Photomediated asymmetric synthesis of (-)-cuparene.

Article abstract of DOI:10.1039/b300815k

Generation of a benzylic quaternary stereocentre via the photomediated cyclisation of a chiral alpha-(aminobutyl)styrene followed by a microwave-assisted Cope elimination has led to a total synthesis of the sesquiterpene (-)-cuparene.

Full text of DOI:10.1039/b300815k

Photomediated asymmetric synthesis of (2)-cuparene†
Richard S. Grainger* and Aslam Patel
Department of Chemistry, King’s College London, Strand, London, UK WC2R 2LS.
E-mail: richard.grainger@kcl.ac.uk; Fax: 44 (0)20 7848 2810; Tel: 44 (0)20 7848 1167
Received (in Cambridge, UK) 21st January 2003, Accepted 3rd March 2003
First published as an Advance Article on the web 31st March 2003
Generation of a benzylic quaternary stereocentre via the
photomediated cyclisation of a chiral a-(aminobutyl)styrene
followed by a microwave-assisted Cope elimination has led
to a total synthesis of the sesquiterpene (2)-cuparene.
of this photochemical precursor (Scheme 2). In the first,
alkylation of pyrrolidine with the known bromide 65 gave a
slow but clean transformation to aminostyrene 8 (Scheme 2).
The presence of a quaternary stereocentre adjacent to an
aromatic ring is a common structural feature in a number of
naturally occurring compounds. Prominent amongst these are
the cuparene and herbertane class of sesquiterpenoids, charac-
terised by the presence of adjacent quaternary carbon atoms on
a cyclopentane ring, but differing in the position of methyl
substitution on the aromatic ring (Fig. 1). Some of these
compounds, particularly those with oxygen functionalities,
show a wide range of biological activity, including potent
antifungal, neurotrophic and inhibition of lipid peroxidation.1
The challenge of establishing a hindered, quaternary ster-
eocentre on a five-membered ring has made the cuparene and
herbertane sesquiterpenoids popular synthetic targets.2 How-
ever, the additional challenge of controlling the absolute
stereochemistry at a sterically congested quaternary ster-
eocentre has meant that far fewer of these approaches can be or
have been rendered asymmetric.3 Our approach to cuparene is
based on the photomediated ring closure of a-(dimethylamino-
butyl)styrene 1 to cyclopentane 2 reported by Lewis et al.
(Scheme 1).4 This reaction is proposed to proceed via electron
transfer from the ground state amine to the singlet excited state
of the styrene, followed by a 1,6-hydrogen transfer and
subsequent cyclisation of the resulting diradical. It occurred to
us that the Lewis cyclisation could be applied to a synthesis of
the cuparene skeleton if such a reaction can tolerate substitution
in both the aromatic ring and connecting chain (3, Scheme 1).
Deamination of the resulting cycloadduct 4 should then furnish
the natural product. Furthermore, the possibility of employing
chiral amines in place of dimethylamine could offer a means to
control the absolute configuration at the new stereocentre, and
hence lead to an asymmetric synthesis of the natural product.
In order to test this hypothesis, pyrrolidine 8 was chosen as an
initial target. We have developed two routes for the preparation
Scheme 1 Photomediated cyclisation of a-(aminobutyl)styrenes.
Scheme 2 Reagents and conditions: (i) acrylonitrile, triton B (40% w/w),
1,4-dioxane, 35 °C, 50 h, 68%; (ii) Ph3PMeBr, KOtBu, toluene, 6 h, 62%;
(iii) CuCl, pyrrolidine, EtOH, reflux then NaBH4, EtOH, 84%; (iv)
pyrrolidine, Na2CO3, EtOH, reflux, 4 d, 95%; (v) hn (265 nm), hexane, rt,
0.005 M, 20 min, 62% of 9; (vi) m-CPBA, CH2Cl2, 83%; (vii) µw (100 W),
DMSO, 200 °C, 1 min, 72% or THF, 60 °C, 8 h, 40%; (viii) H2, 5% Pd/C,
EtOAc, rt, 82%.
Fig. 1 Cuparene and herbertane sesquiterpenoids.
† Electronic supplementary information (ESI) available: experimental
procedures and data including copies of 1H-NMR and 13C-NMR spectra for
all new compounds, and NOESY spectra for compounds 9, 10, 11, 15 and
16. See http://www.rsc.org/suppdata/cc/b3/b300815k/
1072
CHEM. COMMUN., 2003, 1072–1073
This journal is © The Royal Society of Chemistry 2003

Alternatively, 8 could be prepared from 4-methylisobutyr-
ophenone 5 via a three step sequence involving conjugate
addition to acrylonitrile, Wittig reaction, and a one-pot
reductive amination of nitrile 7 via an amidine intermediate.6
Irradiation of 8 with a 400 W medium pressure mercury arc
lamp in an immersion well reactor under high dilution
conditions (0.005 M) cleanly gave a mixture of two cycload-
ducts in a 6+1 ratio as determined by integration of the 1H-NMR
spectrum of the crude reaction mixture. The combined yield of
the two cycloadducts was 100%, and we were able to separate
the major diastereomer 9 in 62% isolated yield by careful
column chromatography. NOESY measurements on both 9 and
10 proved that the major isomer 9 had the pyrrolidine and
aromatic rings trans to one another.‡ Hence we observe the
same sense and slighly higher levels of diastereoselectivity as
compared with the Lewis system (Scheme 1).
Conversion of cycloadduct 9 to cuparene requires a method
for removal of the amine auxiliary. We have developed a three
step sequence to achieve this. Oxidation to the N-oxide 11 was
readily achieved using m-CPBA in dichloromethane.7 The
thermally induced Cope elimination of 11 to produce alkene 12
proved more troublesome, with yields at best 40% (THF, 60 °C,
8 h), coupled with formation of significant amounts of other
unidentifiable byproducts. Control experiments with independ-
ently synthesised N-hydroxypyrrolidine (the byproduct in the
Cope elimination) proved the alkene 12 to be stable under the
reaction conditions and to temperatures up to at least 150 °C in
deuterated DMSO. Eventually we found that performing the
reaction in a microwave reactor greatly improved the yield of
alkene 12. In practice, a solution of the N-oxide in DMSO is
allowed to warm up from 25 °C to 200 °C over a period of one
minute in the chamber of a 100 W focussed microwave reactor
and then allowed to cool to room temperature. Standard work-
up and chromatography gave the alkene 12 in 72% yield.
Finally, hydrogenation of the double bond completed a racemic
synthesis of cuparene.§
The absolute configuration of the major diastereomer 15 was
ultimately established by correlation with the natural product,
obtained via the same three step sequence developed in the
racemic series. Hence oxidation to the amine oxide followed by
Cope elimination and reduction of the double bond gave (S)-
(2)-cuparene with an optical rotation identical to that reported
for the natural product.9
In conclusion, we have synthesised racemic and enatiomer-
ically pure cuparene via the photoelectron transfer initiated
cyclisation of highly substituted a-(aminobutyl)styrenes 8 and
14 respectively. The 2+1 ratio of diastereomers 15 and 16
obtained upon irradiation of 14 establishes for the first time that
a chiral amine can control (albeit with modest levels of
selectivity) the absolute configuration at a hindered quaternary
stereocentre formed in this cyclisation.10
We thank the EPSRC for funding this work (studentship to
AP). The authors also thank the Royal Society for an equipment
grant for the purchase of photochemical apparatus, Dr Nicholas
Leadbeater (King’s College London) for use of a microwave
reactor, and AstraZeneca for additional financial support.
Notes and references
‡ In the NOESY spectrum of the major isomer 9, the methine proton a to
nitrogen shows a cross peak with the aromatic ring protons, but no
correlation with the adjacent methyl group on the cyclopentane ring. In the
NOESY spectrum of the minor isomer 10, the methine proton a to nitrogen
shows a cross peak with the adjacent methyl group on the cyclopentane ring,
and no correlation with the aromatic ring protons. See ESI for details.
In order to render the synthesis asymmetric, we have
incorporated a chiral amine into the cyclisation precursor.
Alkylation of (S)-(+)-2-(methoxymethyl)pyrrolidine 138 with
alkyl bromide 6 provided cyclisation precursor 14 in 77% yield
(Scheme 3). Gratifyingly, irradiation of 14 also resulted in clean
§ Although the double bond is simply reduced in this case, it offers a useful
handle for further elaboration in the synthesis of other members of the
cuparene class of sesquiterpenes. For example, we have also synthesised
(±)-cuparenone as a mixture of a- and g-isomers via allylic oxidation and
double bond reduction.
1
cyclisation (100% by crude H-NMR) to a mixture of all four
possible diastereomers in an approximate 10+5+2+1 ratio by
1H-NMR of the crude reaction mixture. The two major
diastereomers 15 and 16 could be separated by column
chromatography, and were shown by NOESY experiments to
have the amine and the aromatic ring trans to one another, as
observed by Lewis in the case of 2 and by ourselves for 9.
1 A. Matsuo, S. Yuki and M. Nakayama, J. Chem. Soc., Perkin Trans. 1,
1986, 701–710; Y. Fukuyama and Y. Asakawa, J. Chem. Soc., Perkin
Trans. 1, 1991, 2737–2741; Y. Fukuyama, Y. Kiriyama and M.
Kodama, Tetrahedron Lett., 1996, 37, 1261–1264.
2 For recent racemic syntheses see: T.-L. Ho and M.-H. Chang, J. Chem.
Soc., Perkin Trans. 1, 1999, 2479–2482; T. Cohen, T. Kreetha-
dumrongdat, X. Liu and V. Kulkarni, J. Am. Chem. Soc., 2001, 123,
3478–3483; F. G. Favaloro, C. A. Goudreau, B. P. Mundy, T. Poon, S.
V. Slobodzian and B. L. Jensen, Synth. Commun., 2001, 31,
1847–1855.
3 Syntheses of enantiomerically enriched cuparene: H. N. Subba Rao, N.
P. Damodaran and S. Dev, Tetrahedron Lett., 1968, 9, 2213–2214; A.
Abad, C. Agulló, M. Arnó, A. C. Cuñat, M. T. García and R. J.
Zaragozá, J. Org. Chem., 1996, 61, 5916–5919; C. Fuganti and S. Serra,
J. Org. Chem., 1999, 64, 8728–8730; B. R. Aavula, Q. Cui and E. A.
Mash, Tetrahedron: Asymmetry, 2000, 11, 4681–4686.
4 F. D. Lewis, G. D. Reddy and D. M. Bassani, J. Am. Chem. Soc., 1993,
115, 6468–6469; F. D. Lewis, G. D. Reddy, D. M. Bassani, S. Schneider
and M. Gahr, J. Am. Chem. Soc., 1994, 116, 597–605.
5 A. Srikrishna and G. Sundarababu, Tetrahedron, 1991, 47, 481–496.
6 G. Rousselet, P. Capdevielle and M. Maumy, Org. Synth., 1998, 76,
133–141.
7 For a review on the chemistry of amine oxides see: A. Albini, Synthesis,
1993, 263–277.
8 D. Enders and M. Klatt, Synthesis, 1996, 1403–1418.
9 A. Matsuo, N. Nakayama and M. Nakayama, Phytochemistry, 1985, 24,
777–781.
10 For a review on the enantioselective construction of quaternary
stereocentres see: J. Christoffers and A. Mann, Angew. Chem., Int. Ed.,
2001, 40, 4591–4597.
Scheme 3 Reagents and conditions: (i) 6, Na2CO3, EtOH, sealed tube, 2 d,
150 °C, 77%; (ii) hn (265 nm), hexane, rt, 0.01 M, 1 h, 55% 15 (36%
isolated yield); (iii) m-CPBA, CH2Cl2; (iv) µw (100 W), 200 °C, 1 min,
DMSO; (v) 5% Pd/C, H2, EtOAc, rt, 24% over 3 steps.
CHEM. COMMUN., 2003, 1072–1073
1073