Synthesis of C-13 oxidised cuparene and herbertane sesquiterpenes via a Paternò-Büchi photocyclisation-oxetane fragmentation strategy: Total synthesis of 1,13-herbertenediol

Article abstract of DOI:10.1055/s-2004-832813

The intramolecular Paternò-Büchi reaction of 5-aryl-4,4-dimethyl-hex-5-enals has been used to assemble the hindered cyclopentane skeleton of the cuparene and herbertane sesquiterpenes. Regioselective carbon-oxygen bond cleavage in the resulting oxetane is

Full text of DOI:10.1055/s-2004-832813

LETTER
2379
Synthesis of C-13 Oxidised Cuparene and Herbertane Sesquiterpenes via a
Paternò–Büchi Photocyclisation–Oxetane Fragmentation Strategy: Total
Synthesis of 1,13-Herbertenediol
Total
Synthesis
o
i
f
1,13-H
c
e
rbertened
h
iol ard J. Boxall,a Leigh Ferris,b Richard S. Grainger*a
a
Department of Chemistry, King’s College London, Strand, London WC2R 2LS, UK
Fax +44(20)78482810; E-mail: richard.grainger@kcl.ac.uk
b
AstraZeneca, Silk Road Business Park, Macclesfield, Cheshire, SK10 2NA, UK
Received 22 July 2004
duction of nitrile 3, readily prepared in two steps from
4-methylisobutyrophenone 1 as previously described,5
gave the aldehyde 4 in high yield (Scheme 1). Irradiation
of 4 with a 400 W medium pressure mercury arc lamp
(Pyrex filter) initiated an intramolecular Paternò–Büchi
photocyclisation reaction to provide the bicyclo[3.2.0]
oxetane 5 in reasonable yield. The geometry of the cis ring
junction, although ultimately inconsequential from a
target standpoint, was established through NOESY
correlation.
Abstract: The intramolecular Paternò–Büchi reaction of 5-aryl-
4,4-dimethyl-hex-5-enals has been used to assemble the hindered
cyclopentane skeleton of the cuparene and herbertane sesquiter-
penes. Regioselective carbon-oxygen bond cleavage in the resulting
oxetane is applied to the synthesis of 1,13-herbertane diol.
Key words: photochemistry, ring-opening, terpenoids, fused-ring
systems, reduction
The cuparenes and herbertanes (Figure 1) are of a large
family of sesquiterpenes with a diverse array of biological
activity. From a synthetic viewpoint, the challenge of
assembling adjacent quaternary carbon atoms on a cyclo-
pentane ring has given rise to a range of novel strategies,
which continue to be developed.1
In order to reveal the cuparene skeleton embedded in 5,
we turned to the methodology developed by Cohen for the
chemoselective cleavage of carbon-oxygen bonds in oxe-
tanes, employing lithium di-tert-butylbiphenylide
(LDBB) as reductant. In particular, the regioselectivity of
carbon-oxygen bond cleavage in unsymmetrical oxetanes
can be controlled by the presence or absence of a suitable
Lewis acid.6 Indeed, when we subjected oxetane 5 to
LDBB in the presence of Et3Al in THF at –78 °C and al-
lowed the reaction to warm to room temperature, we ob-
served clean formation of the novel 13-hydroxycuparene
6 in 53% yield (unoptimised), along with recovered start-
ing material.7 The regioselectivity of reduction of 6 is con-
sistent with the Cohen model, whereby cleavage of the
bond between oxygen and the more-substituted carbon is
observed in the presence of a Lewis acid. This can be ra-
tionalised by coordination of the Lewis acid to the oxetane
oxygen, followed by a one electron reduction which re-
sults in the cleavage of the C-O bond to initially form the
more stable carbon-centred radical (secondary vs. primary
in this case).
Recent work from the Asakawa group has identified a
number of herbertanes carrying oxygen functionality not
only in the aromatic portion of the molecule, but more un-
usually on the methyl groups attached to the cyclopentane
ring.2 In this letter we report a new strategy for the synthe-
sis of C-13 oxidised herbertane and cuparene sesquiter-
penes based on a novel intramolecular [2+2] Paternò–
Büchi photocyclisation reaction3 of 5-aryl-4,4-dimethyl-
hex-5-enals to assemble the hindered cyclopentane skele-
ton, followed by selective reductive carbon-oxygen bond
cleavage in the resulting oxetane to liberate the C-13
hydroxyl group.4
OH
13
2
1
OH
6
3
7
8
4
5
11
9
Encouraged by this result, we undertook the synthesis of
1,13-herbertenediol as described in Scheme 2. Friedel–
Crafts acylation of 4-methylanisole 7 with isobutyryl
chloride gave the known ketone 8 in good yield.8 Alkyla-
tion of the enolate of 8 with allyl bromide followed by se-
lective hydroboration of the resulting alkene 9 without
reduction of the ketone9 gave the alcohol 10. Attempted
Wittig reaction on 10 failed to produce the desired alkene,
so the alcohol was protected first as a tert-butyldimethyl-
silyl ether 11, which underwent clean olefination to pro-
vide the alkene 12. Silyl deprotection of 12 and Swern
oxidation of the resulting alcohol 13 gave the required
photocyclisation precursor 14 in high yield.10
14
12
10
15
cuparene
herbertene
1,13-herbertenediol
Figure 1 Cuparene and herbertane sesquiterpenes.
As a result of our recent studies on the asymmetric synthe-
sis of cuparene via an a-(aminobutyl)styrene photocycli-
sation reaction,5 we had to hand material which allowed
us to assess the viability of such a strategy. Hence re-
SYNLETT 2004, No. 13, pp 2379–2381
Advanced online publication: 08.09.2004
0
3
.1
1
.2
0
0
4
DOI: 10.1055/s-2004-832813; Art ID: D20604ST
© Georg Thieme Verlag Stuttgart · New York

2380
R. J. Boxall et al.
LETTER
O
O
Photocyclisation of 14 proceeded slowly to give the req-
uisite cis-fused oxetane 15 in 55% yield. This reaction
was accompanied by some secondary photochemical re-
actions, as indicated by the build up of polymeric material
on the sides of the reaction vessel, so was typically
stopped before complete consumption of starting materi-
al.
N
i
ii
1
2
O
N
iii
v
Attempted ring-opening of oxetane 15 under Cohen’s
standard conditions (1.6 equiv LDBB, 2 equiv Et3Al,
THF, –78 °C to r.t.) gave the desired primary alcohol 16
regioselectivity, but in low conversion (38% 16, 6% 17,
35% recovered starting material). Increasing the equiva-
lents of LDBB, triethylaluminium or increasing the reac-
tion time and temperature failed to provide higher yields
of 16, but rather favoured formation of the undesired iso-
mer 17 along with decomposition. Finally, resorting to the
more Lewis acidic diethylaluminium chloride gave rise to
a rapid reaction at –78 °C and complete consumption of
oxetane 15, to provide an approximately 5:2 mixture of
the primary and secondary alcohols 16 and 17, respective-
ly.11 Use of dimethylaluminium chloride in the reductive
ring cleavage of simple oxetanes has previously been re-
ported by Cohen,6 but its application was not pursued be-
cause of the lack of reactivity towards aldehydes of the 5-
membered ring ate complex formed upon rapid cyclisa-
tion of the initial lithium (3-lithiopropoxy)dialkylchloro-
aluminate. However, since protonation of the carbon-
aluminium bond is all that is required in the present in-
stance, it serves its purpose admirably.
3
4
OH
O
iv
H
5
6
Scheme 1 Reagents and conditions: i) Acrylonitrile, triton B (40%
w/w), 1,4-dioxane, 35 °C, 68%; ii) Ph3PMeBr, t-BuOK, toluene, ref-
lux, 77%; iii) DIBAL, CH2Cl2, –78 °C, 85%; iv) 400 W medium pres-
sure mercury arc lamp, immersion well reactor, Pyrex filter, hexane,
0.005 M, r.t., 32 h, 48%; v) THF, Et3Al, LDBB, –78 °C to r.t., 53%
(+12% recovered starting material).
O
O
O
O
O
i
ii
iii
7
8
9
Attempted demethylation of 16 using BBr3 resulted in ex-
tensive decomposition, as previously observed by
Srikrishna et. al. in their synthesis of 1,13-herbertene-
diol.12b However, we have found that demethylation of 16
can be achieved in good yield using an excess of lithium
diphenylphosphide,13,14 providing the natural product
whose data was identical in all respects to that reported in
the literature.2 The overall synthesis of 1,13-herbertene-
diol in 10 steps and 8.6% overall yield from commercially
available material compares favourably with other syn-
theses reported to date.12
OR
OR
O
O
O
O
vi
O
v
vii
10 R = H
12 R = TBDMS
13 R = H
14
iv
11 R = TBDMS
OH
O
O
O
O
OH
H
viii
ix
+
Acknowledgment
We thank the EPSRC and King’s College London for funding, and
AstraZeneca for a CASE award to RJB. RSG thanks AstraZeneca
and Pfizer for additional unrestricted financial support.
15
16
17
x
(+/–)-1,13 herbertenediol
References
Scheme 2 Reagents and conditions: i) AlCl3, ClCOCH(CH3)2,
ClCH2CH2Cl, CH2Cl2, 35 °C, 84%; ii) NaNH2, allyl bromide, tolu-
ene, 50 °C, 74%; iii) Cy2BH, THF, 0 °C, 83%; iv) TBDMSCl, imida-
zole, DMF, r.t., 90%; v) Ph3PCH3Br, t-BuOK, toluene, reflux, 84%;
vi) p-TsOH, THF–H2O (2:1), r.t., 93%; vii) (COCl)2, DMSO, CH2Cl2,
–55 °C to –40 °C then i-Pr2NEt, 98%; viii) 125 W medium pressure
mercury arc lamp, immersion well reactor, Pyrex filter, hexane, r.t.,
0.02 M, 32 h, 55%; ix) LDBB, Et2AlCl, –78 °C, 57% 16 +22% 17;
x) HPPh2, n-BuLi, THF, reflux, 77% (+10% recovered starting
material).
(1) See the following recent examples, and references therein.
(a) Ng, D.; Yang, Z.; Garcia-Garibay, M. A. Org. Lett. 2004,
6, 645. (b) Srikrishna, A.; Rao, M. S. Synlett 2004, 374.
(c) Kuwahara, S.; Saito, M. Tetrahedron Lett. 2004, 45,
5047. (d) Srikrishna, A.; Babu, N. C.; Rao, M. S.
Tetrahedron 2004, 60, 2125. (e) Kita, Y.; Futamura, J.;
Ohba, Y.; Sawama, Y.; Ganesh, J. K.; Fujioka, H. J. Org.
Chem. 2003, 68, 5917. (f) Zhao, X. Z.; Jia, Y. X.; Tu, Y. Q.
J. Chem. Res., Synop. 2003, 54. (g) Paul, T.; Pal, A.; Gupta,
P. D.; Mukherjee, D. Tetrahedron Lett. 2003, 44, 737.
(h) Acherar, S.; Audran, G.; Vanthuyne, N.; Monti, H.
Synlett 2004, No. 13, 2379–2381 © Thieme Stuttgart · New York

LETTER
Tetrahedron: Asymmetry 2003, 14, 2413. (i) Satoh, T.;
Total Synthesis of 1,13-Herbertenediol
2381
diethylaluminium chloride in hexane (1 M, 5.97 mL, 5.97
mmol). Once addition was complete the reaction mixture
was stirred at –78 °C for 1.5 h then quenched with 5% HCl
(5 mL), and the resulting biphasic solution was allowed to
warm to r.t. The aqueous layer was extracted with Et2O (3 ×
15 mL) and the combined organic phases were dried over
MgSO4, and concentrated in vacuo to furnish pale yellow oil.
The crude product was purified by flash column
chromatography (hexane–EtOAc 95:5) to afford first 2-(2¢-
methoxy-5¢-methylphenyl)-2,3,3-trimethylcyclopentanol
17 as a transparent oil (67 mg, 22%); Rf = 0.30 (hexane–
EtOAc 8:2). IR (CDCl3): nmax = 3412, 2945, 1498, 1465 and
1246 cm–1. 1H NMR (360 MHz, CDCl3): d = 7.18 (1 H, d,
J = 1.9 Hz), 7.06–6.98 (1 H, m), 6.80 (1 H, d, J = 8.3 Hz),
5.15 (1 H, s), 3.79 (3 H, s), 2.47–2.32 (1 H, m), 2.29 (3 H, s),
2.09–2.01 (1 H, m), 1.72–1.66 (2 H, m), 1.49–1.44 (1 H, m),
1.29 (3 H, s), 1.16 (3 H, s), 0.64 (3 H, s). 13C NMR (100
MHz, CDCl3): d = 21.3 (q), 21.5 (q), 27.2 (q), 27.6 (t), 28.9
(q), 37.2 (t), 44.1 (s), 54.5 (s), 56.0 (q), 76.4 (d), 112.9 (d),
127.9 (d), 130.2 (s), 130.9 (d), 133.6 (s), 156.4 (s). ESI-MS:
m/z = 271.1664 (C16H24O2Na requires 271.1669). MS (EI):
m/z (%) = 248 (98), 230 (50), 215 (63), 200 (15), 187 (72),
148 (100), 135 (57), 105 (29), 91 (25), 69 (15) and 41 (16).
Followed by [1-(2¢-methoxy-5¢-methylphenyl)-2,2-
Yoshida, M.; Takahashi, Y.; Ota, H. Tetrahedron:
Asymmetry 2003, 14, 281. (j) Nayek, A.; Drew, M. G. B.;
Ghosh, S. Tetrahedron 2003, 59, 5175. (k) Chavan, S. P.;
Kharul, R. K.; Kale, R. R.; Khobragade, D. A. Tetrahedron
2003, 59, 2737. (l) Srikrishna, A.; Rao, M. S. Synlett 2002,
340. (m) Nayek, A.; Ghosh, S. Tetrahedron Lett. 2002, 43,
1313. (n) Srikrishna, A.; Rao, M. S. Tetrahedron Lett. 2002,
43, 151.
(2) Irita, H.; Hashimoto, T.; Fukuyama, Y.; Asakawa, Y.
Phytochemistry 2000, 55, 247.
(3) de la Torre, M. C.; García, I.; Sierra, M. A. J. Org. Chem.
2003, 68, 6611; and references therein.
(4) The combination of a Paternò–Büchi reaction and selective
oxetane cleavage has found notable application in the
synthesis of triquinane natural products. See: Reddy, T. J.;
Rawal, V. H. Org. Lett. 2000, 2, 2711; and references
therein.
(5) Grainger, R. S.; Patel, A. Chem. Commun. 2003, 1072.
(6) Cohen, T.; Mudryk, B. J. Org. Chem. 1991, 56, 5760.
(7) Recently LDBB has been shown to partially reduce certain
electron-deficient aromatic compounds: Donohoe, T. J.;
House, D. J. Org. Chem. 2002, 67, 5015. We have not
observed formation of any Birch-type products, consistent
with the electron-rich nature of the aromatic ring in the
systems under investigation.
dimethylcyclopentyl]-methanol 16 as a pale yellow oil
(168 mg, 57%); Rf = 0.24 (hexane–EtOAc 8:2). IR (CDCl3):
(8) Wang, M.; Liu, S. Z.; Liu, J.; Hu, B. F. J. Org. Chem. 1995,
60, 7364.
(9) Kabalka, G. W.; Yu, S.; Li, N.-S. Tetrahedron Lett. 1997,
38, 5455.
n
max = 3549, 3413, 2952, 1498, 1465, 1240 and 805 cm–1. 1H
NMR (360 MHz, CDCl3): d = 7.08 (1 H, d, J = 1.9 Hz),
7.03–7.00 (1 H, m), 6.77 (1 H, d, J = 8.2 Hz), 4.60 (1 H, dd,
J = 2.7 and 10.5 Hz), 3.77 (3 H, s), 3.49 (1 H, t, J = 10.2 Hz),
2.57–2.52 (1 H, m), 2.28 (3 H, s), 2.17–2.12 (1 H, m), 1.79–
1.73 (2 H, m), 1.58–1.53 (2 H, m), 1.30 (1 H, d, J = 2.0 Hz),
1.10 (3 H, s), 0.67 (3 H, s). 13C NMR (90 MHz, CDCl3): d =
21.3 (q), 21.3 (t), 25.2 (q), 27.8 (q), 35.9 (t), 42.7 (t), 44.5 (s),
55.2 (q), 58.2 (s), 66.3 (t), 111.5 (d), 128.4 (d), 129.9 (s),
130.7 (s), 132.2 (d), 156.6 (s). ESI-MS: m/z = 271.1665
(C16H24O2Na requires 271.1669). MS (EI): m/z (%) = 248
(54), 218 (20), 217 (95), 187 (44), 161 (42), 135 (100), 105
(19), 95 (17), 69 (29) and 41 (11).
(10) Attempted preparation of the photochemical precursor 14
via an analogous sequence of reactions to that described in
Scheme 1 was thwarted by the formation of 19 as the major
product upon attempted Wittig olefination of 18. Alkene 19
is presumably formed by an initial retro-Michael addition
under the reactions conditions to reform ketone 8, which
subsequently undergoes Wittig olefination. The sluggish
reactivity of hindered aryl ketones to the phosphorus ylide
necessitates the high temperature employed, although the
difference in reactivity of 2 and 18 under identical reaction
conditions is notable. To date, we have been unable to achive
methylation of 18 using other methods (Scheme 3).
(12) (a) Fukuyama, Y.; Yuasa, H.; Tonoi, Y.; Harada, K.; Wada,
M.; Asakawa, Y.; Hashimoto, T. Tetrahedron 2001, 57,
9299. (b) Srikrishna, A.; Satyanarayana, G. Tetrahedron
Lett. 2003, 44, 1027. (c) Srikrishna, A.; Rao, M. S. Eur. J.
Org. Chem. 2004, 499.
O
O
O
(13) Aristoff, P. A.; Johnson, P. D.; Harrison, A. W. J. Am. Chem.
Soc. 1985, 107, 7967.
N
Ph3PCH3Br
(14) A solution of HPPh2 (0.182 mL, 1.05 mmol) in THF (3 mL)
was cooled to 0 °C and a solution of n-BuLi in hexane (2.2
M, 0.47 mL, 1.04 mmol) was added dropwise. The orange
solution was stirred at 0 °C for 5 min and then at r.t. for a
further 30 min. A solution of alcohol 16 (40 mg, 0.16 mmol)
in THF (2 mL) was added at r.t. and the resulting solution
was refluxed for 45 min, cooled to 0 °C and quenched with
5% HCl (5 mL). The aqueous layer was extracted with
EtOAc (3 × 15 mL) and the combined organic phases were
dried over MgSO4 and concentrated in vacuo to furnish a
pale yellow oil. The crude product was purified by column
chromatography (hexane–EtOAc 9:1) to afford 1,13-
herbertenediol as a colourless oil (29 mg, 77%). Analytic
data agree with those reported in the literature.2
t-BuOK
toluene
reflux
18
19
Scheme 3
(11) An oven dried Schlenk tube was equipped with a glass stirrer
bar and charged with DBB (1.59 g, 5.97 mmol) and THF (10
mL). Lithium wire (83 mg, 11.90 mmol) was added in one
portion and the mixture was sonicated for 5 min until a rich
turquoise solution was achieved. The resulting solution was
stirred for 5 h at r.t., cooled to –78 °C, and a solution of
oxetane 15 (295 mg, 1.19 mmol) in THF (3 mL) added. The
reaction mixture was then treated dropwise with a solution of
Synlett 2004, No. 13, 2379–2381 © Thieme Stuttgart · New York