Synthesis and 2D-QSAR studies of neolignan-based diaryl-tetrahydrofuran and -furan analogues with remarkable activity against Trypanosoma cruzi and assessment of the trypanothione reductase activity

Article abstract of DOI:10.1016/j.ejmech.2017.08.064

Two series of diaryl-tetrahydrofuran and -furan were synthesised and screened for anti-trypanosomal activity against trypomastigote and amastigote forms of Trypanosoma cruzi, the causative agent of Chagas disease. Based on evidence that modification of a natural product may result in a more effective drug than the natural product itself, and using known neolignan inhibitors veraguensin 1 and grandisin 2 as templates to synthesise simpler analogues, remarkable anti-trypanosomal activity and selectivity were found for 3,5-dimethoxylated diaryl-furan 5c and 2,4-dimethoxylated diaryl-tetrahydrofuran 4e analogues with EC₅₀ 0.01 μM and EC₅₀ 0.75 μM, respectively, the former being 260-fold more potent than veraguensin 1 and 150-fold better than benznidazole, the current available drugs for Chagas disease treatment. The ability of the most potent anti-trypanosomal compounds to penetrate LLC-MK2 cells infected with T. cruzi amastigotes parasite was tested, which revealed 4e and 5e analogues as the most effective, causing no damage to mammalian cells. In particular, the majority of the derivatives were non-toxic against mice spleen cells. 2D-QSAR studies show the rigid central core and the position of dimethoxy-aryl substituents dramatically affect the anti-trypanosomal activity. The mode of action of the most active anti-trypanosomal derivatives was investigated by exploring the anti-oxidant functions of Trypanothione reductase (TR). As a result, diarylfuran series displayed the strongest inhibition, highlighting compounds 5d-e (IC₅₀ 19.2 and 17.7 μM) and 5f-g (IC₅₀ 8.9 and 7.4 μM), respectively, with similar or 2-fold higher than the reference inhibitor clomipramine (IC₅₀ 15.2 μM).

Full text of DOI:10.1016/j.ejmech.2017.08.064

Accepted Manuscript
Synthesis and 2D-QSAR studies of neolignan-based diaryl-tetrahydrofuran and -furan
analogues with remarkable activity against Trypanosoma cruzi and assessment of the
trypanothione reductase activity
Ana Paula Hartmann, Marcelo Rodrigues de Carvalho, Lilian Sibelle Campos
Bernardes, Milene Hoehr de Mores, Eduardo Borges de Melo, Carla Duque Lopes,
Mario Steindel, João Santana da Silva, Ivone Carvalho
PII:
S0223-5234(17)30676-1
10.1016/j.ejmech.2017.08.064
EJMECH 9708
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 11 July 2017
Revised Date: 7 August 2017
Accepted Date: 29 August 2017
Please cite this article as: A.P. Hartmann, M.R. de Carvalho, L.S.C. Bernardes, M.H. de Mores, E.B.
de Melo, C.D. Lopes, M. Steindel, Joã.Santana. da Silva, I. Carvalho, Synthesis and 2D-QSAR studies
of neolignan-based diaryl-tetrahydrofuran and -furan analogues with remarkable activity against
Trypanosoma cruzi and assessment of the trypanothione reductase activity, European Journal of
Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.08.064.
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
Graphical Abstract

ACCEPTED MANUSCRIPT
Synthesis and 2D-QSAR studies of neolignan-based diaryl-tetrahydrofuran and -furan
analogues with remarkable activity against Trypanosoma cruzi and assessment of the
trypanothione reductase activity
Ana Paula Hartmann,a Marcelo Rodrigues de Carvalho,a Lilian Sibelle Campos Bernardes,b
Milene Hoehr de Mores,c Eduardo Borges de Melo,d Carla Duque Lopes,e Mario Steindel,c João
Santana da Silva,d Ivone Carvalhoa*
aSchool of Pharmaceutical Sciences of Ribeirão Preto, University of São Paulo, Av. Café s/n, Ribeirão
Preto, SP 14040-930, Brazil
bDepartment of Pharmaceutical Sciences, Federal University of Santa Catarina, Florianopolis, 88040900
SC, Brazil
cDepartment of Microbiology, Immunology and Parasitology, Federal University of Santa Catarina,
Florianopolis, 88040900 SC, Brazil
dSchool of Pharmacy, Western Parana State University, Rua Universitária, 2069, Cascavel, PR, 85819-
110, Brazil
eRibeirão Preto Medical School, University of São Paulo, Av. Bandeirantes 3900, Ribeirão Preto, SP
14049-900, Brazil
*Corresponding author: carronal@usp.br
Abstract
Two series of diaryl-tetrahydrofuran and -furan were synthesised and screened for anti-
trypanosomal activity against trypomastigote and amastigote forms of Trypanosoma cruzi, the
causative agent of Chagas disease. Based on evidence that modification of a natural product
may result in a more effective drug than the natural product itself, and using known neolignan
inhibitors veraguensin 1 and grandisin 2 as templates to synthesise simpler analogues,
remarkable anti-trypanosomal activity and selectivity were found for 3,5-dimethoxylated
diaryl-tetrahydrofuran 5c and 2,4-dimethoxylated diarylfuran 4e analogues with EC50 0.01 µM
and EC50 0.75 μM, respectively, the former being 260-fold more potent than veraguensin 1 and
150-fold better than benznidazole, the current available drugs for Chagas disease treatment.
The ability of the most potent anti-trypanosomal compounds to penetrate LLC-MK2 cells
infected with T. cruzi amastigotes parasite was tested, which revealed 4e and 5e analogues as
the most effective, causing no damage to mammalian cells. In particular, the majority of the
derivatives were non-toxic against mice spleen cells. 2D-QSAR studies show the rigid central
1

ACCEPTED MANUSCRIPT
core and the position of dimethoxy-aryl substituents dramatically affect the anti-trypanosomal
activity. The mode of action of the most active anti-trypanosomal derivatives was investigated
by exploring the anti-oxidant functions of Trypanothione reductase (TR). As a result,
diarylfuran series displayed the strongest inhibition, highlighting compounds 5d-e (IC50 19.2
and 17.7 μM) and 5f-g (IC50 8.9 and 7.4 μM), respectively, with similar or 2-fold higher than the
reference inhibitor clomipramine (IC50 15.2 μM).
Keywords: Trypanosoma cruzi, Chagas disease, natural neolignans, diaryl-tetrahydrofurans,
diarylfurans, trypanothione reductase, 2D-QSAR.
2. Introduction
Plant-derived natural products represent a valuable source of novel drug leads
because of the chemical diversity found in nature, giving rise to highly active compounds with
selective and specific mechanisms of action [1,2]. Despite the great interest of pharmaceutical
companies in combinatorial chemistry to access the synthetic compound library for HTS (High
Throughput Screen) approaches, the decline of new chemical entities approved by the FDA at
the beginning of the 21st century, has led the research community to reappraise plant-derived
natural product drug discovery as an important approach for the identification of novel
chemotherapeutics agents [3]. Furthermore, the extensive use of natural products by different
cultures for the treatment of various disease conditions make them a valuable supply of new
drugs [1,4,5].
Apart from that, several aspects preclude the successful research, development and
commercialisation of natural products, such as the supply constraints on bioactive
components, ecological and legal considerations for plant accessibility, the isolation processes,
variation of quality and composition (depend on species identity), harvest time, soil
composition, altitude, actual climate, and processing and storage conditions [3,6]. Even though
the resupply of natural products can be achieved by the application of plant cell and tissue
culture, heterologous production or semi-synthesis from isolated precursors. The limited
amount of isolated bioactive natural products available, and the total chemical synthesis of
complex structures containing several functional groups and chiral centers, slow down the
development process, which still requires optimisation of physicochemical properties using
chemical modification approaches [3,7]. Therefore, the chemical simplification and
derivatisation of natural products is an interesting alternative to generate a chemical library, as
2

ACCEPTED MANUSCRIPT
well as to understand the structure–activity relationship for pharmacophore identification. In
the last decade, several natural product extracts have been screened to find more active and
less toxic molecules against Chagas disease, a life-threatening illness caused by the protozoan
parasite Trypanosoma cruzi [8-10]. Chagas disease is endemic in Central and South American
countries, with an estimated 8 million people infected worldwide [11,12]. In 2011, Nakamura
reported 400 species related to approximately 100 families of plant extracts already tested
against different forms of T. cruzi, of which 10% were investigated and their corresponding
active components against the parasite were isolated and characterised, being around 136
compounds assessed [13]. Since then, a brief search at Web of Science [14] revealed about 33
active compounds from plant extracts against T. cruzi [15-26]. The phenotypic screening of
potential anti-trypanosomal compounds from natural products has shown a plethora of
chemical classes, illustrated by alkaloid, flavonoid, catechin and terpene derivatives, including
sesquiterpene lactones, xanthone, oxygenated hydrocarbons and lignans [13,27,28]. In
particular, lignans isolated from Lauraceae and Piperaceae families are of great interest since
the majority of the active components decrease parasite growth of both epimastigote and
trypomastigote forms [29]. Beside lignans, tetrahydrofuran neolignans such as veraguensin (1)
and grandisin (2), isolated from Virola surinamensis and Piper solmsianum, have been
identified as potent inhibitors of trypomastigote Y strain of T. cruzi, producing parasite lysis
with EC50 2.3 and 3.7 µM, respectively [30] (Figure 1). Notably, veraguensin (1) and grandisin
(2) possess 2,5-bis(3,4-dimethoxyphenyl)-ꢀandꢀ2,5-bis(3,4,5-trimethoxyphenyl)-tetrahydro-3,4-
dimethylfuran structures, respectively. In addition, 4-(furan-2-yl) benzenesulfonamide and
symmetrical non-substituted 2,5-bis(3,4-dimethoxyphenyl)furan have been further described
as T. cruzi amastigote growth inhibitors (70 and 65% inhibition at 20 µg/mL) [31]. Our previous
studies regarding the low water solubility of these neolignans have shown the importance of
the heterocycle core to the potency [32,33], considering the distinguished activity displayed by
dihydrofuran derivative 3, the most active of all series with EC50 1.5 µM [32], when comparing
the activities of 1-3 with open chain symmetrical and unsymmetrical diaryl diols and diaryl
diketones precursors, besides symmetrical acetylenic glycols, which preserve the diaryl
polymethoxylated pattern of natural products 1 and 2. These findings encouraged us to
investigate whether simplified unsymmetrical 2,5-diaryl-tetrahydrofuran and -furan derivatives
containing methoxyl substituents in a variable number and at different ring positions, such as
the set compounds 4 and 5, would possess enhanced activity against trypomastigote and
amastigote forms of T. cruzi. Further, 2D-QSAR studies were carried out to identify the
molecular descriptors that may encode structure information of the set molecules in order to
3

ACCEPTED MANUSCRIPT
predict the anti-parasitic responses of new hit derivatives [34-36]. Then, the most active anti-
trypanosomal derivatives were chosen to investigate whether they inhibit Trypanothione
reductase (TR), a key enzyme involved in the parasite redox metabolic pathways.
Figure 1. Neolignans veraguensin (1) and grandisin (2), the synthetic derivative 3 previously
described and target simplified structures of neolignan-based diaryl-tetrahydrofuran and -
furan analogues (4, 5).
2. Results and Discussion
2.1 Chemistry
As shown in scheme 1, the condensation of the dianion of racemic 1-phenylprop-2-yn-
1-ol (6) with a diverse set of commercially available aryl aldehydes 7a-g bearing methoxy
groups at different positions of the aromatic ring afforded key intermediates 1,4-diphenyl-2-
butyn-1,4-diol (8a-g) in good to moderate yields (30-82%) (Scheme 1) [33,37]. Several attempts
to improve the yields led to the protection of the hydroxyl group of precursor carbinol 6 to
prevent the dianion formation under basic conditions. Typically, the treatment of 6 with tert-
butyldimethylsilyl chloride promoted the protection of both hydroxyl and acetylenic groups,
whereas dihydropyran (DHP) under acid conditions [38] gave the mono-O-protected
intermediate in 45% yield. Despite the low to moderate yields obtained during the
condensation reaction between the O-DHP carbinol derivative 6 with a panel of aldehydes 7a-
g, the removal of the O-DHP protecting group using several protocols (p-TsOH, Dowex®50WX4-
50, trifluoro-acetic and acetic acids and 1M HCl) [39-42], proved to be difficult. Therefore, we
performed the condensation reactions with unprotected carbinol 6 for formation of the
1
diastereoisomeric mixtures of 8a-g, which were assessed by H NMR analysis with two
characteristic singlets or doublets around 5.50 ppm, related to the hydrogen attached to the
carbon-linked hydroxyl group (except for the symmetrical intermediate 8a). Furthermore, 13C
NMR revealed clearly the duplicity of signals at 85.0 ppm related to acetylene carbons.
4

ACCEPTED MANUSCRIPT
Subsequent one-pot isomerisation of 1,4-diphenyl-2-butyn-1,4-diol (8a-g)
intermediates with palladium complex Pd2(dba3CHCl3) [tris(dibenzylideneacetone)dipalladium]
and triphenylphosphine [43] was not convenient to afford the corresponding diaryl-diketone
derivative 9a-g, since the yields were very low (6%) and time-consuming (70 h). On the other
hand, an appropriate two-step procedure, involving the alkyne reduction of 8a-g intermediates
followed by hydroxyl group oxidation, was accomplished to produce both 1,4-diphenylbutane-
1,4-diol derivatives 10a-g and 1,4-diphenylbutane-1,4-dione 9a-g intermediates, crucial for the
generation of target diaryl-tetrahydrofurans 4a-g and -furans 5a-g, respectively. As a result,
alkyne reduction with platinum oxide at 10 atm [44] proved suitable for formation of 10a in
92% yield, whilst hydroxyl oxidation with IBX (2-iodoxybenzoic acid) [45] led to the diaryl-
diketone 9a formation in 98% yield, superior to the reaction initially performed with PCC
(pyridinium chlorochromate, 49%) [46]. Therefore, the conditions established for sequential
conversion of 8a into compounds 10a and 9a were successfully applied to produce the
remaining set of compounds.
Scheme 1. Synthesis of 2,5-diaryl-tetrahydrofuran 4a-g and -furan 5a-g derivatives, comprising
methoxyl substituents at different positions of the aromatic ring, starting from the
condensation reaction of the racemic 1-phenylprop-2-yn-1-ol (6) with a diverse set of
commercially available aryl aldehydes 7a-g.
Owing to their significant biological activities [47] and natural occurrence [29,30],
tetrahydrofuran syntheses have been addressed by a wide variety of approaches, the
intramolecular SN2 reaction between a hydroxyl group and a leaving group being a convenient
and practical strategy to afford functionalised cyclic ethers [48]. Therefore, a straightforward
synthesis of a mixture of an equal amount of cis- and trans-2,5-diphenyl-tetrahydrofurans 4a-g
was achieved by cyclisation/dehydration reaction of the diaryl-dihydroxyl derivatives 10a-g,
according to a well-established protocol using trifluoroacetic acid in chloroform [49]. Apart
5

ACCEPTED MANUSCRIPT
from the known compounds 4a [50-55], 4d [56-57] and 4g [51,58-60], obtained by different
procedures, the novel diastereomers were isolated by column chromatography as a single spot
in moderate yield (30–65%) and assigned based on 1H NMR spectra, which showed
characteristic overlapping signals at 1.80–2.50 and 4.80–5.20 ppm related, respectively, to H-
3/H-4 and H-2/H-5 tetrahydrofuran stereoisomers. Additionally, the 13C NMR also revealed the
tetrahydrofuran methylene and methine carbons at 35.0 and 80.0 ppm.
Despite the availability of several metal-catalysed furan procedures [61], the synthesis
of 2,5-diphenylfurans 5a-g was achieved by microwave-assisted Pal-Knorr classical reaction[62]
via cyclisation of 1,4-diphenylbutane-1,4-dione intermediates 9a-g using triflic acid in 20 min,
rather than 68 h previously described [63]. The target furan structures 5c-e were confirmed by
1H and 13C NMR, which showed the signals of the furan ring at 6.30–6.80 ppm and 104.0–
107.0/150.0 ppm, respectively, and the NMR spectra of the known compounds 5a [64,65], 5b
[66], 5f [65] and 5g [66] were in accordance with the literature.
After preparing 2,5-diaryl-tetrahydrofuran 4a-g and -furan derivatives 5a-g, we next
evaluated their anti-trypanosomal activity on trypomastigote forms of Tulahuen using
colorimetric assay as a preliminary screening [67]. Thereafter, we assessed the anti-
trypanosomal activity of the most potent compounds against the infective Y parasite strain,
besides their ability to penetrate the host LLC-MK2 cells and kill the intracellular and
replicative T. cruzi amastigote form [68,69]. The natural compounds veraguensin (1) and
grandisin (2) and the drug benznidazole, currently used to treat Chagas disease, were also
assessed for comparative analysis.
2.2 Anti-trypanosomal activity and cytotoxicity
Trypomastigote forms of T. cruzi Tulahuen strain incubated with monkey kidney cells
were used to measure parasite viability in the presence of the synthesised compounds
(concentrations ranging from 3.75 to 500 µM). Compound concentrations corresponding to
50% anti-trypanosomal activity were calculated based on dose response curves and expressed
as EC50. Furthermore, cytotoxicity assays were performed in mice spleen cells and the
synthesised compounds were tested at concentrations from 1.50 to 250 µM, using Tween 80
as cell death control. The anti-trypanosomal and cytotoxic activities were compiled in Table 1,
besides the selectivity index (SI).
According to Table 1, in the 2,5-diaryl tetrahydrofuran 4a-g series, presenting closer
similarity to veraguensin (1) and grandisin (2), remarkable anti-trypanosomal activity on
6

ACCEPTED MANUSCRIPT
Tulahuen strain and selectivity was achieved by derivative 4e (2,4-OMe) with EC50 0.75 μM,
about twice as high as the activity showed by benznidazole (EC50 1.65 μM). The activity of 4e,
up to 50-fold more potent than the corresponding unsubstituted 4a (EC50 37.31 µM), revealed
a favourable effect of the 2,4-dimethoxy substitution pattern on the aromatic ring in improving
anti-trypanosomal activity. The efficacy also displayed by compound 4f (3-OMe) (EC50 2.29 μM)
and the ability of this series to not damage host cells make compounds 4e and 4f very
promising since they fulfill the TDR criteria for antichagasic agents, based on EC50 of <4.0 μM
and SI of ≥50. Moreover, the remaining methoxylated derivatives were deemed moderately
active (EC50 between 26.21 and 9.07 µM). Interestingly, despite the common 3,4-OMe
substituents that resemble veraguensin (1, EC50 2.67 µM), compound 4d displayed only
moderate efficacy against the trypomastigote parasite (EC50 24.63 µM), being one of the less
favourable modifications introduced in the tetrahydrofuran series, along with the
unsubstituted counterpart 4a (EC50 37.31 µM). A similar effect was achieved for grandisin (2,
EC50 28.59 µM) when compared to the corresponding derivative 4b bearing the 3,4,5-OMe
group, with closer parasite inhibition (EC50 26.21 µM).
Based on the promising results obtained with derivatives 4e and 4f against Tulahuen
strains, we also investigated their activities against the more infective T. cruzi Y strain and
found compound 4e preserves high anti-trypanosomal activity (EC50 1.05 µM) while 4f
exhibited only EC50 34.22 µM. Therefore, the maintenance of a single methoxylated phenyl
ring in the set of compounds 4, rather than two methoxylated phenyl rings, besides the
absence of the 3,4-dimethyl tetrahydrofuran substituents as observed in natural products 1
and 2, has proved to be an interesting simplification strategy applied to this class of
neolignans.
The influence of the number and position of the methoxyl group on the aromatic ring
was also evident in the 2,5-diaryl-furan series 5a-g. Among them, the 3,5-OMe derivative 5c
showed the strongest anti-trypanosomal activity of the current study, with EC50 0.01 µM, 75-
fold better than the corresponding compound 4e (2,4-OMe), the most potent in the
tetrahydrofuran series (EC50 0.75 μM), and more than a 150-fold better than benznidazole
(EC50 1.65 µM); Table 1. Besides the nanomolar activity, the cytotoxicity of 5c (CC50) was not
detectable until 250 µM, resulting in a highly selective agent (SI>250) for further optimisation
and development. In addition, the significant inhibitory effect on the parasite growth displayed
by compounds 5d (3,4-OMe) and 5e (2,4-OMe), with EC50 values of 1.34 and 1.12 µM,
respectively, also revealed the importance of the dimethoxylated substitution pattern on the
7

ACCEPTED MANUSCRIPT
aromatic ring in the 2,5-diaryl-furan series. On the other hand, the trimethoxylated 5b (29.10
µM) and unsubstituted counterparts 5a (21 µM) of this 2,5-diaryl-furan series led to decrease
in inhibitory activity, being less active than the monomethoxylated products 5f (3.34 µM) and
5g (4.75 µM), bearing monomethoxylated groups at the 3 and 4 positions, respectively. The
tests carried out with the more active diarylfurans 5c-g against T. cruzi Y strain confirm the
better profile of compound 5c (EC50 4.07 µM) compared to 5d-g (19.16-35.87 µM), despite the
lower than nanomolar range found with Tulahuen strain (0.01 µM) and reference benznidazole
(0.04 µM) against the same Y strain.
In order to investigate the impact of the tetrahydrofuran/furan central cores on anti-
trypanosomal activities, we also evaluated the anti-trypanosomal and cytotoxic activities of
the precursors related to rigid compounds (8a-g, 9a-g) and flexible (10a-g) open chain
intermediates, having different connection pattern between the substituted and unsubstituted
aromatic rings that can adopt different spatial orientations (Table 1). It was found that among
1,4-diaryl-2-butyn-1,4-diol precursors (8a-g) bearing a rigid alkynyl link, derivatives having
close methoxyl aryl substitution (3,4,5-OMe, 8b; 3,5-OMe 8c and 3,4-OMe, 8d) to grandisin (2)
and veraguensin (1), just in one ring, showed significant activities of 4.88, 4.98 and 8.98 µM,
respectively, whereas others were moderately active (39.11-13.97 µM). This set of compounds
was not cytotoxic in the tested concentrations up to 250 μM. Compared to the distinguished
activities displayed by some rigid 1,4-diaryl-2-butyn-1,4-diol precursors (8b-d), the anti-
trypanosomal activity induced by the corresponding reduced 1,4-diarylbutane-1,4-diol set
(10a-g), containing a flexible central chain linking both aromatic rings, was significantly lower
(19.88 to 193.10 µM). In addition to the decreased activities displayed by 10b (3,4,5-OMe) and
10e (2,4-OMe) derivatives, it was also accompanied by increased toxicity. Oxidation of these
intermediates (10a-g) rendered the constrained 1,4-diarylbutane-1,4-diketone analogues 9a-g,
which highlights 9c (3,5-OMe) and 9e (2,4-OMe), with EC50 1.69/15.91 and 7.79/1.76 µM for
Tulahuen/Y strains, respectively, as the most active of this series, despite lower activities than
those observed for the same methoxylated pattern required for higher activity in the furan (5c)
and tetrahydrofuran (4e) series. Therefore, taking into account the highest potency exhibited
by series 4, 5, 8 and 9, compared to the 1,4-diarylbutane-1,4-diol 10, it should be inferred that
a preferential constrained central core is essential for anti-trypanosomal activity, since in the
1,4-diketones series a specific rigid spatial arrangement might be adopted owing to the
conjugated dienol tautomer formation. Besides, this diketone series does not display
cytotoxicity at concentrations lower than 250 µM.
8

ACCEPTED MANUSCRIPT
Table 1. Anti-trypanosomal activities of novel diaryl-tetrahydrofuran (4a-g) and -furan (5a-g)
derivatives of natural products veraguensin (1) and grandisin (2) and the corresponding
precursors, 8-10(a-g), against T. cruzi Trypomastigote Tulahuen strain, and the most active
compounds against Y strain.
Trypoma Amastigo
Trypomastigote Cytotox
stigote
form of
Y strain
EC50(µM)
te form
of
Tulahuen
strain
SI
CC50/
EC50)
form of
Tulahuen
strain
icity
CC50
(µM)
Compound
EC50 (µM)
EC50(µM)
R1
R2
H
R3
H
R4
H
4a
H
37.31±24.1
26.21±7.3
12.09±1.9
24.63±0.3
0.75±0.6
NC
NC
NC
-
-
-
4b OMe OMe OMe
H
-
-
-
-
4c
OMe
H
OMe
H
H
4d OMe OMe
H
NC
NC
-
-
-
-
1.05±
0.05
19.77±
0.3
H
H
H
OMe
H
H
OMe
4e
4f
34.22±
0.1
125.7±
2.0
NC
NC
-
-
OMe
H
H
H
2.29±1.6
9.07±3.7
4g
OMe
-
-
193.60
±62.3
9.2
-
5a
5b
5c
5d
5e
5f
H
H
H
H
H
21.00±2.4
29.10±5.1
0.01±0.007
1.34±0.9
-
-
OMe OMe OMe
NC
NC
NC
NC
NC
-
-
4.08±
0.3
32.63±
0.6
35.87±
2.6
34.22± 125.60±
0.6 0.2
133.10±
4.0
125.8±
3.2
19.00±
2.7
OMe
H
OMe
H
H
-
-
OMe OMe
H
-
-
H
H
OMe
H
H
OMe
H
1.12±0.2
OMe
3.34±2.6
9

ACCEPTED MANUSCRIPT
19.16± 127.00±
-
5g
H
OMe
H
H
4.75±1.4
NC
0.6
4.5
8a
8b
8c
8d
8e
8f
H
H
H
H
H
H
H
18.67 ± 2.6
4.88 ± 1.5
NC
-
-
-
-
-
-
-
-
-
-
97.11±
5.3
OMe OMe OMe
NC
78.6
±6.4
OMe
H
OMe
H
4.98 ±1.5
8.98 ± 4.5
15.8
-
-
-
-
-
-
NC
NC
OMe OMe
-
H
H
H
OMe
H
H
OMe 37.87 ± 0.4
OMe
H
H
H
39.11 ±6.4
13.97±2.7
NC
NC
-
-
8g
OMe
198.60±
12.1
219.90
±12.7
9a
1.10
H
H
H
H
-
-
OMe OMe OMe
H
H
-
-
-
9b
9c
9d
9e
9f
36.87±3.3
1.69±0.05
35.96±5.7
7.79±0.8
NC
NC
NC
NC
NC
NC
-
-
-
-
-
-
15.91±
3.9
OMe
H
OMe
OMe OMe
H
H
H
OMe
H
-
-
-
-
-
1.76±
1.9
H
H
H
OMe
H
OMe
H
18.31±0.02
15.87±7.2
-
-
9g
OMe
H
10

ACCEPTED MANUSCRIPT
165.90±
22.2
115.20
±38.20
5.0
1.5
-
10a
H
H
H
H
H
H
H
33.16±4.31
77.50±7.9
-
-
-
-
-
-
-
-
10b OMe
10c OMe
OMe OMe
H
OMe
H
67.62±0.8
NC
NC
83.12±11.7
-
OMe
OMe
10d
236.40±
51.9
19.88±6.0
11.9
10e
10f
10g
H
H
H
OMe
H
H
OMe
H
OMe
H
-
-
-
-
-
-
193.10±8.8
68.07±6.7
NC
NC
-
-
OMe
H
NC: Not Cytotoxicity until 250 µM; EC50: Effective Concentration of compounds causing 50% parasite
death, being veraguensin (1): EC50 2.67± 0.5 µM, grandisin (2): EC50 28.59± 0.4 µM and benznidazole:
EC50 1.65± 0.3 µM, using trypomastigote T. cruzi Tulahuen strain. CC50: Cytotoxic Concentration of
compounds causing death to 50% of viable spleen cells (isolated from C57BL/6 mice). SI: Selective Index:
relation between CC50 (µM) (Cytotoxic Concentration of compounds) and EC50 (µM) (Effective
Concentration of compounds against Trypomastigote form of Tulahuen Strain).
To find out whether hit compounds (4e, 4f, 5c-g) might affect not only trypomastigote
forms, but also the replicative T. cruzi amastigote form located inside the host cell, we
performed additional assays using infected LLC-MK2 cells. As a result, the 2,4-dimethoxy-
diaryltetrahydrofuran 4e and 2,4-dimethoxy-diarylfuran 5e were the most potent of the
selected compounds when compared to the benznidazole drug (EC50 9.78 µM), with similar
EC50 values of 19.77 and 19 µM, respectively. These findings are in agreement with the
activities established in the tetrahydrofuran series against trypomastigote parasite, in which
compound 4e was also the most active. Conversely, the most active furan derivative 5c against
T. cruzi trypomastigote displayed only weak activity against the amastigote (EC50 133.1 µM),
about 13-fold lower than benznidazole, while the most potent 5e, identified in the amastigote
assay, was the second most potent against T. cruzi trypomastigote tests (Figure 2).
11

ACCEPTED MANUSCRIPT
100
80
60
40
20
0
100
5c
5d
5e
5f
4e
80
4f
8b
60
9c
5g
BZ
40
20
0
9e
BZ
0
1
2
3
0
1
2
3
log[C]
log[C]
Figure 2. Biological evaluation of the most active compounds identified in the trypomastigote
assays towards amastigote form of T. cruzi, Tulahuen strain.
2.3 2D-QSAR studies
Five molecular descriptors were selected by OPS-PLS methodology[70] and the model
(1) was obtained using 29 compounds with no outliers. Based on three mutually orthogonal
variables with 63.445% of information (LV1: 34.497%; LV2: 14.066%; LV3: 14.882%), obtained
from these descriptors, partial least squares (PLS) was conveniently applied to the model [71].
The obtained and predicted values of each compound are reported in the Supplementary
Material (Table S1).
pEC50ꢀ=ꢀ16.937ꢀ−ꢀ6.916(SpMin3_Bh(e))ꢀ−ꢀ0.878(VE1_B(s))ꢀ+ꢀ61.221(JGI6)ꢀ−ꢀ4.261(MATS3m)ꢀ+ꢀ0.1
52(CATS2D_05_AL) (1)
nꢀ=ꢀ29; R2ꢀ= ꢀ0.679; RMSECꢀ= 0.354; Fꢀ= 17.627; Q2LOOꢀ= 0.507; RMSECVꢀ= 0.439.
The internal validation shows that the model explains and predicts some information,
and fulfills the requirements of literature (R2 > 0.6; Q2LOO > 0.5), and shows, by the result of the
F test (Fcritical= 2.991, pꢀ= 3, and nꢀ–ꢀpꢀ–ꢀ1ꢀ= 25, αꢀ= 0.05), that the model is reliable and
significant. Moreover, the difference between R2 and Q2 (0.172) [72] suggests the model
LOO
does not undergo data over fitting. The model was also approved in the internal quality
robustness tests (LNO cross-validation) and chance correlation y-randomisation (Figure 3) [72-
74].
12

ACCEPTED MANUSCRIPT
Figure 3. Plots of LNO cross-validation (A) and y-randomisation tests (B).
Regarding the external validation, it indicates that Model 1 is realistic and applicable
for prediction purposes. Despite the lower correlation in the cross-validation process, the
external validation of the model showed a good external predictive power, with R2predꢀ= 0.848,
2
2
2
2
RMSEPꢀ= 0.190, average rm (LOO)-scaledꢀ= 0.753, Δrm (LOO)-scaledꢀ= 0.062, |R0 -R'0 |= 0.004,
kꢀ= 0.965 and k'ꢀ= 1.036 [34,75]. Besides the better values of R2pred in comparison with Q2LOO, it
is important to highlight that there is no relationship between internal and external
predictability (i.e., a model with low internal predictability may present high external
predictability, and vice versa); a paradox previously described by Kubinyi [76,77]. The results of
Golbraikh and Tropsha statistics and the RmSquare metrics also confirm the good external
predictability of the model [34,72].
All five selected descriptors are topological descriptors, which encode molecular
information such as size, shape, symmetry, branching, and rigidity, besides atom type and
bond multiplicity [78]. The order of importance of each descriptor is based on absolute values
of their autoscaled coefficients, (|0.647|ꢀxꢀSpMin3_Bh(e) > |0.398|ꢀxꢀVE1_B(s) > |0.369|ꢀxꢀJGI6
> |0.364|ꢀxꢀCATS2D_05_AL > |0.264|ꢀxꢀMATS3m). The meaning of each descriptor and their
corresponding values are shown in Tables S2 and S3, respectively, in the Supplementary
Material.
It is worth noting that the two most important descriptors, (SpMin3_Bh(e) and
VE1_B(s)) are related to molecular electronegativity and intrinsic electropological state [79].
Both may influence the anti-trypanosomal activity according to the position of the methoxyl
group on the aromatic ring, even in those with the same number of methoxyl groups, such as
compounds 8d (3,4-OMe) and 8e (2,4-OMe), 10d (3,4-OMe) and 10e (2,4-OMe). Taking into
account that SpMin3_Bh (e) is obtained based on Burden matrices (B(w)), where the diagonal
elements are atomic carbon-scaled properties (wi/wC),[79] a large number of electronegative
13

ACCEPTED MANUSCRIPT
atoms in the molecule may lead to a decrease of the descriptor value in the model since
trimethoxylated derivatives showed lower activity than the remaining products, with the
exception of 8b (3,4,5-OMe). Alternatively, the electronic effect (σ) of the methoxyl group,
with electron-donating effect at the para-position and electron-withdrawing at the meta-
position[80] may contribute to the high and moderate activities of compounds 4f (3-OMe), 5d
(3,4-OMe), 5f (3-OMe) e 9c (3,5-OMe), along with the most active of the series 5c (3,5-OMe),
containing two methoxyl groups at the meta-position, indicating that the decreased electron
density on the ring is actually relevant for the biological activity, but to a certain extent, which
can be evidenced by the negative sign of the descriptor SpMin3_Bh(e). Similarly, VE1_B(s) is
also obtained from Burden matrix, with close interpretation, and from the sum of the number
of vertices (i.e., atoms) present in an H-depleted molecular graph [79]. This indicates that the
more structures are branched (i.e., the larger the number of methoxyl groups) the greater the
tendency to be less active.
It was also evident that the influence of the number of methoxyl groups on the activity
is not enough to explain the entire model. The position of the substituent also plays an
important role due to the combination of electronic effects on the aromatic ring, as described
by the third descriptor (JGI6), which encodes the total charge transfer between atoms placed
at topological distance 6 (aromatic ring topology). Furthermore, the fourth descriptor
(CATS2D_05_AL) comprises topological distances formed by five bonds (lags) between
hydrogen bond acceptors and lipophilic chemical groups (i.e., two main characteristics of the
methoxyl group). The positive signal of this descriptor in model 1 means the more this
topological feature is present, the greater the tendency of the molecule to belong to the group
of the most active derivatives. Amongst the most active products, it is important to highlight
the prevalence of a methoxyl substitution pattern of R1 and R3, such as in compounds 9c and
5c, or R2 and R4, such as in9e, 4e, 5e and 10e, which is in agreement with the biological results.
Finally, the descriptor (MATS3m) involving atomic mass can be considered the least
important. An increase of the absolute value of this descriptor tends to increase the activity of
derivatives, indicating the tendency of the higher molecular weight compounds to be slightly
more active. In fact, unsubstituted and monosubstituted compounds (9a, 5a, 10a, 4a, 8f, 10g,
10f) represent 41% of the less active compounds (pEC50 < 4.729), while the most active (4e, 4f,
5f, 5g, 9f) make up 29% (pEC50 > 4.729).
2.4 Trypanothione reductase activity
14

ACCEPTED MANUSCRIPT
Although the phenotypic screening of diaryltetrahydrofuran and diarylfuran against the
trypomastigote and amastigote forms of T. cruzi (Tulahuen and Y strains) revealed potent
inhibitors, the identification of the corresponding molecular target are still essential for
subsequent lead optimization process and drug development [12].
Thus, the role played by furan [31] and nitrofuran derivatives, in particular, the redox-
cycling drugs nitrofurazone and nifurtimox [81,82] and the subversive substrates related to
guanidine-containing nitro-furans (under aerobic conditions) [83,84], on inhibition of T. cruzi
Trypanothione reductase (TR) motivated us to investigate the effect of the most active anti-
trypanosomal compounds from our series (IC50 < 12 μM) on this target, taking into account the
unique metabolic pathway and anti-oxidant functions of TR that provides a cellular reducing
environment to protect the parasite against oxidative stress [85].
(A)
(B)
Trypanothione reductase assay
5d
5f
100
50
0
30
20
10
0
5e
5g
Clo
Concentration (µM)
Figure 4. (A) Inhibition of trypanothione reductase by compounds 5d-g and comparison of their
IC50 values with clomipramine (Clo) (B) Corresponding curves of inhibition.
The assessment of TR activity was conducted using a continuous colorimetric
microplate assay (time-dependent TR inhibition of 30 min) based on the capacity of the
enzyme to reduce the trypanothione disulphide substrate (TS2) to the di-thiol product T(SH)2
in the presence of the NAPH cofactor. After incubation, the newly formed T(SH)2 is re-oxidized
back to the original substrate (TS2) by the addition of 5,5'-dithiobis-(2-nitrobenzoic acid)
(DTNB), which releases the chromophore 2-nitro-5-mercaptobenzoic acid (TNB), detected at
412 nm [86]. Thus, the assessment of the prospective diarylfurans and diaryl-tetrahydrofurans
TR inhibitors (50,0 μM concentration) revealed the diarylfuran series as the most active
15

ACCEPTED MANUSCRIPT
inhibitors, highlighting compounds 5d-e (IC50 19.2 and 17.7 μM) and 5f-g (IC50 8.9 and 7.4 μM),
respectively, with similar or 2-fold higher than clomipramine (IC50 15.2 μM) (Figure 4). It was
evident that a monomethoxylated pattern of the diarylfuran core (at either -para or -meta
position) had a significant impact on the TR inhibition.
Unexpectedly, compound 5c that displayed the strongest activity against Tulahuen and
Y strains of T. cruzi proved to be inactive, suggesting that its mechanism of action does not
involve TR activity inhibition. Furthermore, similar results were found for the tetra-hydrofuran
derivatives that showed no TR inhibition at tested concentrations and additional studies are
necessary to elucidate their mechanism of action.
3. Conclusions
Based on natural neolignan products, a series of fourteen novel diaryl-tetrahydrofuran
(4a-g) and -furan derivatives (5a-g) were synthesised as a simplified approach for the
development of more potent anti-trypanosomal agents. The cyclisation/dehydration
intramolecular SN2 reaction between a hydroxyl group and a leaving group was pursued to
produce functionalised cyclic ethers as a mixture of equal amounts of cis- and trans-2,5-
diphenyl-tetrahydrofurans 4a-g. In addition, microwave-assisted acceleration of the Pal-Knorr
classical reaction allowed the preparation of 2,5-diphenylfurans 5a-g in good to moderate
yields.
Preliminary biological results suggested that derivatives with aryl-methoxy groups
linked by a rigid central core, as observed in series 4, 5, 8 and 9, may adopt a favourable
orientation for target interactions in order to contribute with a beneficial entropic effect,
which is not achieved by similar aryl methoxyl groups connected by a flexible link (series 10).
Based on the significant impact of the number and variable positions of the aryl methoxyl
groups on the anti-trypanosomal activities, it should be inferred that the presence of a
dimethoxy substituent pattern in all series had a stronger anti-trypanosomal effect than the
corresponding monomethoxylated ones, which in turn, were better inhibitors than the
trimethoxylated (with exception of compound 8b) and unsubstituted counterparts.
2D-QSAR studies reinforce that the electronic effect of the methoxyl groups and their
corresponding substitution pattern (number and position) on the aromatic ring have a
significant influence on the overall electronic distribution in the final structure, affecting anti-
trypanosomal activities. These findings open up the possibility of further chemical
modifications to obtain more active derivatives, keeping, for instance, methoxyl group at R1
16

ACCEPTED MANUSCRIPT
and varying R2 with similar molecular weight and electron donating groups, such as fluoro,
cyano, trifluoromethyl and chlorine groups.
The identification of the most potent anti-trypanosomal non-chiral derivative in the
diarylfuran series (5c) suggests that the configuration of the 2,5-tetrahydrofuran carbons in
series 4 or even in the corresponding carbons of the rigid alkynyl open chain 8 are not crucial
for parasite growth inhibition. Additionally, the simplified approach that maintained a single
methoxylated aromatic ring rather than two, besides the elimination of two stereogenic
centres (3,4-dimethyl groups of this class of neolignans natural products 1 and 2) afforded
more active derivatives against both Tulahuen and Y strains, along with strong effects on
trypomastigote and amastigote forms of T. cruzi.
Finally, the assessment of trypanothione reductase inhibition using the most active
anti-trypanosomal derivatives, highlight the potential of some diarylfurans to inhibit parasite
growth by targeting its defense mechanism against oxidative stress through TR inhibition. In
summary, this study demonstrates the relevance of simplified neolignan-based derivatives to
act as stronger anti-trypanosomal agents than the parent natural product and, some of them,
as trypanothione reductase inhibitors.
4. Experimental
4.1. General information
All chemicals were purchased as reagent grade and used without further purification and
solvents were dried according to standard procedures [87]. Reactions were monitored by thin
layer chromatography (TLC) on pre-coated silica gel aluminum plates (60 GF254, Merck) with
the indicated eluents. Compounds were visualized under UV light (254 nm) and/or dipping in
ethanol–sulfuric acid (95:5, v/v), followed by heating the plate for a few minutes. Column
chromatography was performed on silica gel 60 (20–63 µm) or on a Biotage Horizon High-
Performance FLASH Chromatography system using 12 mm or 25 mm flash cartridges with the
1
indicated eluents. H and 13C Nuclear magnetic resonance spectra were recorded on Bruker
Advance DRX 300 (300 MHz), DPX 400 (400 MHz) or DPX 500 (500 MHz) spectrometers and
chemical shifts (δ) were expressed in parts per million (ppm), using tetramethylsilane (TMS) as
internal standard. Accurate mass electrospray ionization mass spectra (ESI-HRMS) were
obtained using positive or negative ionization modes on a Bruker Daltonics MicroOTOF II ESI-
TOF mass spectrometer.
17

ACCEPTED MANUSCRIPT
4.2 Synthesis
4.2.1 General procedure for synthesis of derivatives 2,5-diphenyl-tetrahydrofuran, 4a-g.
A solution of 1,4-diphenylbutane-1,4-diol derivatives (10a-g) in CHCl3 was cooled at 0 ºC and
trifluoracetic acid solution (5 eq) in CHCl3 was added drop wise and stirred for 2 h. After that,
the temperature was raised to ambient temperature for complete consumption of starting
material. The reaction mixture was quenched into saturated aqueous NaOH 10% solution and
extracted with dichloromethane. The organic layer was dried over anhydrous MgSO4. The
solvent was removed under reduced pressure and the mixture was purified by flash column
chromatography (silica gel), hexane/ethyl acetate (9:1 v/v).
2,5-diphenyl-tetrahydrofuran (4a)[50-55]
Following procedure described in section 4.2.4, the reaction of 1,4-diphenylbutane-1,4-diol
(10a) (20.3 mg, 0.085 mmol) in CHCl3 (2.0 mL) with TFA (32 µL, 0.425 mmol) in CHCl3 (1.0 mL)
gave the product 4a as a yellow oil, (8.06 mg, 0.036 mmol, 42%). 1H NMR (400 MHz, CDCl3): δ:
1.79-1.97 (4H, m, H-8, H-9); 4.68-4.79 (2H, m, H-7, H-10); 7.24-7.30 (2H, m, H-4); 7.30-7.39 (8H,
m, H-2, H-3, H-5, H-6). 13C NMR (100 MHz, CDCl3): δ: 35.2, 35.9 (C-8, C-9); 74.3, 74.7 (C-7, C-
10); 125.8 (C-2, C-6); 127.5, (C-4); 128.5 (C-3, C-5); 144.5, 144.6 (C-1, C-1’). ESI-HRMS: calcd for
C16H17O [M+H]+ 225.1279; found 225.1242.
2-phenyl-5-(3,4,5-trimethoxyphenyl)tetrahydrofuran (4b)
Following procedure described in section 4.2.4, the reaction of 1-phenyl-4-(3,4,5-
trimethoxyphenyl)butane-1,4-diol (10b) (25.2 mg, 0.076 mmol) in CHCl3 (2.0 mL) with TFA
(29.1 µL, 0.380 mmol) in CHCl3 (1.0 mL) gave the product 4b as a yellow oil, (12.25 mg, 0.039
1
mmol, 52%). H NMR (400 MHz, CDCl3): δ: 1.95-2.08 (2H, m, H-8a, H-9a); 2.40-2.54 (2H, m, H-
8b, H-9b); 3.84 (6H, s, OCH3); 3.88 (3H, s, OCH3); 5.00-5.12 (1H, m, H-7 or H-10); 5.17-5.30 (1H,
m, H-7 or H-10); 6.65 (2H, d, J 3.1 Hz, H-2’, H-6’); 7.26-7.31 (1H, m, H-4); 7.34-7.48 (4H, m, H-2,
H-3, H-5, H-6). ). 13C NMR (100 MHz, CDCl3) ): δ:35.5 (C-8,C-9), 56.0 (C11, C13), 60.8 (C12), 81.3
(C7-C10), 102.6 (C1´, C5´), 124.9 (C1-C5), 127.2 (C3), 127.4 (C2-C4), 137.2 (C3´), 139.3 (C6´),
143.6 (C6), 153,3 (C2´, C4´). ESI-HRMS: calcd for C21H25NaO4 [M+Na]+ 337.1416; found
337.1412.
2-(3,5-dimethoxyphenyl)-5-phenyltetrahydrofuran (4c)
18

ACCEPTED MANUSCRIPT
Following procedure described in section 4.2.4, the reaction of 1-(3,5-dimethoxyphenyl)-4-
phenylbutane-1,4-diol (10c) (20.7 mg, 0.068 mmol) in CHCl3 (2.0 mL) with TFA (26.5 µL, 0.342
1
mmol) in CHCl3 (1.0 mL) gave the product 4c as a yellow oil, (5.7 mg, 0.020 mmol, 30%). H
NMR (400 MHz, CDCl3): δ: 1.85-1.97 (2H, m, H-8a, H-9a); 2.29-2.42 (2H, m, H-8b, H-9b); 3.71
(3H, s, OCH3); 3.73 (3H, s, OCH3); 4.91-5.02 (1H, m, H-7 or H-10); 5.11-5.22 (1H, m, H-7 or H-
10); 6.30 (1H, q, J 2.3 Hz, H-4’); 6.51 (1H, d, J 2.3 Hz, H-2’ or H5’); 6.53 (1H, d, J 2.3 Hz, H-2’ or
H5’); 7.21-7.40 (5H, m, H-2, H-3, H-4, H-5, H-6). 13C NMR (100 MHz, CDCl3): δ: 34.0, 34.3, 35.4
(C-8, C-9); 55.3 (OCH3); 81.1, 81.2, 81.3, 81.4 (C-7, C-10); 99.1 (C-4’); 103.3, 103.9 ( C-2’, C-6’);
125.6, 126.0 (C-2, C-6); 127.2, 127.3 (C-4); 128.3, 128.4 (C-3, C-5); 145.7 (C-1, C-1’); 160.8 (C-3’,
C-5’). ESI-HRMS: calcd for C18H21O3 [M+H]+ 285.1491; found 285.1487.
2-(3,4-dimethoxyphenyl)-5-phenyltetrahydrofuran (4d)[56,57]
Following procedure described in Section 4.2.4, the reaction of 1-(3,4-dimethoxyphenyl)-4-
phenylbutane-1,4-diol (10d) (30.8 mg, 0.102 mmol) in CHCl3 (3.0 mL) with TFA (39.0 µL, 0.510
1
mmol) in CHCl3 (1.0 mL) gave the product 4d as a yellow oil, (18.75 mg, 0.066 mmol, 65%). H
NMR (300 MHz, CDCl3): δ: 1.96-2.06 (2H, m, H-8a, H-9a); 2.40-2.51 (2H, m, H-8b, H-9b); 3.84
(3H, s, OCH3); 3.88 (3H, s, OCH3); 4.99-5.11 (1H, m, H-7 or H-10); 5.24 (1H, dt, J 6.8 Hz, J 7.1 Hz,
H-7 or H-10); 6.65 (2H, d, J 3.1 Hz, H-2’, H4’); 7.26-7.32 (1H, m, H-6’); 7.32-7.45 (5H, m, H-2, H-
3, H-4, H-5, H-6). 13C NMR (75 MHz, CDCl3): δ: 34.2, 35.5, 35.6 (C-8, C-9); 55.8, 55.9 (OCH3);
81.1, 81.2 (C-7, C-10); 109.0, 109.5, 111.0 (C-2’, C-5’); 117.8, 118.1 (C-6’); 125.5, 125.9 (C-2, C-
6); 127.1, 127.2 (C-4); 128.3 (C-3, C-5); 136.1, 135.6, 142.5, 143.7 (C-1, C-1’); 148.2, 149.0 (C-3’,
C-4’). ESI-HRMS: calcd for C18H21O3 [M+H]+ 285.1491; found 285.1488.
2-(2,4-dimethoxyphenyl)-5-phenyltetrahydrofuran (4e)
Following procedure described in section 4.2.4, the reaction of 1-(2,4-dimethoxyphenyl)-4-
phenylbutane-1,4-diol (10e) (9.8 mg, 0.032 mmol) in CHCl3 (1.0 mL) with TFA (13.0 µL, 0.160
1
mmol) in CHCl3 (1.0 mL) gave the product 4e as a yellow oil, (3.98 mg, 0.014 mmol, 45%). H
NMR (400 MHz, CDCl3): δ: 1.82-2.00 (2H, m, H-8a, H-9a); 2.34-2.53 (2H, m, H-8b, H-9b); 3.81
(3H, s, OCH3); 3.82 (3H, s, OCH3); 5.00 (0.5 H, t, J 7.0 Hz, H-7 or H-10); 5.26 (1H, t, J 6.7 Hz, H-7
or H-10); 5.46 (0.5 H, t, J 6.7 Hz, H-7 or H-10); 6.45-6.53 (2H, m, H-3’, H-5’); 7.27-7.53 (6H, m,
H-2, H-3, H-4, H-5, H-6, H-6’). 13C NMR (100 MHz, CDCl3): δ: 33.1, 33.9, 34.3, 35.5 (C-8, C-9),
55.3, 55.4 (OCH3), 80.8, 80.9 (C-7, C-10), 98.3, 98.4 (C-3’), 103.8 (C-5’), 124.0, 124.6 (C-1’),
125.6 (C-2, C-6), 126.4, 126.7 (C-6’), 127.0, 127.2 (C-4), 128.3 (C-3, C-5), 143.2, 144.0 (C-1),
19

ACCEPTED MANUSCRIPT
157.2, 157.3, 159.8, 159.9 (C-2’, C-4’). ESI-HRMS: calcd for C18H21O3 [M+H]+ 285.1491; found
285.1488.
2-(3-methoxyphenyl)-5-phenyltetrahydrofuran (4f)
Following procedure described in section 4.2.4, the reaction of 1-(3-methoxyphenyl)-4-
phenylbutane-1,4-diol (10f) (36.9 mg, 0.135 mmol) in CHCl3 (3.0 mL) with TFA (51.8 µL, 0.677
1
mmol) in CHCl3 (1.0 mL) gave the product 4f as a yellow oil, (13.7 mg, 0.054 mmol, 40%). H
NMR (400 MHz, CDCl3): δ: 1.86-1.96 (2H, m, H-8a, H-9a); 2.33-2.42 (2H, m, H-8b, H-9b); 3.74
(3H, s, OCH3); 4.98 (1H, J 3.0 Hz, J, 4.0 Hz, J 7.0 Hz, H-7 or H-10); 5.16-5.21 (1H, m, H-7 or H-10);
6.72-6.77 (1H, m, H-6’); 6.89-6.97 (2H, m, H-2’, H-4’); 7.18-7.23 (2H, m, H-4, H-5’); 7.26-7.40
(4H, m, H-2, H-3, H-5, H-6). 13C NMR (100 MHz, CDCl3): δ: 34.3, 34.4, 35.6 (C-8, C-9); 55.2, 55.3
(OCH3); 81.1, 81.2, 81.3, 81.4 (C-7, C-10); 111.0, 111.6 (C-2’); 112.6, 112.7 (C-4’); 117.9, 118.3
(C-6’); 125.6, 126.0 (C-2, C-6); 127.2, 127.3 (C-4); 128.4 (C-3, C-5); 129.4 (C-5’); 142.9, 143.6,
144.8,145.5 (C-1, C-1’); 159.7 (C-3’). ESI-HRMS: calcd for C17H18NaO2 [M+Na]+ 277.1204; found
277.1236.
2-(4-methoxyphenyl)-5-phenyltetrahydrofuran (4g)[51,58-60]
Following procedure described in Section 4.2.4, the reaction of 1-(4-methoxyphenyl)-4-
phenylbutane-1,4-diol (10g) (42.2 mg, 0.155 mmol) in CHCl3 (3.0 mL) with TFA (59.3 µL, 0.775
1
mmol) in CHCl3 (1.0 mL) gave the product 4g as a yellow oil, (26.43 mg, 0.104 mmol, 67%). H
NMR (400 MHz, CDCl3): δ: 1.84-1.95 (2H, m, H-8a, H-9a); 2.25-2.42 (2H, m, H-8b, H-9b); 3.72
(3H, s, OCH3); 4.89-4.97 (1H, m, H-7 or H-10); 5.10-5.19 (1H, m, H-7 or H-10); 6.80-6.84 (2H, m,
H-3’, H-5’); 7.15-7.21 (1H, m, H-4); 7.25-7.37 (6H, m, H-2, H-3, H-5, H-6, H-2’, H-6’). 13C NMR
(100 MHz, CDCl3): δ: 34.3, 34.6, 35.6, 35.7 (C-8, C-9); 55.3 (OCH3); 81.1, 81.2 (C-7, C-10); 113.8
(C-3’, C-5’); 125.6, 126.0 (C-2, C-6); 127.0, 127.2 (C-4); 127.3, 127.4 (C-2’, C-6’); 128.4 (C-3, C-5);
134.9, 135.6 (C-1’); 143.1, 143.8 (C-1); 158.9, 159.0 (C-4’). ESI-HRMS: calcd for C17H18NaO2
[M+Na]+ 277.1204; found 277.1218.
4.2.2 General procedure for synthesis of derivatives 2,5-diphenylfuran, 5a-g.
A solution of 1,4-diphenylbutane-1,4-dione derivatives (9a-g) in MeCN was treated with triflic
acid (1.1 equiv). The reaction mixture was stirred for 20 min at 80 ºC (sealed microwave tube)
under microwave irradiation (100 W). After that, the reaction was quenched into saturated
NaHCO3 solution. The organic and aqueous phases were separated and the aqueous phase
extracted with EtOAc (3 x 2.0 mL). The organic layer was dried over anhydrous MgSO4.
20

ACCEPTED MANUSCRIPT
Removal of the solvent under reduced pressure followed by purification by flash column
chromatography (silica gel), hexane/ethyl acetate (9:1 v/v) afforded products (5a-g)
2,5-diphenylfuran (5a)[64,65]
Following procedure described in section 4.2.5, the reaction of 1,4-diphenylbutane-1,4-dione
(9a) (31.5 mg, 0.132 mmol) in MeCN (2.0 mL) with TfOH (12.8 µL, 0.145 mmol) gave the
1
product 5a as a white solid, (21.35 mg, 0.097 mmol, 74%). H NMR (400 MHz, CDCl3): δ: 6.74
(2H, s, H-8, H-9); 7.27 (2H, ddt, J 1.3 Hz, J 1.9 Hz, J 7.4 Hz, H-4); 7.41 (4H, ddt, J 1.3 Hz, J 1.9 Hz,
J 7.4 Hz, H-3, H-5); 7.75 (4H, ddt, J 1.2 Hz, J 1.9 Hz, J 7.3 Hz, H-2, H-6). 13C NMR (100 MHz,
CDCl3): δ: 107.2 (C-8, C-9); 123.7 (C-2, C-6); 127.3 (C-4); 128.7 (C-3, C-5); 130.8 (C-1); 153.4 (C-
7, C-10). ESI-HRMS: calcd for C16H12KO [M+K]+ 259.0525; found 259.0727.
2-phenyl-5-(3,4,5-trimethoxyphenyl)furan (5b)[66]
Following procedure described in section 4.2.5, the reaction of 1-phenyl-4-(3,4,5-
trimethoxyphenyl)butane-1,4-dione (9b) (15.0 mg, 0.046 mmol) in MeCN (1.0 mL) with TfOH
1
(5.0 µL, 0.051 mmol) gave the product 5b as a yellow solid, (4.30 mg, 0.014 mmol, 31%). H
NMR (300 MHz, CDCl3): δ: 3.89 (3H, s, OCH3); 3.96 (6H, s, OCH3); 6.68 (1H, d, J 3.5 Hz, H-8 or H-
9); 6.74 (1H, d, J 3.5 Hz, H-8 or H-9); 6.96 (2H, s, H-2’, H-6’); 7.30 (1H, m, H-4), 7.42 (2H, tapp, J
7.5 Hz, H-3, H-5), 7.74 (2H, m, H-2, H-6). 13C NMR (75 MHz, CDCl3): δ:56.2 (C11, C13), 61.0
(C12), 101.1 (C8, C9), 107.3, 107.0 (C1´, C5´), 123.7 (C1, C5), 126.5 (C3), 127.4 (C6), 128.7 (C2,
C4), 130.7 (C6´),137.7 (C3´),153.2 (C8, C9), 153.6 (C2´, C4´). ESI-HRMS: calcd for C19H19O4 [M+H]+
311.1283; found 311.1265.
2-(3,5-dimethoxyphenyl)-5-phenylfuran (5c) [65]
Following procedure described in section 4.2.5, the reaction of 1-(3,5-dimethoxyphenyl)-4-
phenylbutane-1,4-dione (9c) (27.2 mg, 0.091 mmol) in MeCN (1.5 mL) with TfOH (8.9 µL, 0.100
mmol) gave the product 5c as a yellow solid, (13.16 mg, 0.047 mmol, 52%). 1H NMR (400 MHz,
CDCl3): δ: 3.87 (6H, s, OCH3); 6.41 (1H, t, J 2.3 Hz, H-4’); 6.73 (1H, d, J 3.5 Hz, H-8 or H-9); 6.74
(1H, d, J 3.5 Hz, H-8 or H-9); 6.91 (2H, d, J 2.3 Hz, H-2’, H-6’); 7.28 (1H, dt, J 1.6 Hz, J 7.4 Hz, H-
4); 7.41 (2H, t, J 7.4 Hz, H-3, H-5); 7.74 (2H, dd, J 1.6 Hz, J 7.4 Hz, H-2, H-6). 13C NMR (100 MHz,
CDCl3): δ: 55.4 (OCH3); 99.6 (C-4’); 102.0 (C-2’, C-6’); 107.2, 107.8 (C-8, C-9); 123.8 (C-2, C-6);
127.4 (C-4); 128.7 (C-3, C-5); 130.7 (C-1); 132.5 (C-1’); 153.1, 153.4 (C-7, C-10); 161.1 (C-3’, C-
5’). ESI-HRMS: calcd for C18H16NaO3 [M+Na]+ 303.0997; found 303.0993.
21

ACCEPTED MANUSCRIPT
2-(3,4-dimethoxyphenyl)-5-phenylfuran (5d)
Following procedure described in section 4.2.5, the reaction of 1-(3,4-dimethoxyphenyl)-4-
phenylbutane-1,4-dione (9d) (21.8 mg, 0.073 mmol) in MeCN (1.5 mL) with TfOH (7.1 µL, 0.080
mmol) gave the product 5d as a yellow solid, (15.41 mg, 0.055 mmol, 75%). 1H NMR (300 MHz,
CDCl3): δ: 3.85 (3H, s, OCH3); 3.93 (3H, s, OCH3); 6.54 (1H, d, J 2.4 Hz, H-2’); 6.60 (1H, dd, J 8.6
Hz, J 2.4 Hz, H-6’); 6.74 (1H, d, J 3.4 Hz, H-8 or H-9); 6.88 (1H, d, J 3.4 Hz, H-8 or H-9); 7.24 (1H,
ddt, J 1.2 Hz, J 1.7 Hz, J 7.4 Hz, H-4); 7.39 (2H, tapp, J 7.4 Hz, H-3, H-5), 7.74 (2H, m, H-2, H-6);
7.89 (1H, d, J 8.6 Hz, H-5’). 13C NMR (75 MHz, CDCl3): δ: 56.0 (OCH3); 106.0, 107.3 (C-8, C-9);
111.5 (C-2’); 116.7 (C-5’); 123.6 (C-2, C-6, C-6’); 124.1 (C-1’); 127.2 (C-4); 128.7 (C-3, C-5); 130.9
(C-1); 148.8, 149.2 (C-3’, C-4’); 152.9, 153.4 (C-7, C-10). ESI-HRMS: calcd for C18H16NaO3 [M+Na]+
303.0997; found 303.0999.
2-(2,4-dimethoxyphenyl)-5-phenylfuran (5e)
Following procedure described in section 4.2.5, the reaction of 1-(2,4-dimethoxyphenyl)-4-
phenylbutane-1,4-dione (9e) (22.0 mg, 0.074 mmol) in MeCN (1.5 mL) with TfOH (7.2 µL, 0.081
mmol) gave the product 5e as a yellow solid, (14.40 mg, 0.050 mmol, 68%). 1H NMR (300 MHz,
CDCl3): δ: 3.92 (3H, s, OCH3); 3.98 (3H, s, OCH3); 6.62 (1H, d, J 3.5 Hz, H-8 or H-9); 6.72 (1H, d, J
3.5 Hz, H-8 or H-9); 6.91 (1H, d, J 8.4 Hz, H-5’), 7.22-7.29 (2H, m, H-4, H-3’), 7.32 (1H, dd, J 8.4
Hz, J 2.0 Hz, H-6’), 7.40 (2H, tapp, J 7.5 Hz, H-3, H-5); 7.73 (2H, dd, J 7.5 Hz, J 1.3 Hz, H-2, H-6).
13C NMR (75 MHz, CDCl3): δ: 55.4 (OCH3); 98.7 (C-3’); 104.8, 107.3 (C-8, C-9); 110.2 (C-5’); 113.3
(C-1’); 123.6 (C-2, C-6); 126.6 (C-6’); 126.9 (C-4); 128.6 (C-3, C-5); 131.0 (C-1); 150.0; 151.6 (C-7;
C-10); 156.7, 160.0 (C-2’, C-4’). ESI-HRMS: calcd for C18H16NaO3 [M+Na]+ 303.0997; found
303.0994.
2-(3-methoxyphenyl)-5-phenylfuran (5f)
Following procedure described in section 4.2.5, the reaction of 1-(3-methoxyphenyl)-4-
phenylbutane-1,4-dione (9f) (26.4 mg, 0.098 mmol) in MeCN (1.5 mL) with TfOH (7.2 µL, 0.081
mmol) gave the product 5f as a yellow solid, (15.75 mg, 0.063 mmol, 64%). 1H NMR (400 MHz,
CDCl3): δ: 3.88 (3H, s, OCH3); 6.74 (2H, s, H-8, H-9); 6.83 (1H, dt, J 2.5 Hz, J 8.4 Hz, H-4’); 7.24-
7.34 (4H, m, H-4, H-2’, H-5’, H-6’); 7.40 (2H, tapp, J 7.3 Hz, H-3, H-5); 7.75 (2H, dt, J 1.3 Hz, J 7.3
Hz, H-2, H-6). 13C NMR (100 MHz, CDCl3): δ: 55.3 (OCH3); 107.2, 107.6 (C-8, C-9); 109.2 (C-2’);
112.9 (C-4’); 116.4 (C-6’); 123.7 (C-2, C-6); 127.4 (C-4); 128.7 (C-3, C-5); 129.8 (C-5’); 130.7,
22

ACCEPTED MANUSCRIPT
132.0 (C-1, C-1’); 153.1, 153.4 (C-7, C-10); 159.9 (C-3’). ESI-HRMS: calcd for C17H15O2 [M+H]+
251.1072; found 251.1066.
2-(4-methoxyphenyl)-5-phenylfuran (5g) [66]
Following procedure described in section 4.2.5, the reaction of 1-(4-methoxyphenyl)-4-
phenylbutane-1,4-dione (9g) (24.3 mg, 0.091 mmol) in MeCN (1.5 mL) with TfOH (8.8 µL, 0.100
1
mmol) gave the product 5g as a yellow solid, (12.0 mg, 0.048 mmol, 53%). H NMR (300 MHz,
CDCl3): δ: 3.84 (3H, s, OCH3); 6.60 (1H, d, J 3.5 Hz, H-8 or H-9); 6.71 (1H, d, J 3.5 Hz, H-8 or H-9);
6.94 (2H, d, J 8.9 Hz, H-3’, H-5’); 7.25 (1H, ddt, J 1.2 Hz, J 1.9 Hz,J 7.5 Hz, H-4); 7.39 (2H, tapp, J
7.5 Hz, H-3, H-5); 7.68 (2H, dt, J 1.5 Hz, J 8.9 Hz, H-2’, H-6’); 7.73 (2H, m, H-2, H-6). 13C NMR (75
MHz, CDCl3): δ: 55.3 (OCH3); 105.6, 107.2 (C-8, C-9); 114.2 (C-3’, C-5’); 123.5 (C-2, C-6); 123.9
(C-1’); 125.2 (C-2’, C-6’); 127.1 (C-4); 128.7 (C-3, C-5); 130.9 (C-1); 152.7, 153.4 (C-7, C-10);
159.0 (C-4’). ESI-HRMS: calcd for C17H15O2 [M+H]+ 251.1072; found 251.1056.
4.3 Biological evaluation
Trypomastigote assay
To evaluate the anti-trypanosomal activity, cell-derived trypomastigote forms of T. cruzi,
Tulahuen strain stably expressing the β-galactosidase gene from Escherichia coli (Tulahuen-
lacZ), 5×105 trypomastigotes, were cultured in 96-well plates and incubated for 4 days at 37°C
with the synthesized compounds (products 4-5a-g and intermediates 8-10a-g), natural
products veraguensin (1) and grandisin (2) and benznidazole drug (Sigma-Aldrich) at serial
concentrations in a range between 500 to 3.75 µM. Thereafter, 50 μL of PBS containing 0.5%
of Triton X-100 and 100 μM chlorophenol Red-β-D-galactoside (CPRG-Sigma) were added.
Plates were incubated at 37°C for 4 h and absorbance was read at 570 nm.[67] Results of
parasite viability were measured based on the catalysis of CPRG by β-galactosidase and the
experiments were performed in duplicates. To test the compounds against T. cruzi Y strain,
trypomastigotes were platted in 96 well flat bottom plate at concentration of 1×106 parasites
mL-1 and the more active products on trypomastigote parasite, 4e-f and 5c-g and
intermediates 8b, 9c, were platted in serial concentrations (500 to 3.75 µM). After 24 h, the
parasites were removed and counted in Neubauer chamber as previously described.[68,69]
Amastigote assay
Anti-trypanosomal activities against amastigote form of T. cruzi Tulahuen Lac-Z strain were
performed in vitro. LLC-MK2 cells were resuspended in RPMI 1640 without phenol red (Gibco)
23

ACCEPTED MANUSCRIPT
supplemented with 5% FBS and antibiotics. LLC-MK2 cells were harvested at 2 x 104 cell ×mL-1
in 96 wells plate for 24 h and infected in a parasite/cell ratio 3:1 with the trypomastigote of T.
cruzi Tulahuen strain at 37°C. After 4 hours post-infection, cells were hard washed to remove
the extracellular parasites. The most active compounds against the trypomastigote forms
(products 4e-f, 5c-g and intermediates 8b and 9c), and veraguensin (1) and benznidazole drug
(Sigma-Aldrich) were added at concentrations of 500 µM to 0.24 µM, twenty-four hours after
infection and cultivated at 37°C. After 4 days of culture, the cells were extensive washed and
added 50µL of PBS containing 0.5% of Triton X-100 and 100µM chlorophenol red-β-D-
galactosidase (CPRG, Sigma Aldrich). Plates were incubated at 37°C for 4 h and absorbance was
measured at 570 nm. Results of parasite viability were determined at the base of the catalysis
of CPRG by β-galactosidase. Amastigotes without treatment was considered negative control
and infected cells treated with benznidazole, positive control.
Cytotoxicity assay
The cytotoxicity of compounds (products 4-5a-g and intermediates 8-10a-g), natural products
veraguensin (1) and grandisin (2) and benznidazole drug (Sigma-Aldrich) was evaluated on
spleen cells isolated from C57BL/6 mice at a concentration range of 250 to 1.50 µM. Spleens
were macerated in RPMI 1640 medium (Gibco-BRL Life Technologies, Grand Island, NY) and
incubated for 5 min with red blood cell lysis buffer (one part of 0.17M Tris–HCl and nine parts
of 0.16 M ammonium chloride). The isolated cells were centrifuged at 1500 rpm for 10 min and
resuspended in RPMI medium containing 5% fetal bovine serum (Life Technologies Inc.,
Bethesda, MD) and antibiotics (Sigma Chemical Co., St. Louis). The spleen cells were seeded
flat-bottom 96-well plates at 6.5×106/mL cells/well with different concentrations of the
synthesized compounds at 37°C for 24 h [88,89]. Tween 20 at 1% was used as cell death
positive control and benznidazole drug (Roche) was used as a reference drug. After 24 h, the
cells were incubated with 10 µg/mL propidium iodide (Sigma) and acquired using a
FACSCantoII (Becton-Dickinson Immunocytometry System Inc., San Jose, CA, USA). Data
analysis was performed using FlowJo software (Ashland, Oregon, USA). The experiments were
also performed in triplicates.
Trypanothione reductase assay
Recombinant trypanothione reductase from T. cruzi (TcTR) was expressed in
Escherichia coli BL21DE3 and purified by affinity chromatography. TR assays were performed
as described by Hamilton[86] in 96-well flat bottom micro plates (final volume = 240 µL),
containing TcTR (5 m-unit), HEPES (40 mM pH 7.5), NADPH (Sigma-Aldrich) (0.15 mM) and
24

ACCEPTED MANUSCRIPT
EDTA (1 mM), TS2 (Sigma-Aldrich) (1 μM) and tested compounds (50 µM), diluted in DMSO.
The reaction mixture was pre-incubated at 27ºC for 30 min and after 10μL of DNTB was added,
absorbance was read at 412 nm for 30 min in 5 min interval at 27ºC in a TECAN Infinite M200
micro plate reader. Clomipramine and DMSO 1% were used as positive and negative controls,
respectively. Compounds that inhibited > 50% of TR activity were tested again in serial dilution
(50μM -1.5μM) and IC50 values were calculated by a non-linear regression using the GraphPad
Prism program. All assays were done in triplicate and repeated three times [86].
4.4 Quantitative structure-activity relationships studies
Because most of the compounds are racemic mixtures, a set of more simple molecular
descriptors (as constitutional, topological and molecular descriptors) were obtained using only
the Simplified Molecular Input Line Entry System (SMILES) in the Dragon 6. The absolute
stereochemistry of the derivatives was not considered for generation of the strings. Next, it
was carried out a variable reduction step, to eliminate descriptors with irrelevant or redundant
information. Thus, the following features found in some descriptors were excluded: (i)
constant values; (ii) near-constant values; (iii) standard deviation of less than 0.001; (iv) at least
one missing value; (v) a pair correlation larger than or equal to 0.90; and (vi) absolute
Pearson’s correlation coefficient (|r|) values with the biological activity vector y (in the –log
EC50, or pEC50, format) less than 0.15. The first steps were also carried out in Dragon 6, and the
last step in 2D-QSAR Modeling [90]. A final verification was carried out by visual inspection,
using a resultant matrix with 104 descriptors. Variable selection was performed using the
ordered predictor selection (OPS) algorithm.[49] The models were constructed based on the
partial least squares (PLS). Statistical tools to validate the quality of the fit, significance, and
predicting power of the models were applied in order to evaluate the capacity of the models
to generate reliable predictions of the biological activity under study to find out new active
derivatives [91]. The validation approach comprised the internal and external steps, using 29
and 5 compounds, respectively. The set compounds selected for external validation was
chosen according to their biological activities encompassing a range of 2.423 logaritimic units.
The external validation results were obtained from the Xternal Validation Metric Calculator
program [92]. In order to maintain the quality of the model, compound 5c was removed from
the study for presenting a pEC50 value outside the range of the biological activity variation,
higher than 1.875 logaritimic units in relation to the second more active derivative 4e.
25

ACCEPTED MANUSCRIPT
Acknowledgments
This work was supported by Grants from FAPESP (Fundação de Amparo à Pesquisa do
Estado de São Paulo), CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico),
CAPES (Comissão de Aperfeiçoamento do Pessoal de Nível Superior) and FAPPR (Fundação
Araucária) in Brazil. We are grateful to Vinícius Palaretti and José Carlos Tomaz for NMR and
ESI-HRMS analyses, respectively.
Supplementary data
Supplementary data (synthesis of precursors 8-10 (a-g), 1H, 13C NMR, DEPT-135,
HMQC, HMBC and HRMS (ES+) spectra for all compounds) associated with this article can be
found, in the online version, at ... These data include MOL files of the most important
compounds described in this article.
Rerefences
Acknowledgments
This work was supported by Grants from FAPESP (Fundação de Amparo à Pesquisa do
Estado de São Paulo), CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico),
CAPES (Comissão de Aperfeiçoamento do Pessoal de Nível Superior) and FAPPR (Fundação
Araucária) in Brazil. We are grateful to Vinícius Palaretti and José Carlos Tomaz for NMR and
ESI-HRMS analyses, respectively.
Supplementary data
Supplementary data (synthesis of precursors 8-10 (a-g), 1H, 13C NMR, DEPT-135,
HMQC, HMBC and HRMS (ES+) spectra for all compounds) associated with this article can be
found, in the online version, at ... These data include MOL files of the most important
compounds described in this article.
Rerefences
[1] D.J. Newman, G.M. Cragg, Natural products as sources of new drugs from 1981 to 2014,
Nat. Prod. 79 (2016) 629-661. http://dx.doi.org/10.1021/acs.jnatprod.5b01055
[2] P. M. Cheuka, G. Mayoka, P. Mutai, K. Chibale, The role of natural products in drug
discovery and development against neglected tropical diseases, Molecules 22 (2017) 58.
http://dx.doi.org/10.3390/molecules22010058
[3] A. G. Atanasov, B. Waltenberger, E.M. Pferschy-Wenzig, T. Linder,C. Wawrosch, P. Uhrin, V.
Temml, L. Wang, S. Schwaiger, E. H. Heiss,J. M. Rollinger, D. Schuster, J. M. Breuss, V. Bochkov,
M.D. Mihovilovic, B. Kopp, R. Bauer, V. M. Dirsch, H. Stuppner, Discovery and resupply of
26

ACCEPTED MANUSCRIPT
pharmacologically active plant-derived natural products: A review, Biotechnol. Adv. 33 (2015)
1582–1614. http://dx.doi.org/10.1016/j.biotechadv.2015.08.001
[4] C. Hertweck, Natural products as source of therapeutics against parasitic diseases, Angew.
Chem. Int. Ed. 54 (2015) 1462 –14624. http://dx.doi.org/10.1002/anie.201509828
[5] A.L. Harvey, R.Edrada-Ebel, R.J. Quinn, The re-emergence of natural products for drug
discovery in the genomics era, Nat. Rev. Drug Discov. 14 (2015) 111-129. http://dx.doi.org/
10.1038/nrd4510
[6] D.G.I. Kingston, Modern natural products drug discovery and its relevance to biodiversity
conservation, J. Nat. Prod. 74 (2011) 496–511. http://dx.doi.org/10.1021/np100550t
[7] G.M. Cragg, D.J. Newman, Natural products: a continuing source of novel drug leads,
Biochim.
Biophys.
Acta
1830
(2013)
3670-3695.
http://dx.doi.org/
10.1016/j.bbagen.2013.02.008
[8] F. Annang, G. Pérez-Moreno, R. García-Hernández, C. Cordon-Obras, J. Martín, J. R. Tormo,
L. Rodríguez, N. de Pedro, V. Gómez-Pérez, M. Valente, F. Reyes, O. Genilloud, F. Vicente, S.
Castanys, L. M. Ruiz-Pérez, M. Navarro, F. Gamarro, and D. González-Pacanowska, High-
Throughput Screening platform for natural product–based drug discovery against 3 neglected
tropical diseases: Human African Trypanosomiasis, Leishmaniasis, and Chagas Disease, J.
Biomol. Screen. 20 (2015) 82 –91. http://dx.doi.org/10.1177/1087057114555846
[9] R.J. Neitz, S. Chen, F. Supek, V. Yeh, D. Kellar, J. Gut, C. Bryant, A. Gallardo-Godoy, V.
Molteni, S. L. Roach, A.K. Chatterjee, S. Robertson, A.R. Renslo, M.Arkin, R. Glynne, J.
McKerrow, J.L. Siqueira-Neto, Lead identification to clinical candidate selection: drugs for
Chagas disease, J. Biomol. Screen. 20 (2015) 101-111. http://dx.doi.org/
10.1177/1087057114553103
[10] D.Ndjonka, L.N. Rapado, A.M. Silber, E. Liebau, C. Wrenger, Natural products as a source
for treating neglected parasitic diseases, Int. J. Mol. Sci. 2013, 14, 3395-3439.
http://dx.doi.org/10.3390/ijms14023395
[11] World Health Organization – WHO: http://www.who.int/chagas/en/, accessed in 12th July
2016.
[12] M. C. Field, D. Horn, A. H. Fairlamb, M. A. J. Ferguson, D. W. Gray, K. D. Read, M. De
Rycker, L. S. Torrie, P. G. Wyatt, S. Wyllie, I. H. Gilbert, Anti-trypanosomatid drug discovery: an
ongoing challenge and a continuing need, Nature Reviews Microbiology (2017) 1-15
http://dx.doi.org/10.1038/nrmicro.2016.193.
[13] E. Izumi, T. Ueda-Nakamura, B.P. Dias Filho, V.F. Veiga Junior, C.V. Nakamura, Natural
products and Chagas’ disease: a review of plant compounds studied for activity against
Trypanosoma cruzi, Nat. Prod. Rep. 28 (2011) 809-823. http://dx.doi.org/10.1039/c0np00069h
[14] Web of Science (2017). URL <http://apps.webofknowledge.com>. Accessed 2017 June
[15] V.P. Sulsen, V. Puente, D.Papademetrio, A. Batlle, V.S. Martino, F.M. Frank, M.E.
Lombardo, Mode of action of the sesquiterpene lactones psilostachyin and psilostachyin C on
Trypanosoma cruzi, Plos One 3 (2016) 1-14. http://dx.doi.org/10.1371/journal.pone.0150526
[16] D.S.A. Maciel, V.P. Freitas, G.A.A. Conserva, T.R. Alexandre, S.U. Purisco, A.G. Tempone,
M.S.C. Melhem, M.J. Kato, E.F. Guimarães, J.H.G. Lago, Bioactivity-guided isolation of
laevicarpin, an antitrypanosomal and anticryptococcal lactam from Piper laevicarpu
(Piperaceae), Fitoterapia 111 (2016) 24–28. http://dx.doi.org/10.1016/j.fitote.2016.04.005
[17] N.J. Nwodo, F.B.C. Okoye, D. Lai, A. Debbab, R. Brun, P. Proksch, Two trypanocidal
27

ACCEPTED MANUSCRIPT
dipeptides from the roots of Zapoteca portoricensis (Fabaceae), Molecules 19 (2014) 5470-
5477. http://dx.doi.org/10.3390/molecules19055470
[18] P. Veiga-Santos, V.C. Desotia, N. Mirandaa, T. Ueda-Nakamuraa, B.P. Dias-Filho, S.O. Silva,
D.A.G. Corteza, J.C.P. Mello, C.V. Nakamura, The natural compounds piperovatine and
piperlonguminine induce autophagic cell death on Trypanosoma cruzi , Acta Trop. 125 (2013)
349-356. http://dx.doi.org/10.1016/j.actatropica.2012.11.014
[19] F.M. Frank, J. Ulloa, S.I. Cazorla, G. Maravilla, E.L. Malchiodi, A. Grau, V. Martino, C.
Catalán, and L.V. Muschietti, Trypanocidal activity of Smallanthus sonchifolius: identification of
active sesquiterpene lactones by bioassay-guided fractionation, Evid. Based Complement.
Alternat. Med. (2013) 1-8. http://dx.doi.org/10.1155/2013/627898
[20] A. Rea, A.G. Tempone, E.G. Pinto, J.T. Mesquita, E. Rodrigues, L.G.M. Silva, P. Sartorelli,
J.H.G. Lago, Soulamarin isolated from Calophyllum brasiliense (Clusiaceae) induces plasma
membrane permeabilization of Trypanosoma cruzi and mytochondrial dysfunction, PLoS Negl.
Trop. Dis. 7 (2013) e2556. https://doi.org/10.1371/journal.pntd.0002556
[21] V.P. Sulsen, S.I. Cazorla, F. M. Frank, L.C. Laurella, L.V. Muschietti, C.A. Catalán, V.S.
Martino, E.L. Malchiodi, Natural terpenoids from Ambrosia species are active in vitro and in
vivo against human pathogenic Trypanosomatids, PLoS Negl. Trop. Dis. 7 (2013) e2494.
https://doi.org/10.1371/journal.pntd.0002494
[22] S.S. Grecco, J.Q. Reimão, A.G. Tempone, P. Sartorelli, R.L.O.R. Cunha, P. Romoff, M.J.P.
Ferreira, O.A. Fávero, J.H.G. Lago, In vitro antileishmanial and antitrypanosomal activities of
flavanones from Baccharis retusa DC. (Asteraceae), Exp. Parasitol. 130 (2012) 141-145.
https://doi.org/10.1016/j.exppara.2011.11.002
[23] E. Lozano, P. Barrera, R. Salinas, I.Vega, M. Nieto, C.Tonn, U. Kemmerling, R.A. Mortara,
M. A. Sosa, Sesquiterpene lactones and the diterpene 5-epi-icetexone affect the intracellular
and extracellular stages of Trypanosoma cruzi, Prasitol. Int. 61 (2012) 628-633.
https://doi.org/10.1016/j.parint.2012.06.005
[24] T.R. Morais, P. Romoff, O.A. Fávero, J.Q. Reimão, W.C. Lourenço, A.G. Tempone, A.D.
Hristov, S.M. Di Santi, J.H.G. Lago, P. Sartorelli, M.J.P. Ferreira, Anti-malarial, anti-
trypanosomal, and anti-leishmanial activities of jacaranone isolated from Pentacalia
desiderabilis (Vell.) Cuatrec. (Asteraceae), Parasitol. Res. 110 (2012) 95–101.
https://doi.org/10.1007/s00436-011-2454-9
[25] C. Marín, I. Ramírez-Macías, A. López-Céspedes, F. Olmo, N. Villegas, J.G. Díaz, M.J.
Rosales, R. Gutíerrez-Sánchez, M. Sánchez-Moreno, In vitro and in vivo trypanocidal activity of
flavonoids from Delphinium staphisagria against Chagas Disease, J. Nat. Prod. 74 (2011) 744–
750. https://doi.org/10.1021/np1008043
[26] B.C. Almeida, B.Q. Araújo, A.A. Carvalho, S.D.L. Freitas, D.D.A Maciel, A.J.S. Ferreira, A.G.
Tempone, L.F. Martins, T.R. Alexandre, M.H. Chaves, J.H.G. Lago, Antiprotozoal activity of
extracts and isolated triterpenoids of “carnauba”(Copercinia prunifera) wax from Brazil,
Pharmaceutical
Biology
54
(2016)
3280-3284.
https://doi.org/10.1080/13880209.2016.1224257
[27] T.J. Schmidt, S.A. Khalid, A.J. Romanha, T.M.A. Alves, M.W. Biavatti, R. Brun, F.B. Da Costa,
S.L. de Castro, V.F. Ferreira, M.V.G. de Lacerda, J.H.G. Lago, L.L. Leon, N.P. Lopes, R.C. das
Neves Amorim, M. Niehues, I.V. Ogungbe, A.M. Pohlit, M.T. Scotti, W.N. Setzer, M. de N.C.
Soeiro, M. Steindel, A.G. Tempone, The potential of secondary metabolites from plants as
drugs or leads against protozoan neglected diseases - Part I, Curr. Med. Chem. 19 (2012) 2128-
2175. https://doi.org/10.2174/092986712800229023
28