First total synthesis of (-)-achilleol B: Reassignment of its relative stereochemistry

Article abstract of DOI:10.1021/ol800326n

The first total synthesis of (-)-achilleol B was achieved using a convergent approach with a longest linear sequence of 14 steps. Three key steps were employed, including an enantioselective Robinson annelation for the construction of the bicyclic moiety. The monocyclic synthon was prepared through a Ti(III)-mediated cylization of a chiral monoepoxide obtained via asymmetric dihydroxylation of geranylacetone. The asymmetric preparation of these subunits also permitted us to achieve the enantioselective synthesis of elegansidiol, achilleol A, and farnesiferol B.

Full text of DOI:10.1021/ol800326n

ORGANIC
LETTERS
2008
Vol. 10, No. 9
1723-1726
First Total Synthesis of (-)-Achilleol B:
Reassignment of Its Relative
Stereochemistry
Jesús F. Arteaga, Victoriano Domingo, José F. Quílez del Moral, and
Alejandro F. Barrero*
Departament of Organic Chemistry, Institute of Biotechnology, UniVersity of Granada,
AVda. FuentenueVa, 18071 Granada, Spain
afbarre@ugr.es
Received February 13, 2008
ABSTRACT
The first total synthesis of (-)-achilleol B was achieved using a convergent approach with a longest linear sequence of 14 steps. Three key
steps were employed, including an enantioselective Robinson annelation for the construction of the bicyclic moiety. The monocyclic synthon
was prepared through a Ti(III)-mediated cylization of a chiral monoepoxide obtained via asymmetric dihydroxylation of geranylacetone. The
asymmetric preparation of these subunits also permitted us to achieve the enantioselective synthesis of elegansidiol, achilleol A, and farnesiferol
B.
Achilleol B (1) is a triterpene isolated in 1990 from Achillea
odorata (Asteraceae).1 It possesses a tricyclic skeleton which
could be encompass in a group of triterpenes which could
be described as irregular. Other examples of these irregular
triterpenes are achilleol A (2),2 camelliols C and A (3 and
4),3 preoleanatetraene (5),4 seco-oleananes 6 and 7,5 and
seco-amyrins 8 and 9 (Figure 1).6 All are structurally
characterized by possessing less rings than those usually
found in the oxidosqualene (OS) cyclizations catalyzed by
OS cyclases. Thus, in the case of 2 and 5, their biosynthesis
could be rationalized by considering an incomplete cycliza-
tion of OS7 initiated at the oxirane ring. In this sense, it was
recently proven that the Arabidopsis thaliana OS cyclase
At1g78955 mainly makes camelliol C (3) and minor quanti-
ties of achilleol A and ꢀ-amyrin (10) (0.2%).8 Recently, a
new biosynthetic mechanism toward seco-triterpenes such
as compounds 6-9 has been proposed. This new route
involves a first cyclization and subsequent rearrangements
to the corresponding pentacyclic structure present in the
oleanane or ursane skeletons, followed by a retrocyclization
process starting from a carbenium ion located at C-13 which
(1) (a) Barrero, A. F.; Manzaneda, E. A.; Manzaneda, R. A.; Arseniyadis,
S.; Guittet, E. Tetrahedron 1990, 46, 8161–8168.
(2) Barrero, A. F.; Alvarez-Manzaneda Roldan, E. J.; Alvarez-Man-
zaneda Roldan, R. Tetrahedron Lett. 1989, 30, 3351–3352.
(3) Akihisa, T.; Arai, K.; Kimura, Y.; Koike, K.; Kokke, W. C. M. C.;
Shibata, T.; Nikaido, T. J. Nat. Prod. 1999, 2, 265–268.
(4) Arai, Y.; Hirohara, M.; Ogawa, R.; Masuda, K.; Shiojima, K.; Ageta,
H.; Hsien-Chang, C.; Yuh-Pan, C. Tetrahedron Lett. 1996, 25, 4381–4384.
(5) Roman, L. U.; Guerra-Ramirez, D.; Moran, G.; Martinez, I.;
Hernandez, J. D.; Cerda-Garcia-Rojas, C. M.; Torres-Valencia, J. M.;
Joseph-Nathan, P. Org. Lett. 2004, 2, 173–176.
(7) Xu, R.; Fazio, G. C.; Matsuda, S. P. T. Phytochemistry 2004, 3,
261–291.
(6) Shibuya, M.; Xiang, T.; Katsube, Y.; Otsuka, M.; Zhang, H.; Ebizuka,
Y. J. Am. Chem. Soc. 2007, 5, 1450–1455.
(8) Kolesnikova, M. D.; Wilson, W. K.; Lynch, D. A.; Obermeyer, A. C.;
Matsuda, S. P. T. Org. Lett. 2007, 9, 5223–5226.
10.1021/ol800326n CCC: $40.75
Published on Web 04/02/2008
2008 American Chemical Society

proton located at the interannular carbons. To help clarify
this point, a comparative study of the 13C NMR chemical
shifts of the carbons of the bicyclic system present in 4-6
and 1 was undertaken.
Following this comparison (Table 1), it was suspected that
the actual stereochemistry of the bicyclic system in achilleol
Table 1. 13C Chemical Shifts of the C-D Rings
achilleol
camelliol
preoleana-
seco-C
carbon
B (1)
A (4)
tetraene (5)
oleanane (6)
12
13
14
15
17
18
19
20
27
31.6
133.7
123.9
26.6
31.4
42.3
43.0
31.0
18.6
31.6
133.9
123.9
29.5
31.4
42.3
43.0
31.0
18.7
31.7
133.6
123.9
29.5
31.4
42.3
43.0
31.0
18.6
30.8
134.3
123.5
29.5
31.5
42.9
42.9
31.0
18.8
Figure 1. Irregular triterpenes.
B (1) was cis. With the aim of both confirming this
hypothesis and establishing unambiguously the structure and
absolute stereochemistry of achilleol B, the enantioselective
synthesis of this compound was addressed. Thus, we
anticipated that the tricyclic achillane skeleton might arise
from the coupling of two C15 chiral synthons, namely, A
(nucleophilic where X ) phenylsulfone) and B (electrophilic
with Y ) Br) (Scheme 1).
The syntheses of A and B were projected in accordance
to the absolute configuration found in the normal enantio-
meric series in triterpenes, that is, 3S,5R,17R,18R; the
numbering of the target product was considered. Synthon A
could be disconnected along the olefin bond through a
Horner-Emmons condensation. The enantioselective syn-
thesis of the monocyclic C-13 moiety would involve a radical
carbocyclization of the oxirane intermediate C. This Cp2TiCl-
mediated reaction would lead to the hoped-for methylenecy-
clohexanol with the lateral chain possessing the appropriate
stereochemistry. Further retrosynthesis of oxirane C through
an asymmetric dihydroxylation indicated commercially avail-
able geranylacetone as a suitable starting material. The use
of the premixed catalyst AD-mix-ꢀ would originate, after
selective mesylation and basic treatment, the 3S monoep-
oxide, which after Ti(III)-induced cyclization should lead
to the corresponding 3S-hydroxymonocycle.
would provoke the breaking of the C8-C14 bond.6 In this
sense, a triple rethrocyclization process initiated at the oleanyl
cation would account for the biosynthesis of preoleanatet-
raene (6).
Our interest of this kind of molecule, together with the
discrepancy observed in the interannular ring junction
stereochemistry present in achilleol B and in structures 4–9
prompted us to re-examine the stereochemistry assigned to
achilleol B.1 This stereochemistry was initially described as
trans due to the lack of NOE effect between the methyl and
(9) Barrero, A. F.; Alvarez-Manzaneda, E. J.; Mar Herrador, M.; Alvarez-
Manzaneda, R.; Quilez, J.; Chahboun, R.; Linares, P.; Rivas, A. Tetrahedron
Lett. 1999, 47, 8273–8276.
(10) Caglioti, L.; Naef, H.; Arigoni, D.; Jeger, O. HelV. Chim. Acta 1959,
2557–2570.
(11) Barrero, A. F.; Arseniyadis, S.; Quilez del Moral, J. F.; Herrador,
M. M.; Rosellón, A. Synlett 2005, 789–792.
(12) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;
Hartung, J.; Jeong, K. S.; Kwong, H. L.; Morikawa, K.; Wang, Z. M. J.
Org. Chem. 1992, 10, 2768–2771.
(13) Barrero, A. F.; Quilez del Moral, J. F.; Herrador, M. M.; Sanchez,
E. M.; Arteaga, J. F. J. Mex. Chem. Soc. 2006, 4, 149–156.
(14) Barrero, A. F.; Quilez del Moral, J. F.; Sanchez, E. M.; Arteaga,
J. F. Eur. J. Org. Chem. 2006, 1627–1641.
(15) Tsangarakis, C.; Arkoudis, E.; Raptis, C; Stratakis, M. Org. Lett.
2007, 9, 583–586.
(16) Barrero, A. F.; Cuerva, J. M.; Alvarez-Manzaneda, E. J.; Oltra,
J. E.; Chahboun, Tetrahedron Lett. 2002, 43, 2793–2796.
(17) Yields obtained in the preparation of B synthon were improved
with respect to those obtained in ref 11. In particular, the efficiency in the
generation of the alcohol derivative of 23 by reduction of the corresponding
R,ꢀ-unsaturated ester improves from 79% to 94% yield when DIBALH is
used instead of LAH.
Once we had the required A synthon in hand, the
asymmetric synthesis of related natural products such as
elengasidiol (11),9 its coumarine derivative farnesiferol B
(12),10 and achilleol A 22 could be easily achieved after
straightforward transformations.
(18) Alvarez-Manzaneda, R. Doctoral Thesis, 1989, University of
Granada.
1724
Org. Lett., Vol. 10, No. 9, 2008

Scheme 1
.
Rethrosynthetic Analysis for Achilleol B
Scheme 3
.
Enantioselective Synthesis of (-)-Achilleol A and
(+)-Farnesiferol B
On the other hand, synthon B would derive from an
interesting enantioselective variant of the Robinson annula-
tion using the Nazarov reagent and a chiral enamine.11 In
this sense, the use of 2S-phenylethylamine should permit
epoxide was reacted with 0.2 equiv of Cp2TiCl2 and Mn in
excess (8 equiv) in the presence of 7 equiv of the Ti(III)
regenerator TMSCl-collidine,14 the cyclization took place
efficiently and the monocyclic alcohol was obtained after
acid workup. Only the exocyclic isomer was detected.
Deprotection of the ketyl group led to ketoalcohol 16.
Application of the Horner-Wadsworth-Emmons protocol
to the silyl derivative 17 furnished the requisite two-carbon
homologation. The corresponding unsaturated ester was
reduced with DIBALH to give 18, thus completing the
synthesis of the A synthon. Simple deprotection of the
silylether in 18 led to the natural sesquiterpene (-)-
elegansidiol (11) (Scheme 2).
Scheme 2. Synthesis of A Synthon
The targeted (-)-achilleol B and (+)-farnesiferol B were
efficiently assembled as follows. Alcohol 18 was brominated
with PBr3 to give the allylic bromide 19. Then, alkylation
of the potassium salt of commercial umbelliferone with
bromide 19 afforded the silyl ether derivative of farnesiferol
B which, after TBAF-mediated deprotection, yielded (+)-
farnesiferol B (11). On the other hand, and as anticipated,
the coupling process between 19 and the lithium derivative
from farnesylphenylsulfone (20) proceeded smoothly in 78%
yield. Reductive elimination of the phenylsulfonyl group
using Li-ethylamine and subsequent deprotection of the silyl
group with TBAF afforded (-)-achilleol A (2) in a yield of
77% for the two steps. Although racemic protocols to
farnesiferol B15 and achilleol A16 have been reported, our
synthesis supposes the first enantioselective preparation of
these two natural compounds. These stereocontrolled syn-
theses proved the absolute configuration of both compounds.
In this sense, the stereostructure of natural (+)-farnesiferol
B (12) is the opposite of that described by Caglioti and co-
workers.10
access to the bicyclic system with the appropriate stereo-
chemistry.
Sharpless asymmetric dihydroxylation12 of geranylacetone
led, with acceptable selectivity and efficiency, to diol 13.13
Protection of the carbonyl group led only to 85% of the ketal
14 (based on recovered starting material). Reaction of diol
14 with mesyl chloride and subsequent treatment with base
gave rise to oxirane 15 in 74% global yield. When this
Org. Lett., Vol. 10, No. 9, 2008
1725

The B synthon preparation was previously communicated
by us in our enantioselective synthesis of preoleanatetraene.17
Following a synthetic route parallel to that used with achilleol
A, we proceeded to make allylic bromide 23 react with the
corresponding anion of phenylsulfone 22 to gratifyingly
obtain the desired diasteromeric mixture of sulfones 24 in
good yield. Finally, exposure of 24 to Li-EtNH2 and
subsequent deprotection of the silyl ether led to the formation
of 25. MS and 1H and 13C NMR of our synthetic achilleol B
coincide completely with those of the natural product.
Noteworthy is the existence of a NOE effect between H-18
and the CH3-C17. The sign of the optical rotation [R]D of
synthetic achilleol B (-10.9, c 0.9, CHCl3) matched that of
the natural compound (-8.1, c 1.1, CHCl3).
Scheme 5. Biogenetic Proposal for Achilleol B
All of the above allowed us both to reassign the stereo-
chemistry of the interannular junction of the bicyclic system
in achilleol B and also to establish absolute stereochemistry
of the natural product as that shown is Scheme 4.
in Achillea odorata of achilleol B and the oleanane ꢀ-amyrin
(Scheme 5).18
Scheme 4. Synthesis of Reassigned Achilleol B
In conclusion, the first enantioselective synthesis of
achilleol B has been efficiently completed with a longest
linear sequence of 14 steps. This work has also allowed us
to reassign the structure of this natural triterpene. Further-
more, this asymmetric protocol enabled us to complete the
synthesis of achilleol A, elegansidiol, and farnesiferol B in
10, 8, and 9 steps and 12, 17, and 13% overall yield,
respectively.
Acknowledgment. This research was supported by the
Spanish Ministry of Science and Technology, Projects BQU
2002-03211 and CTQ2006-15575-C02-01.
Supporting Information Available: Experimental pro-
1
cedures and spectroscopic data of new compounds and H
and 13C NMR spectra of 12-18, 22, and those of natural
and synthetic (-)-elegansiol (11), (-)-achilleol (2), and (-)-
achilleol B. This material is available free of charge via the
Internet at http://pubs.acs.org.
With regard to the biosynthetic origin of achilleol B, a
cyclization-rearrangement-retrocondensation6 process is
proposed, a hypothesis that is supported by the coexistence
OL800326N
1726
Org. Lett., Vol. 10, No. 9, 2008