Syntheses of a prenylbisabolane diterpene, a natural insecticide from Croton linearis, and of the bisabolane sesquiterpenes (-)-delobanone and (-)-epi-delobanone

Article abstract of DOI:10.1016/S0040-4020(02)00821-9

An enantioselective first total synthesis of a constituent of Croton linearis, the (-)-7-hydroxy-3,10-prenylbisaboladien-2-one 1, is described as well as the syntheses of the 7-hydroxy-3,10-bisaboladien-2-ones (-)-epi-delobanone (14a) and (-)-delobanone (14b). The model compounds, 7-hydroxy-11-nor-methyl-3-bisabolen-2-one (8a), and 11,15-nor-dimethyl-7-hydroxy-3-bisabolen-2-one (8b), were successfully prepared by opening of the protected carvone epoxide derivative 6 with the appropriate organocuprates. An alternative approach was used for compounds 1 and 14. Thus, these were obtained from homogeranyllithium or homoprenyl Grignard reagent, which reacted successfully with a masked nor-carvone, ketone 11, prepared in four steps from (R)-carvone.

Full text of DOI:10.1016/S0040-4020(02)00821-9

Tetrahedron 58 (2002) 7691–7700
Syntheses of a prenylbisabolane diterpene, a natural insecticide
from Croton linearis, and of the bisabolane sesquiterpenes
(2)-delobanone and (2)-epi-delobanone
*
¨
Olof Smitt and Hans-Erik Hogberg
Chemistry Department of National and Environmental Sciences, Mid Sweden University, SE-851 70 Sundsvall, Sweden
Received 8 May 2002; revised 19 June 2002; accepted 11 July 2002
Abstract—An enantioselective first total synthesis of a constituent of Croton linearis, the (2)-7-hydroxy-3,10-prenylbisaboladien-2-one 1,
is described as well as the syntheses of the 7-hydroxy-3,10-bisaboladien-2-ones (2)-epi-delobanone (14a) and (2)-delobanone (14b). The
model compounds, 7-hydroxy-11-nor-methyl-3-bisabolen-2-one (8a), and 11,15-nor-dimethyl-7-hydroxy-3-bisabolen-2-one (8b), were
successfully prepared by opening of the protected carvone epoxide derivative 6 with the appropriate organocuprates. An alternative approach
was used for compounds 1 and 14. Thus, these were obtained from homogeranyllithium or homoprenyl Grignard reagent, which reacted
successfully with a masked nor-carvone, ketone 11, prepared in four steps from (R )-carvone. q 2002 Elsevier Science Ltd. All rights
reserved.
1. Introduction
a disconnection leading to R-carvone and geraniol as
starting materials (Scheme 1). This synthetic strategy relied
on the regioselective preparation of ketoepoxide 2, which
could then hopefully be opened with a ‘higher order’
geranylcyanocuprate.2,3 To our knowledge, no synthesis of
a prenylbisabolane has yet been reported.
In connection with a project aiming at developing methods
of controlling some forest pest insects, we were interested in
the preparation of heavier analogues of the monoterpene
carvone. One analogue that appeared especially interesting
was the naturally occurring prenylbisabolane diterpene 1, a
constituent of the Jamaican plant Croton linearis (Jacq.),
exhibiting insecticidal properties.1 The proposed structure
of 1 relied on NMR-data (Scheme 1, drawn with the
proposed configuration at C-6, C-7, using the bisabolane
numbering system).1
2. Results and discussion
In order to evaluate the potential of such a synthetic
strategy, we first prepared some model compounds (Scheme
2). Regioselective epoxidation at the electron-rich double
bond of R-carvone with m-CPBA yielded the known
epoxycarvone 2,4 as a 2/3 mixture of diastereomers. Not
Originally, we wanted to develop a biomimetic synthetic
approach to this natural product. This approach consisted of
Scheme 1.
Keywords: isoprenoids; natural products; insecticides.
*
Corresponding author. Tel.: þ46-60-148704; fax: þ46-60-148802; e-mail: hans-erik.hogberg@mh.se
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00821-9

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7692
Scheme 2. Reagents and conditions (yields): (a) m-CPBA, CH2Cl2, 08C (90%), Ref 4; (b) NaBH4, sucrose, H2O, rt; (c) TMSCl, Et3N, DMF, rt (84% from 2,
.90% 1R isomer); (d) for 7a: (1) BuLi, CuI (cat.), THF, 272 to 08C, (2) TsOH (cat.), MeOH, rt (76% over two steps); for 7b: (1) OctylMgBr, CuI (cat.), THF,
2308C to rt, (2) TsOH (cat.), MeOH, rt (74% over two steps); (e) MnO2, CH2Cl2, rt (,60%, 1/1 mixture of diastereomers in both cases).
unexpectedly, addition of a lower order organocuprate
(Bu2CuLi) to the epoxycarvone 2, furnished the undesired
1,4-addition product. In order to ensure a chemoselective
nucleophilic attack on the epoxide moiety rather than on the
conjugated double bond, it was therefore necessary to mask
the ketone. Reduction of epoxycarvone 2 to the correspond-
ing allylic alcohol, followed by silyl protection to give the
TMS ether 6, appeared to be an attractive strategy.
However, the selective 1,2-reduction of the a,b-unsaturated
ketone 2 and the subsequent silylation of the allylic alcohol
were not as straightforward as we had anticipated.
reduction was also highly stereoselective (.90% anti
addition). The propensity of the epoxycarveol 5 to undergo
cyclization to the bottrospicatols 3 and 4 made us directly
protect the alcohol 5 as the TMS ether 6.
Our plan was to make the TMS ether epoxide 6 react with
prenyl- or geranylcuprates either as Gilman-cuprates,8
(R2CuLi), or as higher-order cyanocuprates,9
(R2Cu(CN)Li2). To evaluate the potential of such an
approach, we studied the reactions of the TMS ether
epoxide 6 with some alkyl cuprates prepared from
butyllithium and octylmagnesium bromide. TMS ether
epoxide 6 was successfully reacted with Gilman cuprates,
as well as with higher order cyanocuprates (R2Cu(CN)Li2).
The cuprates were prepared either from butyllithium or
octylmagnesium bromide. When the epoxide 6 was treated
with these cuprates, it yielded the alcohols 7a or 7b,
respectively, after removal of the TMS-group (Scheme 2).
Each of these alcohols was obtained as a mixture of
diastereomers, which was not separated. Manganese dioxide
oxidation of 7a and 7b furnished the model compounds 8a
and 8b, respectively, each as a mixture of diastereomers. If
pure diastereomers should be required, separation of the
precursor diols 7 should be possible.
It is well-known that reduction of a,b-unsaturated carbonyl
compounds with NaBH4 in methanol results in significant
amounts of the saturated alcohol due to 1,4-addition,
followed by reduction of the keto group ultimately giving
a mixture of the saturated and the allylic alcohols. Other
reagents were, therefore explored. Attempted selective
1,2-reduction of epoxycarvone 6 with lithium tri-tert-
butoxyaluminum hydride, (t-BuO)3LiAlH, was unsuccess-
ful. One way to suppress the saturation of the conjugated
double bond is to use the Luche reduction (CeCl3, MeOH).5
Indeed, in our hands this resulted in a chemoselective 1,2-
reduction. Unfortunately, however, this was accompanied
by an intramolecular Lewis acid catalyzed cyclization to
yield a mixture of the known, and previously synthesized,
[3.2.1]-oxabicyclooctane monoterpenes(þ)-bottrospicatol
(3), and (þ)-iso-bottrospicatol (4), shown in Fig. 1.6
Having established that the strategy proposed for the
preparation of the model substances was successful, we
turned to the preparation of the target diterpene 1. First, we
tried to explore the potential of the Grignard reagent
prepared from commercially available geranyl bromide.
This Grignard reagent had been used successfully in a
Li2CuCl4 catalyzed coupling reaction with 2-bromo-1,4-
dimethoxybenzene.10 However, all our attempts to make the
TMS ether epoxide 6 react with the cuprate of geranyl-
magnesium bromide were unsuccessful. The reason for this
might be intramolecular cyclization of the Grignard reagent,
which in turn suffered from steric hindrance.11 If this
reaction had been successful, we had expected problems
concerning the retention of the trans geometry of the allylic
double bond.12 Some efforts have been made to get a better
understanding of allylic cuprates in general and to improve
their reactivity in a reliable way.13 There is, however, still a
lack of general guidelines for the successful preparation and
reaction of complex allylic cuprates. Higher order allylic
Sucrose has rather recently been used as a ‘reactant transfer’
agent in aqueous NaBH4 reductions, causing highly
selective 1,2-reductions of a,b-unsaturated aldehydes and
ketones.7 When we applied this method to the epoxycarvone
2, the unstable allylic alcohol, the epoxycarveol 5, was
obtained (Scheme 2) along with only a minute amount
(,2%) of the undesired saturated epoxyalcohol. The
Figure 1. Bottrospicatol monoterpenes.

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7693
Scheme 3.
cyanocuprates can be prepared from geranyl chloride and
Rieke copper.14 We chose, however, not to use that
method,14 because this method can be expected to lead to
a certain extent of g-attack by the cuprate on the terminal
epoxide group. Instead, a survey of the literature led us to
focus on the preparation of allyl cuprates from the
corresponding allyl stannanes.15 This approach has earlier
been used successfully in total syntheses.3 The higher order
cyanocuprate derived from prenyltributylstannane has been
used successfully in ring-opening of epoxides,15a and
geranyltributylstannane, required in this case, has been
prepared from geraniol with retained E-stereochemistry of
the g-double bond.16
We then decided to test the protection of the alcohol group
of epoxycarveol 5 as the tert-butyldiphenylsilyl ether
(TBDPS ether) 10. It soon became clear that our initial
reaction conditions, which worked well for the preparation
of the TMS ether 6 (DMF, Et3N), were not the appropriate
ones for efficient formation of the other two silyl ethers.
When these, initial, conditions (Scheme 2) were used, the
overall yields of silylated ethers of 5 from carvone were
84% (TMS), 65% (TBDMS) and 51% (TBDPS). The reason
for the decrease of the yield was an unwanted formation of
the silyl ethers of the diastereomeric bottrospicatols, 3 and 4
(Fig. 1). An additional complication was the fact that the
latter products were difficult to separate from the desired
silylated ethers of 5. A change of reaction conditions
(CH2Cl2, imidazole, DMAP) gave a more satisfactory yield
(73% from carvone) of the TBDPS ether epoxide 10.
Subsequent oxidative periodic acid cleavage of this epoxide
furnished the norterpene silyloxyketone 11 as the dominat-
ing stereoisomer (syn vs anti¼94/6) in 60% overall yield
starting from carvone.
In our hands, the higher order geranylcyanocuprate prepared
from geranyltributylstannane failed to react with the TMS
ether epoxide 6. Instead, the latter was recovered
unchanged. In order to investigate whether other allylic
cuprates would react with the epoxide 6, we also prepared
prenyltributylstannane and used it as a starting material for
the preparation of a higher order cyanocuprate,15a which
was subsequently added to the epoxide 6. However, no trace
of the desired silylated precursor of 7 (R¼prenyl) was
detected. Instead unchanged epoxide 6 was the only
compound obtained, despite certain reaction characteristics,
such as color changes having been observed. These
unsuccessful attempts to make the epoxide 6 react with
allylic cuprates led us to abandon the 10Cþ10C disconnec-
tion strategy for the synthesis of the prenylbisabolane 1
(Scheme 1).
Homogeraniol was prepared in more than 99% E-selectivity
18,19
by the excellent procedure described by Kocienski et al.
´
Thus, 4,5-dihydrofuran-2-yllithium was alkylated with
homoprenyl iodide, followed by Ni0-catalyzed ring opening
with methylmagnesium bromide.18 Homogeraniol was then
converted to the corresponding bromide, which was used in
an attempt to prepare the Grignard reagent. Probably side
reactions, such as Wurtz coupling of the Grignard reagent
with the bromide or b-elimination of homogeranyl bromide
induced by the Grignard reagent,20 interfered with the
desired reaction of the Grignard reagent with ketone 11.
Thus only a trace of the desired product, diol 12, could be
detected after deprotection. An alternative choice of the
required nucleophile was the lithium reagent. Such reagents
are readily available from iodoalkanes via a lithium–iodine
exchange using tert-butyllithium at 2788C.21 Using this
method, we converted homogeranyl iodide to homo-
geranyllithium which was reacted with the ketone 11 to
give a diastereomeric mixture of the coupling product
desired,22 the silyl ether protected precursor of 12, in 95%
crude yield, containing the complete carbon backbone of the
target molecule 1.
A second retrosynthetic strategy was then considered,
leading to a C9- and a C11-building block (Scheme 3).
The C9-building block would be derived from carvone after
removal of one carbon atom, whereas the C11-building
block would be a homoallylic metal reagent derived from a
homogeranyl halide. The reaction sequence projected
(Scheme 4) involved a cleavage by periodic acid of a silyl
protected form of epoxycarveol 5 to give the norterpene
ketone 9 (Scheme 3). For that reason, it was necessary to
increase the stability of the silyl protecting group so that it
could withstand the acidic conditions used during the
oxidative cleavage of the epoxide.
The tert-butyldimethylsilyl ether (TBDMS ether) of epoxy-
carveol 5 was first investigated. However, the epoxide
cleavage reaction only gave about 50% yield, due to
deprotection of the alcohol under the acidic conditions.17
Desilylation of this product, followed by chromatographic
separation of the diols obtained gave two 90/10 mixtures of
one (2R,6R,7R or S?)-isomer 12a together with its
(2S,6R,7R or S?)-diastereomer, respectively, followed by a

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7694
Scheme 4. Reagents and conditions (yields): (a) NaBH4, sucrose, H2O, rt; (b) TBDPSCl, imidazole, DMAP, CH2Cl2, rt (73% over two steps); (c) HIO4·2H2O,
Et2O, THF (82%, 94/6 syn/anti ratio); (d) homogeranyllithium, Et2O, then 11, Et2O, 278 to 2108C; (e) TBAF, THF, (80% over two steps), separation (12a vs
12b, 3/2); (f) MnO2, CH2Cl2, rt (89% for 1a, 76% for 1b).
similar mixture of another (2R,6R,7S or R?)-isomer 12b
together with its (2S,6R,7S or R?)-diastereomer (bisabolane
numbering system).
structures shown in Fig. 2. This established the relative
configuration at the stereogenic C-7. The absolute configur-
ation of the starting carvone was R. Hence, compound 12a
was the (2R,6R,7R )-isomer, and 12b was the (2R,6R,7S )-
isomer as shown (Scheme 4). The 3/2 ratio of the diols, 12a
vs 12b, formed from the homogeranyllithium addition after
deprotection, was that expected according to Cram’s rule.
For the determination of the relative configuration of C-7,
bearing the tertiary hydroxyl groups in the two purified
stereoisomers of 12 (Scheme 4), derivatives with a bridging
moiety between the oxygens would be useful. The structures
of such rigid bicyclic diastereomeric compounds probably
could be determinated by NMR. Some diols can form cyclic
silyl-bridged derivatives if they are reacted with dialkylsilyl
reagents.23 Using this method, we prepared the diastereo-
meric [4.3.1] bicyclic silyl derivatives 13 from the separated
diols 12a and 12b by silylation with di-tert-butylsilyl
ditriflate. From NOE measurements (see Fig. 2) it was
unambiguously clear that compounds 13a and 13b had the
Manganese dioxide oxidation of the two separated 90/10-
mixtures (of (2R)- and (2S )-isomers) of 12 from above
furnished the pure (6R,7R )- and (6R,7S)-diastereomers of
hydroxyketones 1a and 1b, respectively. The spectral
properties of both diastereomers were virtually identical
with those of the naturally occurring compound from C.
linearis.1 Although only some of the NMR-shifts observed
for the natural product differed slightly from those obtained
from the synthetic (6R,7R)-diastereomer 1a, the optical
rotations were significantly different. Both of the synthetic
compounds 1a and 1b were 98% enantiomerically pure as
judged by gas chromatography using a chiral stationary
phase (Supelcow b-dex 120). The natural product has
[a ]2D5¼211.4.1 The synthetic isomers 1a and 1b had
[a ]2D5¼24.3 and [a ]2D6¼213.7, respectively, which
strongly indicated that 1b was identical with the natural
product. Furthermore, the 1H NMR spectrum registered for
the natural product was compared with those of the
synthetic isomers 1a and 1b. Whereas an incomplete
match was observed for 1a, the spectrum of 1b matched
that of the natural product establishing that the structure
originally suggested,1 namely 1b, actually was correct.
In order to accomplish the synthesis of the
diastereomeric delobanones, we made homoprenyl-
magnesium bromide and the ketone 11 react (Scheme 5)
Figure 2. Significant NOE’s of the bicyclic di-tert-Bu-silyl derivatives, 13a
and 13b prepared from diols 12a and 12b, respectively. Reaction
conditions: (a) (t-Bu)2Si(OTf)2, 2,6-lutidine, CH2Cl2, 08C.

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7695
cell. Mikrokemi AB, Uppsala, Sweden, carried out the
elemental analyses.
4.2. Synthesis of model compounds 8a and 8b
4.2.1. (1R/S,5R )-2-Methyl-5-[(2R/S )-2-methyloxiran-2-
yl]-1-[(trimethylsilyl)oxy]cyclohex-2-ene (6). Epoxy-
carvone 2 was prepared according to the published
procedure.4 Complementary data of 2: bp 90–
928C/0.7 mbar; 13C NMR d 198.9, 198.8, 144.1, 143.9,
135.7, 135.6, 58.0, 57.9, 53.0, 52.4, 41.4, 40.8, 40.4, 40.0,
27.9, 27.8, 19.1, 18.4, 15.7 (2C); MS (CI) m/z (rel. int.) 167
(MþHþ, 10), 149 (40), 109 (100). Other data were in
agreement with those in the literature.26 The epoxycarvone
2 (5.25 g, 31.6 mmol) was stirred in an aqueous solution of
sucrose (0.8 M, 375 mL), whereupon NaBH4 (96%, 2.31 g,
58.7 mmol) was added in portions over a 5 min period. The
mixture was stirred for 1 h at rt and then extracted with
Et2O. The combined organic phases were dried over
Na2SO4, filtered and Et3N (2 drops) was added. Concen-
tration in vacuo resulted in epoxycarveol 5 as a yellow
unstable oil (5.55 g), which was used in the next step
without further purification. Data of the crude mixture of
major (1R ) isomers: 1H NMR (MeOD (d-4)) d 5.44 (m, 1H),
4.09 (m, 1H), 2.67 (d, J¼4.7 Hz, 1H), 2.54–2.58 (2£d,
J¼4.8 Hz, 1H), 1.87–2.15 (m, 3H), 1.71 (s, br, 3H), 1.32–
1.65 (m, 2H), 1.26 (s, 3H of one diast.), (1.26 (s, 3H of other
diast.)); 13C NMR (MeOD (d-4)) d 138.5, 138.4, 123.7,
123.6, 71.1, 71.0, 60.2, 60.1, 54.2, 53.6, 41.5, 40.8, 36.1
(2C), 28.8, 28.6, 19.4, 19.3, 18.7, 18.1. The crude
epoxycarveol 5 (5.52 g) from above was stirred in DMF
(90 mL), and Et3N (6.54 g, 64.7 mmol, distilled from CaH2)
was added, followed by TMSCl (5.15 g, 47.4 mmol) in
DMF (26 mL). After stirring for 1.5 h at rt, water was added
and the resulting mixture was extracted with Et2O and dried
over Na2SO4. After filtration and concentration, the
resulting oil was purified using MPLC (200 g silica gel,
with a gradient of EtOAc/cyclohexane (0/1!2/3) as eluent).
The title compound (6.35 g, 84% over two steps) was
isolated as a slightly yellowish oil, consisting of a mixture of
diastereomers (,95% purity, of which .90% (1R )-syn-
diastereomers (GC, EC-1 column, FID)), which was
subjected to the next step without further purification. IR
(neat, KBr) 2960, 1455, 1350, 1250, 1100, 1065, 905,
840 cm21. Other data of the mixture of the two major (1R)-
syn-isomers: 1H NMR d 5.45 (m, 1H), 4.19 (m, 1H), 2.54–
2.65 (m, 2H), 1.89–2.10 (m, 3H), 1.66 (d, br, J¼1.6 Hz,
3H), 1.47–1.53 (m, 2H), 1.27 (s, 3H of one diast.), (1.26 (s,
3H of other diast.)), 0.15 (s, 9H); 13C NMR d 137.0, 136.9,
123.0, 122.7, 70.9, 70.7, 58.9, 58.7, 53.6, 52.6, 40.3, 39.3,
35.6, 35.4, 27.9, 27.7, 19.5, 19.4, 18.6, 17.5, 0.3 (6C); MS
(EI) m/z (rel. int.) 240 (Mþ, 5), 223 (15), 209 (35), 181 (60),
151 (10), 93 (35), 73 (100); MS (CI) m/z (rel. int.) 241
(MþHþ, 4), 223 (5), 183 (20), 151 (50), 93 (100).
Scheme 5. Reagents and conditions (yields): (a) Et2O, 08C to rt; (b) TBAF,
THF, (21% over two steps), separation (7R vs. 7S, 1/1); (c) MnO2, CH2Cl2,
rt.
to give (2)-7-epi-delobanone (14a) and (2)-delobanone
(14b) (bisabolane numbering system), after deprotection,
chromatographic separation of the diastereomeric products
and subsequent MnO2 oxidation in the same manner as
described (Scheme 4) above. The enantiomer of compound
14b, (þ)-delobanone, has been isolated from the root of
Lindera triloba, and the structure has been unambiguously
proven by NMR studies.24 Both (þ)-7-epi-delobanone (14a)
and (þ)-delobanone (14b) have been synthesized before.25
Our method appeared to offer better control of the
number of products formed, and of their stereochemical
purity. The Grignard reaction used by us (Scheme 5) was
not as clean and efficient, however, as the organolithium
method used for the homogeranyl elongation, described
above (Scheme 4).
3. Conclusion
In conclusion, we have explored the utility of alkyl and
allylic cuprates, regarding regioselectivity, functional group
compatibility and reactivity. We have made a regioselective
transformation of carvone to silyl ether protected carveol
epoxides and further transformed these to analogues of
carvone. With this strategy we prepared the two separated
(2)-diastereomers (1a and 1b) of the prenylbisabolane
diterpene 1, and the two separated (2)-diastereomers of the
sesquiterpene delobanone (14a and 14b).
4. Experimental
4.1. General
Commercially available chemicals were used as received.
The ee of R-carvone was 98.0%. Dry THF and Et2O were
obtained by distillation under argon from potassium
benzophenone ketyl and from LiAlH4, respectively. The
4.2.2. (6R,7R/S)-7-Hydroxy-11-nor-methyl-3-bisabolen-
2-one (8a). Dry THF (5 mL) was added to CuI (0.34 g,
1.8 mmol) under argon. The stirred mixture was cooled to
2728C (bath temperature) and a solution of BuLi in hexanes
(12 mL, 1.6 M, 19 mmol) was added dropwise during a
5 min period. After 5 min additional stirring, the epoxysilyl
ether 6 (3.45 g, 14.4 mmol) was added dropwise. The
mixture was allowed to reach 08C during 2.5 h and NH4Cl
boiling points were not corrected. 1H (250.13 MHz) and 13
C
(62.9 MHz) NMR spectra were obtained with a sample
temperature of 258C, using CDCl3 as solvent (unless
otherwise stated) and TMS as internal reference. Mass
spectroscopy was carried out using a GC–MS (ion trap
detector) in EI or CI (CH3CN as chemical ionization gas)
mode. Optical rotation values were measured with a 1 dm

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7696
(aq, sat.) was added. The resulting mixture was extracted
with Et2O. The combined organic phases were washed with
water, followed by brine. Drying (Na2SO4), filtration and
evaporation of the solvents afforded (2R/S,6R,7R/S )-7-
hydroxy-11-nor-methyl-2-[(trimethylsilyl)oxy]-3-bisabol
ene (3.59 g) as a slightly yellowish oil, which was subjected
to the next step without further purification. Data of the
crude mixture of major (–OTMS¼2R ) isomers: 1H NMR d
5.50 (m, 1H), 4.21 (m, 1H), 1.85–2.02 (m, 2H), 1.69 (s, br,
3H), 1.20–1.58 (m, 10H), 1.16, (s, 3H of one diast.), (1.12
(s, 3H of other diast.)), 0.89–0.99 (m, 4H), 0.18 (s, 9H); 13C
NMR d 136.6, 136.3, 123.8, 123.6, 73.8, 73.7, 71.4 (2C),
42.3, 41.7, 40.3, 39.3, 34.4, 33.8, 32.5 (2C), 27.1, 26.2, 24.3,
23.4, 23.2, 22.9, 22.7 (2C), 19.5 (2C), 14.1 (2C), 0.3 (6C);
MS (EI) m/z (rel. int.) 298 (Mþ, 5), 280 (30), 265 (10), 209
(50), 181 (100), 137 (35). The crude monosilylated diol
(3.59 g) from above was stirred in MeOH (60 mL) with
TsOH (0.17 g, 1.0 mmol) at rt. After 15 h the MeOH was
evaporated and the resulting oil was purified using flash
chromatography (80 g silica gel, EtOAc/cyclohexane (3/7)
as eluent), affording the diol 7a (2.47 g, 76% over two steps)
as a slightly yellowish viscous oil, which was a mixture of
diastereomers. This mixture was subjected to the next
synthetic step without further purification. Oven tempera-
ture 2008C/0.85 mbar (Kugelrohr distillation); IR (neat,
KBr) 3385, 2930, 2850, 1450, 1375 cm21. Other data of the
mixture of the major (2R)-syn-isomers: 1H NMR d 5.49 (m,
1H), 4.17 (m, 1H), 1.88–2.26 (m, 3H), 1.76 (s, br, 3H),
1.20–1.55 (m, 10H), 1.15 (s, 3H of one diast.), (1.12 (s, 3H
of other diast.)), 0.89 (t, J¼7 Hz, 3H); 13C NMR d 136.6,
136.4, 123.9, 123.7, 74.0, 73.9, 71.0 (2C), 42.0, 41.8, 40.2,
39.7, 34.3, 33.8, 32.5, 32.4, 27.0, 26.3, 24.3, 23.6, 23.3,
23.0, 22.7 (2C), 18.9, 18.8, 14.1 (2C); MS (EI) m/z (rel. int.)
227 (MþHþ, 2), 193 (10), 137 (55), 115 (35), 109 (45), 93
(90), 79 (100). The diol 7a (0.85 g, 3.8 mmol) was stirred in
a mixture of CH2Cl2 (10 mL) and MnO2 (90%, 5.0 g,
52 mmol, dried at 1408C, stored in a desiccator before use)
at rt for 2.5 days. The mixture was filtered through a pad of
celite, which was rinsed with CH2Cl2. The solvent was
evaporated, and the resulting oil was purified using MPLC
(15 g silica gel, with a gradient of EtOAc/cyclohexane
(0/1!2/3) as eluent). This resulted in the title compound
(0.51 g, 61%) as a slightly yellowish viscous oil, which was
a ,1/1 mixture of diastereomers: oven temperature
1758C/0.6 mbar (Kugelrohr distillation); IR (neat, KBr)
3465, 2935, 2870, 1660, 1450, 1370 cm21; 1H NMR d 6.77
(m, 1H), 2.57 (m, 1H), 2.05–2.43 (m, 4H), 1.78 (m, 3H),
1.40–1.54 (m, 2H), 1.20–1.40 (m, 6H), 1.17 (s, 3H of one
diast.), (1.16 (s, 3H of other diast.)), 0.89 (t, J¼7 Hz, 3H);
13C NMR d 200.5, 200.3, 145.4, 145.1, 135.2 (2C), 73.3
(2Cs), 44.0, 43.9, 40.2, 40.0, 39.5, 39.0, 32.4 (2C), 27.1,
26.7, 24.2, 24.1, 23.3, 23.2, 22.6 (2C), 15.6 (2C), 14.0 (2C);
MS (EI, one of the isomers, the other similar) m/z (rel. int.)
207 (Mþ–OH, 5), 153 (10), 135 (10), 115 (40), 110 (100),
95 (70). Anal. Calcd for C14H24O2: C, 75.0; H, 10.8. Found:
C, 74.6; H, 10.8.
reagent was added dropwise to CuI (1.1 g, 5.8 mmol) in dry
THF (77 mL) at 2308C under Ar, followed by addition of
the epoxysilyl ether 6 (10 g, 42 mmol). The resulting
mixture was stirred over a period of 15 h while allowed to
reach rt. NH4Cl (aq, sat.) was added and the resulting
mixture was extracted with Et2O. Drying (Na2SO4),
filtration and evaporation of the ether afforded
(2R/S,6R,7R/S )-11,15-nor-dimethyl-7-hydroxy-2-[(trimeth-
ylsilyl)oxy]-3-bisabolen (15.3 g) as a slightly yellowish oil.
This oil was subjected to the next step without further
purification. Thus the crude monosilylated diol was
deprotected using the procedure described for 7a (see
Section 4.2.2). This afforded the diol 7b as a mixture of
diastereomers, which was subjected to the next step without
further purification. MnO2 oxidation of the diol 7b using the
same procedure as that described for the oxidation of 7a (see
Section 4.2.2.), resulted in the title compound as a slightly
yellowish viscous oil and as a ,1/1 mixture of dia-
stereomers: oven temperature 2208C/0.4 mbar (Kugelrohr
distillation); IR (neat, KBr) 3470, 2925, 2855, 1665, 1455,
1
1370 cm21; H NMR d 6.77 (m, 1H), 2.56 (m, 1H), 2.05–
2.43 (m, 4H), 1.76 (d, br, J¼1.0 Hz, 3H), 1.40–1.54 (m,
2H), 1.18–1.38 (m, 16H), 1.16 (s, 3H of one diast.), (1.15 (s,
3H of other diast.)), 0.88 (t, J¼7 Hz, 3H); 13C NMR d 200.6,
200.5, 145.6, 145.3, 135.1 (2C), 73.1 (2C), 44.1, 44.0, 40.2,
40.0, 39.5, 39.0, 31.9 (2C), 30.3 (2C), 29.6 (4C), 29.3 (2C),
27.2, 26.7, 24.1 (2C), 23.6 (2C), 22.7 (2C), 15.6 (2C), 14.1
(2C); MS (EI, one of the isomers, the other similar) m/z (rel.
int.) 281 (MþHþ, 3) 264 (5), 171 (20), 153 (5), 135 (5), 110
(100), 95 (45). Anal. Calcd for C18H32O2: C, 77.1; H, 11.5.
Found: C, 77.0; H, 11.6.
4.3. Synthesis of prenylbisabolanes: 1a, 1b and
bisabolanes: 14a, 14b
4.3.1. (1R)-1-[(3R/S)-3-{(tert-Butyldiphenylsilyl)oxy}-4-
methylcyclohex-4-enyl]ethanone (11).
A mixture of
epoxycarveol isomers 5 (1.58 g), prepared as described for
6 (see Section 4.2.1.), was dissolved in CH2Cl2 (40 mL),
and imidazole (1.34 g, 19.7 mmol) was added, followed by
DMAP (0.30 g, 2.45 mmol) and TBDPSCl (2.88 g,
10.5 mmol). The mixture was stirred at rt for 19 h, water
was added and the resulting mixture was extracted with
CH2Cl2. The organic phase was washed with brine and dried
over Na2SO4 and the solvent was evaporated. The resulting
oil was purified using flash chromatography (205 g silica
gel, with EtOAc/cyclohexane (400 mL, 1/39 followed by
800 mL, 1/19) as eluent) affording the epoxysilyl ether 10
(2.76 g, 73%), as a 93/7 mixture of the syn–anti
diastereomers (GC, EC-1 column, FID) as a colorless
viscous oil. [a]2D4¼240.8 (c 0.98, CHCl3); IR (neat, KBr)
3070, 3045, 2930, 2855, 1425, 1110, 1060, 740, 700 cm21
.
1
Other data for the mixture of major (1R )-syn-isomers: H
NMR d 7.65–7.75 (m, 4H), 7.33–7.46 (m, 6H), 5.39 (m,
1H), 4.25 (m, 1H), 2.38–2.47 (m, 2H), 1.80–1.93 (m, 2H),
1.65–1.75 (m, 3H), 1.46–1.26 (m, 3H), 1.11 (s, 3H of one
diast.), (1.10 (s, 3H of other diast.)), 1.07 (s, 9H); MS (EI)
m/z (rel. int.) 407 (MþHþ, 2), 389 (5), 349 (100), 271 (30),
255 (50), 199 (85), 151 (5), 133 (35), 93 (40); MS (CI) m/z
(rel. int.) 407 (MþHþ, 5), 151 (100), 133 (50), 93 (95).
Periodic acid, HIO4·2H2O (3.27 g, 14.3 mmol) was dis-
solved in THF (110 mL) and stirred at 08C (bath
temperature). The epoxysilyl ether 10 (4.88 g, 12.0 mmol)
4.2.3. (6R,7R/S )-11,15-nor-Dimethyl-7-hydroxy-3-bisa-
bolen-2-one (8b). Dry Et2O (31 mL) was added to
activated magnesium turnings (2.7 g, 0.11 mol) under
argon. 1-Bromooctane (15 g, 78 mmol, dried over K2CO3,
distilled before use) was added dropwise under gentle
reflux. After refluxing for 15 min the resulting Grignard

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7697
1
from above, dissolved in Et2O (25 mL) was added, and the
mixture was stirred for 1 h at 08C to give a milky white
solution, to which NaHCO3 (50 mL, aq, sat.) was added.
After stirring for an additional 15 min, the mixture was
filtered through a pad of celite, which was rinsed with
Et2O. The filtrate was extracted with Et2O and the combined
organic phases were washed with water, Na2S2O3 (5% aq),
brine and dried over MgSO4. Concentration in vacuo
afforded an oil, which was purified using flash chromato-
graphy (150 g silica gel, with Et2O/pentane (300 mL, 1/19
followed by 600 mL, 1/9) as eluent). This afforded the title
compound (3.86 g, 82%) as a 94/6 mixture of the (3R )-syn-
and (3S)-anti-diastereomers (GC, EC-1 column, FID) and
as a colorless viscous oil. [a ]2D3¼285.4 (c 1.06, CHCl3); IR
(neat, KBr) 3070, 3045, 2930, 2855, 1715, 1425, 1110,
1060, 740, 700 cm21. Other data of the major (3R )-syn-
isomer: 1H NMR d 7.67–7.75 (m, 4H), 7.33–7.47 (m, 6H),
5.40 (m, 1H), 4.29 (m, 1H), 2.38 (m, 1H), 2.00–2.21 (m,
2H), 1.93 (s, 3H), 1.88 (m, 1H), 1.72 (d, br, J¼1.3 Hz, 3H),
1.56 (ddd, J¼12.4, 12.4, 9.8 Hz, 1H), 1.07 (s, 9H); 13C
NMR d 210.0, 137.5, 136.1 (2C), 136.0 (2C), 134.5, 133.7,
129.7, 129.6, 127.6 (2C), 127.5 (2C), 121.9, 71.8, 47.0,
34.8, 27.6, 27.1 (3C), 27.0, 20.2, 19.5; MS (EI) m/z (rel. int.)
392 (Mþ, 1), 335 (100), 199 (35), 139 (10), 119 (25), 77
(15); MS (CI) m/z (rel. int.) 393 (MþHþ, 1), 137 (100).
Anal. Calcd for C25H32O2Si: C, 76.5; H, 8.2. Found: C,
76.9; H, 8.3.
colorless oil: bp 65–678C/0.4 mbar; H NMR d 5.06–5.14
(m, 2H), 3.11 (t, J¼7.4 Hz, 2H), 2.58 (dt, J¼7.4, 7.2 Hz,
2H), 1.97–2.12 (m, 4H), 1.68 (d, br, J¼1.1 Hz, 3H), 1.61
(2£s, 2£3H); 13C NMR d 138.1, 131.6, 124.1, 123.0, 39.6,
32.4, 26.5, 25.7, 17.7, 16.3, 6.1; MS (EI) m/z (rel. int.) 278
(Mþ, 2), 235 (20), 151 (10), 123 (35), 109 (10), 95 (50), 81
(30), 69 (100). Other data were in agreement with those
described earlier.18a
4.3.5. (2R/S,6R,7R )-7-Hydroxy-3,10-prenylbisabola-
dien-2-ol (12a) and (2R/S,6R,7S)-7-hydroxy-3,10-prenyl-
bisaboladien-2-ol (12b). Homogeranyl iodide (2.34 g,
˚
8.34 mmol, stored over K2CO3, 3 A molecular sieves) in
dry Et2O (50 mL, degassed with argon) under argon was
stirred at 2788C (bath temperature). t-BuLi (10.3 mL,
1.7 M in pentane, 17.5 mmol) was added dropwise. After
the temperature had slowly been raised to 2358C during
1.5 h, the mixture was cooled to 2788C and the ketone 11
(2.53 g, 6.44 mmol) in dry Et2O (25 mL) was added
dropwise. The temperature was slowly raised to 2358C
during a 2.5 h period and NH4Cl (aq, sat.) and Et2O were
added. The organic phase was separated and the aqueous
phase was extracted with Et2O. The combined organic
phases were washed with brine, dried over Na2SO4 and the
solvents were evaporated. The resulting oil was filtered
through a plug of silica gel using EtOAc/cyclohexane (1/9
followed by 1/4) as eluent. This afforded (2R/S,6R,7R/S )-2-
[(tert-butyldiphenylsilyl)-oxy]-7-hydroxy-3,10-prenylbisa-
boladien (3.32 g) as a colorless viscous oil and as a 3/2
mixture of diastereomers (regarding epimers at the exo-
cyclic hydroxyl group). This mixture was used in the
following step without further purification. Data of the
crude mixture of major (–OTBDPS¼2R) isomers: 1H NMR
d 7.66–7.75 (m, 4H), 7.32–7.45 (m, 6H), 5.42 (m, 1H),
4.98–5.13 (m, 2H), 4.28 (m, 1H), 1.71–2.12 (m, 15H), 1.68
(s, br, 3H), 1.60 (s, br, 3H) 1.54 (s, br, 3H), 1.18–1.32 (m,
2H), 1.08 (s, 9H), 1.08 (s, 3H of one diast.), (1.07 (s, 3H of
other diast.)); MS (EI) m/z (rel. int.) 545 (MþHþ, 1), 450
(5), 433 (5), 375 (15), 347 (20), 199 (55), 177 (100), 135
(25), 121 (55), 107 (25), 93 (50), 85 (30); MS (CI) m/z (rel.
int.) 177 (100), 85 (15). Tetrabutylammonium fluoride,
TBAF (15 mL, 1.0 M in THF, 15 mmol), was added to the
diastereomeric mixture of the monosilylated diols from
above (2.98 g, 5.47 mmol) in THF (50 mL) at 08C (bath
temperature). After stirring at rt overnight, the main part of
the solvent was evaporated off and water was added. The
mixture was extracted with Et2O and the combined organic
phases were washed with NaHCO3 (aq, sat.), brine, dried
over Na2SO4 and concentrated. The resulting oil was
purified using MPLC (70 g silica gel, with a gradient of
EtOAc/cyclohexane (0/1!3/2) as eluent). At first, this
furnished the diol 12a (0.75 g), then a mixture of
diastereomers (0.21 g, ,1/1) and at last the diol 12b
(0.45 g) all as colorless viscous oils (80% yield over two
steps).
4.3.2. (3E )-4,8-Dimethylnona-3,7-dien-1-ol (homo-ger-
aniol). This was prepared according to the published
procedure.18 The crude intermediate 5-(4-methylpent-3-
enyl)-2,3-dihydrofuran was used directly in the following
step without further purification. Thus homogeraniol was
obtained in 70% total yield with more than 99% isomeric
purity (GC, EC-1 column, FID). Complementary data: 13C
NMR d 139.0, 131.7, 124.1, 119.9, 62.4, 39.8, 31.5, 26.6,
25.7, 17.7, 16.2; MS (EI) m/z (rel. int.) 125 (60), 109 (10),
95 (15), 81 (35), 69 (100); MS (CI) m/z (rel. int.) 169
(MþHþ, 25), 151 (55), 125 (50), 109 (25), 95 (100), 81 (30),
69 (20).
4.3.3. (6E)-9-Bromo-2,6-dimethylnona-2,6-diene (homo-
geranyl bromide). This compound was prepared according
to the published procedure in 90% yield.27 Complementary
data: bp 93–948C/4 mbar; IR (neat, KBr) 2965, 2925, 1440,
1
1375, 1265, 1205, 1110 cm21; H NMR d 5.05–5.17 (m,
2H), 3.34 (t, J¼7.3 Hz, 2H), 2.57 (dt, J¼7.2, 7.0 Hz, 2H),
1.97–2.12 (m, 4H), 1.68 (d, br, J¼1.0 Hz, 3H), 1.63 (s, 3H),
1.60 (s, 3H); 13C NMR d 138.6, 131.6, 124.0, 120.9, 39.6,
32.8, 31.7, 26.5, 25.7, 17.7, 16.3; MS (EI) m/z (rel. int.) 231
(MþHþ, 5), 229 (4), 217 (10), 215 (10), 189 (50), 187 (50),
123 (30), 109 (10), 95 (40), 81 (30), 69 (100).
4.3.4. (6E )-9-Iodo-2,6-dimethylnona-2,6-diene (homo-
geranyl iodide). Homogeranyl bromide (4.59 g,
19.9 mmol) was added dropwise to a mixture of NaI
(7.64 g, 51.0 mmol) in acetone (40 mL, dried over K2CO3
before use). The mixture was stirred at rt for 4 h whereupon
the milky white suspension was concentrated in vacuo.
Water was added, and the mixture was extracted with
Et2O. The combined organic phases were dried over MgSO4
and the solvents were evaporated off to leave an oil which
on distillation gave the title iodide (5.16 g, 93%) as a
Data of 12a: Rf 0.40 (EtOAc/cyclohexane (3/2) as eluent);
[a ]2D4¼þ0.48 (c 0.84, MeOH); IR (neat, KBr) 3375, 2965,
2920, 1450, 1375, 1105, 1035 cm21. Other data of the major
(2R )-isomer: 1H NMR d 5.50 (m, 1H), 5.04–5.18 (m, 2H),
4.17 (m, 1H), 1.88–2.20 (m, 10H), 1.76 (s, br, 3H), 1.68 (d,
br, J¼0.8 Hz, 3H), 1.62 (s, br, 3H), 1.60 (s, br, 3H), 1.48–
1.58 (m, 2H), 1.35 (m, 1H), 1.17 (s, 3H); 13C NMR d 136.3,

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7698
135.6, 131.5, 124.2 (2C), 123.9, 74.1, 71.0, 42.3, 39.7, 39.4,
34.4, 26.7, 26.3, 25.7, 24.0, 22.3, 18.9, 17.7, 16.0; MS (EI)
m/z (rel. int.) 307 (MþHþ, 1), 289 (5), 270 (15), 255 (10),
195 (5), 177 (40), 135 (30), 121 (45), 109 (50), 93 (95), 79
(70), 69 (100), 55 (25); MS (CI) m/z (rel. int.) 289 (Mþ–OH,
5), 271 (5), 215 (10), 195 (20), 177 (100), 119 (15), 109
(10), 95 (15). Anal. Calcd for C20H34O2: C, 78.4; H, 11.2.
Found: C, 78.1; H, 11.3.
2,6-Lutidine (98 mg, 0.91 mmol) followed by (t-Bu)2-
Si(OTf)2 (0.11 mL, 0.30 mmol) were added to a stirred
solution of the diol 12b (36 mg, 0.12 mmol) in CH2Cl2
(1.1 mL) at 08C (bath temperature) under argon. The
solution was stirred at rt for 2 h, NaHCO3 (5% aq) was
added and the mixture was extracted with CH2Cl2
(3£10 mL). The combined organic phases were dried over
Na2SO4, filtered and concentrated. The resulting oil was
purified by chromatography (Et2O/pentane (1/13) as eluent).
This gave the silyl derivative 13b (32 mg, ,60%) as a
viscous oil: Rf 0.77 (EtOAc/cyclohexane (3/2) as eluent); 1H
NMR (acetone (d-6)) d 5.41 (m, 1H), 5.06–5.19 (m, 2H),
4.56 (m, 1H), 2.50 (dddd, J¼11.9, 5.5, 2, 2 Hz, 1H), 1.83–
2.14 (m, 9H), 1.78 (s, br, 3H), 1.66 (s, br, 3H), 1.61 (s, br,
3H), 1.59 (s, br, 3H), 1.42–1.51 (m, 2H), 1.33 (m, 1H), 1.08
(s, 3H), 1.07 (s, 9H), 1.04 (s, 9H).
Data of 12b: Rf 0.32 (EtOAc/cyclohexane (3/2) as eluent);
[a ]2D4¼þ14.8 (c 0.79, MeOH). Other data of the major,
(2R )-isomer: 1H NMR d 5.47 (m, 1H), 5.04–5.17 (m, 2H),
4.18 (m, 1H), 2.25 (m, 1H), 1.78–2.16 (m, 9H), 1.76 (s, br,
3H), 1.68 (d, br, J¼1.0 Hz, 3H), 1.62 (s, br, 3H), 1.60 (s, br,
3H), 1.48–1.59 (m, 2H), 1.35 (m, 1H), 1.13 (s, 3H); 13C
NMR d 136.6, 135.6, 131.5, 124.2 (2C), 123.7, 74.0, 71.1,
42.1, 39.9, 39.7, 33.8, 27.1, 26.7, 25.7, 23.4, 22.1, 18.8,
17.7, 16.0.
4.3.9. [4.3.1] Bicyclic bis(tert-butyl)silyl derivative (13a).
Similar treatment (see Section 4.3.8.) of the diol 12a
furnished the title silyl derivative 13a: Rf 0.72 (EtOAc/
4.3.6. (6R,7R )-Hydroxy-3,10-prenylbisaboladien-2-one
(1a). The diol 12a (0.27 g, 0.88 mmol) was stirred in a
mixture of CH2Cl2 (11 mL) and MnO2 (2.26 g, 23 mmol,
90%, dried at 1408C, stored in desiccator before use) at rt for
2 h. The mixture was filtered through a pad of celite,
concentrated, and the resulting oil was purified using flash
chromatography (8 g silica gel, EtOAc/cyclohexane (3/7) as
eluent). This gave the title compound 1a (0.24 g, 89%), as a
slightly yellowish viscous oil: Rf 0.52 (EtOAc/cyclohexane
(3/2) as eluent); oven temperature 2158C/0.3 mbar (Kugel-
rohr distillation); [a ]D25¼24.3 (c 1.01, CHCl3); IR (neat,
1
cyclohexane (3/2) as eluent); H NMR (acetone (d-6)) d
5.42 (m, 1H), 5.06–5.21 (m, 2H), 4.57 (m, 1H), 2.39 (dddd,
J¼11.8, 5.5, 2, 2 Hz, 1H), 1.89–2.26 (m, 9H), 1.78 (d, br,
J¼1.2 Hz, 3H), 1.66 (d, br, J¼0.7 Hz, 3H), 1.61 (s, br, 3H),
1.59 (s, br, 3H), 1.42–1.52 (m, 2H), 1.32 (m, 1H), 1.11 (s,
3H), 1.07 (s, 9H), 1.02 (s, 9H).
4.3.10. (6R,7R )-7-Hydroxy-3,10-bisaboladien-2-one
(14a), (2)-7-epi-delobanone and (6R,7S )-7-hydroxy-
3,10-bisaboladien-2-one (14b), (2)-delobanone. Dry
Et2O (5 mL) was added to activated magnesium turnings
(0.15 g, 6.2 mmol) under argon. 5-Bromo-2-methylpent-2-
ene (1.0 g, 6.1 mmol) was added dropwise under gentle
reflux. After refluxing for 15 min the mixture was cooled to
08C (bath temperature) and the ketone 11 (1.8 g, 4.6 mmol)
in dry Et2O (10 mL) was added dropwise. The resulting
mixture was stirred at rt for 2.5 h, followed by addition of
NH4Cl (aq, sat.), and the mixture was extracted with
Et2O. The combined organic phases were washed water,
brine and dried over MgSO4 and the solvents were
evaporated. The resulting oil was filtered through a plug
of silica gel using EtOAc/cyclohexane (1/9 followed by 1/4)
as eluent. This gave (2R/S,6R,7R/S)-2-[(tert-butyldiphenyl-
silyl)oxy]-7-hydroxy-3,10-bisaboladiene (1.1 g) as a color-
less viscous oil. This mixture of diastereomers was taken to
the next step without further purification. Data of the crude
mixture of major (–OTBDPS¼2R ) isomers: d 7.65–7.75
(m, 4H), 7.30–7.44 (m, 6H), 5.41 (m, 1H), 5.01 (m, 1H),
4.28 (m, 1H), 1.50–2.12 (m, 10H), 1.67 (s, br, 3H), 1.54 (s,
br, 3H), 1.14–1.36 (m, 3H), 1.08 (s, 9Hþ3H of one diast.),
(1.08 (s, 3H of other diast.)); MS (EI) m/z (rel. int.) 477 (Mþ,
2), 337 (80), 259 (20), 199 (100), 181 (15), 139 (35), 121
(75), 93 (55), 77 (35); MS (CI) m/z (rel. int.) 139 (95), 121
(35), 95 (100). The mixture of diastereomers of the crude
monosilylated diols from above was desilylated as
described for (2R/S,6R,7R/S )-2-[(tert-butyldiphenyl-silyl)-
oxy]-7-hydroxy-3,10-prenylbisaboladiene (see Section
4.3.5.). MPLC (70 g silica gel, with a gradient of
EtOAc/cyclohexane (0/1!3/2) as eluent) first furnished
(2R/S,6R,7R)-7-hydroxy-3,10-bisaboladien-2-ol (0.10 g),
then a mixture of diastereomers (0.02 g), and at last
(2R/S,6R,7S)-7-hydroxy-3,10-bisaboladien-2-ol (0.11 g),
both as colorless viscous oils (21% yield over two steps).
1
KBr) 3460, 2970, 2925, 1660, 1450, 1375, 1110 cm21; H
NMR d 6.78 (m, 1H), 5.04–5.17 (m, 2H), 2.39–2.61 (m,
2H), 1.94–2.35 (m, 9H), 1.78 (m, 3H), 1.68 (d, br,
J¼0.9 Hz, 3H), 1.62 (s, br, 3H), 1.60 (s, br, 3H), 1.47–
1.59 (m, 3H), 1.17 (s, 3H); 13C NMR d 200.2, 145.3, 136.0,
135.2, 131.6, 124.2, 123.8, 73.4, 44.3, 39.7 (2C), 39.6, 26.7,
26.6, 25.7, 23.9, 22.2, 17.7, 16.1, 15.6; MS (EI) m/z (rel.
int.) 304 (Mþ, 1), 287 (5), 269 (5), 195 (10), 177 (55), 135
(35), 121 (50), 110 (90), 109 (100), 108 (15), 107 (40), 95
(50), 81 (50), 69 (65), 55 (15); MS (CI) m/z (rel. int.) 305
(MþHþ, 10), 287 (75), 269 (100), 177 (80), 135 (50), 109
(45). Anal. Calcd for C20H32O2: C, 78.9; H, 10.6. Found: C,
78.6; H, 10.6.
4.3.7. (6R,7S )-Hydroxy-3,10-prenylbisaboladien-2-one
(1b). Similar treatment (see Section 4.3.6.) of the diol 12b
furnished the title compound 1b, as a slightly yellowish
viscous oil: Rf 0.54 (EtOAc/cyclohexane (3/2) as eluent);
[a ]D26¼213.7 (c 0.76, CHCl3), lit.1 [a ]2D5¼211.37 (c 1,
CHCl3); 1H NMR d 6.75 (m, 1H), 5.04–5.17 (m, 2H), 2.63
(m, 1H), 1.93–2.45 (m, 10H), 1.78 (m, 3H), 1.68 (d, br,
J¼0.9 Hz, 3H), 1.62 (s, br, 3H), 1.60 (s, br, 3H), 1.47–1.58
(m, 3H), 1.19 (s, 3H); 13C NMR d 200.4, 145.0, 136.0,
135.3, 131.6, 124.2, 123.8, 73.3, 44.4, 39.7, 39.5, 39.0, 27.2,
26.6, 25.7, 24.0, 22.2, 17.7, 16.1, 15.6; MS (EI) m/z (rel.
int.) 304 (Mþ, 5), 287 (5), 269 (5), 195 (10), 177 (50), 135
(35), 121 (50), 110 (100), 109 (95), 108 (20), 107 (40), 95
(65), 81 (60), 69 (90), 55 (20); MS (CI) m/z (rel. int.) 305
(MþHþ, 10), 287 (70), 269 (80), 177 (100), 135 (75), 109
(65). Quadrupol MS (HP 5973) data were very similar to the
MS data reported for the natural product for both isomers.1
4.3.8. [4.3.1] Bicyclic bis(tert-butyl)silyl derivative (13b).

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7699
Data of the (2R/S,6R,7R)-isomer: Rf 0.34 (EtOAc/cyclo-
hexane (3/2) as eluent); [a]2D3¼þ2.4 (c 0.78, MeOH); IR
(neat, KBr) 3375, 2970, 2920, 1450, 1375, 1110,
1035 cm21. Other data of the major, (2R )-isomer: d 5.50
(m, 1H), 5.13 (m, 1H), 4.17 (m, 1H), 1.86–2.22 (m, 6H),
1.76 (s, br, 3H), 1.69 (d, br, J¼0.8 Hz, 3H), 1.62 (s, br, 3H),
1.48–1.58 (m, 3H), 1.35 (m, 1H), 1.17 (s, 3H); 13C NMR d
136.3, 132.0, 124.3, 123.9, 74.0, 71.0, 42.3, 39.4, 34.4, 26.3,
25.7, 24.0, 22.4, 18.9, 17.7; MS (EI) m/z (rel. int.) 238 (Mþ,
1), 221 (5), 202 (25), 187 (15), 135 (10), 127 (50), 119 (20),
109 (100), 94 (55), 79 (60), 69 (45), 55 (30); MS (CI) m/z
(rel. int.) 221 (Mþ–OH, 5), 203 (55), 147 (10), 127 (100),
119 (10), 109 (10), 95 (5).
Financial support from the Mid Sweden University, the
Foundation for Strategic Environmental Research, MIS-
TRA and the Swedish Science Research Council is grate-
fully acknowledged.
References
1. Alexander, I. C.; Pascoe, K. O.; Manchard, P.; Williams, L. A.
D. Phytochemistry 1991, 30, 1801.
2. E.g. (a) Huynh, C.; Derguini-Boumechal, F.; Linstrumelle, G.
Tetrahedron Lett. 1979, 20, 1503. (b) Beau, J.-M.; Aburaki, S.;
Pougny, J.-R.; Sinay¨, P. J. Am. Chem. Soc. 1983, 105, 621.
(c) Brockway, C.; Kocienski, P.; Pant, C. J. Chem. Soc.,
Data of the (2R/S,6R,7S)-isomer: Rf 0.26 (EtOAc/cyclo-
hexane (3/2) as eluent); [a]2D4¼þ6.0 (c 0.78, MeOH). Other
ˆ
Perkin Trans. 1 1984, 875. (d) Larcheveque, M.; Petit, Y.
1
data of the major (2R )-isomer: H NMR d 5.47 (m, 1H),
Tetrahedron Lett. 1987, 28, 1993. (e) Mitani, M.; Iwaki, T.
J. Chem. Res., Synop. 1998, 498. For a review on allylic
organometallics, see: (e) Yamamoto, Y.; Asao, N. Chem. Rev.
1993, 93, 2207.
5.12 (m, 1H), 4.17 (m, 1H), 2.25 (m, 1H), 1.85–2.11 (m,
5H), 1.76 (s, br, 3H), 1.69 (d, br, J¼0.9 Hz, 3H), 1.62 (s, br,
3H),. 1.47–1.55 (m, 3H), 1.35 (m, 1H), 1.13 (s, 3H); 13C
NMR d 136.6, 132.0, 124.3, 123.7, 73.9, 71.1, 42.1, 39.9,
33.8, 27.1, 25.7, 23.4, 22.1, 18.8, 17.7; MS (EI) and (CI)
similar to the 7R isomer.
´
´
3. E.g. (a) Marco, J. A.; Carda, M.; Gonzalez, F.; Rodrıguez, S.;
Murga, J.; Falomir, E. An. Quim. 1995, 91, 103. (b) Hayward,
M. M.; Roth, R. M.; Duffy, K. J.; Dalko, P. I.; Stevens, K. L.;
Guo, J.; Kishi, Y. Angew. Chem., Int. Ed. Engl. 1998, 37, 192.
(c) Murakami, T.; Shimizu, T.; Taguchi, K. Tetrahedron 2000,
56, 533.
The diols (2R/S,6R,7R )-7-hydroxy-3,10-bisaboladien-2-ol
and (2R/S,6R,7S )-7-hydroxy-3,10-bisaboladien-2-ol were
oxidized separately using the procedure described for 12a
(see Section 4.3.6.). This afforded (2)-7-epi-delobanone
(14a) and (2)-delobanone (14b).
4. Baldwin, J. E.; Broline, B. M. J. Am. Chem. Soc. 1982, 104,
2857.
5. Gemal, A. L.; Luche, J.-L. J. Am. Chem. Soc. 1981, 103, 5454.
6. (a) Noma, Y.; Nishimura, H.; Hiramoto, S.; Iwami, M.;
Tatsumi, C. Agric. Biol. Chem. 1982, 46, 2871. (b) Nishimura,
H.; Hiramoto, S.; Mizutani, J.; Noma, Y.; Furusaki, A.;
Matsumoto, T. Agric. Biol. Chem. 1983, 47, 2697. (c) Noma,
Y.; Nishimura, H. Agric. Biol. Chem. 1987, 51, 1845. Note that
the value of 13C d for C1 of (þ)-bottrospicatol given in this
paper is probably wrong. Most probably, it should read 139.6
instead of 131.6.
4.3.11. (2)-7-epi-Delobanone (14a). Rf 0.50 (EtOAc/
cyclohexane (3/2) as eluent); [a]2D3¼27.0 (c 0.63, dioxane),
1
lit.25 [a ]2D2¼þ10.6 of slightly impure ent-14a; H NMR d
6.78 (m, 1H), 5.12 (m, 1H), 2.38–2.60 (m, 2H), 1.99–2.34
(m, 5H), 1.78 (m, 3H), 1.69 (d, br, J¼1.0 Hz, 3H), 1.62
(s, br, 3H), 1.49–1.57 (m, 3H), 1.17 (s, 3H); 13C NMR d
200.2, 145.3, 135.2, 132.4, 123.9, 73.4, 44.2, 39.7, 39.6,
26.7, 25.7, 23.9, 22.3, 17.7, 15.6; MS (EI) m/z (rel. int.)
237 (MþHþ, 1), 219 (25), 201 (5), 135 (5), 110 (60), 109
(100), 95 (25), 81 (15), 69 (25), 55 (10); MS (CI) m/z
(rel. int.) 237 (MþHþ, 5), 219 (100), 201 (25), 135 (5), 95
(5).
7. Denis, C.; Laignel, B.; Plusquellec, D.; Le Marouille, J.-Y.;
Botrel, A. Tetrahedron Lett. 1996, 37, 53.
8. Gilman, H.; Jones, R. G. J. Org. Chem. 1952, 17, 1630.
9. (a) Lipshutz, B. H.; Wilhelm, R. S.; Floyd, D. M. J. Am. Chem.
Soc. 1981, 103, 7672. For reviews on organocuprates, see:
(b) Lipshutz, B. H.; Wilhelm, R. S.; Kozlowski, J. A.
Tetrahedron 1984, 40, 5005. (c) Lipshutz, B. H.; Sengupta,
S. Org. React. (NY) 1992, 41, 135. (d) Taylor, R. J. K.
Synthesis 1985, 364. (e) Krause, N.; Gerold, A. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 187.
4.3.12. (2)-Delobanone (14b). Rf 0.52 (EtOAc/cyclo-
hexane (3/2) as eluent); [a ]2D4¼211.4 (c 0.69, dioxane),
lit.24 [a]2D3¼þ10.2 (c 0.998, dioxane) of ent-14b; IR (neat,
1
KBr) 3460, 2970, 2925, 1660, 1450, 1375, 1110 cm21; H
NMR d 6.76 (m, 1H), 5.11 (m, 1H), 2.62 (m, 1H), 1.99–2.37
(m, 6H), 1.78 (m, 3H), 1.69 (d, br, J¼1.0 Hz, 3H), 1.62 (s,
br, 3H), 1.48–1.56 (m, 3H), 1.18 (s, 3H); 13C NMR d 200.4,
145.0, 135.3, 132.3, 123.9, 73.3, 44.4, 39.5, 39.0, 27.2, 25.7,
24.0, 22.3, 17.7, 15.6; MS (EI) m/z (rel. int.) 237 (MþHþ,
1), 219 (5), 135 (5), 110 (70), 109 (100), 95 (35), 81 (20), 69
(35), 55 (10); MS (CI) m/z (rel. int.) 237 (MþHþ, 1), 219
(100), 201 (60), 135 (15), 123 (30), 95 (20).
10. Kad, G. L.; Khurana, A.; Singh, V.; Singh, J. J. Chem. Res.
1999, 164.
11. (a) Felkin, H.; Umpleby, J. D.; Hagaman, E.; Wenkert, E.
Tetrahedron Lett. 1972, 2285. For a review on intramolecular
metallo-ene reactions, see: (b) Oppolzer, W. Angew. Chem.,
Int. Ed. Engl. 1989, 28, 38.
12. Hutchinson, D. A.; Beck, K. R.; Benkeser, R. A.; Grutzner,
J. B. J. Am. Chem. Soc. 1973, 95, 7075.
13. E.g. (a) Lipshutz, B. H.; Ellsworth, E. L.; Dimock, S. H.;
Smith, R. A. J. J. Org. Chem. 1989, 54, 4977. (b) Lipshutz,
B. H.; Ellsworth, E. L.; Dimock, S. H.; Smith, R. A. J. J. Am.
Chem. Soc. 1990, 112, 4404. (c) Lipshutz, B. H.; Hackmann,
C. J. Org. Chem. 1994, 59, 7437.
Acknowledgments
We thank Mr Martin Henriksson for technical assistance in
the synthesis of model compound 6b and Dr Lawrence A. D.
Williams for providing a copy of the original H NMR
spectrum of the natural product 1b isolated from C. linearis.
14. (a) Stack, D. E.; Dawson, B. T.; Rieke, R. D. J. Am. Chem.
Soc. 1991, 113, 4672. (b) Stack, D. E.; Dawson, B. T.; Rieke,
R. D. J. Am. Chem. Soc. 1992, 114, 5110. (c) Stack, D. E.;
1

¨
O. Smitt, H.-E. Hogberg / Tetrahedron 58 (2002) 7691–7700
7700
Klein, W. R.; Rieke, R. D. Tetrahedron Lett. 1993, 34, 3063.
(d) Rieke, R. D.; Klein, W. R.; Wu, T.-C. J. Org. Chem. 1993,
58, 2492.
20. Dodd, D. S.; Oehlschlager, A. C. J. Org. Chem. 1992, 57,
2794.
21. (a) Bailey, W. F.; Punzalan, E. R. J. Org. Chem. 1990, 55,
5404. (b) Negishi, E.; Swanson, D. R.; Rousset, C. J. J. Org.
Chem. 1990, 55, 5406.
15. (a) Lipshutz, B. H.; Crow, R.; Dimock, S. H.; Ellsworth, E. L.
J. Am. Chem. Soc. 1990, 112, 4063. (b) Lipshutz, B. H.; Ung,
C.; Elworthy, T. R.; Reuter, D. C. Tetrahedron Lett. 1990, 31,
4539.
22. For similar reactions see: (a) Marcos, I. S.; Oliva, I. M.; Diez,
D.; Basabe, P.; Lithgow, A. M.; Moro, R. F.; Garrido, N. M.;
Urones, J. G. Tetrahedron 1995, 51, 12403. (b) Coates, R. M.;
Jin, A. Q. J. Org. Chem. 1997, 62, 7475.
16. Weigand, S.; Bru¨ckner, R. Synthesis 1996, 475.
17. For a similar observation see: Muto, S.; Nishimura, Y.; Mori,
K. Eur. J. Org. Chem. 1999, 2159.
´
18. (a) Kocienski, P.; Wadman, S.; Cooper, K. J. Org. Chem.
´
1989, 54, 1215. (b) Kocienski, P. J.; Pritchard, M.; Wadman,
23. Corey, E. J.; Hopkins, P. B. Tetrahedron Lett. 1982, 23, 4871.
24. Takeda, K.; Sakurawi, K.; Ishii, H. Tetrahedron 1971, 27,
6049.
S. N.; Whitby, R. J.; Yeates, C. L. J. Chem. Soc., Perkin Trans.
1 1992, 3419.
25. Harwood, L. M.; Julia, M. Tetrahedron Lett. 1980, 21, 1743.
26. (a) Mori, K.; Igarashi, Y. Liebigs Ann. Chem. 1988, 93.
(b) Arnone, A.; DesMarteau, D. D.; Novo, B.; Petrov, V. A.;
Pregnolato, M.; Resnati, G. J. Org. Chem. 1996, 61, 8805.
27. Oehlschlager, A. C.; Singh, S. M.; Sharma, S. J. Org. Chem.
1991, 56, 3856.
19. Alternative synthesis of homogeraniol: (a) Kobayashi, M.;
Valente, L. F.; Negishi, E. Synthesis 1980, 1034. (b) Leopold,
E. J. Organic Syntheses; Wiley: New York, 1990; Collect.
Vol. VII. pp 258–263.