Discovery of Marinoquinolines as Potent and Fast-Acting Plasmodium falciparum Inhibitors with in Vivo Activity

Article abstract of DOI:10.1021/acs.jmedchem.8b00143

We report the discovery of marinoquinoline (3H-pyrrolo[2,3-c]quinoline) derivatives as new chemotypes with antiplasmodial activity. We evaluated their inhibitory activities against P. falciparum and conducted a structure-activity relationship study, focusing on improving their potency and maintaining low cytotoxicity. Next, we devised quantitative structure-activity relationship (QSAR) models, which we prospectively validated, to discover new analogues with enhanced potency. The most potent compound, 50 (IC₅₀^3d7 = 39 nM; IC₅₀^K1 = 41 nM), is a fast-acting inhibitor with dual-stage (blood and liver) activity. The compound showed considerable selectivity (SI > 6410), an additive effect when administered in combination with artesunate, excellent tolerability in mice (all mice survived after an oral treatment with a 1000 mg/kg dose), and oral efficacy at 50 mg/kg in a mouse model of P. berghei malaria (62% reduction in parasitemia on day 5 postinfection); thus, compound 50 was considered a lead compound for the discovery of new antimalarial agents.

Full text of DOI:10.1021/acs.jmedchem.8b00143

Subscriber access provided by Kaohsiung Medical University
Article
Discovery of Marinoquinolines as Potent and Fast-Acting
Plasmodium falciparum Inhibitors with In Vivo Activity
Anna Caroline Campos Aguiar, Michele Panciera, Eric Francisco Simão dos Santos, Maneesh Kumar
Singh, Mariana Lopes Garcia, Guilherme Eduardo de Souza, Myna Nakabashi, José Luiz Costa,
Célia R.S. Garcia, Glaucius Oliva, Carlos Roque Duarte Correia, and Rafael Victorio Carvalho Guido
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00143 • Publication Date (Web): 07 Jun 2018
Downloaded from http://pubs.acs.org on June 8, 2018
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 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 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.
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 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Marinoquinolines as Potent and Fastꢀ
Acting Plasmodium falciparum Inhibitors with
In Vivo Activity
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
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‡
2‡
2
Anna Caroline Campos Aguiar , Michele Panciera , Eric Francisco Simão dos Santos ,
3
,4
1
1
Maneesh Kumar Singh , Mariana Lopes Garcia , Guilherme Eduardo de Souza ,
4
5
3
1
Myna Nakabashi , José Luiz Costa, Célia R.S. Garcia , Glaucius Oliva ,
2
1
Carlos Roque Duarte Correia *, Rafael Victorio Carvalho Guido *
1
Sao Carlos Institute of Physics, University of Sao Paulo, Av. Joao Dagnone, 1100
Jardim Santa Angelina, São Carlos, SP, 13563ꢀ120, Brazil
2
Institute of Chemistry, State University of Campinas, Josue de Castro St., Campinas, SP,
13083ꢀ970, Brazil
3 Faculty of Pharmaceutical Sciences, University of Sao Paulo, Av. Prof. Lineu Prestes, 580
Cidade Universitária, São Paulo, SP, 05508ꢀ900, Brazil
4
Department of Physiology, University of Sao Paulo, Rua do Matão 101, travessa 14,
São Paulo, SP, 05508ꢀ090, Brazil
5
Faculty of Pharmaceutical Sciences, State University of Campinas, Rua Oswaldo Cruz,
º andar, bloco F3, Cidade Universitária, Campinas, SP, 13083ꢀ859, Brazil
2
‡
These authors contributed equally to the work
*Corresponding authors: (RVCG) rvcguido@ifsc.usp.br; (CRDC) roque@iqm.unicamp.br
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 98
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
ABSTRACT
We report the discovery of marinoquinoline (3Hꢀpyrrolo[2,3ꢀc]quinoline) derivatives as new
chemotypes with antiplasmodial activity. We evaluated their inhibitory activities against P.
falciparum and conducted a structureꢀactivity relationship study, focusing on improving their
potency and maintaining low cytotoxicity. Next, we devised quantitative structure activity
relationship models (QSAR), which we prospectively validated, to discover new analogues
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
3d7
K1
with enhanced potency. The most potent compound, 50 (IC50 = 39 nM; IC50 = 41 nM), is
a fastꢀacting inhibitor with dualꢀstage (blood and liver) activity. The compound showed
considerable selectivity (SI > 6410), an additive effect when administered in combination
with artesunate, excellent tolerability in mice (all mice survived after an oral treatment with a
1000 mg/kg dose) and oral efficacy at 50 mg/kg in a mouse model of P. berghei malaria
(62% reduction in parasitemia on day 5 postꢀinfection); thus, compound 50 was considered a
lead compound for the discovery of new antimalarial agents.
2
ACS Paragon Plus Environment

Page 3 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Malaria is one of the most prevalent parasitic infections in the world. The disease affects 216
million people each year, and in 2016, it was estimated to have caused 445,000 deaths. In
humans, malaria is caused by Plasmodium falciparum, Plasmodium knowlesi, Plasmodium
malariae, Plasmodium ovale, and Plasmodium vivax, of which P. falciparum is responsible
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
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
for highest mortality, particularly in children under 5 years old in Africa.
The human malaria parasite has a complex life cycle that includes several developmental
stages in the human host and the mosquito vector, which tremendously contributes to the
2–5
challenge of eradicating malaria. Disease control still relies on chemotherapy; however,
drug resistance is a limiting factor, and the emergence of artemisinin resistance in southeast
6–9
Asia and China is a serious threat to the global control of P. falciparum malaria. Therefore,
the discovery and development of new and efficacious drugs for the treatment and prevention
of malaria are urgently needed to both overcome resistance and meet the challenge of
10–13
eradicating malaria.
Due to the limitations of available chemotherapies, several efforts
have focused on expanding the range of new antimalarial compounds to thwart parasite
14–23
growth.
In this context, ideal new drugs should (1) not exhibit crossꢀresistance to existing
drugs, (2) be administered as a single dose, (3) prevent transmission (active against the sexual
24
stages of the parasite), and (4) provide chemoprotection (active against liver stages).
Marinoquinolines (MQ) were first isolated from the marine gliding bacteria Rapidithrix
25
thailandica (MQ A) and exhibit acetylcholinesteraseꢀinhibiting activity. Subsequently,
some new MQ derivatives (MQ BꢀF) were isolated from another gliding bacteria,
Ohtaekwangia kribbensis (Figure 1). All natural MQs contain the 3Hꢀpyrrolo[2,3ꢀc]quinoline
core and exhibit moderate antiprotozoal activity against the chloroquineꢀresistant P.
falciparum K1 lineage (IC = 1.7 – 15 µM) and Trypanosoma cruzi (IC = 22 – 53 µM) as
50
50
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 98
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
6
well as cytotoxic activity against the tumor cell lines L929, MCFꢀ7 and KBꢀ3ꢀ1. Recently,
other natural marinoquinoline analogs were isolated from bacterial strains of the species
Mooreia alkaloidigena and Catalinimonas alkaloidigena. Five novel natural compounds were
isolated from these cell cultures (MQ GꢀK). MQ I showed activity against the bacterial
strain CNJꢀ912 (Pontibacillus sp.), whereas MQ J and MQ K, were tested against
chloroquineꢀresistant strain of P. falciparum (K1) and showed IC50 values of 3.2 and 6.4 ꢁM,
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
7,28
respectively.
Based on these findings, which indicate that marinoquinolines have
promising biological activity against the malaria parasite, here, we report the use of an
integrated approach, including assessments of structureꢀactivity relationships (SAR),
quantitative structureꢀactivity relationships (QSAR), mechanistic assessments, parasitological
profiling and in vivo evaluation, to identify a frontrunner compound as a lead candidate for
the discovery of new antimalarial agents.
RESULTS
Structureꢀactivity Relationships
Chemistry. Effective routes were developed for the synthesis of several marinoquinoline
derivatives from readily available starting materials or from intermediates synthesized as
indicated in Schemes 1aꢀ1c and Equations 1aꢀ1c. The pinacol borane reagent 3ꢀ(4,4,5,5ꢀ
tetramethylꢀ[1,3,2]dioxaborolanꢀ2ꢀyl)ꢀ1ꢀ(triisopropylꢀsilanyl)ꢀ1Hꢀpyrrole 2 was synthesized
2
9
on the gram scale by adapting procedures reported by Buchwald and Billingsley, and
30,31
Muchowski and coworkers,
and an overall yield of 58% was obtained in three steps. For
the synthesis of iodobenzenes 25 and 26, anilines 23 and 24 were first converted into their
corresponding aryl diazonium salts, followed by iodination with NaI to provide the desired
3
2
iodobenzenes 25 and 26 at 82% and 73% yields, respectively (Equation 1a). The key
4
ACS Paragon Plus Environment

Page 5 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
intermediate aryl pyrroles 3, 28 and 31 were synthesized using SuzukiꢀMiyaura crossꢀ
coupling reactions between pyrrole boronate ester 2 and the respective iodoꢀ and bromoꢀaryl
compounds 1, 25 and 26, producing moderate to good yields (Schemes 1a and 1b). The
nitroꢀcontaining aryl pyrrole 30, which was obtained by another SuzukiꢀMiyaura coupling
reaction followed by removal of the TIPS protecting group by TBAF, as indicated in Scheme
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
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
1c, was reduced by catalytic hydrogenation to provide the pyrroloaniline 31. After obtaining
pyrroloanilines 3, 28 and 31, several marinoquinolines analogues were then synthesized using
modified PictetꢀSpengler reactions employing aliphatic and aromatic aldehydes. Two Pictetꢀ
Spengler protocols were devised to improve the effectiveness of the cyclization reactions.
Methods A (using trifluoroacetic acid) and B (using benzotriazole and pꢀtoluene sulfonic
acid) generated marinoquinoline analogues 4ꢀ22 (Scheme 1a), 32ꢀ44 (Scheme 1c), and 48ꢀ50
(Equation 1c) in moderate to good isolated yields. Finally, the bromoꢀmarinoquinolines 45ꢀ47
were prepared by a regioselective bromination reaction of marinoquinolines 19, 33 and 35
using NBS, as indicated in Scheme 1b.
Scheme 1: General scheme for the preparation of marinoquinolines (MQ).
Scheme 1a: Preparation of the marinoquinolines using Suzuki Coupling and Pictet-Spengler ring closure
Bpin
NH2
Br
c
H N
2
N
+
N
N
H
4 - 22
TIPS
N
R
1
2
3
H
4
- methyl
- isobutyl
- benzyl
- ethyl
- phenyl
- 4-methoxy
14 - 4-nitro
15
5
6
7
8
9
-
4-amino
1
1
1
1
2
6 - 4-dimethylamino
7 - 4 - CO H
2
8 - 4- CO Me
2
9 - indole-3-yl
0 - 5-bromo-indol-3-yl
10 - 3,5-dimethoxy
11 - 3,4,5-trimethoxy
1
1
2 - 4-chloro
3 - 4-bromo
21 - 6-bromo-indol-3-yl
22 - indol-2-yl
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 98
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
Equation 1a: Preparation of the Iodobenzenes
NH2
I
NO2
NO2
d
R
R
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
23: R= Br
25: R= Br
24: R= OMe
26: R= OMe
Scheme 1b: Preparation of the brominated marinoquinoline using Suzuki Coupling and Pictet-Spengler ring closure
Br
Br
Br
I
Bpin
NO2
N
f
c
+
N
H N
N
O N
2
2
H
TIPS
Br
N
H
N
H
N
29
25
2
27
28
H
Scheme 1c: Preparation of the methoxylated marinoquinoline using Suzuki Coupling and Pictet-Spengler ring closure
MeO
MeO
MeO
I
Bpin
NO
2
a, b
e
c
+
N
N
H N
N
O N
2
2
H
TIPS
OMe
R
N
N
H
26
2
30
H
31
32 - 44
3
3
3
3
2 - isobutyl
3 - methyl
4 - ethyl
39 - aminophenyl
40
-
4-dimethylamino phenyl
4
4
4
4
1 - 4-carboxyphenyl
2 - 4-CO Me-phenyl
3 - 4-chlorophenyl
4 - benzyl
5 - phenyl
2
36 - 3,5-dimethoxy
37 - 3,4,5-trimethoxy
8 - nitrophenyl
3
6
ACS Paragon Plus Environment

Page 7 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Equation 1b: Regioselective bromination of marinoquinolines 19 and 33
R
R
Br
g
N
N
N
N
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
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
H
H
R'
R'
19: R= H, R' = indole-3-yl
33: R= OMe, R' = methyl
35: R= OMe, R' = phenyl
45: R= H, R' = indole3-yl
46: R= OMe, R' = methyl
47: R= OMe, R' = phenyl
Equation 1c: Preparation of the theoretically predicted marinoquinolines derivatives
MeO
MeO
4
4
8 - 4-N-phenyacetamida
9 - indole-3-yl
c
N
50 - 4-tert-butylphenyl carbamate
H N
N
2
H
R
N
H
31
48 - 50
a
Reaction conditions: (a) Pd(OAc) , SPhos, K PO , nꢀBuOH, H O, 80 °C, 12 h; (b) TBAF
2
3
4
2
(
1.0 M), THF, 25 °C; (c) Corresponding aldehyde, CH
.OEt , (ii) NaI, I ; (e) 10% Pd/C, MeOH, H
MeOH, reflux, 1 h.; (g) NBS, THF/H O 4:1, 25 °C, 15 min.
2
Cl
2
, TFA, 25 °C, 1 h.; d) (i) tꢀ
BuONO, BF
3
2
2
2 2 2
, 25 °C; (f) SnCl .2 H O, HCl(conc),
2
3
3
Biological Activity. We used the SYBR green method to perform the primary screening
and showed that natural marinoquinolines 4 and 6 were not active against the P. falciparum
D7 strain at the highest concentration evaluated (IC50 > 10 µM) (Table 1). However, 5
3
showed antiplasmodial activity in the low micromolar range (IC50 = 4 µM, LE = 0.44).
Moreover, the cytotoxic effect of 5 was evaluated against a human hepatic cell (HepG2)
using the MTT assay, and no toxic activity was observed against HepG2 cells (EC50 > 250
µM and SI > 70). Based on these results, we selected 5 as a privileged scaffold for the SAR
investigation.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 98
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 initial optimization efforts were focused on the R1 substituents. Replacing the isobutyl
substituent with an ethyl (7) or phenyl (8) group provided slightly less potent compounds,
with IC50 values of 7.3 and 9.9 µM, respectively, whereas the low cytotoxicity profile was
retained. Despite the reduced potency of compound 8, the selectivity index was > 75,
suggesting that this compound displayed an attractive safety profile. Subsequently, the SAR
of the 4ꢀphenyl substituent was explored, as shown in Table 2. Various substituents, such as
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 2
4ꢀOMe (9), 3,5ꢀOMe (10), 3,4,5ꢀOMe (11), Cl (12), Br (13), NO (14), NH (15), NMe (16),
COOH (17), and COOMe (18), were tolerated at the 4ꢀphenyl substituent of the
marinoquinoline; however, none of them improved potency. On the other hand, compound
16, a dimethylaniline derivative, showed a significant improvement in the selectivity index
34
(
SI > 104). Due to structural similarities between indoloquinoline and marinoquinoline
derivatives, we tested whether the addition of an indole substituent improved the inhibitory
activity of the series. Thus, an indole derivative was introduced in the 4ꢀposition of the
marinoquinoline to explore the potential of this moiety for improving potency (19ꢀ22) (Table
3). In general, the replacement was tolerated but did not enhance the inhibitory activity of the
corresponding 4ꢀphenyl derivatives.
After initial tentative optimization of the R1 groups, the reference compound 5 proved to be
the most active compound of this set. We then turned our attention to the R2 substituent, with
the goal of improving potency while maintaining cytotoxicity at a moderate level (SI > 50)
(Table 4). Introducing a methoxy group in the 7ꢀposition of the marinoquinoline ring in 5
yielded compound 32, which exhibited improved potency (IC50 = 0.5 ꢁM) and maintained the
ligand efficiency (LE = 0.46) and low cytotoxicity (SI > 70) compared with the hit compound
5
. Subsequent analogues were obtained using the 7ꢀmethoxyꢀlinked substituent at R2 (Table
4). Overall, the 7ꢀmethoxyꢀsubstituted derivatives were 2ꢀ to 18ꢀfold more active than the 7ꢀ
hydrogen analogues, leading to the discovery of submicromolar P. falciparum inhibitors. For
8
ACS Paragon Plus Environment

Page 9 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
example, compounds 40 (IC50 = 0.4 ꢁM; LE = 0.37) and 43 (IC50 = 0.6 ꢁM; LE = 0.39)
showed 15ꢀ and 18ꢀfold increases in the inhibitory activity compared to the related
compounds 16 (IC50 = 5.8 ꢁM; LE = 0.33) and 12 (IC50 = 10.5 ꢁM; LE = 0.35), respectively.
Additionally, the most potent inhibitors, 32, 40 and 43, consistently exhibited acceptable SI
values (> 50).
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
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
2Dꢀ and 3DꢀQSAR Studies
With the aim of obtaining insights into the structureꢀactivity relationships underlying this
series, we devised both 2Dꢀ and 3DꢀQSAR analyses to systematically describe the
antiplasmodial activity of our series at the structural level and to prospectively test our
hypotheses. The 2Dꢀ and 3DꢀQSAR analyses were performed using the hologram QSAR
35
36
(HQSAR)
and CoMFA methods, respectively. From the original data set, 31
marinoquinoline derivatives were selected as members of the training set for model
generation, and the other 3 inhibitors were selected as members of the test set for external
model validation. The compounds of both the training and test sets were selected according to
the structural diversity and coverage on the range of the pIC50 values. A structure similarity
map analysis confirmed that both training and test sets composition included compounds
representative of the whole data set (Figure S1). Hence, the training and test sets were
appropriated for the purpose of QSAR model development. HQSAR descriptors and CoMFA
molecular fields were used as independent variables, whereas the pIC50 values were used as
dependent variable in the PLS regression analyses to build QSAR models. The same training
set was employed for all QSAR analysis, and the internal and external predictive ability of
2
2
the models was assessed by their q and predictive r values (external test set predictions),
respectively.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 98
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
HQSAR employs a specialized fragment fingerprint, the molecular holograms, and PLS to
37
generate fragmentꢀbased structureꢀactivity relationships. The studies require specific values
to be selected for parameters, such as the distinction, size and type of the fragments, and the
length of the hologram, to build predictive models. Hence, 14 HQSAR models were
constructed using different combinations of fragment distinction and the default fragment
size of 4ꢀ7. The best model used a combination of atoms, bonds, connections and hydrogen as
fragment distinction parameters (Supporting Information Table S1). Next, the model was
selected for further investigations of the influence of fragment size on the internal predictive
power. The statistical quality of models was not improved as a function of
increasing/decreasing the fragment size (2–5, 3–6, 5–8, 6–9, and 7–10) (Supporting
Information Table S2). The best HQSAR model comprised the following parameters:
fragment distinction A/B/C/H, fragment size 4–7, hologram length 53; and statistical
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
parameters: q = 0.50, r = 0.74, SEE = 0.23, and N = 3 (Figure 2A and 2B). The model was
submitted to leaveꢀmanyꢀout (LMO) crossꢀvalidation employing 10 or 5 groups (LMO10 or
LMO , respectively) of compounds as a more rigorous test for assessing the stability and
5
statistical significance of the model. The results confirmed the stability and reliability of the
best
HQSAR
model
2
2
(q
LMO5 = 0.48 ± 0.04; q LMO10 = 0.49 ± 0.03, Table S4). Moreover, the Receiver Operating
Characteristic (ROC) analysis was used to confirm the performance of the model (Figure S2).
The area under the curve (AUC) for the best model was 0.74, which indicated a reasonable
discriminating power. The predictive power of the best HQSAR model was assessed by
predicting the inhibitory activity of 3 test set compounds (18, 20 and 36). This step evaluates
the true predictive power of QSAR models by calculating the biological activity of
38
compounds that were not included in the model training step. The results of the external
evaluation are shown in Table 5, and the graphical results for the predicted versus
1
0
ACS Paragon Plus Environment

Page 11 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
experimental activities of both the training set and test set are displayed in Figure 2A. The
2
satisfactory agreement between experimental and predicted values and r pred (0.85) indicate
that the best HQSAR model exhibited good predictive capability for external compounds.
In addition to predicting properties, HQSAR models provided contribution maps that were
useful for the identification of the impact of fragments on biological activity. In this map,
atoms represented in red and orange indicate negative contributions to the biological activity,
atoms in yellow and green indicate positive contributions, and atoms in white represent
neutral contributions. The contribution map for the most potent compound of the series, 40
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
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
(IC50 = 0.4 µM ꢀ pIC50 = 6.4), suggests that substitutions at the 4ꢀpositiion of the phenyl
substituent and the presence of a pyrrole ring on the marinoquinoline moiety improve
inhibitory activity Figure 2C).
HQSAR maps are useful tools for indicating which groups exhibit the greatest contributions
to the inhibitory activity; however, these maps do not provide insights into the
physicochemical properties of the new substituents. Thus, we investigated this series using
QSAR 3D to obtain further insights into antiplasmodial activity and generate new analogues
with enhanced inhibitory activity. The 3DꢀQSAR analyses were performed using
36
comparative molecular field analysis (CoMFA). The dataset of 34 molecules was
structurally aligned to the lowest energy conformer of compound 40 using the database
alignment protocol in SYBYL X. Molecular alignments and the maximal common
substructure of marinoquinoline derivatives used for structural superimposition and
alignment are shown in Figure 3A. The initial CoMFA model was derived for the same
training set compounds used in the HQSAR analysis (31 inhibitors) using the default
parameters of grid spacing and lattice point weight. However, the model showed an
2
2
unsatisfactory crossꢀvalidation coefficient (model 1, q = 0.41; SEP = 0.34; r = 0.65; SEE =
0.26, Supporting Information Table S3). Thus, we applied a region focusing protocol to
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 98
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
obtain models with improved internal consistency and predictive power. Table S3
summarizes the statistical parameters for the models constructed. The best CoMFA model
was built with a focusing region weight by Stdev*Coefficients of 1.5 and grid spacing of 1 Å
2
2
(
model 10, q = 0.60, SEP = 0.30, r = 0.76, SEE = 0.23). Moreover, we assessed the stability
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
and performance of the CoMFA model by applying the leaveꢀmanyꢀout (LMO) crossꢀ
validation and ROC analysis, respectively, as described for HQSAR model validation. Again,
2
2
the results demonstrated the consistency of the best model (q
= 0.57 ± 0.04; q
=
LMO10
LMO5
0.59 ± 0.02, Table S4), and a reasonable discriminating power (AUC = 0.78, Figure S2). The
values of the statistical parameters suggest that the model displayed internal consistency and
reasonable predictive ability. Then, the best model was selected for the external validation by
predicting the pIC50 of the 3 test set compounds that were not used for QSAR model
development (Table 5). Figure 3B presents a plot of experimental versus predicted pIC50
2
values. The r
value calculated using this method was 0.92, suggesting good predictive
pred
power for external compounds.
CoMFA models provide structural information as contour maps, thereby allowing the
identification of favorable and unfavorable regions for bulky, electropositive and
electronegative substituents. Briefly, the green regions in the steric contour maps indicate
areas where bulky substituents may enhance biological activity, while the yellow contours
indicate areas where bulky substituent are detrimental for activity. In case of the electrostatic
contour maps, regions in blue and red indicate electropositive and electronegative
substituents associated with increased activity, respectively. The contour maps of the most
potent compound, 40 (pIC50 = 6.4), are presented in Figure 3C. As shown in the maps, bulky
substituents in the 4ꢀposition of the phenyl substituent and in the 7ꢀposition of the
marinoquinoline moiety improve biological activity. These data are highly consistent with the
HQSAR maps, suggesting that substitutions at the 4ꢀposition of the phenyl ring improve
1
2
ACS Paragon Plus Environment

Page 13 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibitory activity. Moreover, electropositive substituents at the 2ꢀ and 4ꢀpositions of the
phenyl substituent and electronegative substituents in the 7ꢀposition of the marinoquinoline
moiety might increase the inhibitor potency. According to the structural and physicochemical
features provided by QSAR models, new compounds were synthetized and assessed.
Prospective Validation of the QSAR Models
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
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
To prospectively assess our hypothesis and verify the SAR data for marinoquinoline
derivatives, we designed three new compounds based on the 2Dꢀ and 3DꢀQSAR maps and
used both HQSAR and CoMFA models to predict their activities as P. falciparum inhibitors
(Table 6). This strategy is particularly relevant because our model was generated from a
dataset containing a limited number of molecules (i.e., 34 compounds) and range of pIC50
values (i.e., 2 orders of magnitude). Our design was focused on substitutions of compound 40
to increase potency while retaining low cytotoxicity. Thus, bulky and electronegative
substituents such as carbamate and acetamide were introduced in the 4ꢀposition of the phenyl
substituent and an indole group was introduced in the 4ꢀposition of the marinoquinoline
moiety. The predicted inhibitory activities of compounds 48–50 were in the highest potency
activity range in both HQSAR and CoMFA models (predicted pIC50 values ≥ 6.0) (Table 5).
Therefore, we then synthetized and assessed these compounds. For 48, the predicted
biological activity was overestimated. Based on the experimental data, compound 48 was
poor inhibitor of P. falciparum (IC > 10 µM). On the other hand, for compounds 49 and 50,
5
0
the evaluated inhibitory activities matched the predictions reasonably well (Figures 2A and
B). Interestingly, the evaluated inhibitory potencies for compounds 49 (IC50 = 0.061 µM;
LE = 0.42) and 50 (IC50 = 0.039 µM; LE = 0.36) were higher than the predicted values (Table
). According to the results of the cytotoxicity assay, compound 49 was toxic to HepG2 cells
in the micromolar range (EC = 20 µM); however, the selectivity index was reasonable (SI =
3
5
50
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 98
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
328). On the other hand, compound 50 exhibited very low cytotoxicity and a remarkably high
selectivity index (SI > 6400). In summary, the prospective assessment of our hypothesis
additionally strengthened the external validation of the models and confirmed the SAR
underlying this series. Hence, the models have limited predictive power, however, they did
capture the essential SAR features which gave us useful insights for the discovery of new
derivatives with enhanced inhibitory potency.
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
Marinoquinoline Derivatives are FastꢀActing Inhibitors with Additive Effects in
Combination with Artesunate, are Active Against Resistant Strain, but have no Activity
on Male and Female Gametocytes
To better characterize the antiplasmodial activity of the series, the most potent and selective
compounds (i.e., 32, 43 and 50) were selected as representatives for this series for further
evaluation using inocula with rings, schizonts or asynchronous as predominant starting P.
falciparum forms. As shown in Table 7, the calculated IC50 values against asynchronous,
rings, trophozoites, and schizont were equivalent to the values observed for the ring stages,
suggesting that the compounds are potent inhibitors of all erythrocytic forms of P.
falciparum. Next, to determine the speed of action of the marinoquinoline derivatives, we
incubated synchronized cultures (presenting at least 90% of parasites at ring stage) with the
compounds at a concentration 10ꢀfold greater than IC50 values and observed the
morphological changes under the microscope at 8, 24, 32 and 42 h postꢀsynchronization
(Figure 4A). Compounds 32, 43 and 50 showed activity toward the early ring stages,
inducing alterations in P. falciparum morphology between 8 and 16 h after incubation. These
data suggest a fastꢀacting mechanism in which the young forms of P. falciparum in the
intraerythrocytic cycle are susceptible to the effects of the compounds. Additionally, we
1
4
ACS Paragon Plus Environment

Page 15 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
assessed the potential drug interactions of these molecules with artesunate to inhibit P.
falciparum in vitro using concentrationꢀresponse assays. The assay aimed to elucidate the
potential benefits and limitations of candidate molecules administered in combination with
39
antimalarial drugs. The concentration of the marinoquinoline derivatives corresponding to
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
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 IC values applied in combination with artesunate was first evaluated in a fixed ratio of
50
1:1; then, the fractional inhibitory concentration index (FIC index) was determined. The FIC
index was calculated for each fixed ratio. An FIC index = 1 indicates an additive drug
interaction, an FIC index < 1 indicates a synergistic interaction, and an FIC index > 1
40
indicates an antagonistic interaction. As shown in Table 8, the FIC indexes for the selected
inhibitors ranged from 0.9 ± 0.1 to 1.3 ± 0.1, suggesting an additive effect when applied in
combination with artesunate. Additionally, the isobologram plot for the representative
inhibitor 50 confirmed the additive interaction with artesunate (ΣFIC = 0.8 ± 0.2) (Figure
4B).
We selected the most potent compound, 50, and evaluated its inhibitory activity toward the P.
falciparum K1 strain, a multidrugꢀresistant strain that does not respond to the antimalarial
drugs chloroquine, pyrimethamine and sulfadoxine. Compound 50 was active at nanomolar
concentrations and effectively maintained its potency when tested against the resistant cell
K1
line (IC50 = 41 nM) (Figure 4C).
To further evaluate the antimalarial potential of this series, we evaluated the in vitro killing
profile by determining the capability of treated parasites to undergo new invasions. The 3D7
P. falciparum strain was cultivated in the presence of compound 50 at a 10x IC50
concentration for 24 and 48 h. The untreated parasites and standard antimalarials chloroquine,
pyrimethamine, atovaquone and artesunate were used as a control. Compound 50 exerted
effects on parasite viability that were comparable to the fastꢀacting agents chloroquine and
artesunate (Figure 4D and 4E). These results are highly consistent with the results of speed of
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 98
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
action studies, which indicated that compound 50 displayed inhibitory activity toward the
parasite at a very young stage in the intraerythrocytic cycle (Figure 4A). Since few
compounds are effective against the gametocyte stage of the parasite and thus prevent disease
41
transmission, further experiments were conducted using compound 50 to test whether the
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
compound was active against late stage (IV–V) gametocytes in a P. falciparum dual gamete
4
2
formation assay (DGFA). Compound 50 showed very low activity against male (inhibition
= 7 ± 1% at 1 ꢁM) and female (inhibition = 2.0 ± 0.4% at 1 ꢁM) gametocytes.
We also tested the in vitro prophylactic potential of compound 50 against the liver stages of
P. berghei (PbEEF), which precede symptomatic bloodꢀstage infection, using a luciferaseꢀ
43
expressing transgenic P. berghei line. The inhibitory potency of compound 50 against
PbEEF was similar to the potency determined for the erythrocytic form of P. falciparum,
Pb
namely, the compound was active in the nanomolar range against both forms (IC50 = 76
Pf
nM; IC50 = 39 nM, respectively). Moreover, the activity of compound 50 toward PbEEF
and is relation to HepG2 cytotoxicity indicates that the compound is highly selective for P.
berghei exoꢀerythrocytic stages (SI = 50), thereby emphasizing the safety profile of this
series.
Marinoquinoline Derivatives Interfere with Heme Species Sequestration
Resistance to antimalarial drugs is associated with mutations in the chloroquine resistance
transporter (PfCRT) and multidrug resistant protein (PfMDR1), both of which are located at
44,45
the digestive vacuole (DV) membrane.
The DV is a bilayer membrane compartment in P.
46
falciparum where host hemoglobin is metabolized. In the DV, free heme is sequestered as
inert crystalline hemozoin (Hz). In most instances, chloroquine (CQ) accumulates inside the
DV and interferes with Hz formation, leading to parasite death. Mutations in PfCRT impair
1
6
ACS Paragon Plus Environment

Page 17 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CQ accumulation, resulting in drug resistance. Concentrationꢀdependent assays designed to
monitor heme species (Hb, free heme and Hz) in both CQꢀsusceptible (CQS) and CQꢀ
47,48
resistant Plasmodium strains have shown distinct distribution patterns.
We performed a
calorimetric estimation of heme species following treatment with compound 50 to investigate
if this representative compound also possesses a similar mechanism of action as
aminoquinolines. In this assay, cellular fractionation was performed to collect the heme
species in the presence of pyridine and their distributions were determined
spectrophotometrically. Notably, Fe(III)PPIX forms a lowꢀspin complex with pyridine and
exhibits a very peculiar spectral property in UVꢀvisible spectroscopy, thus allowing the
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
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
49
estimation of heme species. Ringꢀstage parasites were incubated with compound 50 at 20–
160x IC50 doses for 32 h and then the heme fractionation assay was conducted. A final
analysis was conducted to estimate both the percentage of total heme and amount of heme in
the iron (Fe) form in single cells (Figure 6). Parasites of the control group showed a
significant reduction in levels of free heme (Figure 6A and 6D) and a simultaneous increase
in Hz accumulation (Figure 6B and 6E). On the other hand, parasites treated with compound
50 exhibited the opposite results for the distribution of total heme. Significant sequestration
of Hb was observed at higher concentrations (Figure 6C and 6F). Our results showed the
concentrationꢀdependent distribution of the percentage of heme species in parasites and
quantitation of the levels of the hemeꢀiron form per cell, thereby indicating that compound 50
interfered with hemoglobin metabolism. Nonetheless, the required concentration to inhibit
hemozoin formation was 100 times higher than the IC50 value, suggesting that this product is
not the main target of marinoquinolines.
Marinoquinoline Derivatives are Well Tolerated and Active in In Vivo Model
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 98
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
Motivated by the in vitro results, we tested the in vivo activity of representative
marinoquinoline derivatives (32, 43 and 50). We initially conducted in vivo toxicity assays to
assess unpredicted toxic effects of this series. The in vivo toxicity of compounds 32, 43 and
50 was evaluated in mice receiving the compounds administered by gavage or intraperitoneal
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
50
injections (3 mice per group). The inhibitors were administered at a single dose of 1000
mg/kg, and the survival rate was evaluated for 30 days of observation. The compounds were
well tolerated and did not display apparent signs of toxicity; all mice survived for up to 30
days (the last day of observation). Moreover, all animals gained weight in the same
proportion as untreated controls. Next, we tested the in vivo activity of marinoquinoline
51
inhibitors in P. bergheiꢀinfected mice (Figure 7A). Three infected mice were treated with
50 mg/kg of an inhibitor via the oral route (po) or intraperitoneal (ip) injections for three
consecutive days after infection. Parasitemia was evaluated on days 5, 7 and 9 postꢀinfection.
Compounds that reduced parasitemia by greater than 50% on day 5 postꢀinfection were
considered active. The antimalarial chloroquine (CQ) was used as positive control (20
mg/kg). Compound 50 reduced parasitemia by 60% on day 5 and 55% on day 9 postꢀ
infection via both administration routes (Figure 7B and 7C). Additionally, animals treated
with compound 50 showed increased survival (25 ± 3 days) compared to the untreated
controls (19 ± 5 days) (Table 9). Conversely, compounds 32 and 43 were not active.
To shed some light on the lack of in vivo activity for compounds 32 and 43, we conducted an
investigation of the metabolic stability of compounds 32, 43 and 50 in rat liver microsomes
(Figure 8). Compounds 32 and 43 displayed poor metabolic stability, similar to the verapamil
control, with levels of compounds remaining after 30 min of incubation as low as 24%, 15%
and 10%, respectively. On the other hand, nearly half of the initial amount of compound 50
was still detected after a 30ꢀmin incubation with rat liver microsomes, indicating that this
compound displayed a relatively higher metabolic stability. The poor metabolic stability and
1
8
ACS Paragon Plus Environment

Page 19 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Pf
Pf
modest in vitro potency of compounds 32 (IC50 = 0.5 µM) and 43 (IC50 = 0.6 µM) may
have contributed to the lack of in vivo activity (Figure 7B and 7C). On the other hand, the
combination of the structural and biological properties of compound 50, such as increased
Pf
metabolic stability, the enhanced in vitro potency (IC50 = 0.039 µM) and the presence of a
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
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
Pf
labile carbamate group, which generates compound 39 (IC50 = 1.3 µM) as a potential
pharmacologically active metabolite following hydrolysis, favorably contributed to the
observed in vivo efficacy (Figure 7B and 7C).
DISCUSSION AND CONCLUSIONS
In the present study, we used a combination of organic synthesis, SAR studies and molecular
modeling strategies to synthesize and identify marinoquinoline derivatives as new lead
candidates for antimalarial drug development. The best compounds showed very low
cytotoxicity, with an SI > 6140, and in vitro inhibitory activity against the erythrocytic form
of the parasite (sensitive and resistant strains) in the nanomolar range. Furthermore, these
compounds showed a fastꢀacting mechanism and potential additive effect when administered
in combination with artesunate. The representative marinoquinoline compound (50) interferes
with heme species sequestration at a concentration 100 times higher than the evaluated IC50
value, thereby suggesting that hemozoin metabolism is not the main target of
marinoquinolines. Moreover, compound 50 showed excellent correlations with the potencies
against asexual and hepatic (which do not display hemozoin metabolism) forms, indicating
that the mechanism of action is shared by the liver and blood forms. Inhibitors with dual
actions against liver and blood forms have been discovered for dihydroorotate dehydrogenase
52
53
(DHODH), cytochrome bc1 complex (cyt bc1), phosphatidylinositolꢀ4ꢀOH kinase
54
55
(
PI4K), and cyclic amine resistance locus (CARL) . These enzymes are validated
molecular targets for new antimalarials and attractive starting points to investigate the
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 98
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
mechanism of action underlying this series. In summary, the most promising compound
within this series (50) is potent against sensitive and resistant strains, well tolerated, and a
fastꢀacting inhibitor of parasite growth, with noticeable in vitro activity against both the early
stage of intraerythrocytic cycle and liver stages of development, and it has modest in vivo
activity. The discovery of compound 50 as a new chemotype will enable us to rationally
design derivatives with improved drugꢀlike properties aimed at developing the next
generation of antimalarial drugs.
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
EXPERIMENTAL SECTION
1. Chemistry. Synthesis of compounds
All reagents were commercially available and were used without further purification. Nuclear
magnetic resonance spectra were recorded in CDCl , (CD ) CO or MeOD solutions.
3
3 2
1
13
Hydrogen and Carbonꢀ13 spectra ( H and C NMR) were recorded at Bruker spectrometers
1
operating at 250, 400, 500, 600 MHz for H, using tetramethylsilane (TMS) as internal
standard. Data are reported as follows: chemical shift (δ), multiplicity, integrated intensity
and coupling constant (J) in Hertz. HRMS analysis was performed using an Agilent 6550
AccurateꢀMass QꢀTOF LC/MS system with Agilent Jet Stream technology for electrospray
ionization. Flash column chromatography was performed using silica gel (230ꢀ400 mesh) and
mixtures of EtOAc:hexanes; DCM:MeOH; EtOAc:MeOH. Thin layer chromatography (TLC)
®
was performed using silica gel employing Merck Silica gel 60 F254 plates. Airꢀ and
moistureꢀsensitive reactions were conducted in oven dried glassware under dry nitrogen
atmosphere, using freshly distilled solvents from the appropriate desiccant agent. Melting
points (mp) were determined on a capillary melting point apparatus, and are uncorrected. The
2
0
ACS Paragon Plus Environment

Page 21 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
chemical purity of compounds 32, 43 and 50 were confirmed by quantitative H NMR
(qNMR), with the values determined as 99,0%, 99,6%, and 98,9% respectively. All the other
1
compounds tested were confirmed to be ≥95% by H NMR.
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
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
General Produce A: Synthesis of 1ꢀiodoꢀ2ꢀnitrobenzene compounds from
2
9
nitroanilines: To a roundꢀbottomed flask containing boron trifluoride diethyl etherate (9.95
mL, 78.53 mmol) at ꢀ30 °C, it was added 2ꢀnitroaniline (18.67 mmol) dissolved in 50 mL of
THF and 90% tertꢀbutyl nitrite (8.17 mL, 68.72 mmol) dissolved in 50 mL of THF. The
reaction mixture was allowed to warm to ꢀ5 °C when 100 mL of diethyl ether was added and
the mixture was allowed to stir at ꢀ5 °C for 30 min until a solid precipitated. The solid was
filtered and washed with ether to provide the intermediate aryldiazonium tetrafluoroborate,
which was used in the next step without further purification.
The crude aryldiazonium salt was then added in one portion to a 500 mL round bottomed
flask containing a stir bar, potassium iodide (4.40 g, 26.53 mmol), iodine (3.37 g, 13.27
mmol), and 75 mL of acetonitrile. The reaction was allowed to stir at room temperature for
3
0 min., followed by the addition of 150 mL of a saturated aqueous solution of sodium
thiosulfate and 150 mL of CH Cl . The mixture was stirred for 5 min, the layers were
2
2
separated, and the combined organic layers were dried with anhydrous MgSO
was removed in vacuum to give the corresponding aryl iodides.
4
. The solvent
General Produce B: SuzukiꢀMiyaura Couplings: A roundꢀbottomed flask was charged
with a mixture of Pd(OAc) (0.013 g, 0.059 mmol, 0.5 mol%), SPhos (0.048 g, 0.119 mmol,
mol%), K PO (10.13 g, 47.70 mmol), pyrroleꢀborane 2 (5.00 g, 14.31 mmol), the
2
1
3
4
corresponding 2ꢀhaloanilines (11.92 mmol), and 17 mL of nꢀBuOH, 6.85 mL of H O. The
2
resulting mixture was then stirred at 80 °C for 12 h. The mixture was cooled to r.t. and
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 98
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
extracted with three portions (50 mL) of ethyl acetate. The organic layer was separated,
washed with brine, dried over sodium sulfate, filtered and the solvent was removed under
reduced pressure. The crude product was purified by column chromatography on silica gel
using 5 to 10% ethyl acetate:hexanes to provide analytical samples of the corresponding
arylpyrroles.
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
General procedure C: preparation of marinoquinolines analogues by PictetꢀSpengler
Reaction:
Method A: To a stirred solution of the aryl pyrroles (0.5 mmol) in 3.5 mL CH Cl it was
2
2
added the corresponding aldehyde (0.6 mmol) and TFA (0.25 mmol). The reaction mixture
was submitted to vigorous stirring for 1 h. Next, the crude reaction mixture was transferred to
a separatory funnel followed by addition of saturated NaHCO
3
until pH = 9 and extracted
with three portions of CH Cl (10 mL). The combined organic phases were dried over
2
2
anhydrous sodium sulfate, filtrated, and the filtrate was concentrated under reduced pressure.
The residue was then purified by flash chromatography using EtOAc/Hexanes,
2 2
EtOAc/CH Cl or EtOAc/MeOH as eluent to give the corresponding marinoquinolines in
good yields. Analytical samples were obtained from a middle cut from the purification
process by flash chromatography.
5
6
Method B: To a stirred solution of the aryl pyrroles (0.5 mmol) and benzotriazole (0.5
mmol) dissolved in 2.0 mL in dry toluene it was added the corresponding aldehyde (0.6
mmol), followed by addition of pꢀTsOH (10 mol %). The reaction mixture was then stirred
2 2
overnight at room temperature. After 12 h, the reaction mixture was diluted with CH Cl and
extracted with water. The organic layer was washed with brine and dried over anhydrous
sodium sulfate. The solvent was concentrated under reduced pressure and purified by flash
2
2
ACS Paragon Plus Environment

Page 23 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2 2
chromatography using EtOAc/Hexanes, EtOAc/CH Cl or EtOAc/MeOH as eluent to give
the corresponding marinoquinolines. Analytical samples were obtained from a middle cut
from the purification process by flash chromatography.
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
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
General procedure D: preparation of 1ꢀbromomarinoquinolines analogues: To a solution
2
of corresponding marinoquinoline (0.3mmol) in THF/H O 4:1 (10 mL) it was added NBS
(0.3 mmol). The reaction mixture was stirred vigorously at room temperature for 15 min.
Next, water (5 mL) and AcOEt (20 mL) were added and the reaction mixture extracted with
two portions of AcOEt. The organic layers were combined, dried over anhydrous sodium
sulfate, and concentrated under reduced pressure. The crude product was purified via flash
column chromatography on silica gel using hexaneꢀethyl acetate as eluent to afford the
corresponding products.
2ꢀ(1Hꢀpyrrolꢀ3ꢀyl)aniline (3): Prepared using the general procedure B starting from 2ꢀ
bromoaniline (1) (4.10 g, 23.85 mmol) to afforded 2ꢀ(1ꢀ(triisopropylsilyl)ꢀ1Hꢀpyrrolꢀ3ꢀ
1
yl)aniline (7.04 g, 94% yield) as a yellow oil. H NMR (400 MHz, CD
Cl
2 2
) δ 7.22 (dd, J =
8
6
1
.1, 1.4 Hz, 1H), 7.02 (td, J = 7.7, 1.6 Hz, 1H), 6.98 – 6.97 (m, 1H), 6.88 (t, J = 2.4 Hz, 1H),
.74 (dd, J = 11.3, 4.4 Hz, 1H), 6.51 (dd, J = 2.6, 1.4 Hz, 1H), 3.98 (s, 1H), 1.50 (sept, J =
13
4.7, 7.4 Hz, 3H), 1.14 (d, J = 7.5 Hz, 18H). C NMR (100 MHz, CD Cl ) δ 144.51, 130.07,
2
2
127.41, 125.40, 124.52, 122.90, 122.81, 118.81, 115.82, 111.10, 18.20, 12.28. HRMS for
+
C
19
H
30
N
2
Si (M + H) calcd 315.2251, found 315.2269.
An oven dried 100 mL flask was charged with the crude product prepared in the previous
reaction (7.04 g, 22.38 mmol). The flask was capped with a rubber septum and then
evacuated and backfilled with nitrogen. Anhydrous THF (90 mL) was added via syringe,
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 98
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
followed by the addition of 1.0 M TBAF in THF (33.57 mL, 33.57 mmol) at room
temperature. After 3 h, it was added 50 mL of water to quench the reaction followed by
2 2
extraction with three portions of 50 mL of CH Cl . The combined organic layers were dried
over anhydrous sodium sulfate and concentrated under reduced pressure. The crude product
was purified via flash column chromatography on silica gel (10 ꢀ 40% EtOAc/Hexanes) to
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
afforded 2ꢀ(1Hꢀpyrrolꢀ3ꢀyl)aniline 3 (3.01 g, 85% yield) as a green oil. H NMR (250 MHz,
CDCl
1H); 6.82ꢀ6.67 (m, 3H); 6.39 (m, 1H); 3.89 (br s, 2H). C NMR (62.5 MHz, CDCl
43.6, 129.8, 127.1, 122.3, 121.6, 122.3, 121.6, 118.6, 118.4, 116.1, 115.5, 108.4. HRMS for
3
) δ 8.32 (br s, 1H); 7.22 (dd, 1H, J=7.5 and 1.3); 7.04 (td, 1H, J=7.8 and 1.3); 6.87 (m,
1
3
3
): δ
1
+
C H N (M + H) calcd 159.0917, found 159.0918.
10
10
2
4ꢀMethylꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (Marinoquinoline A ꢀ 4). Prepared using general
procedure C (Method A) from compound 3 and acetaldehyde. The crude product was purified
by chromatography on silica gel (40% EtOAc/Hexanes) to provide the compound
marinoquinoline A (4) (47.3 mg, 52% yield) as a white solid. Mp 235ꢀ236 ºC; lit. 233ꢀ235
57
ºC.
1
H NMR (500 MHz, (CD
3 2
) CO) δ 11.18 (br s, 1H), 8.24 – 8.20 (m, 1H), 8.02 – 7.98 (m, 1H),
13
7
.58 (d, 1H, J = 2.5), 7.55 – 7.45 (m, 2H), 7.12 (d, 1H, J = 3.0), 2.83 (s, 3H). C NMR (125
MHz, (CD ) CO) δ 146.9, 143.7, 129.8, 128.4, 127.2, 126.1, 125.7, 124.2, 123.7, 102.0, 21.2.
3 2
+
HRMS for C H N (M + H) calcd 183.0917, found 183.0917.
12
10
2
4ꢀisobutylꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (Marinoquinoline B ꢀ 5): Prepared using general
procedure C (Method A) from compound 3 and isovaleraldehyde. The crude product was
purified by chromatography on silica gel (30% EtOAc/Hexanes) to provide marinoquinoline
1
B (5) (61.7 mg, 55% yield) as a white solid. Mp 205ꢀ206 ºC; lit. 205ꢀ207 ºC. H NMR (500
2
4
ACS Paragon Plus Environment

Page 25 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MHz, (CD
3
)
2
CO) δ 11.16 (br s, 1H), 8.25 – 8.21 (m, 1H), 8.04 – 8.01 (m, 1H), 7.56 (t, 1H, J
= 2.5), 7.54 – 7.47 (m, 2H), 7.12 (dd, 1H, J = 2.5 and 1.5), 3.07 (d, 2H, J = 7.0), 2.45
1
3
(septuplet, 1H, J = 7.0), 1.00 (d, 6H, J = 6.5). C NMR (125 MHz, (CD
3 2
) CO) δ 150.1,
143.7, 130.1, 129.9, 128.6, 127.0, 126.0, 125.6, 124.1, 123.7, 101.9, 44.1, 29.0, 23.0. HRMS
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
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
+
for C H N (M + H) calcd 225.1386, found 225.1397.
15
16
2
4ꢀbenzylꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (Marinoquinoline C ꢀ 6): Prepared using general
procedure C (Method A) from compound 3 and 2ꢀphenylacetaldehyde. The crude product
was purified by chromatography on silica gel (30% EtOAc/Hexanes) to provide the
marinoquinoline C (6) (76.2 mg, 59% yield) as a white solid. Mp 180ꢀ181 ºC; lit. 179ꢀ181 ºC.
1
H NMR (500 MHz, (CD
3
)
2
CO) δ 11.05 (br s, 1H), 8.26 – 8.22 (m, 1H), 8.08 – 8.04 (m, 1H),
7
2
.56 – 7.49 (m, 3H), 7.42 (d, 2H, J = 7.0), 7.24 (t, 2H, J = 7.0), 7.18 – 7.11 (m, 2H), 4.56 (s,
1
3
H). C NMR (125 MHz, (CD
3
)
2
CO) δ 148.9, 143.7, 139.9, 130.2, 129.7, 129.3, 129.3,
+
129.1, 127.4, 127.0, 126.2, 125.9, 124.2, 123.7, 102.1, 42.6. HRMS for C H N (M + H)
18
14
2
calcd 259,1230, found 259.1240.
4ꢀethylꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (7): Prepared using general procedure C (Method A)
from compound 3 and propanal. The crude product was purified by chromatography on silica
gel (40% EtOAc/Hexanes) to provide the marinoquinoline 7 (59.8 mg, 61% yield) as a
1
yellow solid. Mp 184ꢀ186 ºC; lit. 185ꢀ186 ºC. H NMR (500 MHz, (CD ) CO) δ 11,12 (sl,
3
2
1H), 8.24 – 8.20 (m, 1H), 8.04 – 8.00 (m, 1H), 7.56 (t, 1H, J = 3.0), 7.54 – 7.47 (m, 2H), 7.12
13
(dd, 1H, J = 3.0 and 2.0), 3.21 (qt, 2H, J = 7.5), 1.45 (t, 3H, J = 7.5). C NMR (125 MHz,
(CD
3
)
2
CO) δ 151.5, 143.7, 130.0, 129.0, 128.6, 126.9, 126.0, 125.6, 124.1, 123.7, 102.0,
+
28.1, 12.6. HRMS for C H N (M + H) calcd 197.1073, found 197.1079.
13
12
2
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 98
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
4ꢀphenylꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (8): Prepared using general procedure C (Method A)
from compound 3 and benzaldehyde. The crude product was purified by chromatography on
silica gel (25% EtOAc/Hexanes) to provide the marinoquinoline 8 (64.7 mg, 53% yield) as a
1
white solid. Mp 211ꢀ212 ºC; lit. 212ꢀ214 ºC. H NMR (500 MHz, (CD ) CO) δ 11.09 (br s,
3
2
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
1H), 8.34 – 8.30 (m, 1H), 8.16 – 8.12 (m, 1H), 8.12 – 8.08 (m, 2H), 7.65 (t, 1H, J = 2.5), 7.62
1
3
–
3 2
7.51 (m, 5H), 7.24 (dd, 1H, J = 2.5 and 1.5). C NMR (125 MHz, (CD ) CO) δ 148.1,
144.6, 140.5, 131.4, 131.0, 130.6, 130.3, 130.3, 129.0, 128.9, 128.9, 127.3, 127.2, 125.0,
+
124.5. HRMS for C17
H
12
N
2
(M + H) calcd 245.1073, found 245.1085.
4ꢀ(4ꢀmethoxyphenyl)ꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (9): Prepared using general procedure C
(Method A) from compound 6 and 4ꢀmethoxybenzaldehyde. The crude product was purified
by chromatography on silica gel (25% EtOAc/Hexanes) to provide the marinoquinoline 9
1
(
86.4 mg, 63% yield) as a white solid. Mp 173ꢀ175 ºC; lit. 173ꢀ174 ºC. H NMR (600 MHz,
(CD ) CO) δ 11.09 (br s, 1H), 8.3 – 8.27 (m, 1H), 8.12 – 8.09 (m, 1H), 8.06 (distorted d, 2H,
3
2
J~9.0), 7.63 (t, 1H, J = 3.0), 7.59 – 7.52 (m, 2H), 7.21 (dd, 1H, J = 3.0 and 1.8), 7.14 – 7.10
13
(
distorted d, 2H, J ~ 8.4), 3.89 (s, 3H). C NMR (150 MHz, (CD
3
2
) CO) δ 160.6, 146.2,
1
5
43.1, 131.3, 130.0, 129.6, 129.3, 127.2, 127.2, 125.6, 125.2, 123.2, 122.8, 114.0, 101.2,
+
14 2
4.9. HRMS for C18H N O (M + H) calcd 275.1179, found 275.1180.
4ꢀ(3,5ꢀdimethoxyphenyl)ꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (10): Prepared using general
procedure C (Method A) from compound 3 and 3,5ꢀdimethoxybenzaldehyde. The crude
product was purified by chromatography on silica gel (30% EtOAc/Hexanes) to provide
marinoquinoline 10 (115.6 mg, 76% yield) as a brown solid. Mp 182ꢀ183 ºC; lit. 183ꢀ184 ºC.
1
H NMR (500 MHz, (CD ) CO) δ 11.18 (br s, 1H), 8.33 – 8.28 (m, 1H), 8.16 – 8.11 (m, 1H),
3
2
7.64 (distorted t, 1H, J ~ 2.5), 7.61 (m, 2H), 7.23 (dd, 1H, J = 3.0 and 2.0), 7.19 (d, 2H, J =
2
6
ACS Paragon Plus Environment

Page 27 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
3
3 2
2.0), 6.63 (t, 1H, J = 2.5), 3.87 (s, 6H). C NMR (125 MHz, (CD ) CO) δ 162.1, 147.4,
143.7, 141.7, 130.6, 130.2, 128.2, 126.5, 126.4, 124.3, 123.8, 107.4, 102.1, 101.8, 55.7.
+
HRMS for C19
H
16
N
2
O
2
(M + H) calcd 305.1285, found 305.1298.
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
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
4
ꢀ(3,4,5ꢀtrimethoxyphenyl)ꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (11): Prepared using general
procedure C (Method A) from compound 3 and 3,4,5ꢀtrimethoxybenzaldehyde. The crude
product was purified by chromatography on silica gel (35% EtOAc/Hexanes) to provide the
marinoquinoline 11 (120 mg, 72% yield) as a yellow solid. Mp 200ꢀ201 ºC; lit. 199ꢀ201 ºC.
1
H NMR (400 MHz, CD OD) δ 8.60 – 8.55 (m, 1H), 8.26 – 8.22 (m, 1H), 8.21 (d, 1H, J =
3
2
.8), 7.88 (quintet of doublets, 2H, J = 7.6 and 1.6), 7.56 (d, 1H, J = 2.8), 7.34 (s, 2H), 4.00
13
(
s, 6H); 3.93 (s, 3H). C NMR (100 MHz, CD
3
OD) δ 155.5, 145.0, 142.8, 139.3, 136.2,
1
34.6, 131.0, 129.5, 126.8, 125.9, 125.5, 123.1, 121.5, 108.3, 105.1, 61.3, 57.0. HRMS for
+
C
20
H
18
N
2
O
3
(M + H) calcd 335.1390, found 335.1396.
4ꢀ(4ꢀchlorophenyl)ꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (12): Prepared using general procedure C
Method A) from compound 3 and 4ꢀchlorobenzaldehyde. The crude product was purified by
(
chromatography on silica gel (20% EtOAc/Hexanes) to provide the marinoquinoline 12 (90.6
1
mg, 65% yield) as a yellow solid. Mp 181ꢀ182 ºC; lit. 182ꢀ184 ºC. H NMR (500 MHz,
(
CD ) CO) δ 11.15 (s, 1H), 8.35 – 8.30 (m, 1H), 8.15 – 8.09 (m, 3H), 7.67 (t, 1H, J = 2.5),
3 2
1
3
7
3 2
.64 – 7.56 (m, 4H), 7.25 (dd, 1H, J = 3.0 and 2.0). C NMR (125 MHz, (CD ) CO) δ 146.0,
143.8, 138.5, 135.4, 131.2, 130.7, 130.5, 129.7, 128.4, 128.0, 126.7, 126.6, 124.3, 123.8,
+
102.3. HRMS for C17
H
11ClN
2
(M + H) calcd 279.0683, found 279.0692.
4
ꢀ(4ꢀbromophenyl)ꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (13): Prepared using general procedure C
(
Method A) from compound 3 and 4ꢀbromobenzaldehyde. The crude product was purified by
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 98
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
chromatography on silica gel (20% EtOAc/Hexanes) to provide the marinoquinoline 13 (96.9
1
mg, 60% yield) as a yellow solid. Mp 168ꢀ169 ºC; lit. 168ꢀ170 ºC. H NMR (600 MHz,
(CD
3 2
) CO) δ 11,16 (sl, 1H), 8.34 – 8.30 (m, 1H), 8.15 – 8.11 (m, 1H), 8.04 (distorted d, 2H, J
~
8.4), 7.76 (distorted d, 2H, J ~ 8.4), 7.55 (t, 1H, J = 2.4), 7.62 – 7.57 (m, 2H), 7.25 (dd, 1H,
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
13
J = 3.0 and 1.8). C NMR (150 MHz, (CD ) CO) δ 145.2, 142.9, 138.0, 131.8, 130.6, 129.7,
3
2
2
129.6, 127.6, 127.1, 125.8, 125.8, 123.4, 123.0, 122.8, 101.4. HRMS for C17H11BrN (M +
+
H) calcd 323.0178, found 323.0185.
4ꢀ(4ꢀnitrophenyl)ꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (14): Prepared using general procedure C
(Method A) starting from compound 3 and 4ꢀnitrobenzaldehyde. The crude product was
purified by chromatography on silica gel (20% EtOAc/Hexanes) to provide the
marinoquinoline 14 (89.7 mg, 62% yield) as a yellow solid. Mp 220ꢀ222 ºC; lit. 220ꢀ221 ºC.
1
3 2
H NMR (600 MHz, (CD ) CO) δ 11,33 (sl, 1H), 8.41 (d, 2H, J = 8.7), 8.37 – 8.33 (m, 3H),
8.18 – 8.14 (m, 1H), 7.70 (t, 1H, J = 3.0), 7.65 – 7.60 (m, 2H), 7.28 (dd, 1H, J = 3.0 and 1.8).
1
3
C NMR (150 MHz, (CD ) CO) δ 149.0, 145.8, 144.9, 143.7, 130.9, 130.8, 130.7, 128.9,
3
2
+
1
28.0, 127.1, 127.0, 124.6, 124.5, 123.9, 102.4. HRMS for C17
90.0924, found 290.0919.
11 3 2
H N O (M + H) calcd
2
4ꢀ(3Hꢀpyrrolo[2,3ꢀc]quinolinꢀ4ꢀyl)aniline (15): Prepared by catalytic reduction of 4ꢀ(4ꢀ
nitrophenyl)ꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline (14) by the following procedure: To a stirred
solution of (4ꢀ(4ꢀnitrophenyl)ꢀ3Hꢀpyrrolo[2,3ꢀc]quinoline) (115.7 mg, 0.4 mmol) in MeOH
(4.0 mL) it was added 42.5 mg of Pd/C 10% (10 mol% of Pd). The reaction mixture was
purged with nitrogen for about 10 minutes and then maintained under H atmosphere (balloon
2
of hydrogen) for 6 h. The reaction mixture was then filtered through a short pad of Celite and
washed with three portions of MeOH (10 mL). The filtrate was concentrated under reduced
2
8
ACS Paragon Plus Environment

Page 29 of 98
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pressure and the residue was purified by flash chromatography (5% MeOH/CH
2
Cl
2
) to
1
furnish marinoquinoline 15 (89.19 mg, 86% yield) as a yellow solid. Mp 193ꢀ194 ºC. H
NMR (500 MHz, CDCl
3
) δ 7.00 (s, 1H), 6.27 (t, J = 6.7 Hz, 2H), 5.87 (d, J = 8.3 Hz, 2H),
13
5.69 – 5.58 (m, 2H), 5.48 (s, 1H), 5.18 (s, 1H), 4.93 (d, J = 8.4 Hz, 2H), 1.95 (s, 2H). C
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
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
NMR (125 MHz, DMSO) δ 149.99, 146.72, 142.31, 129.62, 128.80, 128.56, 127.55, 126.42,
+
125.49, 124.61, 122.86, 122.54, 113.66, 100.99. HRMS for C17
H
13
N
3
(M + H) calcd
260.1182, found 260.1170.
N,Nꢀdimethylꢀ4ꢀ(3Hꢀpyrrolo[2,3ꢀc]quinolinꢀ4ꢀyl)aniline (16): Prepared using general
procedure C (Method A) starting from compound 3 and 4ꢀ(dimethylamino)benzaldehyde.
The crude product was purified by chromatography on silica gel (40% EtOAc/Hexanes) to
provide marinoquinoline 17 (107.6 mg, 75% yield) as a red solid. Mp 172ꢀ173 ºC, lit 173ꢀ175
1
ºC. H NMR (600 MHz, (CD
3
)
2
CO) δ 11.05 (br s, 1H), 8.26 (dd, 1H, J = 7.8 and 1.2), 8.09
(dd, 1H, J = 7.8 and 0.6), 8.01 (d, 2H, J = 9.0), 7.60 (d, 1H, J = 2.4), 7.57 – 7.48 (m, 2H),
1
3
7
1
1
.18 (d, 1H, J = 2.4), 6.88 (d, 2H, J = 8.4), 3.03 (s, 6H). C NMR (150 MHz, (CD ) CO) δ
3 2
52.2, 147.7, 144.1, 130.4, 130.3, 129.9, 128.1, 127.7, 127.4, 126.3, 125.6, 124.0, 123.6,
+
12.9, 102.0, 40.4. HRMS for C19
H
17
N
3
(M + H) calcd 288.1495, found 288.1496.
4ꢀ(3Hꢀpyrrolo[2,3ꢀc]quinolinꢀ4ꢀyl) benzoic acid (17): Prepared using general procedure C
(Method A) starting from compound 3 and 4ꢀformylbenzoic acid. The crude product was
purified by chromatography on silica gel (40% EtOAc/Hexanes) to provide marinoquinoline
1
1
7 (76.3 mg, 53% yield) as a yellow solid. Mp 223ꢀ224 ºC. H NMR (500 MHz, DMSO) δ
13.32 (s, 1H), 12.59 (s, 1H), 8.52 (d, J = 6.4 Hz, 1H), 8.24 (d, J = 8.3 Hz, 2H), 8.22 – 8.19
13
(
m, 1H), 8.16 (d, J = 8.3 Hz, 2H), 8.04 (s, 1H), 7.82 – 7.73 (m, 2H), 7.49 (s, 1H). C NMR
100 MHz, DMSO) δ 166.9, 143.67, 137.45, 133.85, 132.55, 131.79, 129.77, 127.84, 127.10,
(
2
9
ACS Paragon Plus Environment