Preparation and antitrypanosomal activity of secochiliolide acid derivatives

Article abstract of DOI:10.1016/j.bmcl.2013.06.064

Secochiliolide acid (1) isolated from the Patagonian shrub Nardophyllum bryoides, was used as a scaffold for the preparation of a series of nine derivatives. Compound 1 and its derivatives were tested against Trypanosoma cruzi epimastigotes grown in liquid media. It was first observed that secochiliolide acid (1) inhibited the proliferation of the parasites, with an IC₅₀ of 2 μg/mL. Six of the synthesized derivatives were also active with IC₅₀'s between 2 and 7 μg/mL which are comparable to that of the commercial drug benznidazole (2.5 μg/mL). These results indicate that the carboxyl group is not essential for the bioactivity of 1, while the presence of the tetrasubstituted exocyclic double bond seems to be important. Moreover, the presence of the furan and spirolactone rings is not essential for the bioactivity per se, but is important in combination with other structural fragments present in the molecule.

Full text of DOI:10.1016/j.bmcl.2013.06.064

Bioorganic & Medicinal Chemistry Letters 23 (2013) 4964–4967
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Preparation and antitrypanosomal activity of secochiliolide acid
derivatives
Gastón E. Siless a, , Esteban Lozano b, , Marianela Sánchez a, Marcia Mazzuca c, Miguel A. Sosa b,d
,
Jorge A. Palermo a,
⇑
a UMYMFOR-CONICET – Departamento de Química Orgánica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria,
Pab. 2 (1428), Buenos Aires, Argentina
b Laboratorio de Biología y Fisiología Celular ‘Dr. Francisco Bertini’, Instituto de Histología y Embriología, CONICET, Facultad de Ciencias Médicas,
Universidad Nacional de Cuyo, Mendoza, Argentina
c Dpto. de Química, Facultad de Ciencias Naturales, Universidad Nacional de la Patagonia San Juan Bosco, Km 4 (9000), Comodoro Rivadavia, Chubut, Argentina
d Instituto de Ciencias Básicas (ICB), Universidad Nacional de Cuyo, Mendoza, Argentina
a r t i c l e i n f o
a b s t r a c t
Article history:
Secochiliolide acid (1) isolated from the Patagonian shrub Nardophyllum bryoides, was used as a scaffold
for the preparation of a series of nine derivatives. Compound 1 and its derivatives were tested against
Trypanosoma cruzi epimastigotes grown in liquid media. It was first observed that secochiliolide acid
Received 22 April 2013
Revised 12 June 2013
Accepted 17 June 2013
Available online 3 July 2013
(1) inhibited the proliferation of the parasites, with an IC50 of 2
were also active with IC50’s between 2 and 7 g/mL which are comparable to that of the commercial drug
benznidazole (2.5 g/mL). These results indicate that the carboxyl group is not essential for the bioactiv-
lg/mL. Six of the synthesized derivatives
l
l
Keywords:
ity of 1, while the presence of the tetrasubstituted exocyclic double bond seems to be important. More-
over, the presence of the furan and spirolactone rings is not essential for the bioactivity per se, but is
important in combination with other structural fragments present in the molecule.
Ó 2013 Elsevier Ltd. All rights reserved.
Secochiliolide acid
Trypanosoma cruzi
Nardophyllum bryoides
trypanocidal activity
Trypanosoma cruzi is a monoflagellate parasite that causes Cha-
gas’ disease, one of the major parasitic diseases in Latin America.
These parasites go through various stages between vertebrate
and invertebrate hosts.1–4 The flagellate non-infective stage epim-
astigotes have been the most used for initial testing of drugs, be-
cause they can be grown in vitro and their life cycle is well
known.5 Although many efforts have been made for decades to find
effective drugs against this parasite, the results are still frustrating.
The main cause of failure for most of the compounds with trypan-
ocidal activity is the occurrence of significant side effects on the
patients, making them non-effective during the chronic phase of
the disease.6,7 Benznidazole and Nifurtimox have been compounds
widely used against Chagas’ disease, although their use is re-
stricted because of their side effects.7–9 Therefore, the search for
new compounds with trypanocidal activity remains in the interest
of many laboratories. In this endless search, one of the strategies is
the screening of plant natural products, because of their abundance
and structural variety in nature, and that in some cases they can be
obtained in large quantities.6,10
In a previous Letter, we reported the isolation of secochiliolide
acid (1) and other compounds from the Patagonian shrub
Nardophyllum bryoides, and a preliminary study on their cytotoxic
activity.11 Secochiliolide acid was first isolated by Bohlmann and
co-workers from Nardophyllum lanatum and the related species
Chiliotrichum rosmarinifolium.12 As was usual at that time, the
authors only described the isolation and structure elucidation,
and no bioactivity studies were performed on the compounds.
Secochiliolide acid has a rare seco ent-halimane structure, with a
spirolactone and a furan ring as characteristic moieties, and offers
several sites for possible structural modifications. These structural
features, together with the fact that 1 is a major metabolite in the
ethanolic extract of N. bryoides and can be isolated in large
amounts, make this compound an interesting scaffold for the prep-
aration of derivatives. In particular the spirolactone moiety is pres-
ent insome natural trypanocidal compounds such as psylostachyin,
a sesquiterpene lactone, identified as one of the active components
of Ambrosia tennuifolia, and previously isolated from other plants of
the same genus,13,14 which made secochiliolide acid an interesting
compound to test for this type of activity. This hypothesis was re-
warded by initial promising results, which led to the preparation of
a series of derivatives that are described herein, as well as their ef-
fects on T. cruzi epimastigotes. Some of these compounds were ac-
⇑
Corresponding author.
E-mail address: palermo@qo.fcen.uba.ar (J.A. Palermo).
These authors contributed equally to this work.
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.06.064

G. E. Siless et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4964–4967
4965
tive against the parasites at low IC50’s, comparable to that of
benznidazole.
by elimination of hydrazoic acid from an initially formed acylazide
to yield a ketene, which then reacts with t-butanol to produce the
ester 3.16 A possible hypothesis on the formation of compounds 3
and 4 is outlined in Scheme 2 (Supplementary data). In order to
prevent the formation of compound 3, toluene was used as solvent
instead of t-butanol, which enabled the obtention of the isocyanate
5. Treatment of crude 5 with base (KOH) in acetonitrile/H2O
yielded, instead of the expected primary amine, the dimeric urea
6 (75%). The formation of this interesting symmetric derivative
can be explained by the attack of the initially formed amine, acting
as a good nucleophyle, to the isocyanate 5. This attack is faster than
the corresponding by the hydroxyl group.
The incorporation of p-benzoquinone as a structural fragment,
usually produces compounds with great bioactivity. In a previous
publication we reported the preparation of norcholane-p-benzo-
quinone hybrids by means of a Barton decarboxylation reaction,
which was inspired in a synthesis of avarone.17–19 Following the
same strategy, the Barton ester of 1 was prepared by reaction with
N-hydroxy-2-thiopyridone in the usual way. Irradiation of the Bar-
ton ester 7 with a 300 W tungsten lamp generated the alkyl radical
produced by decarboxylation of 1, which was trapped by an excess
of p-benzoquinone to yield the adduct 8, which was reductively
desulfurized with Raney Ni in CH2Cl2 to yield hydroquinone 9 in
very good yield. Compound 9 was finally oxidized to the quinone
10 with MnO2.
A simplified protocol was established for the isolation of 1 in a
preparative scale which took advantage of the acidity of the
compound, minimizing the chromatographic steps and avoiding
the use ofHPLC for the final purification. As in our previous work,
the crude ethanolic extract of fresh N. bryoides was concentrated
to an aqueous suspension and then partitioned between MeOH:H2-
O (9:1) and cyclohexane to yield lipophilic and polar subextracts.
The latter was concentrated under reduced pressure and then par-
titioned between EtOAc and 10% aqueous NaOH. The basic fraction
was then acidified with HCl and extracted with EtOAc. This final or-
ganic fraction contained mainly 1 and flavonoids, and the final
purification of the desired compound was achieved by Sephadex
LH-20 permeation and flash column chromatography (Supplemen-
tary data).
The structural modifications performed on 1 were focused on
the carboxyl group and the double bond. The modifications on
the carboxyl group include the introduction of functional groups
bearingnitrogen atoms by means of the Curtius rearrangement,
and the coupling with a benzoquinone moiety by means of a Bar-
ton decarboxylation reaction. The modifications performed on the
double bond include a diasteroselective epoxidation and a bromo-
cyclization reaction. All these modifications are depicted in
Scheme 1.
Transformation of the carboxyl group into an acylazide would
lead, by a Curtius rearrangement, to the obtention of the corre-
sponding isocyanate, which in turn could be attacked by a variety
of nitrogen nucleophiles, yielding aza-nor derivatives of 1. Reaction
of 1 with diphenylphosphorazidate (DPPA),15 a mild azide donor,
and t-butanol in an one-pot attempt to obtain the t-butylcarba-
mate, gave a mixture of three products: the desired product 2
(19%), the t-butyl ester 3 (17%) and the carbamoylazide 4 (24%).
The presence of compound 3 as a by-product is interesting since
it is not easily prepared by other techniques, and can be explained
The epoxidation of the double bond of 1 would give access to a
variety of additional derivatives, while the stereochemical rela-
tionships of the substituents in the six-membered ring of com-
pound 1 would probably influence the diasteroselectivity of the
reaction. Epoxidation with MCPBA was performed both on 1 and
its methyl ester, which in turn was prepared by reaction of 1 with
CH2N2 in ether. Reaction of the methyl ester of 1 with MCPBA in
CH2Cl2 at 0° proceeded with complete conversion, yielding a 5:1
mixture of diasteromers, 11 and 12. Both diasteromers could be
purified by HPLC, and completely characterized. The major product
Scheme 1. Reagents and conditions: (a) DPPA, Et3N, tBuOH, reflux 2 h; (b) DPPA, Et3N, dry toluene, reflux 3 h; (c) KOH, CH3CN:H2O; (d) DCC, CH2Cl2, 0°; (e) CH2Cl2, 0°, h
m
(300 W), 300; (f) Raney Ni, DME, reflux 200; (g) MnO2, Et2O, rt 200; (h) (i) CH2N2, ether, (ii) mCPBA, CH2Cl2, 0°; (i) (i) NBS, NaHCO3, CHCl3:EtOH (3:1), rt, (ii) HCl, acetone, rt.

4966
G. E. Siless et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4964–4967
steric effects. The use of a hydroxyl group to induce stereoselectiv-
ity has been previously reported.20
Encouraged by the previous results, a bromonium-catalyzed
cyclization was performed on compound 1. In this case, it was ex-
pected that the carboxyl side chain on the
a-face would attack
from the opposite face of a cyclic bromonium ion to give a lactone
as final product. Bromination of 1 with NBS/NaHCO3 in CHCl3:-
MeOH (3:1) gave lactone 13 as the main product. The spectroscopic
analysis of 13 (MS, NMR) showed that the molecule had no bro-
mine atoms, and that an isopropenyl group had been formed. This
structure is consistent with a reaction mechanism in which anini-
tially formed bromonium ion on the less hindered b-face was at-
tacked from the opposite face by the carboxylate ion formed in
the basic medium to yield an intermediate cis-lactone (which could
not be detected), which then eliminates HBr to give compound 13.
Compound 13, named secochiliolide lactone, had been previously
isolated as a natural product by Bohlmann together with secochi-
liolide acid 1 from N. lanatum.12 The easy formation of compound
13 from 1 suggests a possible biosynthetic pathway which can re-
late both compounds.
Figure 1. Effect of secochiliolide acid on the growth of T. cruzi epimastigotes.
Parasites were incubated with the compound at the indicated concentrations.
DMSO used as solvent did not affect the growth of parasites.
11 had the epoxide at the b-face, which could be explained by the
combined steric effects of the side chain and Me-17, which are
both located at the
a-face. However, when the same reaction
was performed on the acid 1, the diasterofacial selectivity was lost,
yielding a 1:1 mixture of diasteromeric epoxides. This may be
explained if the free carboxyl group directs the epoxidation on
The activity of 1 and several derivatives was tested against T.
cruzi epimastigotes grown in liquid media.21 It was first observed
that secochiliolide acid (1) inhibited the proliferation of parasites,
the
a-face, which would counterbalance the above mentioned
partially at concentrations between 1 and 5 lg/mL, and completely
Figure 2. Effect of semisynthetic derivatives of secochiliolide acid on the growth of epimastigotes. The points on each curve represent the means of parasite concentration in
three different experiments.

G. E. Siless et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4964–4967
4967
Table 1
Trypanocidal evaluation of secochiliolide acid and derivatives
Compd
IC50 g/mL)
S.E.M.
1
2
3
4
5
6
9
10
11
13
Benznidazole
(
l
2
0.24
2
0.3
N.A.
4
0.27
N.A.
5
0.29
7
0.22
2
0.2
N.A.
5
0.32
2.5
0.24
The values represent the mean SEM from three independent experiments.
at 10 lg/mL (Fig. 1). The estimated IC50 for compound 1 was 2 lg/
Acknowledgments
mL, which is comparable with that of benznidazole. In order to
identify the active groups of this compound, the previously de-
scribed semi-synthetic derivatives of 1 were tested, (Fig. 2) and
their IC50’s are shown in Table 1. Six of the semi-synthetic deriva-
tives (2, 4, 6, 9,10 and 13) displayed activity against parasites in the
same concentration range as benznidazole, while compounds 3, 5
and 11 did not affect the growth of parasites. Some of the synthetic
intermediates such as 7 and 8 were not tested. Among the semi-
synthetic derivatives, only compounds 2 and 10 had a similar try-
Research at the University of Buenos Aires was supported by
Grants from CONICET (PIP No. 516-2009), UBACYT (X163
Programación 2008–2010 and 20020100100123, Prog. 2011–
2014), and ANPCYT (PICT 2010-1808). G.S. thanks CONICET for
adoctoral fellowship. This study was also supported, in part, by
the Grant 06/J413 from SECTyP—Universidad Nacional de
Cuyo—Argentina. We thank Mr. T. Sartor for his valuable technical
assistance.
panocidal power (IC50 ꢀ2
l
g/mL) than secochiliolide acid, while
the other compounds were less active (IC50 >5
l
g/mL).
Supplementary data
The results obtained in the present study indicate that secochi-
liolide acid 1 is a promising model compound against T. cruzi, since
its IC50 was similar to that of benznidazole. The fact that this com-
pound can be easily obtained in fairlylarge amounts from an abun-
dant natural source, together with its relatively easy purification
contribute to its importance as a new natural alternative to the
highly toxic drugs actually in use. The results obtained from the
semi-synthetic derivatives indicate that the carboxyl group is not
essential for the bioactivity since its replacement by several other
groups (quinone 10, hydroquinone 9, t-butylcarbamate 2) yielded
products that showed comparable activity or only a slight decrease
in IC50. Quite surprisingly, the t-butyl ester 3, which can be meta-
bolically converted to 1 by hydrolysis, was inactive. A significant
result is the comparison of the almost identical growth curves
Supplementary data (experimental procedures, Scheme 2, spec-
troscopic data and copies of NMR spectra of all new compounds)
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.bmcl.2013.06.064.
References and notes
1. Moloney, A. Lancet 2009, 374, 1490.
2. Morris, K. Lancet Infect. Dis. 2009, 9, 468.
3. The Lancet Editorial Lancet 2009, 373, 1820.
4. The Lancet Neurology Lancet Neurol. 2009, 8, 501.
5. De Souza, W. Curr. Pharm. Des. 2002, 8, 269.
6. Sepúlveda-Boza, S.; Cassels, B. K. Planta Med. 1996, 62, 97.
7. Duschak, V. G. Recent Pat. Anti-Infect. Drug Discovery 2011, 6, 216.
8. Amato Neto, V. Mem. Inst. Oswaldo Cruz 1999, 94, 337.
9. Maya, J. D.; Repetto, Y.; Agosin, M.; Ojeda, J. M.; Tellez, R.; Gaule, C.; Morello, A.
Mol. Biochem. Parasitol. 1997, 86, 101.
10. Salem, M. M.; Werbovetz, K. A. Curr. Med. Chem. 2006, 13, 2571.
11. Sánchez, M.; Mazzuca, M.; Veloso, M. J.; Fernández, L. R.; Siless, G.; Puricelli, L.;
Palermo, J. A. Phytochemistry 2010, 71, 1395.
12. Jakupovic, J.; Banerjee, S.; Bohlmann, F.; King, R. M.; Robinson, H. E. Tetrahedron
1986, 42, 1305.
13. Sülsen, V. P.; Frank, F. M.; Cazorla, S. I.; Anesini, C. A.; Malchiodi, E. L.; Freixa, B.;
Vila, R.; Muschietti, L. V.; Martino, V. S. Antimicrob. Agents Chemother. 2008, 52,
2415.
14. Mabry, T. J.; Miller, H. E.; Kagan, H. B.; Renold, W. Tetrahedron 1966, 22, 1139.
15. Liang, H. Synlett 2008, 2554.
16. Shiori, T. TCIMail 2007, 134.
17. Barton, D. H.; Bridon, D.; Zard, S. Tetrahedron 1987, 43, 5307.
18. Ling, T.; Poupon, E.; Rueden, E.; Kim, S. H.; Theodorakis, E. A. J. Am. Chem. Soc.
2002, 124, 12261.
19. Siless, G. E.; Knott, M. E.; Derita, M. G.; Zacchino, S. A.; Puricelli, L.; Palermo, J. A.
Steroids 2012, 77, 45.
for compounds 1, 9 and 10 obtained at a concentration of 10 lg/
mL, which inall cases indicate a complete inhibition of the parasite
proliferation at that concentration, in spite of their final different
IC50
.
On the other hand, the presence of the tetrasubstituted exocy-
clic double bond seems to be necessary for the bioactivity, since
epoxidation produced an inactive derivative (11). Compound 13,
which also arises from a transformation on this double bond, is
considerably less active than 1. But, taking into account that the
methyl ester group is still present in 11, and the complex transfor-
mation that produced compound 13, the loss of activity may not be
attributed solely to the transformation of the double bond. The fact
that the inactive compounds that were prepared still retain other
important structural fragments such as the spirolactone and the
furan ring, suggest that the observed activity of compound1 may
arise by a combination of the different structural fragments pres-
ent in the molecule. Further modifications of 1 will be performed
in order to validate these observations.
20. Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. Rev. 1993, 93, 1307.
21. Lozano, E.; Barrera, P.; Tonn, C.; Nieto, M.; Sartor, T.; Sosa, M. A. Parasitol. Int.
2012, 61, 275.