt (not isolated,
39
Balz–Schiemann reaction ) in 70% yield, starting with known 3ꢀaminochloroquinolines 13,
1
4, Scheme 2. The introduction of iodine in position C(4) was accomplished using low
temperature generation of intermediate lithio species followed by low temperature iodination
4
0
thereof, affording the iodine regioisomers (17 and 18, R=Cl) in (4:1)ꢀratio. It is interesting
to note that when R=H, the lithiation/iodination sequence afforded only 3ꢀfluoroꢀ4ꢀ
iodoaminoquinoline (19) in 70% yield. The synthesis of the aminoquinoline part was
concluded by Buchwald–Hartwig or nucleophilic amination (41ꢀ77%, Scheme 2). The yields
of amination reactions at C(2) and C(4) varied; the crossꢀcoupling reaction proceeded
smoothly except on two occasions when possibly steric congestion of fluorine and a short
spacer lowered the reaction yields (compounds 20 and 68 in Supporting information). All
compounds screened for antiplasmodial activity were fully characterized and were found to be
>95% pure (HPLC).
ACS Paragon Plus Environment

Page 9 of 65
Journal of Medicinal Chemistry
8
1
2
3
4
5
6
7
8
9
48 % aq HBF , NaNO ,
4
2
o
a. LDA, THF, 4 h, -85 C
1,2-dichlorbenzene,
NH2
F
o
o
2
h, 110 C
b. I
2
, THF, 2 h, -78 C
R
N
R
N
R=Cl, 50 %
R=H, 70 %
1
3: R = Cl
14: R = H
15: R = Cl, 70 %
16: R = H, 97 %
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Buchwald-Hartwing
amination/nucleophilic
substitution
F
I
F
NHR
1
R
N
41-77 %
(H)Cl
N
I-C(4) : I-C(2)
4 : 1
20-21, 25-26, 32, 38-39, 45,
1
7 : 18 (R=Cl)
separated isomers
9: R=H
68-76
1
Scheme 2. Synthesis of 7ꢀchloroꢀ3ꢀfluoroꢀ4ꢀaminoquinolines. R : (Nꢀalkylꢀ2ꢀ(adamantꢀ1ꢀ
yl)alkanamine); 5ꢀ(diethylamino)pentanꢀ2ꢀyl, 3ꢀ(diethylamino)propanꢀ1ꢀyl.
1
Plasmodium asexual blood stages
In vitro antiplasmodial activity
Synthesized compounds were tested in vitro for their antiplasmodial activity against three P.
falciparum strains: D6 (CQ susceptible (CQS) strain), W2 (CQ resistant (CQR) strain), and
TM91C235 (Thailand, a multidrugꢀresistant (MDR) strain), by measuring the inhibition of
incorporation of radiolabeled hypoxanthine by drugꢀtreated parasites, as previously reported.
CQ, mefloquine (MFQ) and artemisinin (ART) were used as positive controls. The
4
1
antiplasmodial activity of all compounds tested is given in Table S1 (Supporting Information).
Twentyꢀone of these compounds were more potent than CQ and MFQ (IC50) against the CQS
strain D6, and twentyꢀthree were more active against MDR C235 strain than MFQ. Of 49
compounds more active than CQ against the CQR strain W2, 26 were ≥10ꢀfold more active
against the given strain. In addition, three compounds, 9, 33, and 34, were more active than
ART against all three P. falciparum strains tested.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 65
9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The in vitro activity data of the most potent compounds with the adamantane carrier is
selected to Table 1. Detailed analysis revealed several interesting SAR issues within the series
differing in linker length:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-
In general, a series with shortest linker (m=1, Scheme 1) was more active than the
ones with m=2, 3 against all three P. falciparum strains (compounds 9, 10, 33, 34). All four
compounds had excellent selectivity indices against HepG2 and RAW264.7 cells
(SIHepG2/D6= 4,188; SI RAW 264.72/D6 = 1,886ꢀ2,901), indicating that they could constitute
good candidates for in vivo evaluation. The introduction of methyl groups in the αꢀ
positions to nitrogens (in an attempt to subdue possible metabolic hydroxylation of the
respective αꢀcarbons) afforded remarkably more active antiplasmodials, e.g., 9, 10, 33, 34
vs. 1 (for chemistry and activity details of 1 see ref. 37); however, the screening in human
and mouse liver microsomal preparations indicated that only 33 and 34 are acceptably
metabolically stable (MLM). Compounds 33 and 34 (m=1, n=2) appeared to be 2.5ꢀ6 times
more stable than those with shorter linkers, m=1, n=1 (Table 1). In addition, the position of
the methyl group in NꢀC(α)ꢀC(α’)ꢀNꢀC(α’’) linker also had a favorable effect on activity:
3 3 3
antiplasmodials 9, 33 (CH ꢀC(α)) vs. 10, 34 (CH ꢀC(α’), Table 1) and 11 (CH ꢀC(α’’),
Table S1).
Table 1. In vitro antiplasmodial activity
Toxicity
f
HepG2
In vitro antimalarial
h
i
a,b
MLM
HLM
SI HepG2
(
RAW
activity (P.
g
(min)
(min) (RAW) /D6
Compound
MW
falciparum, IC , nM)
264.7)
5
0
IC , nM
50
c
d
e
D6
1
C235
W2
9
3
83.97
1
9
3
(2901)
(3771)
10
10
11
13
(2901)
(1886)
1
0
383.97
2
10
ACS Paragon Plus Environment

Page 11 of 65
Journal of Medicinal Chemistry
1
0
1
2
3
4
5
6
7
8
9
H
N
N
H
N
Cl
2
2
3
98.00
4
15
22
(3161)
11
16
60
790
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
3
398.00
10
17
10
10
3045
2317
3967
60
60
305
136
2
2
4
5
363.54
135
185
60
13
21
4
15.99
81.53
135
60
197
259
238
157
29
3
j
k
(79 )
(78 )
2
2
6
9
3
4
12.02
12.02
21
23
22
24271
57
49
1156
30
4
7
19
25
3090
1612
441
32
3
3
1
2
412.02
51
165
282
430.01
198
342
488
3549
18
3
3
3
4
3
98.00
98.00
1
6
1
7
4
7
(2910)
25127
60
60
57
27
(2910)
4188
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 65
1
1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
3
6
7
412.02
13
7
13
102
171
546
14
13
126
203
394
26
24271
1590
3212
60
60
1867
227
22
377.57
430.01
395.56
426.05
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
38
143
154
7
3
9
3
42
13273
60
60
1896
4
4
3
4
426.05
391.59
444.04
9
47
81
2598
1593
2608
289
133
9
12
286
590
393
547
45
288
CQ
MFQ
ART
319.87
378.31
282.33
12
16
9
139
36
456
5
13
7
a
3
Antiplasmodial IC50 values (nM) ([ H]hypoxanthine incorporation method) for isolates and
clones of P. falciparum. All in vitro experiments were performed as technical triplicates with
R2 within 0.96–1. CQ susceptible P. falciparum African D6 clone. P. falciparum multidrug
resistant C235 strain (Thailand). CQ resistant P. falciparum Indochina W2 clone.
Hepatocellular carcinoma. Rat Macrophage Cell Line. Mouse liver microsomes. Human liver
microsomes. Mouse hepatocytes. Human hepatocytes.
b
c
d
e
f
g
h
i
j
k
ACS Paragon Plus Environment

Page 13 of 65
Journal of Medicinal Chemistry
1
2
1
2
3
4
5
6
7
8
9
Based on the above findings we turned our attention to antiplasmodials with methyls
at C(α) position in the series with extended linkers:
The examination of the series with m=2; n=1, n=2 (Scheme 1, Ad derivatives), revealed that
antiplasmodials with a 7ꢀchloroꢀ4ꢀaminoquinoline moiety (22, 23, 24, 36, 37) had acceptable
SI (136ꢀ1,867), but we found low SI values for fluorine derivatives 25, 38 (29, 22,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
respectively), as well as considerable in vivo toxicity issues for compound 39 (vide infra). In
this series, good activity against CQS D6, CQR W2 and MDR C235 strains was afforded by
7ꢀchloro derivatives 22, 23, and 36 (4ꢀ22 nM, Table 1). In order to directly compare the
influence of chlorine at C(7), desꢀchloroaminoquinolines 24 and 37 were prepared along with
their respective fluoro derivatives 26 and 39. We found that omitting the ClꢀC(7) substituent
did not affect the in vitro activity of 24 and 37 against the CQS D6 strain (IC50=17 nM, 7 nM,
respectively, Table 1; c.f. compounds 35, 44, Table S1). However, the activity against the
CQR W2 and MDR C235 strains was significantly impaired; all desꢀchloro compounds were
8
ꢀ86ꢀfold less active than their 7ꢀchloroꢀ4ꢀaminoquinoline analogs against the MDR C235
strain, and 10ꢀ24ꢀfold less active against the CQR W2 strain (23 vs. 24; 36 vs. 37; 33 vs. 35;
42 vs. 44, Table 1 and Table S1). The introduction of a fluorine atom in position C(3)
failed to improve the in vitro activity. All FꢀC(3) derivatives with the adamantane carrier, 25,
8, 26, 39, 32, 45 (Table 1) and 20, 21 (Table S1), were significantly less active than their
3
C(3)ꢀH isosteres. On the nM scale they could be considered virtually inactive. The lack of in
vitro activity of C(3) fluorine derivatives was extended to CQ derivative 74, which is ca. 34ꢀ
fold less active than CQ against the CQS D6 strain (Table S1).
The thiophene moiety was investigated on several occasions as an aromatic
3
4, 38
contributor to the aminoquinoline pharmacophore.
As part of this report we analyzed the
effect of the benzothiophene contributor to the aminoquinoline moiety on antiplasmodial
activity (Table S1, Table 2). Two series of this class were prepared: C(2)ꢀ and C(3)ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 65
1
3
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
substituted benzothiophenes. As a general remark, we note that benzothiophenes did not
improve the in vitro activity of aminoquinolines, and that the whole series was less potent
38
than the 7ꢀchloroꢀ4ꢀaminoquinolineꢀthiopheneꢀbenzene combination. C(2)ꢀsubstituted
benzothiophenes were found less active in vitro than their C(3) isomers, with only 52 being as
active as CQ and MFQ against P. falciparum strains D6 and C235, respectively (Table S1).
Of twelve C(3) substituted benzothiophenes, five antiplasmodials were more active than CQ
against the CQS strain D6, and six were more active than MFQ against the MDR strain C235.
Compounds without a chlorine atom at the quinoline C(7) position (54, 62, 66), were much
less active than the whole benzothiophene series against all three P. falciparum strains. The
introduction of fluorine and cyano group at positions C(5) and C(6) (benzothiophene
numbering) did not appreciably influence the benzothiopheneꢀaminoquinoline hybrids’ in
vitro activity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In vivo bloodꢀstage efficacy studies
The compounds presented in Table 2 were examined for their in vivo efficacy in a mouse
model. A modified Thompson test was employed to determine the blood schizonticidal
efficacy of 14 compounds. In addition, in a separate host toxicity study, groups of five healthy
mice were dosed with 160 mg/kg/day × 3 days of 34, 33, 23, 24, 25, 42, 74, 73, 75, 58, 59, 63,
without any overt clinical manifestations of toxicity. This was monitored by observations of
individual mouse behavior and appearance two times a day for 31 days. Compound 58 was
toxic at 160 mg/kg/day, and was therefore tested at 80 mg/kg/day, at which concentration it
did not exibit any signs of toxicity.
Compounds were administered po at given doses on days 3, 4, and 5 after infection
6
42
using the Thompson test in C57Bl6 female mice infected with 1× 10 P. berghei parasites.
The compounds with an ethylene linker (1, 34, 33) cleared the parasites on D6. However,
ACS Paragon Plus Environment

Page 15 of 65
Journal of Medicinal Chemistry
1
4
1
2
3
4
5
6
7
8
9
recrudescence occurred on days 10ꢀ14 and mice succumbed to malaria later on. Only at the
higher dose of 160 mg/kg/day did the compound 33 cure 2 mice, while the minimal active
dose (MAD) for 1, and 34 was estimated at 40 and 80 mg/kg/day, respectively. Next, we
tested the series of compounds with a propylene linker: 23, 24, 25, and 26. The 7ꢀchloroꢀ4ꢀ
aminoquinoline isosteres 23 and 25 were tested in a doseꢀdependent manner and a minimal
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
curative dose (MCD) of 80 mg/kg/day for both compounds was estimated. Although none of
the two compounds showed curative effect at the lower doses employed, a small difference
was noticed: aminoquinoline 23 was more potent than 25, exhibiting MAD even at 10
mg/kg/day, and clearance with recrudescence at 40 and 20 mg/kg/day, while 25 was
ineffective at 20 and 10 mg/kg/day (not shown). The analogs of 23 and 25, isosteres 24 and
26, were selected for examining the in vivo 4ꢀaminoquinoline activity dependence on the
chlorine substituent at C(7), at 80 mg/kg/day. As shown in Table 2, none of the compounds
cured infection. However, isostere 24 (HꢀC(3)) was more active than 26 providing clearance
with recrudescence at the administered concentration.
a,b
Table 2. In vivo antimalarial activity (Thompson test).
Mice
dead/ day
died
Mean
survival time
(MST, day)
Mice alive on
day 31/total
Compound
1
mg/kg/day
Parasitemia
D6: negative, clearance. Two
mice positive at D31
40
3/15-24
2/5
0/3
>21.4
15.3
D3: 3 mice 0.4%-0.5%; D6: 3
mice negative; D10: 2 mice
negative, 1 mouse 0.2%;
D14%: 3 mice 1.6%-3.6%
1/14,
1/15, 1/17
34
80
D6: mice negative; D24: 3
mice 2.7%-9.6%; D28: 1
mouse 67.4%. D31: 2 mice
negative
c
33
160
2/24, 1/28
2/5
27.6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 65
1
5
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
D3: 5 mice 0.7%-1.2%; D6: 5
mice negative; D17: 5 mice
negative; D31: 5 mice
negative
1
60
5/5
5/5
0/5
>31
>31
17.2
D3: 5 mice 0.4%-0.5%; D6: 5
mice negative; D17: 5 mice
negative; D31: 5 mice
negative
80
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
D3: 5 mice 0.3%-2.4%; D6: 5
2
3
mice negative; D10: 5 mice
negative; D14: 5 mice 0.2%-
1.2%; D17: 2 mice 2.1%-4.6%
1/16,
2/17, 2/18
4
0
D3: 4 mice 0.5%-3.5%; D6: 4
mice negative; D10: 4 mice
2/14,
1/15, 1/18
2
0
0
0/4
0/5
0/5
15.2
13
0
.2%-0.4%; D14: 2 mice 1%-
.9%; D17: 1 mouse 3.5%
3
D3: 5 mice 0.4%-1.6%; D6: 5
mice 0.18%-0.5%; D10: 5
mice 1.6%-8.9%
1
/11,
1
3
/13, 1/15
D3: 5 mice 0.2%-1%; D6: 5
mice negative; D10: 3 mice
negative, 2 mice 0.2%-0.9%;
D14: 5 mice 1.4%-2.5%
2
4
80
2/14, 3/16
15.2
D3: 5 mice 0.3%-0.5%; D6: 1
mouse ex., 4 mice negative;
D17: 4 mice negative; D31: 4
mice negative
160
1/7
4/5
6/6
26.2
>31
D3: 6 mice 0.3%-1%; D6: 6
mice negative; D17: 6 mice
negative; D31: 6 mice
negative
8
0
2
5
D3: 5 mice 0.5%-3%; D6: 5
mice 1%-4.7%; D10: 5 mice
3.1%-16.3%; D14: 3 mice
2/12,
1/21,
1/23, 1/24
4
0
0/5
0/5
18.4
12.8
5
.6%-23%; D21: 2 mice
7.5%-62.4%
3
D3: 5 mice 0.4%-1.4%; D6: 5
mice 1.3%-26.8%; D10: 3
mice 3.2%-25%;
2
1
/8, 1/12,
/13, 1/23
26
80
D14→D21: 1 mouse
7.7%→35%
D6: 5 mice negative; D14: 4
mice negative, 1 mouse
0.3%; D17: 3 mice
recrudescence, 1 mouse
negative
1/17,
1/19,
1/21,
160
0/5
0/5
21.4
13.6
4
2
1/24, 1/26
D6-D14: 0.4%-52.3%
4
0
1/12, 4/14
ACS Paragon Plus Environment

Page 17 of 65
Journal of Medicinal Chemistry
1
6
1
2
3
4
5
6
7
8
9
D3: 5 mice 0.5%-1.5%; D6: 5
mice negative; D10: 2 mice
negative, 3 mice 0.1%-0.4%;
D14: 5 mice 0.16%-3.4%;
D17: 2 mice 6.9-10%
1/15,
1/16,
1/17, 2/18
7
7
4
3
160
0/5
16.8
D3: 5 mice 0.3%-1.3%; D6: 5
mice negative; D10: 5 mice
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3/14,
1/15, 1/16
160
160
80
0/5
0/6
0/4
14.6
20.2
14
0
.2% -0.4%; D14: 5 mice 3%-
7
%
D3: 6 mice 0.1%-1.5%; D6: 6
mice 0.2%-2.6%; D10: 5 mice
1.7%-4.9%; D14: 4 mice
1/9, 1/12,
2/22,
1/27, 1/29
7
5
9
.6%-23.5%; D21: 4 mice 52-
8%; D28: 1 mouse 60%.
5
D3: 4 mice 0.2%-0.7%; D6: 4
mice 0.2%-0.3%; D10: 4 mice
5
5
8
9
4/15
0
.4%-0.6%; D14: 4 mice
.3%-2.7%
1
D3: 5 mice 0.9%-2.3%; D6: 5
mice negative; D10: 5 mice
negative; D14: 5 mice
160
1/18, 1/21
3/5
3/4
26.4
negative; D17: 5 mice 0.3%-
4
.4%; D21: 4 mice 2.3%-
3
.8%; D31: 3 mice positive
D3: 4 mice 0.4%-0.9%; D6: 4
mice negative; D10: 3 mice
negative, 1 mouse 0.2%;
160
D14: 4 mice 0.2%-1.2%; D17:
1/27
30
4
mice 0.5%-1.7%; D21: 4
mice 1%-37%; D31: 3 mice
positive
63
D3: 5 mice 0.1%-1.7%; 5 mice
negative; D10: 5 mice 0.2%-
0
.4%; D14: 5 mice 0.6%-
3.5%; D17: 5 mice 1.2%-
.5%; D21: 5 mice 2.4%-
80
1/27, 3/31
1/5
5/5
3/5
30.2
>31
6
1
5.2%; D31: 1 mouse
positive
D3: 5 mice 0.4%-0.9%; D6-
D31: 5 mice negative
160
D3: 5 mice 0.5%-0.9%; D6: 5
mice negative; D10: 5 mice
negative; D14: 3 mice
negative, 2 mice 0.2%; D17: 3 1/17, 1/18
mice negative, 1 mouse 4%;
D21: 3 mice negative; D31: 3
mice negative.
CQ
80
25.6
Infected
controls
All mice died on day 6-8
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 65
1
7
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Groups of five P. berghei (ANKA strain) infected mice were treated with aminoquinolines suspended in 0.5%
hydrohyethylcellulose-0.1% Tween 80 administered po once per day on days on 3-5 postinfection. Mice alive
b
on day 31 with no parasites in a blood film were considered cured. All compounds were tested in separate
c
experiments (mouse groups) for toxicity. None exerted any toxic effect when dosed 3× 160 mg/kg/day. Body
temperatures were recorded to predict death according to IACUC protocols and parasitemia was captured
o
prior to euthanizing the animal. The animal euthanasia point was 35.5 C.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Next, we examined CQꢀ, and AQꢀ13ꢀfluoro derivatives 74 and 73, respectively, at the
nonꢀtoxic dose of 160 mg/kg/day. Both compounds led to clearance with recrudescence, with
no cure even at such a high dose.
Three benzothiophenes were examined for their efficacy at the selected doses (Table
2
), and since 5ꢀfluorobenzothiophene 58 was toxic at the higher dose it was tested at 80
mg/kg/day with no cure or survival. Its FꢀC(6) isomer 59 was examined at the 160 mg/kg/day
dose, and exhibited clearance with recrudescence and survival of 3/5 animals, with MST >26
days (D31: 3 mice positive). The higher homolog of 58, compound 63, also provided
clearance with recrudescence with an MST of 30 days, and survival of 3/4 animals. However,
a dramatic difference emerged between the two homologs at 80 mg/kg/day. While 58
provided no clearance, homolog 63 led to clearances with recrudescence, 1/5 survival and
twice the MST of 58.
Inhibitory activity of Plasmodium liver stages
In vitro LS activity
In an attempt to develop multistage antimalarials, the possible LSꢀactivity of our
aminoquinoline blood stage antiplasmodials was investigated. The evaluation of the LS
activity of our steroidal and nonꢀsteroidal tetraoxanes was also performed by measuring the
luminescence intensity in human hepatoma cell line (Huhꢀ7) cells infected with a firefly
4
3
luciferaseꢀexpressing P. berghei line, as previously described.
ACS Paragon Plus Environment

Page 19 of 65
Journal of Medicinal Chemistry
1
8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Activity of compounds 23, 25, 36, 38, and 69 against P. berghei liver stages. Antiꢀ
P. berghei activity (infection scale in Relative Luminescence Units (RLU), bars) and toxicity
to Huh7 cells (cell confluency scale in Relative Fluorescence Units (RFU), circles) are shown
at several concentrations. The infection load of human hepatocellular carcinoma (Huhꢀ7) cells
was determined by bioluminescence measurements of cell lysates 48 h after infection with
luciferaseꢀexpressing P.berghei parasites.
Initialy, we evaluated the aminoquinoline bloodꢀstage antiplasmodials with an adamantane
carrier 23, 36, 25, 38, 69 for LS activity using primaquine as a positive control (Table 3). The
percentage of Huh7 cell confluency was used as a measure of relative cell toxicity (Figure 2).
Table 3: Liver stage activity of 4-aminoquinoline and
tetraoxane antiplasmodials.
Compound
% LS
infection
(5 µM)
LS IC50
(µM)
LS in vivo,
(mg/kg/day)
b
c
a
23
0.1
1-2.5
1
8
6 (50);
(80)
25
13 (2µM)
0.31±0.14
6
3
3
9
6
8
~2
8
1-5
2.5-5
1-2.5
1-5
3
74
~35
~40
31
7
6
7
1-5
7
<1
78
0.55
0.33 ±0.05
40 (50);
9 (100)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 65
1
9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
PQ
CQ
62 (10µM)
25
ca. 9
2
3
25
ART
ATQ
>75
ca. 12
22 nM
2
5
a
1
0 000 Huh7 cells/well seeded 24 h before infection. 10
0
00 PbA-LuciGFPcon spz/well. Confluency measured by a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
fluorimetric assay. Infection was measured by a
bioluminescence assay.
b
All dose-responses employed at least 5 dosages.
c
Measurement of liver load by qRT-PCR or by
luminescence measurement. Mean of 2 independent
experiments. All mice were infected by i.v. injection of
1
0,000 P. berghei sporozoites. Control groups were treated
by oral gavage with vehicle (equivalent% of DMSO, in
sunflower oil). Experimental groups were treated by oral
gavage administration of 25 and 78 at the given doses.
The in vitro results presented in Table 3 indicate that 7ꢀACQꢀAd antiplasmodials 23, 25, 36,
8, 69 all exhibit relatively high potency against the liver forms of the parasite. LS inhibitory
3
activity was assessed at seven concentrations (10 µM – 0.1 µM) and several compounds
showed a doseꢀdependent behavior (Figure 2). Table 3 and Figure 2 show that at 5 µM
concentration, homologs 23 and 36 appear to be potent inhibitors of LS infection (0.1%, 8%,
respectively). However, their activity dropped at lower concentrations yielding IC values (1ꢀ
50
25
2
.5 µM, 2.5ꢀ5 µM), comparable to that of AQꢀ13, CQ. We then explored the effect of the Fꢀ
C(3) aminoquinoline substituent on LS inhibitory activity by assaying another homolog pair
of fluoro isosteres 25 (←23) and 38 (←36). It was found that fluorine at C(3) significantly
enhanced the activity of 25 in comparison to 23 (IC50= 0.31 µM vs. IC50 = 1ꢀ2.5 µM,
respectively), and of 38 in comparison to 36 (IC =1ꢀ2.5 µM vs. IC = 2.5ꢀ5 µM,
50
50
respectively). Significant in vitro inhibitory activity (38%) of 25 was observed even at 125
nM concentration.To the best of our knowledge, the 3ꢀfluoro compound 25 is the first 4ꢀ
aminoquinoline that exerts evident and significant in vitro activity against Plasmodium LS
(
Figure 2). The effect of the fluorine on Plasmodium LS stage is additionally indicated by
enhanced activity of 2ꢀaminoquinoline 69 (a regioisomer of 25, LS infection of 69 at 5 µm
ACS Paragon Plus Environment

Page 21 of 65
Journal of Medicinal Chemistry
2
0
1
2
3
4
5
6
7
8
9
~2%, IC = 1ꢀ5 µM), and the activities of CQ fluoro derivative 74 and its 2ꢀaminoquinoline
50
analog 76 (both IC = 1ꢀ5 µM), Table 1, Figure 2. For comparison, LS activity of CQ is ~9
50
25
µM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
23
24
Tetraoxanes were coupled to pyrimidineꢀnitriles and primaquine, and the
respective hybrids were found to be active against LS Plasmodium infection in vitro. During
our work on steroidal and nonꢀsteroidal tetraoxane molecules, we exposed their unexpected
stability to various reaction conditions (e.g., stability to LAH, pH1.6), and during in vitro
metabolism studies we noticed that tetraoxane moiety was resistant to metabolic
44, 45, 46
transformations.
Therefore, we decided to evaluate the activity of these compounds
during the LS of parasite development. Tetraoxanes evaluated here have proven to be
excellent intraꢀerythrocytic antimalarials without any observed toxicity (Figure S1, Table
4
4, 45, 46, 47, 48, 49
S2).
The structures of the two most LSꢀactive steroidal tetraoxanes are given in Figure 3,
and their respective activities are presented in Table 3.
O
AcO
Z
O
O
OAc
O
H
O
7
7: Z = OCH3
n
7
8: Z = NHPr
Figure 3. Liver stage active tetraoxanes.
The two tetraoxanes differing only in their side chain exhibited pronounced inhibition
potencies with nꢀPr amide 78 being ca. 3ꢀfold more active than the respective methyl ester 77
IC50= 0.33 µM vs IC50= <1 µM, respectively). While methyl ester 77 was as active as
(
artemether, the in vitro potency of nꢀPr amide 78 was comparable to the only LSꢀactive
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 65
2
1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
5
peroxide, artemisone (IC50~ 0.04 µM ). Although the activity of other examined tetraoxanes
was lower than that of 77 and 78 (Table S2), the overall results confirm that the tetraoxane
moiety itself is probably capable of killing hepatic parasites without the aid of the 8ꢀ
aminoquinoline moiety, which would lead to primaquineꢀlike in vitro LS inhibition.
Tetraoxane 78 was also found to be moderately active when tested in vitro against
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
stage IVꢀV P. falciparum gametocytes of the 3D7elo1ꢀpfs16ꢀCBG99 transgenic strain.
Compound 78 inhibited gametocytes viability with IC50 = 1.16 ± 0.37 µM (mean of three
different experiments in duplicate). Control drugs employed were DHA (IC50= 0.44 ± 0.29
µM) and epoxomicin (IC50= 11.8 ± 4.2 nM).
In vivo LSꢀactivity of 7ꢀACQ and tetraoxane molecules 25 and 78.
To obtain a better insight into LS activity of our most active aminoquinoline 25 and
tetraoxane 78, we examined their in vivo activity against P. berghei LS. Thus, two groups of
Figure 4. Luminescenceꢀbased measurement of liver parasite loads after oral administration
of compound 25 measured 48 h after infection and plotted as percentage of the infection
levels in control mice.
ACS Paragon Plus Environment

Page 23 of 65
Journal of Medicinal Chemistry
2
2
1
2
3
4
5
6
7
8
9
five C57BL/6J mice were treated orally with 25 (DMSO+sunflower oil) at 50 mg/kg and 80
mg/kg doses during 3 consecutive days (Dꢀ1, D0, D+1, two independent experiments each).
Mice were infected with luciferaseꢀexpressing P. berghei sporozoites on D0, followed by
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
assessment of liver parasite load 48 h after infection by luminescence, and of the appearance
of parasites in the blood, disease symptoms, and survival. Our experiments employing 50
mg/kg and 80 mg/kg of the compound showed an 84% and 92% decrease of the P. berghei
liver load relative to controls, respectively (Figure 4).
In a similar experiment tetraoxane 78 administered at 50 mg/kg and 100 mg/kg
reduced the P. berghei liver load 60% and 91%, respectively (Figure 5, Table 3).
Figure 5. Parasite liver load in C57BL/6 mice measured 48 h after infection by qRTꢀPCR
of P. berghei 18S rRNA, normalized to hypoxanthineꢀguanine phosphoribosyltransferase
(Hprt) and plotted as percentage of the infection levels in control mice.
Discussion
The effect of numerous substituents at C(5), C(6), and C(7) of aminoquinoline
51
antiplasmodials is well explored. However, the antimalarial data on C(3) substituted
5
2, 53, 54
aminoquinolines are scarce,
and the introduction of several alkyl/aryl groups at C(3) as
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 65
2
3
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the logical consequence of the slightly improved activity of purified sontochin over that of
chloroquine, revealed that 4ꢀaminoquinolines with C(3) alkyl/aryl substituents
55
(
“pharmachins”) are generally more active against CQR strains than CQ in vitro. We sought
to investigate the effect of a fluorine substituent at aminoquinoline position C(3) on βꢀhematin
formation, and on in vitro and in vivo blood stage activity; in addition we investigated the
activity of our inhibitors against P. berghei LS.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
As part of our wider investigation of 4ꢀaminoquinolineꢀadamantane combination as
37
pharmacophoreꢀcarrier symbiosis, aminoquinoline 42 was first tested in vivo due to its
favorable microsomal metabolic stability and excellent SI data (Table 1, Table S1). Its rather
5
6
high logPow (QPlogPow 5.67, experimental 5.87) was assisted by favorably predicted
ꢀ5
apparent MDCK cell permeability QPPMDCK=1.29 × 10 cm/s, rather low predicted
QPlogHERG toxicity IC50= ꢀ5.076 (8.40 µM), and its MTD was estimated at >160 mg/kg (no
host toxicity in mice up to 160 mg/kg/day × 3 days). At 160 mg/kg/day, 42 showed excellent
suppression activity on D6ꢀD14 (Table 2). However, the suppression activity was maintained
only with 1/5 mice on D17 and recrudescence occurred in 3/5 cases. MST was estimated at
2
1.4 days without any survival. Screening the compounds containing the ethylene bridge
m=1, Fig. 1) with lower logPow 1, 33, 34 (Table 2), showed that only 33 cured 2/5 mice at
60 mg/kg/day. Next, we focused on compounds with a propylene linker (m=2, Fig. 1) and
(
1
with FꢀC(3) substituents. As shown in Scheme 2 all FꢀC(3) derivatives with an adamantane
carrier, 25, 38, 26, 39, 32, 45 (Table 1) and 20, 21 (Table S1), were significantly less active in
vitro than their HꢀC(3) isosteres. Homologs 23 and 36 possessed excellent potency across all
three P. falciparum strains, displayed excellent microsomal metabolic stability (Table 1) and
were therefore tested in vivo; in addition, their fluorine isosteres 25 and 38 were also chosen
for examination in a P. berghei challenge assay, despite the lower in vitro activity and low
microsomal stability of 25. The respective desꢀchloro derivatives 24, 26, 37 and 39 were also
ACS Paragon Plus Environment

Page 25 of 65
Journal of Medicinal Chemistry
2
4
1
2
3
4
5
6
7
8
9
prepared in order to explore the influence of ClꢀC(7) substituent at the aminoquinoline moiety
in a P. berghei challenge test. First, the compounds were tested for toxicity in mice at doses of
160 mg/kg/day × 3 days. To our surprise, compounds 36, 37, 38, 39 (all m=2, n=2; Scheme 1)
were all found toxic regardless of their respective substitution pattern at C(3) and C(7), which
is in sharp contrast to compounds 42 (m=3, n=2), 23, 24, 25, 26 (m=2, n=1), and other
compounds listed in Table 2. It is interesting to note that in vivo toxicity results could not be
predicted based on in vitro toxicity parameters (c.f. RAW 264.7, HepG2 activity, Table 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As indicated above, isosteres 23 and 25 cured all mice from bloodꢀstage malaria in the
Thompson test when dosed with 160, 80 mg/kg/day. The difference between the two became
evident at lower doses with isostere 23 providing clearance on D6 even at 20 mg/kg/day.
Fluoro derivative 25, which demonstrated cures of blood stage malaria, was found very active
against LS P. berghei infection. The activity difference (and lack of mouse toxicity) between
the two isosteres warranted further investigation of their MOA. Thus, elementary information
about possible mechanism of action was obtained by submitting compounds 23 and 25, 24
and 26, fluoro AQꢀ13 derivative 73, FꢀCQ derivative 74 and its isomer 76, and CQ as
57
standard, to a βꢀhematin inhibitory activity assay (BHIA). The assay screens the interference
of potential antimalarials with the heme detoxification process employed by P. falciparum.
The results obtained indicated that the introduction of fluorine at the C(3) position of 4ꢀ
aminoquinolines greatly enhances the inhibition of βꢀhematin formation – 23 (1.84) vs 25
(
0.61); 24 (1.22) vs 26 (0.61); CQ (1.23) vs 74 (1.07) all IC values, c.f. Table S1. The
50
influence of the FꢀC(3) substituent is consistent: it significantly enhances the BHIA and
indicates that such compounds would strongly interfere with heme detoxification process
within FV. However, this observation did not completely match with the compounds’
antimalarial activity: all tested fluoro derivatives exhibited slightly lower in vivo activity with
respect to their H isosteres (Table 2). Additional information on the influence of amine
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 65
2
5
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
regiochemistry on BHIA is obtained from the activity of 76, the 2ꢀaminoalkyl isomer of FꢀCQ
4. The lower activity of 76 as compared to 74 (IC =2.43 vs IC =1.07, respectively)
7
5
0
50
indicates that the interference of alkylaminoquinolines with the hematin formation probably
depends on the position of the alkylamino side chain as well.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The introduction of the FꢀC(3) substituent had a profound effect on the respective
dipoles (Table S1), e.g., 23 (6.04) vs. 25 (3.10), logP, 23 (3.85) vs. 25 (3.25), and pKa values.
Experimentally determined pKa values given in Table S1 clearly indicate that all examined
fluoro derivatives are monoprotonated at physiological pH, and to a great extent in FV.
Therefore, it appears that while 7ꢀACQ 23 interacts with hematin as diprotonated species, its
fluoro isostere 25 is expected to be mostly monoprotonated in FV, as well as within the liver
hepatocytes, where it exerts considerable inhibition of intraꢀhepatocytic parasites (for
distribution diagrams see Supporting information). Since the fluoro derivative 25 showed
remarkable in vivo activity both against LS and in asexual blood stages, we checked this
compound for hepatocyte metabolic stability followed by metabolite profiling. The intrinsic
6
6
clearance of 8.9 µL/min/10 cells and 8.7 µL/min/10 cells (human (h), mouse (m),
respectively) was estimated, therefore, significant stability in hepatocytes resulted − t1/2 = 78
min, 79 min (h, m), contrary to the reported poor metabolic stability when incubated with
hepatic microsomes (Table 1). The amount of remaining parent compound 25 at 120 min is
32% (h) and 37% (m). Metabolite identification exposed the monohydroxylation product as
the main one, and the product of dihydroxylation as minor ꢀ after 60 min: 25 : 25+16 : 25+32
5:26:1 (h); 25 : 25+16 : 25+32 = 33:26:1 (m). No Phase II products were detected. Further
=
3
metabolite analysis, isolation of the main metabolite and its testing will be carried out.
Arrhythmia can be induced by blockage of hERG channels by certain antimalarial
5
8
59
agents such as lumefantrine and quinidine. Both, 23 and 25 were submitted to hERG test
and IC50 > 5 µM was estimated for both. At 5 µM concentration (highest concentration tested)
low inhibition was found – 23, 11.7%; 25, 8.54% (positive control quinidine IC = 2.67 µM).
50
ACS Paragon Plus Environment

Page 27 of 65
Journal of Medicinal Chemistry
2
6
1
2
3
4
5
6
7
8
9
The Ames test used to assess the mutagenic potential of 23 afforded positive results:
antiplasmodial 23 was negative at 500 µg/mL against S. typhimurium strains TA98 and
TA100 (+ S9, ꢀ S9).
Current data on isosteres 23 and 25 indicate that both 4ꢀaminoquinolines with
adamantane carrier are nonꢀtoxic and nonꢀmutagenic antiplasmodials, they are relatively
metabolically stable in microsomal preparations (23) and in hepatocyte preparations (25).
Both compounds are efficient in vivo against P. berghei asexual RBC stages and a possible
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
indication on their MOA was obtained from BHIA experiments, which showed that all fluoro
ꢀACQꢀAd derivatives were ca. 2ꢀfold more active than their hydrogen isosteres.
7
Benzothiophene antiplasmodials were also found to be more active than CQ. This might
suggest that our compounds exert their activity on asexual RBC stages by interfering with the
P. falciparum detoxification process within the FV. Having in mind the pronounced twoꢀstage
activity of 25 and the lack of host toxicity in mice for >31 days (indicating the nonꢀtoxicity of
monohydroxylation metabolite too), a strategy for the elucidation of LS MOA will be
developed.
In vitro LS activities of tetraoxanes with cholic acid carrier appear to be higher than
the simple cyclohexylidene analogs and their IC fall within ca. 1ꢀ5 µM (Table S2). The
50
highest activity was observed for Nꢀpropyl amide 78 (IC50= 0.33 µM) and the proof of
concept was obtained after in vivo experiments with 78 reducing the P. berghei liver load by
91% at 100 mg/kg. Their mechanism of action is not clear at present; however, since the only
pharmacophore present is the tetraoxane moiety, we speculate that peroxideꢀgenerated
radicals could be involved in parasite elimination.
Asexual P. falciparum blood stage infection is followed by the development of sexual
parasite forms, gametocytes, which are transmitted to mosquitoes to continue their
development. Each round of schizogony is followed by differentiation of merozoites (ca. 1%)
into the male and female gametocytes through stages I–V. Tetraoxane 78 was found to
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 65
2
7
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
moderately inhibit the viability of P. falciparum late gametocytes (stage IVꢀV) with IC =
5
0
1.16 ± 0.37 µM, which is only 3.5ꢀfold higher than the IC50 against LS. Our results indicate
that tetraoxane 78 constitutes a good starting material for upgrading this pharmacophore
against LS and late stage gametocytes, in addition to its established asexual erythrocytic stage
activity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Conclusion
The syntheses and antiplasmodial activities of variously substituted 4ꢀaminoquinolines
coupled to adamantane carrier were described. The compounds exhibited pronounced in vitro
and in vivo activity against P. berghei in the Thompson test. The series with the
benzothiophene carrier showed lower activity against D6, C235, and W2 P. falciparum strains
in comparison to the adamantane series, and did not afford any cure in vivo. Tethering
fluorine atom to C(3) position of 4ꢀaminoquinoline yielded fluoroaminoquinolines that
constitute intraꢀhepatocytic parasite inhibitors, with compound 25 having IC50 = 0.33 µM.
Compound 25 afforded up to 92% reduction of the in vivo P. berghei liver load at 80 mg/kg
dose. Testing our peroxide antimalarials as inhibitors of LS infection revealed that the
tetraoxane pharmacophore itself is also a powerful LS P. berghei parasite inhibitor (78: IC50
.33 µM). A 91% reduction of the parasite liver load in mice was achieved at 100 mg/kg. To
the best of our knowledge, compounds 25 and 78 are the first examples of an 4ꢀ
=
0
aminoquinoline and a tetraoxane liver stage inhibitor, respectively. Neither of the compounds
was toxic in vivo and both possessed good metabolic stability. Their mode of action in the LS
is currently unknown. However, in the case of 78, it could be anticipated that it is likely that
peroxideꢀgenerated radicals are involved in parasite elimination since the only pharmacophore
present is a tetraoxacyclohexane moiety. Compound 25 and its hepatocyte monohydroxylated
metabolite are not toxic, as confirmed by challenging the host, 25 is not mutagenic, and has a
ACS Paragon Plus Environment

Page 29 of 65
Journal of Medicinal Chemistry
2
8
1
2
3
4
5
6
7
8
9
good hERG profile. The results warrant focusing on the MOA of aminoquinoline 25 and its
monohydroxylated metabolite, as well as of tetraoxane 78 against Plasmodium LS.
Finally, the compounds exhibiting their action through two (25), or more (78) parasite
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
stages have good potential to be developed into efficient inhibitors to add to already existing
3
antimalarial drugs. The results presented here open the door to new LSꢀactive chemotypes.
Experimental section
Chemistry. Melting points were determined on a Boetius PMHK apparatus and were not
corrected. IR spectra were taken on a ThermoꢀScientific Nicolet 6700 FTꢀIR diamond crystal.
1
13
H and C NMR spectra were recorded on a Varian Geminiꢀ200 spectrometer (at 200 and 50
MHz, respectively), and a Bruker Ultrashield Advance III spectrometer (at 500 and 125 MHz,
respectively) in the indicated solvent (vide infra) using TMS as the internal standard.
Chemical shifts are expressed in ppm (δ) and coupling constants (J) in Hz. ESI–MS (HRMS)
spectra of the synthesized compounds were acquired on a Agilent Technologies 1200 Series
instrument equipped with Zorbax Eclipse Plus C18 (100 × 2.1 mm i.d. 1.8 ꢁm) column and
DAD detector (190ꢀ450 nm) in combination with a 6210 TimeꢀofꢀFlight LC/MS instrument in
positive and negative ion modes. The samples were dissolved in MeOH (HPLC grade). The
selected values were as follows: capillary voltage 4 kV; gas temperature 350°C; drying gas 12
ꢀ1
L min ; nebulizer pressure 45 psig; fragmentator voltage: 70 V. Mass spectral analyses were
performed using electrospray ionization in positive ion mode on a Surveyor separations
module coupled to a ThermoFinnigan TSQ AM triple quadrupole mass spectrometer.
Compounds were analyzed for purity using an Agilent 1200 HPLC system equipped with
Quat Pump (G1311B), Injector (G1329B) 1260 ALS, TCC 1260 (G1316A) and Detector
1260 DAD VL+ (G1315C), and Waters 1525 HPLC dual pump system equipped with an
Alltech Select degasser system, and a dual k 2487 UV–Vis detector. LKB 5060ꢀ006 Micro
ACS Paragon Plus Environment