Improvement of antimalarial activity of a 3-alkylpiridine alkaloid analog by replacing the pyridine ring to a thiazole-containing heterocycle: Mode of action, mutagenicity profile, and Caco-2 cell-based permeability

Article abstract of DOI:10.1016/j.ejps.2019.105015

The development of new antimalarial drugs is urgent to overcome the spread of resistance to the current treatment. Herein we synthesized the compound 3, a hit-to?lead optimization of a thiazole based on the most promising 3-alkylpyridine marine alkaloid analog. Compound 3 was tested against Plasmodium falciparum and has shown to be more potent than its precursor (IC₅₀ values of 1.55 and 14.7 μM, respectively), with higher selectivity index (74.7) for noncancerous human cell line. This compound was not mutagenic and showed genotoxicity only at concentrations four-fold higher than its IC₅₀. Compound 3 was tested in vivo against Plasmodium berghei NK65 strain and inhibited the development of parasite at 50 mg/kg. In silico and UV–vis approaches determined that compound 3 acts impairing hemozoin crystallization and confocal microscopy experiments corroborate these findings as the compound was capable of diminishing food vacuole acidity. The assay of uptake using human intestinal Caco-2 cell line showed that compound 3 is absorbed similarly to chloroquine, a standard antimalarial agent. Therefore, we present here compound 3 as a potent new lead antimalarial compound.

Full text of DOI:10.1016/j.ejps.2019.105015

Accepted Manuscript
Improvement of antimalarial activity of a 3-alkylpiridine alkaloid
analog by replacing the pyridine ring to a thiazole-containing
heterocycle: Mode of action, mutagenicity profile, and Caco-2
cell-based permeability
Daniel Silqueira Martins Guimarães, Letícia Silveira de Sousa
Luz, Sara Batista do Nascimento, Lorena Rabelo Silva, Natália
Rezende de Miranda Martins, Heloísa Gonçalves de Almeida,
Vitória de Souza Reis, Sarah El Chamy Maluf, Alexandre Budu,
Juliane Aparecida Marinho, Clarice Abramo, Adriana
Karaoglanovic Carmona, Marina Goulart da Silva, Gisele
Rodrigues da Silva, Victor Matheus Kemmer, Anna Paola Butera,
Renato Márcio Ribeiro-Viana, Marcos Leoni Gazarini, Clébio
Soares Nascimento Júnior, Luciana Guimarães, Fabio Vieira dos
Santos, Whocely Victor de Castro, Gustavo Henrique Ribeiro
Viana, Cristiana Ferreira Alves de Brito, Fernando de Pilla Varotti
PII:
DOI:
Article Number:
Reference:
S0928-0987(19)30279-9
https://doi.org/10.1016/j.ejps.2019.105015
105015
PHASCI 105015
To appear in:
European Journal of Pharmaceutical Sciences
Received date:
Revised date:
Accepted date:
10 June 2019
11 July 2019
20 July 2019
Please cite this article as: D.S.M. Guimarães, L.S. de Sousa Luz, S.B. do Nascimento, et
al., Improvement of antimalarial activity of a 3-alkylpiridine alkaloid analog by replacing
the pyridine ring to a thiazole-containing heterocycle: Mode of action, mutagenicity
profile, and Caco-2 cell-based permeability, European Journal of Pharmaceutical Sciences,
https://doi.org/10.1016/j.ejps.2019.105015
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may

be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Improvement of antimalarial activity of a 3-alkylpiridine alkaloid analog by
replacing the pyridine ring to a thiazole-containing heterocycle: mode of action,
mutagenicity profile, and Caco-2 cell-based permeability.
Daniel Silqueira Martins Guimarãesa, Letícia Silveira de Sousa Luzb, Sara Batista do
Nascimentob, Lorena Rabelo Silvab, Natália Rezende de Miranda Martinsb, Heloísa
Gonçalves de Almeidac, Vitória de Souza Reisc, Sarah El Chamy Malufd, Alexandre
Budud, Juliane Aparecida Marinhoe, Clarice Abramoe, Adriana Karaoglanovic
Carmonad, Marina Goulart da Silvab, Gisele Rodrigues da Silvaf, Victor Matheus
Kemmeri, Anna Paola Buterai, Renato Márcio Ribeiro-Vianah, Marcos Leoni Gazarinig,
Clébio Soares Nascimento Júniorc, Luciana Guimarãesc, Fabio Vieira dos Santosb,
Whocely Victor de Castrob, Gustavo Henrique Ribeiro Vianab, Cristiana Ferreira Alves
de Britoa and Fernando de Pilla Varottib*
aInstituto René Rachou, Fundação Oswaldo Cruz (FIOCRUZ),
1715 Augusto de Lima Avenue, Belo Horizonte, MG 30190-009, Brazil.
bNúcleo de Pesquisa em Química Biológica, Universidade Federal de São João Del-
Rei – Campus Centro Oeste, 400 Sebastião Gonçalves Coelho Street, Divinópolis, MG,
35501-296, Brazil.
cUniversidade Federal de São João del-Rei, Campus Dom Bosco, 74 Dom Helvécio
Square, São João del Rei, MG, 36301-160, Brazil.
dUniversidade Federal de São Paulo, Departamento de Biofísica, 669 Pedro de Toledo
Street, São Paulo, SP, 04039-032, Brazil.
eNúcleo de Pesquisas em Parasitologia, Universidade Federal de Juiz de Fora, José
Lourenço Kelmer Street, Juiz de Fora, MG 36036-900, Brazil.
fUniversidade Federal de Ouro Preto, Departamento de Farmácia, Campus Morro do
Cruzeiro, w/n, Bauxita, Ouro Preto, MG 35400-000, Brazil.
gUniversidade Federal de São Paulo, Departamento de Biociências, 136 Silva Jardim
Street, Santos, SP 11015-020, Brazil.
hUniversidade Tecnológica Federal do Paraná, Departamento Acadêmico de Química
(DAQUI) Londrina, PR, 6036-370, Brazil.
iUniversidade Estadual de Londrina, Departamento de Química, Londrina, PR, 86057-
970, Brazil
* Corresponding author
Daniel Silqueira Martins Guimarães: danielmartinsguimaraes@gmail.com
Letícia Silveira de Sousa Luz: leticia_slux@hotmail.com
Sara Batista do Nascimento: sara.bnas2@hotmail.com
Lorena Rabelo Silva: lorenarabelo6@gmail.com

ACCEPTED MANUSCRIPT
Natália Rezende de Miranda Martins: naty-ande@hotmail.com
Heloísa Gonçalves de Almeida: heloisa.almeidag@outlook.com
Vitória de Souza Reis: souzareisvitoria@gmail.com
Sarah El Chamy Maluf: sarah.maluf08@gmail.com
Alexandre Budu: alexandre.budu@unifesp.br
Juliane Aparecida Marinho: jumarinho28@yahoo.com.br
Clarice Abramo: clarice.abramo@ufjf.edu.br
Adriana Karaoglanovic Carmona: ak.carmona@unifesp.br
Marina Goulart da Silva: marina@ufsj.edu.br
Gisele Rodrigues da Silva: giselersilva@ufop.edu.br
Marcos Leoni Gazarini: marcos.gazarini@unifesp.br
Clébio Soares Nascimento Júnior: clebiojunior@gmail.com
Luciana Guimarães: lucianaguimaraes8@gmail.com
Renato Márcio Ribeiro-Viana: renatoviana@utfpr.edu.br
Anna Paola Butera: paolabut@uel.br
Victor Matheus Kemmer: victorkemmer@gmail.com
Fabio Vieira dos Santos: fabiosantos@ufsj.edu.br
Whocely Victor de Castro: whocely@ufsj.edu.br
Gustavo Henrique Ribeiro Viana: viana@ufsj.edu.br
Cristiana Ferreira Alves de Brito: cristiana.brito@fiocruz.br
Fernando de Pilla Varotti: varotti@ufsj.edu.br
Declarations of interest: none

ACCEPTED MANUSCRIPT
Abstract
The development of new antimalarial drugs is urgent to overcome the spread of
resistance to the current treatment. Herein we synthesized the compound 3, a hit-to-lead
optimization of a thiazole based on the most promising 3-alkylpyridine marine alkaloid
analog. Compound 3 was tested against Plasmodium falciparum and has shown to be
more potent than its precursor (IC50 values of 1.55 and 14.7 µM, respectively), with
higher selectivity index (74.7) for noncancerous human cell line. This compound was
not mutagenic and showed genotoxicity only at concentrations four-fold higher than its
IC50. Compound 3 was tested in vivo against Plasmodium berghei NK65 strain and
inhibited the development of parasite at 50 mg/kg. In silico and UV-vis approaches
determined that compound 3 acts impairing hemozoin crystallization and confocal
microscopy experiments corroborate these findings as the compound was capable of
diminishing food vacuole acidity. The assay of uptake using human intestinal Caco-2
cell line showed that compound 3 is absorbed similarly to chloroquine, a standard
antimalarial agent. Therefore, we present here compound 3 as a potent new lead
antimalarial compound.
Keywords
Plasmodium falciparum, 3-alkylpyridine marine alkaloid analogs, antiplasmodial
activity, thiazole, malaria, ferriprotoporphyrin-IX.
Abbreviations
3-APA: 3-alkylpyridine marine alkaloid
ACTs: Artemisinin-based Combination Therapies
ADMET: absorption, distribution, metabolism, excretion and toxicity
AO: Acridine Orange
AS: Artesunate
CBMN: Cytokinesis-Block Micronucleus
CQ: Chloroquine
DAPI: 4',6-diamidino-2-phenylindole
DFT: Density Functional Theory
DMSO: Dimethyl sulfoxide
ELISA: Enzyme-Linked Immunosorbent Assay
FBS: Fetal Bovine Serum

ACCEPTED MANUSCRIPT
Fe(III)PPIX: Ferriprotoporphyrin-IX
HBSS:Hanks' Balanced Salt solution
HPLC: High-performance liquid chromatography
HRMS: High Resolution Mass Spectrometry
HRPII: Histidine-Rich Protein II
IC50: Inhibitory concentration 50%
IR: Infrared
MMS: Methyl methanesulfonate
MQ: Mefloquine
MTT: 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
NDI: Nuclear Division Index
NMR: Nuclear magnetic resonance
OD: Optical Density
RBC: Red Blood Cell
SD: Standard Deviation
SI: Selectivity Index
PBS: Phosphate Buffered Saline
PMSF: phenylmethylsulfonyl fluoride
TM: Transport Medium
TMS: tetramethylsilane
1. Introduction
More than 3 billion people live in areas with potential risk of malaria
transmission, the most important mosquito-borne infectious disease (Ferrao et al., 2018;
Walker et al., 2018). The World Health Organization estimated that, in 2017 occurred
throughout the world 219 million malaria cases resulting in approximately 435,000
deaths. Five species of Plasmodium commonly infect humans, P. falciparum, P. vivax,
P. ovale, P. malariae and P. knowlesi. Among these species P. falciparum is the most
virulent being responsible for more than 90% of the cases and deaths (WHO, 2018). In
the absence of an efficient vaccine, malaria treatment relies on Artemisinin-based
Combination Therapies (ACTs) (Martin et al., 2018; Nsanzabana, 2019). However, the

ACCEPTED MANUSCRIPT
burden and widespread resistance of parasites to the current antimalarial drugs impair
the elimination of the disease (Martin et al., 2018). In this scenario, the need for the
development of novel effective drugs against resistant P. falciparum strains is of
paramount importance (Mishra et al., 2017).
The determination of new active antimalarial compounds is not always associated
with the description of their potential targets (Lobo et al., 2018; Thakur et al., 2018).
The lack of knowledge about the mode of action of a new antimalarial compound can
negatively affect preclinical tests and hinder the employment of rational strategies to
improve compound potency and selectivity (Skinner-Adams et al., 2016).
Taking into account all these requirements, our research group has designed and
synthesized new 3-alkylpyridine marine alkaloid (3-APA) analogs with antiplasmodial
activity (Barbosa et al., 2017; Hilário et al., 2011; Ribeiro-Viana et al., 2016). Based on
rational approaches, combining in silico and biophysical data, we have demonstrated
that these 3-APA analogs were able to form stable complexes with ferriprotoporphyrin-
IX (Barbosa et al., 2017; Ribeiro-Viana et al., 2016).
The inhibition of hemozoin formation in malarial parasites remains an attractive
target for the development of new antimalarial drugs (Ashley and Phyo, 2018). During
the erythrocytic stage, malaria parasites use amino acids resulting from the breakdown
of hemoglobin (Hb) present in red blood cells for its growth; however, the free heme
(ferriprotoporphyrin-IX) generated in this process is highly toxic to the parasite.
Therefore, heme polymerization is a detoxification strategy adopted by the malaria
protozoan to convert toxic heme into a non-toxic polymerized crystal called hemozoin
(Egan, 2008). This pathway is vital for parasite survival and the inhibition of heme
polymerization results in the death of the protozoan (Singh et al., 2014).

ACCEPTED MANUSCRIPT
In the specific context of malaria, heterocycles play a critical role in their
antiplasmodial activity according to a recent literature review (Kalaria et al., 2018).
Many heterocyclic compounds have high antimalarial activity and better ADMET
properties in pre-clinical studies. Furthermore, we recently reported that the pyridine
ring is critical to the antiplasmodial activity of 3-APA analogs (Barbosa et al., 2017;
Ribeiro-Viana et al., 2016). However, the pyridine heterocycle is known to be toxic to
living organisms (Lodha et al., 2008; Mudliar et al., 2008; Padoley et al., 2006).
Due to practical, patient compliance, cost-effectiveness, and safety reasons
antimalarial drugs are preferably developed as solid oral dosage forms. It means that the
drug compound must fulfill physicochemical requirements to achieve adequate
bioavailability, which is driven by dissolution, permeability, solubility, absorption,
metabolism, and excretion (Zhu et al., 2017).
However, many potential new antimalarial candidates described in drug discovery
process find a pitfall in such properties impairing or simply preventing the drug
candidate to be administered orally (Mishra et al., 2017). To overcome these limitations,
this work describes the optimization of a previously synthesized 3-APA which has its
antimalarial activity enhanced by replacing the pyridine ring to a thiazole heterocycle
providing a rational scaffold for new antimalarials candidates.
2. Material and Methods
2.1. Chemistry
Reagents and solvents were purchased as reagent grade from Sigma-Aldrich and used
without further purification. NMR spectra were recorded using Bruker Avance DRX-
200. Chemical shifts are reported as δ (ppm) downfield from tetramethylsilane (TMS),

ACCEPTED MANUSCRIPT
and the J values are reported in Hz. IR spectra were recorded using a Shimadzu
IRAffinity-1 Fourier transform spectrometer. High-resolution mass spectrometry
(HRMS) was recorded using an ESI micrOTOF-QII Bruker mass spectrometer. Column
chromatography was performed with silica gel 60, 70–230 mesh (Merck, Darmstadt,
Germany). Interaction studies were carried out in a UV-vis spectrophotometer
(Shimadzu, Model UV-2600) equipped with a thermoregulator (Peltier, Model S-1700)
at 25 °C, using 1 cm path cuvettes.
2.2. Synthetic routes
2.2.1. Synthesis of 10-((4,5-dihydrothiazol-2-yl)thio)decan-1-ol 3
Scheme 1: Synthetic route to 10-(4,5-dihydrothiazol-2-yl)thio)decan-1-ol 3) (1).
2.2.2. Synthesis of 10-bromo-decan-1-ol 2
Hydrobromic acid (1.2 mL 48% aqueous solution, 9.5 mmol) was added to a
suspension of 1,10-decanediol 1 (1.5 g, 8.6 mmol) in 30 mL of toluene in a round-
bottomed flask equipped with a Dean-Stark trap and a cooler. The mixture was refluxed
for 16h. After cooling, the solvent was removed under reduced pressure. The residue
obtained was chromatographed on silica gel using hexane/ethyl acetate (8:2) to yield the
pure compound 2.
Yield 74%; yellow oily product: IR (KBr): = 3325, 2925, 2850, 1450, 1250,
1050, 700, 625, 550 cm-1. ¹H NMR (200 MHz, CDCl3): δ 1.20-1.40 (m, 12H), 1.45-1.60

ACCEPTED MANUSCRIPT
(m, 2H), 1.84 (qn, J= 6.8Hz, 2H), 3.39 (t, J= 6.8Hz, 2H), 3.62 (t, J= 6.4Hz, 2H) ppm.
¹³C NMR (50 MHz, CDCl3) δ: 25.67, 28.11, 28.69, 29, 32, 29.42, 32.73, 32.77, 33.98,
62.99 ppm.
2.2.3. Synthesis of 10-(4,5-dihydrothiazol-2-yl)thio)decan-1-ol) 3
A 25 mL round-bottomed flask containing a mixture of 10-bromedecan-1-ol (0.42
mmol), 2-thiazoline-2-thiol (0.46 mmol) and potassium hydroxide (0.46 mmol) in 10
mL of ethanol/water (3:2) was placed in a CEM Discover microwave oven (open vessel
mode). Microwave irradiation of 100 W was used, and the temperature was ramped
from 25 °C to 60 °C. Once 60 °C was reached, in around 3 min, the reaction mixture
was held at this temperature for 15 min, under stirring. All temperatures were measured
externally by an IR sensor. The reaction mixture was cooled to room temperature and
then extracted with ethyl acetate (3×15 mL). The organic extracts were combined and
dried (Na2SO4), and the solvents were removed under reduced pressure. The residue
obtained was chromatographed on silica gel using hexane/ethyl acetate (6:4) to yield the
pure compound 3.
Yield 97%; yellow oily product: ¹H NMR (200 MHz, CDCl3): δ 1.15-1.4 (m,
12H), 1.4-1.6 (m, 2H), 1.6-1.7 (m, 2H), 3.08 (t, J=7.2Hz, 2H), 3.36 (t, J = 7.4Hz, 2H),
3.65 (t, J =5.2Hz, 2H), 4.20 (t, J =7.6 Hz, 2H) ppm. ¹³C NMR (50 MHz, CDCl3): δ
25.63, 28.60, 28.94, 29.13, 29.25, 29.36, 32.66, 32.72, 35.16, 62.79, 64.21, 166.16 ppm.
HRMS (m/z): [M + H]+ = 276.1451 (theoretical 276.1378).
2.3.
In silico analysis
Density Functional Theory (DFT) calculations were performed in order to obtain
structures and binding energies for the process involving two compounds, the
compound 3 and their active site, the ferriprotoporphyrin-IX heme group [Fe(III)PPIX],

ACCEPTED MANUSCRIPT
in 1:1 molar ratio. In the present work, we have used the same theoretical methodology
(UB3LYP/GEN) as described in our recent papers (Barbosa et al., 2017, 2018; Ribeiro-
Viana et al., 2016). The UB3LYP binding energy (∆E) related to this linking process
was obtained. All theoretical calculations were performed using the Gaussian 09
quantum mechanical package.
2.4.
Heme Binding Experiments
2.4.1. DMSO/HEPES Buffer (40% v/v)
1.1906g of HEPES was dissolved in 5 mL of distilled water. This solution was
transferred to a 250 mL volumetric flask containing 100 mL of DMSO. The remaining
volume was filled with distilled water up to 250 mL and the pH was adjusted to 7.4.
2.4.2. Hematin Stock solution (0.0026 mol/L)
17.2 mg of hemin (Sigma-Aldrich) was transferred to a 10 ml volumetric flask
and filled with DMSO/HEPES buffer (pH 7.4).
2.4.3. Hematin analysis solution (5 µmol/L)
48 µL of hematin stock solution was transferred to a 25 mL volumetric flask
and filled with DMSO/HEPES buffer (pH 7.4). This solution was prepared daily.
2.4.4. Compound 3 stock solution (68.97 mmol/L)
15.2 mg of compound 3 was weighted in an eppendorf and dissolved in 800 uL
of DMSO.
2.4.5. Compound 3 analysis solution (8.10 mmol/L)
94 µL of compound 3 stock solution was mixed with 706 µL of hematin
analysis solution (SAhm) in an eppendorf.
2.4.6. Compound 3-hematin Titration experiment

ACCEPTED MANUSCRIPT
Two cuvettes were used to perform the titration assay, blank and test. Blank
cuvette was filled with 2.5 mL of DMSO/HEPES buffer and test curvet with 2.5 mL of
hematin analysis solution. To the test solution were added incremental volumes of
compound 3 analysis solution, under magnetic stirring. After each hematin solution
addition, one minute was holding on before the scanning in the UV-Vis range (200 nm
to 500 nm) was performed. After each reading, the ƛmax was recorded as 401 nm. All
the experiments were performed in triplicate. Curve fitting was performed using
GraphPad Prism 5.01 software.
2.5. Biological evaluation
2.5.1. In vitro schizonticidal activity against P. falciparum
The P. falciparum chloroquine-resistant (W2) strain was maintained in continuous
culture using human red blood cells in RPMI 1640 medium supplemented with human
plasma. Human red blood cells and human plasma were provided by the Foundation of
Hemotherapy and Hematology of Minas Gerais (Fundação Hemominas). The parasites
were synchronized using sorbitol treatment (Lambros and Vanderberg, 1979) and the
parasitemias were microscopically evaluated with Giemsa-stained blood smears. The
antimalarial activity was determined using an anti-HRPII assay (Noedl et al., 2002).
Infected red blood cells were incubated on a 96-well plate at 0.05 % parasitemia and
1.5% hematocrit. Different concentrations of the drugs were added in triplicate and
twelve drug-free wells were used as controls (six frozen after 24 h as the HRPII
background). After incubation (72 h), the plate was frozen, thawed twice and an ELISA
using anti-HRPII antibodies was performed. The results were expressed as the mean of
the half-maximal inhibitory dose (IC50) of three assays with different drug
concentrations performed in triplicate, compared with drug-free controls. Curve fitting

ACCEPTED MANUSCRIPT
was performed using Origin Pro 8.0 software (Origin Lab. Corporation, Northampton,
MA, USA). Chloroquine (Sigma-Aldrich) was used as antimalarial standard.
2.5.2. In vitro cytotoxicity evaluation
The noncancerous human lung fibroblast cell line WI-26-VA4 (ATCC CCL-95.1)
was used to assess the cell viability after each chemical treatment employing the MTT
colorimetric assay (Carmichael et al., 1987). Briefly, 1ꢀ×ꢀ106 cells were seeded on 96-
well plates in RPMI 1640 medium supplemented with fetal bovine serum (FBS) and
penicillin-streptomycin antibiotics. Then, plates were incubated overnight at 37 °C, 5%
CO2, followed by the treatment with each compound solubilized in DMSO. Negative
control groups were constituted of cells without treatment. Five serial dilutions (from
100 to 0.01 µg.mL-1) were made from a stock solution (10ꢀmg.mL−1) using RPMI
supplemented with 1% FBS. After 48ꢀh of incubation, cell viability was evaluated by
discarding the medium and adding 100 μL of MTT 0.05 %, followed by 3ꢀh of
incubation. Then, the supernatant was discharged, and the insoluble formazan product
was dissolved in DMSO. The optical density (OD) was measured using a microplate
spectrophotometer at 570 nm. The OD in untreated control cells was defined as 100 %
cell viability. All assays were performed in triplicate. The selectivity index (SI) was
calculated as SI = IC50 WI-26-VA4 / IC50 P. falciparum.
2.5.3. Alkaline Comet Assay
The alkaline comet assay (single-cell gel electrophoresis assay) was performed
according to Olive and Banáth (2006) with adaptations. Briefly, RKO-AS45-1 (ATCC
CRL-2579) cells were seeded in 24-well plates (2 x 105 cells/well) in complete medium
and the treatments were performed after 24 h. The cells were exposed to different
concentrations of the compound 3 (3.6 µM, 7.25 µM, and 14.55 µM) for 3 h in culture

ACCEPTED MANUSCRIPT
medium without serum. The cells of the positive control group were treated with methyl
methanesulfonate (MMS - 120 µM) (Sigma-Aldrich) and the negative control group
was treated with PBS (phosphate buffered saline). The quantification of chromosomal
damages was achieved by visual scoring, based on the comets classification from 0 (no
damage) to 4 (maximum damage) (Collins, 2004). For each treatment, 100 comets were
analyzed and the calculation of the scores was performed employing the equation Score
= 0(C0) + 1(C1) + 2(C2) + 3(C3) + 4(C4), where C0–C4 are the numbers of comets in
each classification of damage. Three independent experiments were performed and a
mean of the scores obtained was calculated for each treatment. In the statistical analysis,
ANOVA (Analysis of Variance) was performed followed by the Tukey-Kramer
multiple comparisons post-test with a significance level of 0.05 (de Oliveira et al.,
2017).
2.5.4. Cytokinesis-Block Micronucleus (CBMN) Assay
To assess the potential of compound 3 to induce chromosomal mutations in vitro,
the cytokinesis-block micronucleus (CBMN) assay was performed in the RKO-AS45-1
(ATCC CRL-2579) human cell line, with and without metabolic activation (S9
fraction). The procedures were carried out as described by Fenech (2007) with
adaptations. Briefly, 2.5 x 105 cells/well were seeded in 24-well plates in complete
medium and the treatments were performed for 24 h. MMS (400 μM) was used as
positive control and the negative control group received culture medium without serum
as treatment. Compound 3, diluted in culture medium without serum, were evaluated in
three different concentrations (3.6 µM, 7.25 µM, and 14.55 µM) and three independent
experiments were conducted under these conditions, with or without metabolic
activation. After 3 h of treatment, cells were washed and fresh complete medium
containing cytochalasin-B (Sigma-Aldrich) (3.0 µg.mL-1) was added for 24 h. Next,

ACCEPTED MANUSCRIPT
cells were processed for the slides confection. Slides were stained with DAPI (4',6-
diamidino-2-phenylindole – 1 µg.mL-1) (Sigma-Aldrich) diluted in PBS for cytogenetic
analysis under a fluorescent microscope (Zeiss, Axioscope A1) with an excitation filter
of 365 nm and a barrier filter of 445/450 nm. In a blind test, one thousand binucleated
cells were analyzed for each treatment and cells containing 1–3 micronuclei were scored
(Gomes et al., 2011), following the criteria for the identification of micronuclei (MNs)
described by Titenko-Holland et al., 1997. For statistical analysis, ANOVA was
performed followed by the Tukey's Multiple Comparison post-test with a significance
level of 0.05. The influence of compound 3 on cell proliferation was assessed by
calculating the Nuclear Division Index (NDI) in the same slides prepared for the CBMN
assay. Three hundred cells with well-preserved cytoplasm were counted using
fluorescence microscopy, as described above. The NDI was calculated according to
Eastmond and Tucker (1989) and Fenech (2007), using the equation NDI = (M1 +
2(M2) + 3(M3) + 4(M4))/N, where M1–M4 are the numbers of cells with 1, 2, 3 and 4
nuclei, respectively, and N is the total number of analyzed cells.
2.5.5.Live Cell Imaging
Synchronous trophozoites of P. falciparum (W2 strain) were marked with the
lysosomotropic probe acridine orange (AO) (Sigma-Aldrich) ( as described by Bennett
and colleagues (2004) with modifications. Briefly, the culture with 10 % parasitemia
was centrifuged for 5 min at 9,000 x g and resuspended in RPMI without phenol red.
The RBC number was determined using a Neubauer chamber and it was adjusted to
1×107 RBC.mL-1 of RPMI medium supplemented with 5 µM AO. The sample was
incubated for 40 min at 37 ºC. After that, cells were washed three times and 500 µL
transferred to a four-chamber petri dish (Cell view glass - Greiner Bio-one).
Fluorescence images was acquired using a multiphoton scanning microscope (LSM 780

ACCEPTED MANUSCRIPT
Zeiss, Germany) with parameters: 488 nm exciting wavelength and collecting
fluorescence emission between 560-615 nm. AO fluorescence localization and intensity
were measured before and after a 3 minute-treatment with compound 3 and CQ.
Experiments were performed in triplicate and each parasite cell was analyzed
independently.
2.5.6. In vivo evaluation of the antimalarial activity
For in vivo assays, female Swiss mice, 4 to 6 weeks of age provided by the Center
for Reproductive Biology of Universidade Federal de Juiz de Fora (Ethics committee
approval process nº 043/2016) were kept in micro isolated cages and inside ventilated
shelves (ALESCO), maintained at controlled temperature (22 ± 2 ºC), 60-90 %
humidity under a 12 h light/dark cycle with free access to food and water.
The suppressive test was performed according to Peters (1965), with
modifications. The animals were randomly divided, and after weighing, the inoculum of
0.2 mL solution was adjusted to 5×106 infected erythrocytes/mL P. berghei NK65 in
RPMI medium (Cultilab) supplemented with 10 % fetal bovine serum (Sigma-Aldrich).
After 4 h of the infection, mice were treated by gavage with 0.1 mL of the compound 3
and the association of Artesunate (Sigma-Aldrich) (5 mg/kg) and Mefloquine (Sigma-
Aldrich) (10 mg/kg) diluted in 5 % dimethyl sulfoxide (DMSO) and filtered water for 4
days. The compounds were tested at doses of 50, 20 and 10 mg/kg. The parasitaemia
was determined by Giemsa-stained smears with blood collected from caudal vein at the
5th, 7th, 9th and 12th-day post-infection (d.p.i). Mice survival was evaluated for 30 days.
Inhibition of parasite multiplication (IPM) was calculated by the difference between the
mean of the parasitaemia of the group not treated by the mean of the tested group,
divided by the mean of that from the untreated group, expressed as a percentage
(Andrade-Neto et al., 2003).

ACCEPTED MANUSCRIPT
2.5.7.Caco-2 cell culture and in vitro uptake
Human intestinal Caco-2 cells provided by American Type Culture Collection
ATCC (Rockville, MD, USA) were cultured for 21 days with Dulbecco´s Modified
Eagle Medium (DMEM) high glucose supplemented with 10 % of fetal bovine serum, 1
% of nonessential aminoacids, 100 U.mL-1 penicillin G, 100 mg.mL-1 streptomycin,
1.25 mg.mL-1 amphotericin B, and 10 mM sodium pyruvate in tissue culture test plate.
All of them from Sigma Chemical Company (St. Louis, MO). The atmosphere was kept
with 90–95 % relative humidity, 37 °C, 5 % CO2 and 9 5% air.
In the experiment day, DMEM was removed and the cells were incubated with
compound 3 (15, 30 and 60 min.) and chloroquine for 60 min. After incubation the
solutions were removed, cells were washed three times with ice-cold transport medium
o
(TM) consisted of Hank´s balanced salt solution pH 7.4 (0-4 C). The cells were lysed
with 400 µL of NaOH (1M) and sonicated for 2 minutes. The suspension was
neutralized with HCl (1M), sonicated for additional 2 minutes, and centrifuged at 22,000
x g for 10 min at 15 °C. The supernatant (400 µL) was transferred for vials and 75 µL
were injected in the HPLC for quantification. The pellets were sonicated in an Unique ®
DES500 ultrasonic cell disruptor (Indaiatuba, Brazil) and the protein content was
measured by the Bradford method (Bradford, 1976). Sonication was conducted under ice
bath using three cycles of 30 sec. on at 30% of potency and 10 sec. off.
Stock solutions of compound 3 and chloroquine at 10,000 µM were prepared in
ethanol and transport medium (TM), respectively, whereas, the work solutions (100
µM) of both compounds were prepared in TM.
Compound 3 and chloroquine detection were performed by high pressure liquid
chromatograph. Briefly, separation of the compounds was achieved using a 150 x 4.00
mm column packed with 5 μm ODS (octadecylsilane) particles (NST®, São Carlos,

ACCEPTED MANUSCRIPT
Brasil) at 40 °C. For the compound 3 analysis the mobile phase consisted of acidified
water (pH 3.2) and methanol (25:75). The flow rate was 1.2 mL/min and the detection
occurred at 232 nm. The chloroquine was detected at 245 nm using as mobile phase
acetonitrile and aqueous solution (pH 3.4) containing 1-heptanesulfonic acid (20 mM)
and diethylamine (0.07%) (30:70, v/v) at a flow rate of 1.0 mL/min, according to the
method described by Yonemitsu (2005). In both cases the pH of the aqueous phase was
previously adjusted with phosphoric acid.
The uptake studies were conducted in triplicate and the results (ng/mg of protein)
expressed as mean ± sd. Compound 3 and chloroquine intracellular concentration were
compared using One-Way ANOVA followed by the test of Tukey for multi-
comparison. A value of p < 0.05 was considered significant.
3. Results and Discussion
3.1. Heterocycle change
In our previous work (Barbosa et al., 2017) we reported the synthesis of a series
of 3-APA analogs and the identification of a hit compound 4 among them with the
lowest IC50 values of the series, and the highest SI (Figure 1).
Figure 1. Previously synthesized 3-APA analog 4 and the optimized compound 3 with the moiety derived
from 3-pyridinopropanol and the new thiazole ring. The pyridine and thiazole rings are highlighted in red.

ACCEPTED MANUSCRIPT
Therefore, compound 4 was selected as a starting point to lead optimization. The
thiazole core was selected for pyridine replacement, supported by successfully reported
cases of new potential synthetic antimalarials compounds with enhanced activity
(Kumawat, 2017; Sharma et al., 2019).
A recent report described the antimalarial activity of several compounds
containing imidazopyridine heterocycle with low toxicity (Baker et al., 2017). A
subsequent study demonstrated that the substitution of imidazopyridine heterocycle by a
thiazole core resulted in a potent scaffold with nanomolar range activity against P.
falciparum intraerythrocytic stages (Penzo et al., 2019).
Herein, we have focused our development on the removal of the potentially toxic
pyridine ring represented by the commercially 3-pyridinopropanol moiety used to
synthesize the 3-APA analog and change it to a thiazole heterocycle (Figure 1).
However, this alteration on the structure of the compound 4 could affect the
molecular recognition of heme group. To examine this possibility, we carried out in
silico and UV-vis analysis.
3.2. In silico Thiazoline-heme binding
The ferriprotoporphyrin-IX [Fe(III)PPIX] interaction with compound 3 has been
investigated. The optimized geometry for the complex, in which the nitrogen atom of
thiazole ring binds to Fe(III) ion, is depicted in Figure 2. The calculated UB3LYP/GEN
binding energy is -12.5 kcal.mol-1. It can be observed that the formation of 1:1
compound 3 complex with Fe(III)PPIX is energetically favored, as also observed for
pyridine ring in our previous works (Barbosa et al., 2017, 2018; Ribeiro-Viana et al.,
2016). In both cases the nitrogen atom of heterocyclic ring is the responsible for the
binding. Finally, the DFT (UB3LYP/GEN) theoretical analysis indicated that, based on

ACCEPTED MANUSCRIPT
the energetics of binding process, the heme group can be considered a target for the
compound 3.
Figure 2. UB3LYP optimized geometry for the compound 3-[Fe(III)PPIX] complex in a 1:1 molar ratio.
3.3. Hematin binding studies
Hematin binding studies using UV-vis titration were performed to identify the
behavior of compound 3 in the presence of hematin and confirm it as a target. This
experiment provides an important qualitative data of the formation of supramolecular
complex. The complex formation is observed by the pattern of the isotherm formed after
the data analysis. The data collected in this work, exhibited a different pattern of curve
from the others compounds synthesized by our group (Figure 3).
The absorbances during the addition of compound 3 showed three different
moments: a small decrease in the beginning, followed by increase of the observed
absorbance and finally a trend of a plateau. Our previous synthesized molecules showed
different isotherm pattern (Barbosa et al., 2017, 2018; Ribeiro-Viana et al., 2016). It
was observed a decrease in absorbance values as the concentrations of added ligand
were increased, indicating that those complexes have lower molar absorption than

ACCEPTED MANUSCRIPT
hematin. In this particular case, a plateau is not formed soon after the decrease of the
absorbances, but after an increase of it. This kind of pattern suggests a different mode of
interaction of this molecule with hematin. Simulation of isotherm in different scenarios
via Supramolecular website (Thordarson, 2011; Hibbert and Thordarson, 2016) showed
a ternary complex in which the first ligand-host complexation had lower molar
absorptivity, in comparison to free hematin. In addition, it was observed a second ligand
binding to the complex with higher molar absorptivity. Therefore, these data suggest
that the replacement of pyridine ring by thiazoline group in compound 3 kept the
hematin as a target with lower affinity in solution, and changed the binding mode.
Figure 3. Titration of heme (5 µmol/L) with increasing concentration of compound 3 (0−1,9 mmol/L) at
A401 nm.
3.4. In vitro antiplasmodial activity
The compound 3 was tested against P. falciparum W2 strain and showed an IC50
value of 1.55 µM (Table 1), which is more than 9-fold lower than the IC50 of its
precursor 4 (14.7 µM) (Barbosa et al., 2017). The cytotoxicity evaluation showed that,
both the precursor and the compound 3, exhibited similar IC50 against human WI26VA-
4 cell line (99.1 and 115.8 µM, respectively). The choice of chemical starting points for
the development of new potential antimalarial candidates is crucial for success of

ACCEPTED MANUSCRIPT
clinical studies (Mishra et al., 2017). In this sense, a set of specific criteria for defining
hits and leading compounds were proposed in the development of new drug candidate
against malaria. (Katsuno et al., 2015). For in vitro assays, compound 3 exhibits
potency around 1–2 μM and the SI is greater than 10-fold against a human cell line.
To examine the safety of compound 3 we performed genetic toxicological studies.
3.5. Genotoxicity evaluation
Cytokinesis-Block Micronucleus (CBMN) Assay showed that compound 3 did not
induce the formation of micronuclei compared to the control (PBS). This assay is
required for drug development to determine mutagenicity by detecting both clastogenic
and aneugenic agents (Kirsch-Volders et al., 2014). The Alkaline Comet assay, that
detects DNA breaks, was also performed and three concentrations were evaluated.
Compound 3 was only genotoxic at 7.25 µM and 14.55 µM but not at 3.6 µM. This
result indicates that even though at some point the compound induces DNA strand
breaks, the concentration required is more than 4.5 times higher than the IC50 for P.
falciparum. Nevertheless, these primary DNA damages were not fixed as mutations as
verified in CBMN assay. In studies performed with and without metabolic activation
with the S9 fraction, the exposition to compound 3 did not induce a significant increase
in the frequency of micronucleated cells (Table 1, Figure 4).
The evaluation of the genotoxicity of compounds with pharmacological potential
is an important step in the pathway of a drug registration process, once gene mutations
and chromosomal aberrations are related with degenerative diseases and heritable
disorders (Ji et al., 2017). In this way, results presented herein are very promising in the
toxicological assessment of this drug prototype once the assays performed showed that
DNA damages were induced only in concentrations much higher than the IC50 obtained
in P. falciparum. In addition to data collected in human cell line, the cells were able to

ACCEPTED MANUSCRIPT
repair the damage, preventing the fixation of mutations. It is important to highlight that
new antimalarial medicines must be safe in specific patient groups, such as: i) pregnant
women, ii) youngest children, iii) and patients with co-morbidities, like HIV (Burrows
et al., 2017). Moreover, a previous study of our study group (Gonçalves et al., 2014)
showed that the compound 4 was mutagenic in the CBMN assay carried out with the
same cell line used in present work.
Figure 4. Score values obtained for Compound 3 in Alkaline Comet Assay and frequency of binuclated
cells with micronuclei observed in CBMN assay, with (+S9) and without (-S9) metabolic activation in
vitro. * Statistically significant (p<0.05) when compared to negative control (0 µM).

ACCEPTED MANUSCRIPT
Table 1. Antiplasmodial, cytotoxicity and mutagenicity of compound 3 compared to
Chloroquine
IC50 (µM ± S.D.)
SI Comet (*µM) CBMN +S9/-S9
RKOAS45-1
P. falciparum WI26VA-4
1.55 ± 0.27
0.38 ± 0.09
115.8 ± 5.7 74.7
>100 >250
7.25
NT
NM/NM
NT
3
Chloroquine
S.D. = Standard Deviation; SI = Selectivity Index; NM = not mutagenic for every concentration tested;
NT = not tested; *Lowest concentration in which genotoxicity was detected
3.6. Analysis of food vacuole pH alterations
After the in vitro analysis was performed, and the in silico model revealing
insigths of compound 3 mode of action, our goal has been to probe infected red blood
cells with AO to determine food vacuole as a potential intracellular target to the
compound 3. Food vacuole is the organelle in Plasmodium responsible for hemoglobin
digestion and highly dependent of internal pH (Singh et al., 2014). The treatment with
compound 3 and CQ occurred for 3 min. Both compounds were tested at their IC50 (1.55
µM for 3 and 0.38 µM for CQ). Figure 5 shows the decrease in fluorescence intensity
after treatment with both compounds.

ACCEPTED MANUSCRIPT
Figure 5. Confocal imaging of P. falciparum W2 loaded with Acridine Orange before and after the
addition of Chloroquine (0.38 µM) and compound 3 (1.55 µM). Food vacuole alkalinization was
evaluated for 180 seconds (3min) after addition of the compounds. Scale bar 10µm.
The drop of AO fluorescence in food vacuole selected area is observed after the
addition of compound 3 in a graph of fluorescence intensity versus time (Figure 6).
Figure 6. Fluorescence measurement of acridine orange in food vacuole selected areas by confocal
microscopy. Chloroquine (CQ) and compound 3 were added at the beginning of the measurement and AO
fluorescence was evaluated for 180 seconds. Fluorescence intensities (arbitrary fluorescence units - AFU)
represent at least three different cell preparations. The red arrow indicates the moment of CQ and
compound 3 addition.
Acridine orange accumulates inside acid organelles, such as lysosome and the
parasite’s food vacuole and it emits red fluorescence (>560 nm) and little green
fluorescence (500-530 nm) when inside nucleus and cytoplasm (El Chamy Maluf et al.,
2016). Confocal analysis showed that the compound 3 interferes in the proton
homeostasis (H+), since the fluorescence of the digestive vacuole decreased with the
treatment, making this organelle less acidic. Therefore, the differences between the
initial fluorescence (f0) and final fluorescence (f1), using Student's t-test, were

ACCEPTED MANUSCRIPT
evaluated to compare the mean fluorescence difference between the two treatments. No
statistical difference between the treatments was observed.
3.7. In vivo antimalarial assays
Considering that the IC50 value obtained to P. falciparum is now under 2 µM, as
suggested by Katsuno (2015), we decided to proceed to test the compound 3 for its in
vivo antimalarial activity. Three treatment doses were tested (50, 20 and 10 mg/kg)
during four days on previously P. berghei NK65 infected swiss mice. At the fifth day of
the experiment, 50 mg/kg of compound 3 reduced almost 65 % of the parasitaemia
compared to the control group. However, the reduction did not persist on the following
days. The doses of 10 and 2 0mg/kg reduced 19.2 and 26.5 % of the parasitaemia on the
first day post-treatment, respectively, however, this reduction disappeared at the ninth
day and beyond. The drugs used as standard antimalarials were artesunate and
mefloquine, with doses of 5 and 10 mg/kg, respectively. This treatment reduced at least
90 % of the parasitemia at the 4 days evaluated (Table 2, Figure 7).
Table 2. Parasitemia and percentage reduction at different days post inoculation on
swiss female mice
Parasitaemia ± SD (% Reduction) at day after inoculation
Dose
Compound
(mg/kg)
5th
7th
9th
12th
50
0.26 ± 0.21a,b (64.4) 1.63 ± 0.91 (24.5) 1.23 ± 0.26 (38.2)
9.7 ± 4.42 (8.69)
3
20
10
0.53 ± 0.25 (26.5)
1.4 ± 0.31(35.3)
2.34 ± 0.90 (0)
3.07 ± 1.7 (0)
12.25 ± 6.42 (0)
15.95 ± 7.86 (0)
0.77 ± 0.72a (92.7)
10.62 ± 7.24
0.59 ± 0.49 (19.2) 1.5 ± 0.62 (30.4)
AS/MQ
Water
5/10
-
0.06 ± 0.03a (91.3) 0.10 ± 0.05a(95.3) 0.18 ± 0.10a (90.9)
0.73 ± 0.29 2.16 ± 1.08 1.99 ± 0.76
AS= Artesunate MQ=Mefloquine SD=Standard Deviation

ACCEPTED MANUSCRIPT
a = p-value < 0.05 (compared to water-treated group)
b = p-value > 0.05 (compared to AS/MQ-treated group)
The parasitemia of infected mice treated with 50 mg/kg of compound 3 was
statistically different from water-treated group (Mann-Whitney test, p-value <0.05).
Therefore, no statistical difference was observed comparing the treatment with
compound 3 (50 mg/kg) and standard antimalarials on day 5 (Mann-Whitney test, p-
value >0.05).
Figure 7. Percentage of parasitemia reduction at days 5, 7, 9 and 12 after the beginning of the treatment.
Compound 3 50mg/kg/day(▲); 20mg/kg/day (●); 10mg/kg/day(○) and AS 5mg/kg/day + MQ
10mg/kg/day (■).
The overall mortality in all infected groups is shown in Figure 8. No statistical
difference was found between treatments. Therefore, at the end of the experiment, 80 %
of the mice survived on the group treated with 50 mg/kg of compound 3, as well as the
AS/MQ group. Moreover, no weight loss was observed on treated mice, indicating that
the gavage did not wound the stomachs of the mice. The group treated with artesunate
and mefloquine did gain weight, probably due to the fast recovering from the infection
(data not shown).

ACCEPTED MANUSCRIPT
Figure 8. Survival of P. berghei infected mice after 4-day oral treatment. Mice survival were evaluated
for 30 days. Compound 3, 50mg/kg/day (▲);20mg/kg/day (●);10mg/kg/day (○); AS/MQ, 5mg/kg/day of
AS + 10mg/kg/day of MQ (■), and non-treated controls (CT) (Δ).
Drug discovery projects, like Medicines for Malaria Venture (MMV) or Drugs for
Neglected Diseases initiative (DNDi), established that a new clinical candidate
molecule should have a high in vivo efficacy (106-fold parasite reduction) with an oral
and single dose (Burrows et al., 2013). Despite compound 3 did not exhibit this pattern
of action, it is important to mention that the in vivo tests were carried out with a model
of rodent malaria, instead a Plasmodium falciparum-infected severe combined
immunodeficient (SCID) mouse model. SCID is considered the gold pattern to establish
antimalarial in vivo results (Kaushansky et al., 2014). Although, parasite clearance is not
the only criteria adopted in the checkpoint for the development of a new antimalarial.
Other parameters like: i) easy of synthesis with low cost, ii) good oral bioavailability,

ACCEPTED MANUSCRIPT
iii) drug metabolism, and iv) good membrane permeability are properties that can lead
to success in final phases of clinical trials (Gayvert et al., 2016).
In this context, in order to determine the membrane permeability of compound 3,
we performed Caco-2 permeability assay.
3.8. Compound 3 uptake by Caco-2 cell line
Both compound 3 and CQ were evaluated for their Caco-2 cell line uptake. The
uptake of compound 3 increased linearly according to the incubation time and after 60
min no statistical difference was observed when compared to CQ (Figure 9). These
results suggest that compound 3 is a good candidate for oral administration.
Figure 9. Uptake of compound 3 (100 µM) and chloroquine (100 µM) by Caco-2 cells (n=3) after
o
incubation at 37 C in Hank´s balanced salt solution pH 7.4. * p< 0.05, ** p<0.01 when compared to the
CQ. Mean ± SD. CQ – Chloroquine
The permeability and uptake characteristics of CQ in Caco-2 cell model was
systematically investigated (Augustijns, 1996). After 60 min of incubation in the
presence of CQ (100 µM), the uptake was about 24 µg per mg of protein. This result is

ACCEPTED MANUSCRIPT
about 30-fold higher than that found in our study. Such discrepancies can be attributed
to the differences in culture conditions and cell density (Bentz et al., 2013).
4.
Conclusions
In conclusion, we optimized the development of a previously reported 3-APA hit
compound to a new thiazole lead. Compound 3 has a potent antimalarial activity with a
great selectivity index. This compound is not mutagenic and is capable of inhibiting P.
berghei in vivo development. We also determined that the uptake of compound 3 by
Caco-2 cells is similar to CQ and it acts within the parasite’s food vacuole, as predicted
by in silico and UV-vis studies. Thus, compound 3 presents a promising lead for
optimization as an antimalarial new drug with high selectivity, low toxicity, good
membrane permeability. Ongoing studies are focused on to establish the in vivo PK/PD
profile of compound 3.
Acknowledgments
This work was supported by the Brazilian Agencies: CNPq, FAPEMIG, FAPESP
and CAPES. We thank the Multiuser Multiphoton Confocal Microscopy Laboratory
(INFAR-UNIFESP) for the access to confocal microscopy devices. The authors also
thank the Cellular Biology Service from the Ezequiel Dias Foundation for donation of
the human cell lines.
References
Andrade-Neto, V.F., Brandão, M.G.L., Stehmann, J.R., Oliveira, L. a., Krettli, A.U.,
2003. Antimalarial activity of Cinchona-like plants used to treat fever and malaria
in Brazil. J. Ethnopharmacol. 87, 253–256. https://doi.org/10.1016/S0378-