Total synthesis and anti-leishmanial activity of R-(-)-argentilactone

Article abstract of DOI:10.1016/S0040-4039(01)01559-3

Starting from the commercially available D-glucal (2), the naturally occurring α,β-unsaturated δ-lactone argentilactone (1a) has been synthesized. The important step is the configuration inversion at C-6 by the Mitsunobu reaction following the sugar-non-sugar-sugar strategy. The synthetic argentilactone has been tested against Leishmania mexicana for bioactivity.

Full text of DOI:10.1016/S0040-4039(01)01559-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 7401–7403
Total synthesis and anti-leishmanial activity of
R-(−)-argentilactone
Muhammad Saeed,a Thomas Ilg,b Miriam Schick,a Muhammad Abbasa and Wolfgang Voeltera,*
aAbteilung fu¨r Physikalische Biochemie, Physiologisch-Chemisches Institut der Universita¨t Tu¨bingen, Hoppe-Seyler-Straße 4,
72076 Tu¨bingen, Germany
bMax-Planck-Institut fu¨r Biology, Corrensstraße 38, 72076 Tu¨bingen, Germany
Received 15 June 2001; accepted 20 August 2001
Abstract—Starting from the commercially available
D
-glucal (2), the naturally occurring a,b-unsaturated d-lactone argentilactone
(1a) has been synthesized. The important step is the configuration inversion at C-6 by the Mitsunobu reaction following the
sugar–non-sugar–sugar strategy. The synthetic argentilactone has been tested against Leishmania mexicana for bioactivity. © 2001
Elsevier Science Ltd. All rights reserved.
R-(−)-Argentilactone (1a), an a,b-unsaturated d-lac-
tone, was originally isolated in 1977 from the rhizomes
of Aristolochia argentina Gris.1 This compound is the
major constituent of the essential oil and hexane extract
of Annona haematantha2 and also present in the
methanolic extract of the leaves of Chorisia crispiflora.3
Biological studies on 1a have revealed high anti-
leishmanial2 and cytotoxic activity against P-388 mouse
leukaemia cells.3 Moreover, it can also be used as
starting material for the synthesis of interesting
pheromones.4 These features prompted some research
groups to undertake the total synthesis of this
compound.5–8
sugar from
lem was solved by converting commercially available
-glucal (2) into the optically active furan glycol 3a10
L-series is needed. This stereochemical prob-
D
followed by inversion of configuration of the asymmet-
ric center by the Mitsunobu reaction.11 Although the
authors used 2 equiv. of the reagent system (PPh3,
DEAD, benzoic acid) to produce the crystalline diben-
zoate 4a in 92% yield,10 we found that the use of 1
equiv. of the reagents gave the monobenzoate 4b12 with
the free primary hydroxyl group in more than 80%
yield containing the dibenzoate 4a in less than 5% yield.
Verification that the inversion was stereospecific was
obtained from the fact that the optical rotation of 3b
was identical in magnitude but opposite in sign to that
of the furandiol 3a. Compound 3b was prepared by the
hydrolysis of 4b and 4a.
Recently, we have reported the first total synthesis of
non-natural enantiomer of argentilactone (1b) from
glucose9 (Fig. 1). Extending our research for the synthe-
sis of a,b-unsaturated d-lactones based on cheap and
easily available sugar templates, we describe here the
total synthesis of R-(−)-argentilactone (1a). As the
stereochemistry of 1a at C-6 (pyran numbering) is R, a
Thus, the commercially available
D
-glucal (2) was con-
verted into the furandiol 3a under the conditions
(HgSO4, 0.002 M H2SO4) defined by Gonzalez and
co-workers13 in 90% yield. Inversion of the configura-
9
8
7
10
11
O
O
6
12
13
1
5
2
O
O
4
3
1a
1b
Figure 1. R-(−)-Argentilactone (1a) and its non-natural enantiomer (1b).
Keywords: synthesis; argentilactone; Mitsunobu reaction; biological activity; Leishmania mexicana.
* Corresponding author. Fax: +49 7071 293348; e-mail: wolfgang.voelter@uni-tuebingen.de
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01559-3

7402
M. Saeed et al. / Tetrahedron Letters 42 (2001) 7401–7403
tion of the secondary alcohol in 3a was carried out via
the Mitsunobu reaction11 by using 1.1 equiv. amounts
of each, PPh3, DEAD, and benzoic acid in dry THF.
Benzoylation of the secondary hydroxyl group was
indicated by the low-field absorption (l 6.19 ppm) of
of 6 into the
L
-hexenulose 7 (Scheme 2). This was
accomplished smoothly by oxidation of 6 with NBS.16
The hemiacetal 7, having the required stereochemistry
at C-6 in the target argentilactone 1a, was obtained as
a 1:1 mixture of anomers. Compound 7 was protected
at the anomeric position to get the glycoside 8 as a 2:3
anomeric mixture, favouring the a-anomer as the major
product.
1
H-2 in the H NMR spectrum.
The free hydroxyl group was then protected as tert-
butyl diphenyl silyl ether (TBDPSCl, DMF, imida-
zole)14 in quantitative yield to afford 5. Removal of the
benzoyl group was tested with MeOH/H2O/Et3N, how-
ever, the yield was low and most of the starting mate-
rial was recovered back even after stirring at reflux for
more than 36 h. Use of lithium aluminium hydride was
also not successful because it removed the TBDPS
group15 to give 3b. Finally, catalytic amounts of sodium
in methanol were used to afford the monohydroxyfuran
derivative 6, protected at the primary position with
TBDPS group (Scheme 1).
The ketone 8 was reduced with NaBH4 in methanol at
−40°C in the presence of CeCl3·7H2O to give the allylic
alcohol 9, which was mesylated (MsCl, Py, 5°C) in
excellent yield to produce 10. The latter was treated
with lithium triethylborohydride (Superhydride®) to
produce the C-5-deoxygenated product 11. The TBDPS
group was removed with tetra-n-butylammonium
fluoride in quantitative yield and the generated primary
alcohol 12 was oxidized under Swern oxidation (oxalyl
chloride, DMSO, Et3N, −78°C) to produce the alde-
hyde 13 in 90% yield. Wittig reaction of 13 with hexyl-
triphenylphosphonium bromide gave the olefin 14 in
the desired cis configuration in 78% yield as was indi-
Having a successful preparation of the furan alcohol 6
from
D
-glucal (2), the next step was the transformation
HO
O
a
b
HO
OH
OR
O
O
OH
HO
OR1
2
3a
4a R, R1 = Bz
c
4b R = H, R1 = Bz
3b R, R1 = H
5 R = TBDPS, R1 = Bz
d
e
f
6 R = TBDPS, R1 = H
Scheme 1. (a) H2O/H+, HgSO4, rt, 3 h; (b) PPh3, DEAD, benzoic acid, THF, 2 h; (c) BzCl, Py, ether; (d) MeOH:Et3N:H2O
(5:1:4), overnight at rt; (e) TBDPSCl, DMF, imidazole, 0°C to rt, 2 h; (f) NaOMe, MeOH.
TBDPSO
TBDPSO
RO
TBDPSO
O
O
a
O
c
e
6
O
OR
OEE
OEE
7 R = H
8 R = EE
9 R = H
10 R = Ms
b
d
11
HO
OHC
O
O
O
h
f
g
OEE
OEE
OEE
14
12
13
O
i
O
1a
Scheme 2. (a) NBS, NaHCO3, NaOAc·3H2O, THF:H2O (4:1), 0°C, 30 min; (b) ethyl vinyl ether, PPTS, CH2Cl2, rt, 2 h; (c)
CeCl3·7H2O, NaBH4, MeOH, −40°C, 30 min; (d) MsCl, Py, 5°C, (e) LiBHEt3, THF, 40°C, 3 h; (f) tetrabutylammonium fluoride,
THF, under nitrogen, 1 h; (g) oxalyl chloride, DMSO, Et3N, CH2Cl2, −78°C, 30 min; (h) hexyltriphenylphosphonium bromide,
n-BuLi, THF, −50°C; (i) H2O2, MoO3 (catalytic amounts), and then Et3N, Ac2O.

M. Saeed et al. / Tetrahedron Letters 42 (2001) 7401–7403
7403
References
1. Priestap, H. A.; Bonafede, J. D.; Ru´veda, E. A. Phyto-
chemistry 1977, 16, 1579–1582.
2. Waechter, A. I.; Ferreira, M. E.; Fournet, A.; Rojas de
Arias, A.; Nakayama, H.; Torres, S.; Hocquemiller, R.;
Cave´, A. Planta Med. 1997, 63, 433–435.
3. Matsuda, M.; Endo, Y.; Fushiya, S.; Endo, T.; Nozoe, S.
Heterocycles 1994, 38, 1229–1232.
4. Sala, L. F.; Cravero, R. M.; Signorella, S. R.; Gonza´lez-
Sierra, M.; Ru´veda, E. A. Synth. Commun. 1987, 17,
1287–1297.
5. Fehr, C.; Galindo, J.; Ohloff, G. Helv. Chim. Acta 1981,
64, 1247–1256.
6. O’Connor, B.; Just, G. Tetrahedron Lett. 1986, 27, 5201–
5202.
7. Carretero, J. C.; Ghosez, L. Tetrahedron Lett. 1988, 29,
2059–2062.
8. Chidambaram, N.; Satyanarayana, K.; Chandrasekaran,
S. Tetrahedron Lett. 1989, 30, 2429–2432.
9. Saeed, M.; Abbas, M.; Khan, K. M.; Voelter, W. Z.
Naturforsch. 2000, 56b, 325–328.
10. Hauser, F. M.; Ellenberger, S. R.; Ellenberger, W. P.
Tetrahedron Lett. 1988, 29, 4939–4942.
11. For a review on the Mitsunobu reaction, see: Mitsunobu,
O. Synthesis 1981, 1–28.
Figure 2. The effect of synthetic argentilactone on the growth
of L. mexicana.
12. It seems that the reaction involves the intermediate 15 that
undergoes an SN2 reaction with the benzoate ion.
1
cated by the coupling constant J7,8=10.2 Hz in the H
NMR spectrum of the final compound. The anomeric
1
mixture 14 produced a very complex H NMR spec-
O
trum and the assignments were not carried out. Finally,
compound 14 was oxidized by H2O2 in the presence of
catalytic amounts of MoO3 to generate the target argen-
O
O
PPh3
O
+δ
1
tilactone 1a. All physical data (MS, H and 13C NMR,
O
optical rotation, etc.) were found identical in all respects
with those reported for the natural argentilactone.1
15
13. Gonzalez, F.; Lesage, S.; Perlin, A. S. Carbohydr. Res.
1975, 42, 267–274.
14. Hanessian, S.; Lavallee, P. Can. J. Chem. 1975, 53,
2975–2977.
15. Rajashekhar, B.; Kaiser, E. T. J. Org. Chem. 1985, 50,
5480–5484.
16. Harris, J. M.; Keranen, M. D.; O’Doherty, G. A. J. Org.
Chem. 1999, 64, 2982–2983.
17. Leishmania (L) mexicana LV4 (MNYC/BZ/62/M379,
LEM280) promastigotes were grown in semi-defined
medium 79, supplemented with 4% heat inactivated fetal
calf serum at 26–27°C. To preserve their virulence, L.
mexicana promastigotes were isolated from mouse lesion
amastigotes at 8–12 week intervals. For toxicity, cultures
were seeded at 105/ml promastigotes with the addition of
either 5 or 10 mg/ml argentilactone (dissolved in 5 mg/ml
in dimethylsulphoxide) or an equivalent volume of solvent.
After the initial lag phase (ꢀ90 h), the promastigotes
density was determined at 24 h intervals by triplicate
counting of the cell numbers (for reference, see: Brun, R.;
Scho¨nenberger, M. Acta Trop. 1979, 36, 289–292.)
18. Neal, R. A. In The Leishmaniasis in Biology and Medicine;
Peter, W.; Killick-Kendrick, R., Eds.; Academic Press:
New York, 1987; Vol. 2.
Anti-leishmanial activity of 1a
In the present study, the in vitro activity of synthetic
argentilactone on the growth on L. mexicana was also
assessed17 (Fig. 2). At 5 mg/ml, synthetic argentilactone
leads to marked growth retardation in promastigote
cultures, while at 10 mg/ml the parasites are unable to
proliferate and die within 2–3 days, thus showing
unequivocally that argentilactone is, indeed, the pharma-
cologically active compound of the above mentioned
plant extracts. The efficiency of argentilactone against in
vitro-cultured leishmania promastigotes is comparable to
that of the mainly used clinical anti-leishmania drug,
sodium stibogluconate.18 This indicates that argentilac-
tone may be used as a lead compound for the develop-
ment of new anti-leishmania drug.
Acknowledgements
We would like to thank the DAAD for a Ph.D. scholar-
ship to M.S.