Biomimetic rearrangements of simplified labdane diterpenoids

Article abstract of DOI:10.1016/j.tet.2010.05.052

Simplified ethyl-substituted labdane diterpenoids 14 and 19 have been synthesised from (+)-sclareolide (18). Biomimetic rearrangements of these compounds, involving stereospecific 1,2-alkyl and hydride shifts, have been carried out by treatment with a variety of Lewis and protic acids. Halimane compounds, such as 34 and simple dehydration products such as 3, 32 and 33 have been formed either selectively or as mixtures depending on the reaction conditions. However, further rearrangement to clerodane products such as 1 and 2 was not observed, indicating a high degree of enzymatic control for the in vivo formation of these natural products.

Full text of DOI:10.1016/j.tet.2010.05.052

Tetrahedron 66 (2010) 6321e6330
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Biomimetic rearrangements of simplified labdane diterpenoids
*
*
Jonathan H. George, Mark McArdle, Jack E. Baldwin , Robert M. Adlington
Department of Chemistry, Univeristy of Oxford, Chemistry Research Laboratory, Mansfield Road, Oxford, OX1 3TA, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Simplified ethyl-substituted labdane diterpenoids 14 and 19 have been synthesised from (þ)-sclareolide
(18). Biomimetic rearrangements of these compounds, involving stereospecific 1,2-alkyl and hydride
shifts, have been carried out by treatment with a variety of Lewis and protic acids. Halimane compounds,
such as 34 and simple dehydration products such as 3, 32 and 33 have been formed either selectively or
as mixtures depending on the reaction conditions. However, further rearrangement to clerodane prod-
ucts such as 1 and 2 was not observed, indicating a high degree of enzymatic control for the in vivo
formation of these natural products.
Received 19 March 2010
Received in revised form 11 May 2010
Accepted 13 May 2010
Available online 20 May 2010
Keywords:
Biomimetic synthesis
Labdanes
Ó 2010 Elsevier Ltd. All rights reserved.
Clerodanes
1. Introduction
The labdanes and clerodanes1 are large families of diterpenoid
natural products, mainly isolated from plant sources, featuring the
basic hydrocarbon skeletons as shown in Figure 1.
Figure 2. Clerodane and labdane diterpenoids isolated from Mycale sp.
1
4
9
2
3
10
8
7
6
5
labdane
clerodane
Figure 1. The labdane and clerodane core structures.
Labdane and clerodanes with related structures are often
isolated from the same organism, which suggests a biogenetic link
between the two families. For example, the hydrocarbons 1, 2 and 3
were co-isolated from a southern Australian marine sponge, Mycale
sp., in 1997 (Fig. 2).2 Although obtained as an inseparable mixture,
the structures of 1, 2 and 3 were assigned by NMR studies.
Clerodane natural products appear to be biosynthetically related
to the labdanes via
a series of methyl and hydride shifts
(Scheme 1).3 The trans-clerodanes 8 and 9 might be produced from
labdane carbocation 4 via concerted 1,2-hydride and methyl shifts
* Corresponding authors. Tel.: þ44 01865 275626; e-mail addresses: jack.bald-
win@chem.ox.ac.uk (J.E. Baldwin), robert.adlington@chem.ox.ac.uk (R.M.
Adlington).
Scheme 1. Proposed biosynthesis of clerodanes from labdane carbocation 4.
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.05.052

6322
J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
(sequence a). The cis-clerodanes 10 and 11, however, cannot be
formed by a concerted process, and their biosynthesis must pre-
sumably follow a series of migrations via intermediate 5, which at
this point could lead to cis or trans compounds depending on which
C-4 methyl group migrates. In the synthesis of cis-clerodanes, for
a steroelectronic shift of C-18 to C-5 (sequence b) a conformational
change in the ring would be required. The final carbocations of the
rearrangement pathways, 6 and 7, can lose a proton from either C-3
(to give endocyclic tri-substituted alkenes) or C-19 (to give
exocyclic di-substituted alkenes).
Evidence for the proposed biosynthetic rearrangement is
derived from isotopic labelling experiments,4 and the pathway is
also supported by the isolation of several partially rearranged
labdane compounds such as chettaphanin I5 (12) from Adenocha-
laena siamensis, and salmantic acid6 (13), which contains a hali-
mane-type structure featuring an internal D5,10 olefin, from Cistus
laurifolius (Fig. 3).
CO2H
O
O
10
5
Scheme 2. Retrosynthetic analysis for synthesis of labdane 14 from (þ)-sclareolide.
O
OH
MeO2C
12: chettaphanin
13: salmantic acid
Figure 3. Partially rearranged labdane natural products.
2. Results and discussion
Despite the strong biogenetic link between labdane and
clerodane diterpenoid natural products, an in vitro biomimetic
synthesis of a clerodane skeleton from a labdane has never been
achieved. The aim of this project was to synthesise a simplified
labdane structure, such as 14, in order to fully investigate
the possibility of inducing the proposed series of 1,2-methyl
and hydride shifts under acid catalysis to give the clerodane
ring systems of 1 and 2. It was anticipated that the use of an
unfunctionalised labdane such as 14 would represent the
best chance to induce the biomimetic cascade reaction, as there
are no functional groups that could interfere with the rear-
rangement, or trap out early carbocation intermediates as has
been previously observed.7 The retrosynthetic plan for the
synthesis of the labdane 14 from (þ)-sclareolide 18 is outlined in
Scheme 2. Compound 14 could be formed by hydrogenation of
alkene 15, which could in turn be derived from aldehyde 16 via
a Wittig-type olefination. Aldehyde 16 could be formed by
oxidation of the diol 17, which has been previously synthesised
from the readily available chiral starting material (þ)-sclareolide
in three steps.8
Scheme 3. Possible concerted rearrangement of a labdane with an axial hydroxyl
group at CeX.
2.1. Synthesis of simplified labdane substrates
The starting point for the synthesis of ladbane 14 was the
commercially available terpenoid (þ)-sclareolide 18, which was
converted into diol 17 via a three-step sequence, which modified
the known procedure of Kuchkova et al.8 (Scheme 4). Addition of
MeLi to 18 at ꢀ78 ꢁC gave a ketone, which underwent Baeyer/Vil-
liger oxidation on treatment with in situ generated trifluoroper-
acetic acid to give an acetate. Basic hydrolysis of this intermediate
then gave 17, which was subjected to a Swern oxidation to give
aldehyde 16 in 78% yield.11 However, efforts to methylenate this
aldehyde to give alkene 15 proved to be problematic. Firstly, an
attempted Wittig olefination of 16 with methylenetriphenyl-
phosphorane, generated in situ from n-butyllithium and methyl-
triphenylphosphonium bromide, was unsuccessful, with the
a,b-unsaturated aldehyde 20 being formed as the main product.
Labdane 14 contains an equatorial hydroxyl group at C-8, but the
axially substituted alcohol 19 might also be synthesised by an
analogous route, using 8-epi-sclaerolide as the starting material.
This compound can be obtained from (þ)-sclaerolide by isomer-
isation of the lactone ring junction under strongly acidic condi-
tions.9 The axially substituted tertiary alcohol 19 might be more
pre-disposed towards the biomimetic rearrangement as the CeO
bond would be anti-periplanar to the migrating CeH bond
Further attempts to methylenate the labile b-hydroxy aldehyde 16,
including use of the Nysted reagent12 and the Oshimal/Lombardo
reaction,13 were also unsuccessful.
However, a Corey/Füchs reaction14 of aldehyde 16 with CBr4 and
Ph3P in CH2Cl2 at 0 ꢁC successfully gave the geminal dibromide 21
in 88% yield, with no elimination by-products being formed
(Scheme 5). In this process, the phosphorus ylide is generated
under relatively non-basic conditions, so the base-labile b-hydroxy
(Scheme 3). Related 1,2-methyl and hydride shifts of both 3
delanol and estradiols have shown a stereoelectronic requirement
b
-frie-
aldehyde 16 does not undergo elimination. Treatment of the
dibromoolefin 21 with n-BuLi then generated the terminal alkyne
22 in 65% yield. Hydrogenation of terminal alkyne 22 gave the
desired ethyl labdane alcohol 14 in quantitative yield.
for colinearity between the migrating group and the
*
s antibonding
orbital of the CeO bond.10

J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
6323
O
O
O
O
O
O
OH
OH
1. MeLi, Et2O, -78 °C
2. (CF3CO)2O, 35% H2O2,
NaHCO3, CH2Cl2, 0 °C
3. 10% KOH in MeOH, rt
H2SO4,
HCO2H, rt, 4 h
8
8
(80%)
(49% over 3 steps)
reference 8
H
H
H
H
i-Pr2NH, Bu2Mg,
MoOPH, THF, -78 °C
to rt, 24 h
17
18
23
18
(COCl)2,
O
(78%)
DMSO, Et3N,
CH2Cl2, -78 °C
O
HO
H
O
HO
HO
O
O
O
O
OH
n-BuLi, Ph3PCH3Br,
+
THF, 0 °C
(30%)
H
H
H
H
25 (13%)
24 (48%)
20
16
DIBAL-H, Et2O,
0 °C, 40 h
various
methylenation
protocols
OH
OH
OH
HO
H
HO
O
OH
OH
NaIO4, H2O, THF,
0 °C, 20 min
O
+
(94%)
H
H
H
15
28
26 (28%)
27 (55%)
Scheme 4. Attempted synthesis of alkene 15 from (þ)-sclareolide.
Scheme 6. Synthesis of aldehyde 28 from (þ)-sclareolide.
24 in 48% yield. In addition to the formation of 24, some unreacted
starting material was also recovered along with an undesired side
product 25 resulting from competitive nucleophilic addition of the
enolate intermediate to 24. The stereochemistry of 24 is consistent
with the expectation that the MoOPH would preferentiallyapproach
the enolate from the less hindered a-face. Reduction of 24 with
DIBAL-H gave a mixture of triol 26 (28%) and lactol 27 (55%). The triol
26 could be easily converted to the desired aldehyde 28 by oxidative
cleavage of the 1,2-diol functional group with NaIO4 in THF/H2O.
The lactol 27 initially seemed like an unusable side-product, as
re-subjection of it to a variety of hydride sources failed to reduce it to
the diol. 1H NMR studies of 27 indicated that the hydroxy groups are
cis to each other, (J¼5.0 Hz), which explains the lack of reactivity to
DIBAL-Handotherhydridesourcesasastablechelatewithametalion
couldpresumablybeformed. However, giventhatthehydroxygroups
of lactol 27 are cis to each other, it was reasoned that oxidative
cleavage of this compound to give a formate ester would be possible.
Indeed, treatment of 27 with NaIO4 gave 29 in 66% yield (Scheme 7).
Scheme 5. Synthesis of ethyl labdane 14.
The first step in producing the epimeric labdane 14 was to induce
inversion of the C-8 oxygen substituent of (þ)-sclareolide 18
(Scheme 6). This was carried out according to the procedure of
Quideau et al.,9 with exposure of (þ)-sclareolide to a mixture of
sulfuric and formic acids at room temperature giving 8-epi-sclar-
eolide 23 in 80% yield after crystallisation from hexane. The ease of
this inversion of configuration at C-8 is attributed to the release of
1,3-diaxial interaction between the C-8- and C-10-methyl groups in
18, as well as the formation of a more stable cis-fused lactone.
Application of a similar three-step dehomologation procedure to
18/17 was unsuccessful with 8-epi-slareolide as the starting ma-
terial because the more stable cis-fused lactone is less susceptible to
nucleophilic attack by MeLi, and so the ring-opened products were
formed in very low yields as under the reaction conditions depro-
tonation was presumably the favoured outcome. An alternative,
more lengthy pathway for the dehomologation of 23 was therefore
implemented. Firstly, the lactone of 23 was
a
-hydroxylated by
Scheme 7. Oxidative cleavage of cis-lactol 27.
treatment with base and MoO5$pyridine$HMPA (MoOPH)15 to give

6324
J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
Hydrolysis of 29 with Na2CO3 in MeOH then gave aldehyde 28 in 57%
yield and the -unsaturated aldehyde 20 in 19% yield.
difference whether the axial or equatorial labdane alcohol was
used. 3 and 33 could not be separated by column chromatography.
Finally it was found that when the temperature was raised to 0 ꢁC
and the reaction time extended to 5 h, compound 34, which has
a halimane-type skeleton, was the sole product formed (as a single
diastereomer). The 13C NMR of this compound showed two qua-
ternary signals at 136.8 and 132.5 ppm characteristic of the D5,10
internal carbonecarbon double bond.
a,b
Aldehyde 28 was expected to be as sensitive towards base-in-
duced elimination as the epimeric aldehyde 16, if not more so as the
hydroxyl substituent at C-8 is now axial. However, application of
a similar Corey/Füchs reaction protocol to that used earlier was
successful, with dibromoolefin 30 being formed in 90% yield
(Scheme 8). Treatment with n-BuLi then gave terminal alkyne 31,
which was hydrogenated to give the saturated labdane 19 in 97%
yield over two steps.
Further reactions with BF3$OEt2 gave no further rearrangement
beyond the halimane structure 34.
A proposed mechanistic
reasoning for the reaction of 14 or 19 with BF3 at ꢀ78 ꢁC is shown in
Scheme 10. It is thought that the initially formed 3ꢁ carbocation at
C-8, 35, undergoes proton loss from the exo-methyl group, thus
giving rise to the observed exo-methylene compound 32, which can
be trapped as the kinetic product by quenching the reaction after
2 min. Further reaction of this alkene with BF3 can generate the
carbocation 36, which can then lose a proton to give the more
stable tri- and tetra-substituted alkenes 3 and 33.
BF3
H
OH
BF3.Et2O
CH2Cl2, -78 °C
8
H
H
35
14 or 19
- H+
Scheme 8. Synthesis of ethyl labdane alcohol 19.
BF3
BF3
2.2. Acid-catalysed rearrangements of labdanes
H
H
32
Labdanes 14 and 19 were initially treated with the strong Lewis
acid BF3$Et2O in CH2Cl2 (Scheme 9). It was envisaged that this
reagent would complex with the hydroxyl group of the labdane
derivatives, generating either a partial positive charge at C-8 or
a full carbocation, which might then rearrange to give a clerodane
skeleton. Firstly, 14 and 19 were separately stirred with 5 equiv of
BF3$Et2O in CH2Cl2 at ꢀ78 ꢁC for 2 min before being quenched with
water. TLC analysis of the crude product showed formation of a very
non-polar product. NMR analysis showed that this product was the
exocyclic alkene 32, when the reaction was carried out with both 14
and 19 as the starting material. The 1H NMR spectrum of this
compound showed distinctive singlet peaks at 106.1 and 148.4 ppm
indicative of an exo-methylene group. When the reactions were left
at ꢀ78 ꢁC for a longer period of time (20 min), NMR analysis of the
crude product showed formation of a mixture of alkenes 3 and 33,
both of which feature an endocyclic double bond. Again it made no
36
loss of H+ and C-BF3
bond protonolysis
+
H
H
3
33
Scheme 10. Mechanism of BF3-mediated rearrangement of 14 and 19 at ꢀ78 ꢁC.
At higher temperature, the alkene 33 can go through further
rearrangement by reaction with BF3 to generate a carbocation
centred at C-9, 37 (Scheme 11). This species can undergo
BF3.Et2O
CH2Cl2, 0 °C
9
BF3
BF3
H
H
33
37
stereospecific
1,2-Me shift
loss of H+ and C-BF3
bond protonolysis
10
10
BF3
5
H
38
34
Scheme 9. BF3-mediated rearrangement of labdanes 14 and 19.
Scheme 11. Mechanism of BF3-mediated rearrangement of 33 at 0 ꢁC.

J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
6325
a stereospecific 1,2-methyl shift to form a carbocation centred at C-
10, 38, which can lose a proton to form the
5,10 double bond of the
halimane skeleton. Importantly, the intermediacy of alkene 33 in-
dicates that the rearrangements of 14 and 19 to 34 is not a con-
certed process, but proceeds via a series of rapidly interconverting
carbocations.
D
mCPBA, CH2Cl2,
rt, 1 h
O
O
+
:
34
40
41
TMSOTf was used as an alternative Lewis acid reagent, and in
this case reaction of 14 and 19 with 2 equiv of TMSOTf in CH2Cl2 at
0 ꢁC for 24 h formed mainly the halimane compound 34, as
observed by NMR analysis of the crude product.
Given these results and our proposed mechanism for the rear-
rangement, the stereochemistry at C-8 and C-9 in the halimane
structure 34 is clearly not derived from a concerted, stereospecific
set of 1,2-methyl and hydride shifts. Rather, the stereochemistry at
1.4
1
Scheme 13. Epoxidation of 34 with mCPBA.
substituent to C-10 could then give tertiary carbocation 44, which
could be trapped by trapped by H2O to form tertiary alcohol 45.
Lewis-acid induced fragmentation of 45 could then give the tetra-
substituted alkene 46, the major reaction product. Alternatively,
loss of a C-1 proton from 43 gives diene 47 as the minor product.
C-8 is due to attack of BF3 at the a-face of 33, opposite the axial Me
group at C-10. To verify this stereochemical outcome, confirmation
of the cis C-8, C-9 dimethyl configuration of 34 was sought. Thus,
NOE NMR experiments were attempted on 34, but due to the
unfunctionalised nature of the hydrocarbon, the signals of the two
methyl peaks at C-4, the methyl peak at C-8 and the methyl and
ethyl peaks at C-9 were heavily overlapped in a variety of deuter-
ated solvents, and the results were inconclusive. Attempted
ozonolysis of 34 did not give the expected diketone product 39.
Instead, a single product was isolated in 63% yield; 13C spectrum of
this compound showed quaternary carbon signals at 68.8 and
69.9 ppm, which suggested that the epoxide 40 had been formed.
The high resolution mass spectrum of 40 showed a molecular ion of
formula C16H28ONaþ, which supported this assignment. The 1H and
13C NMR showed that the epoxide 40 was formed as a single
stereoisomer, which indicates a surprising degree of stereo-
selectivity in this oxidation process. Epoxidation of 34 had signifi-
cantly altered the chemical shifts of the methyl and ethyl peaks and
a NOESY NMR experiment could be carried out on 40, which
showed that the Me groups attached to C-8 and C-9 are syn to each
other and that the epoxide is anti to the C-9 ethyl substituent. It
appears that the alkene 34 is too hindered to undergo ozonolysis
due to the difficulty in forming the primary ozonide by a cycload-
dition process. Instead ozone acts as an electrophilic oxygen source
in the formation of the epoxide 40 (Scheme 12). Similar reactions of
hindered alkenes with ozone are known.16
BF3.Et2O, CH2Cl2,
9
0 °C, 1 h
O
O
BF3
42
40
stereospecific
1,2-Et shift
9
10
43
44
- H
+ H2O, - H
H
O
1
BF3
45
47 (21%)
fragmentation, H
O
O3,CH2Cl2/MeOH
(10:1), -78 °C, 10 min;
then Me2S, -78 °C to rt
9
9
8
8
O
(63%)
4
40
34
46 (55%)
ozonolysis
Scheme 14. BF3-mediated rearrangement of epoxide 40.
O
No further rearrangement of the halimane structure 34 was
observed during the BF3-mediated reactions, which indicates that
34 represents a thermodynamic sink for the overall reaction.
Conversion of halimane 34 to clerodanes 1 or 2 is possibly hindered
by formation of unfavourable 1,3-diaxial interactions between the
C-5 and C-9 methyl substituents. The use of harsher reagents (such
as strong protic acids) and higher temperatures to induce further
rearrangement was also investigated. Stirring 14 and 19 with
Amberlyst-15 in toluene at room temperature for 2 h gave an
inseparable mixture of alkene products (Scheme 15). Heating the
reaction at 60 ꢁC for 24 h gave the halimane 34 as the main product
(approx. 90% yield), with traces of other alkene by-products. Similar
results were obtained when treating 14 and 19 with a mixture of
acetic acid and concd HCl at 120 ꢁC for 24 h. The protic acid induced
rearrangements did not yield pure 34 (approx. 90% yield) as the
BF3$Et2O reaction had done, despite the higher temperature and
O
39
Scheme 12. Attempted ozonolysis of 34dunexpected formation of epoxide 40.
That the ozonolysis product 40 was the epoxide was further
confirmed by reaction of alkene 34 with mCPBA in CH2Cl2. This gave
a 1.4:1 inseparable mixture of 40 and its diastereomer 41 in 90%
overall yield (Scheme 13).
Treatment of epoxide 40 with BF3$Et2O was then carried out in
an attempt to induce migration of one of the C-4 methyl groups to
C-5. However, the only isolated products were tetra-substituted
cyclohexene 46 (55%) and diene 47 (21%) (Scheme 14). In this case,
attack of BF3 at the epoxide appears to generate the allylic carbo-
cation 43, via 42. Stereospecific 1,2-migration of the C-9 ethyl

6326
J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
periplanar to the equatorial eOMs group of the intermediate
mesylate 48 is situated on the C-8 methyl group. When the more
hindered labdane alcohol 19 was heated at 100 ꢁC, alkenes 3 and 33
were formed by a similar E2 elimination reaction, with in this case
the anti-periplanar H atoms at C-7 and C-9 of mesylate 49 being
removed.
Amberlyst 15,
toluene, rt, 2 h
OH
+
+
H
H
H
H
33
32
3
14 or 19
Amberlyst 15,
60 °C, toluene, 24 h
or
AcOH, conc.
HCl, 120 °C, 24 h
3. Conclusion
The ethyl-substituted labdane alcohols 14 and 19 have been
synthesised in an efficient manner from (þ)-sclareolide.
Attempts to induce a full rearrangement of these compounds
with Lewis or protic acids to give clerodane derivatives was not
possible, with either halimanes such as 34 or simple dehydration
products such as 3 being formed instead. The fact that labdane 19
(with an axial hydroxyl group at C-8) and 14 (which has an
equatorial hydroxyl group at C-8) display identical reactivity
towards acidic reagents, combined with the intermediacy of
alkenes such as 32 and 33, indicates that the 1,2-hydride and
methyl shifts that are observed proceed via an unconcerted
mechanism involving an equilibrium of rapidly interconverting
carbocations. In nature, the labdane to clerodane rearrangement
presumably occurs under enzymatic control, with a carbocation
equilibrium being formed inside an enzyme pocket. The C-4
intermediary cation, that is, the direct precursor of the cler-
odanes must be formed at the end of the rearrangement cascade.
Thus, stabilisation of this cation by the enzyme must be required,
34
Scheme 15. Reactions of 14 and 19 with protic acids.
longer reaction time, which indicates that strong Lewis acids
such as BF3 are better reagents for this type of transformation.
The results so far indicate that the rearrangement of the
labdanes 14 and 19 is occurring via a series of non-concerted 1,2-
hydride and Me-shifts, even when the leaving group is anti-peri-
planar to the CeH bond. A dynamic equilibrium of carbocations is
thus formed, with the major alkene product being dictated by
thermodynamic stability of the molecule.
Up to this point, no difference had been observed in the
reactions of the two epimers 14 and 19. However, when the two
compounds were separately treated with mesyl chloride in 2,6-
lutidine at 0 ꢁC the
a non-polar product after 24 h, while the
a
-labdane alcohol 14 was found to react to give
b
-ladbane alcohol 19
showed no reaction. This is presumably because 19 has a more
hindered axial hydroxyl group (due to 1,3-diaxial interactions).
The product of the reaction of 14 with MsCl was found to be the
exo-methylene compound 32 (Scheme 16). This alkene is selec-
tively formed via an E2 mechanism as the only H atom anti-
either through
p-cation interactions with aryl groups of side-
chains or through electrostatic interactions with negatively
charged residues.17
In order to execute a successful labdane to clerodane rear-
rangement, there needs to be a driving force for the second
1,2-methyl shift to prevent the reaction terminating at the
halimane stage. Given that the halimane skeleton appears to be
inherently more stable than the clerodane structure, a reaction
system in which the C-4 centred carbocation precursor to the
clerodanes is stabilised relative to the other carbocations in the
rearrangement pathway would need to be designed. Trapping
out this stabilised carbocation under kinetic control could thus
form the clerodane ring system. One possibility for this would be
to use a synthetic labdane precursor such as 50, with a silicon
substituent at C-3 (Scheme 17). Treatment of this compound
with a Lewis acid should preferentially generate the carbocation
52 via 51, with the carbocation centre at C-4 being stabilised by
the CeSi bond at C-3.
H
H
H
MeSO2Cl,
2,6-lutidine,
0 °C, 24 h
OSO2Me
OH
8
(98%)
H
H
14
48
H
32
MeSO2Cl,
2,6-lutidine,
100 °C, 24 h
OH
OSO2Me
9
7
H
H
H
H
19
49
loss of C(9)-H
loss of C(7)-H
H
H
3
33
Scheme 17. Possible formation of the clerodane ring system of 2 via Si-stabilised
carbocation 52.
Scheme 16. Mesylation and subsequent E2 elimination of labdanes 14 and 19.

J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
6327
ꢃ
4. Experimental section
208.2; HRMS calculated for C15H26O2 [MD ] 238.1933, found
238.1934.
4.1. General experimental procedures
4.1.2. (1R,2R,4aS,8aS)-1-(2,2-Dibromo-vinyl)-2,5,5,8a-tetramethyl-
decahydro-naphthalen-2-ol (21). Ph3P (6.42 g, 24.5 mmol) was
added to a solution of CBr4 (4.06 g, 12.2 mmol) in CH2Cl2 (25 mL) at
0 ꢁC and the resulting mixture was stirred at 0 ꢁC for 30 min. A so-
lutionofaldehyde16 (1.46 g, 6.12 mmol)inCH2Cl2 (25 mL)wasadded
and the mixture was stirred at 0 ꢁC for 2 h. The reaction mixture was
diluted with Et2O, and the organic layer was washed with brine and
dried with MgSO4. Evaporation of the organic solvent gave a residue,
which was purified by flash chromatography (silica gel, hexanes/
EtOAc, 8:1) to yield the dibromide 21 (2.12 g, 5.38 mmol, 88%) as
a white solid. Data for 21: Rf 0.55 (hexanes/EtOAc, 4:1); mp 75e77 ꢁC;
All reactions were carried out under an atmosphere of nitro-
gen in glassware that had been dried at 100 ꢁC overnight, unless
otherwise specified. Proton magnetic resonance spectra were
recorded on Brüker DPX200 (200 MHz), Brüker DQX400
(400 MHz) and Brüker AVC500 (500 MHz) spectrometers at am-
bient temperatures. Chemical shifts (dH) are reported as a ratio of
parts per million (ppm) and are referenced to the residual sol-
vent peak. Coupling constants (J) are reported in Hertz (Hz) and
recorded to ꢂ0.5 Hz. Proton spectra assignments are supported
by 1He1H COSY and HSQC where necessary. The order of citation
is (i) multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; m,
multiplet and br s, broad singlet), (ii) coupling constant (J)
quoted in Hertz (Hz) to the nearest 0.5 Hz, (iii) number of
equivalent nuclei (by integration). Carbon magnetic resonance
spectra were recorded on Brüker DQX400 (400 MHz) and Brüker
AVC500 (500 MHz) spectrometers at ambient temperatures.
Chemical shifts (dH) are reported as a ratio of parts per million
(ppm) and are referenced to the residual solvent peak. Carbon
spectra assignments are supported by DEPT analysis and 1He13C
correlations where necessary. Mass spectra were recorded using
Bruker MicroTOF (ES mode) and Micromass GCT (EI mode)
spectrometers. Only molecular ions (Mþ), fragments from mo-
lecular ions and other major peaks are reported. Spectra recorded
on a Micromass Autospec spectrometer (EI mode) and Field
Ionisation (FI) measurements are accurate to ꢂ0.5 ppm. Infrared
spectra were recorded on a Bruker Tensor 27 FT-IR spectrometer
as a thin film between NaCl plates (oils), and as KBr disks (solids).
[a]
22 þ7.0 (c 1, CHCl3); IR (KBr disc): 3439 (w br), 2886 (s), 2725 (m),
D
1744(w),1461(s),1377(s),1312(m) cmꢀ1;1HNMR(400 MHz, CDCl3):
d
¼0.81 (s, 3H), 0.90 (s, 3H), 0.92 (s, 3H), 0.99 (dd, J¼12.0, 2.5 Hz, 1H),
1.08 (m, 1H),1.16 (m,1H),1.23 (s, 3H),1.30 (m, 1H),1.37e1.62 (m, 6H),
1.70 (m, 1H), 1.93 (dt, J¼13.0, 3.5 Hz, 1H), 2.29 (d, J¼11.0 Hz, 1H), 6.45
(d, J¼11.0 Hz,1H); 13C NMR (100 MHz, CDCl3)
¼16.2,18.3, 20.2, 21.6,
d
24.8, 33.3, 33.4, 38.9, 39.6, 41.8, 43.2, 55.7, 65.9, 73.1, 91.2,136.3; HRMS
ꢃ
calculated for C16H26Br2O [MD ] 394.0331, found 394.0310.
4.1.3. (1R,2R,4aS,8aS)-1-Ethynyl-2,5,5,8a-tetramethyl-decahydro-
naphthalen-2-ol (22). A solution of n-butyllithium (1.6 M in hex-
ane, 10.1 mL, 16.1 mmol) was added to a solution of dibromide 21
(2.12 g, 5.38 mmol) in anhydrous THF (25 mL) at ꢀ78 ꢁC and the
resulting mixture was stirred at this temperature for 45 min. The
reaction mixture was then quenched by the addition of saturated
aqueous NH4Cl solution (25 mL). Et2O (100 mL) was added, the
organic layer was separated and the aqueous layer was extracted
with Et2O (2ꢄ50 mL). The combined organics were then washed
with water and brine, dried over Mg2SO4 and filtered. Concen-
tration under reduced pressure gave a residue, which was puri-
fied by flash chromatography (silica gel, hexanes/EtOAc, 10:1) to
Absorption maxima (vmax) are reported in wavenumbers (cmꢀ1
)
and only selected peaks are reported. The following abbrevia-
tions are used: w, weak; m, medium; s, strong and br, broad.
Melting points (mp) were recorded on a Reicher/Koffler block
apparatus and are uncorrected. Optical rotations were obtained
on a Perkin/Elmer 241 polarimeter. Thin layer chromatography
(TLC) was performed using Merck aluminium foil backed plates
pre-coated with silica gel 60 F254 (1.05554). Visualisation was
effected by staining with ceric ammonium molybdate, (CAM),
followed by heating. Retention factors (Rf) are reported to ꢂ0.05.
Flash chromatography was performed using ICN silica 32e63,
60 Å.
give alkyne 22 as a colourless oil (0.82 g, 3.50 mmol, 65%). Data
22
for 22: Rf 0.40 (silica gel, hexanes/EtOAc, 4:1); mp 91e93 ꢁC; [
a]
D
þ3.0 (c 1, CHCl3); IR (KBr disc): 3553 (s), 3442 (br s), 3296 (s),
3270 (s), 2924 (s), 2098 (w), 1386 (m) cmꢀ1 1H NMR (400 MHz,
CDCl3):
;
d
¼0.79 (s, 3H), 0.86 (m, 1H), 0.87 (s, 3H), 0.97 (s, 3H), 0.98
(s, 1H), 1.11e1.28 (m, 2H), 1.33 (s, 3H), 1.36e1.49 (m, 3H),
1.55e1.69 (m, 2H), 1.86 (m, 1H), 1.92 (dt, J¼12.5, 3.5 Hz, 1H), 2.12
(br s, 1H), 2.21 (d, J¼2.5 Hz, 1H), 2.29 (d, J¼2.5 Hz, 1H); 13C NMR
(100 MHz, CDCl3):
d
¼16.0, 18.5, 19.9, 21.4, 25.1, 33.2, 33.3, 37.9,
4.1.1. (1R,2R,4aS,8aS)-2-Hydroxy-2,5,5,8a-tetramethyl-decahydro-
naphthalene-1-carbaldehyde (16)11. To a solution of DMSO (2.61 mL,
36.5 mmol) in CH2Cl2 (30 mL) was added oxalyl chloride (1.58 mL,
18.6 mmol) at ꢀ78 ꢁC and the reaction mixture was stirred at
ꢀ78 ꢁC for 30 min. A solution of diol 17 (1.68 g, 6.98 mmol) in
CH2Cl2 (30 mL) was then added and the reaction mixture was
stirred at ꢀ78 ꢁC for 30 min. Et3N (10.9 mL, 78.3 mmol) was then
added and the reaction mixture was allowed to warm to room
temperature over 30 min. The reaction mixture was then diluted
with ice water and extracted with CH2Cl2. The organic layer was
washed with brine, dried over MgSO4 and filtered. Evaporation of
the organic solvent under reduced pressure gave a residue, which
was purified by flash chromatography (silica gel, hexanes/EtOAc,
5:1) to give aldehyde 16 (1.30 g, 5.45 mmol, 78%) as a colourless oil.
Data for 16: Rf 0.25 (silica gel, hexanes/EtOAc, 4:1); mp 61e63 ꢁC;
40.5, 40.8, 42.0, 54.8, 57.8, 71.8, 73.7, 82.9; HRMS calculated for
ꢃ
C16H26O [MD ] 234.1984, found 234.1984.
4.1.4. (1R,2R,4aS,8aS)-1-Ethyl-2,5,5,8a-tetramethyl-decahydro-
naphthalen-2-ol (14). A catalytic amount of 10% Pd on carbon
(70 mg) was added to a solution of 22 (689 mg, 2.94 mmol) in
anhydrous MeOH under a nitrogen atmosphere at room temperature.
Thereactionflask wasthenpurgedwithhydrogengas and left stirring
at room temperature for 24 h. The resulting solution was filtered
through Celite and the filtrate was evaporated to give 14 as a white
solid (700 mg, 2.94 mmol, 100%). Data for 14: Rf 0.45 (silica gel, hex-
anes/EtOAc, 4:1); mp 78e79 ꢁC; [
a]
D
22 ꢀ13.2 (c 1, CHCl3); IR (KBr disc):
3418 (br s), 2927 (s), 2869 (s), 1465 (m), 1386 (m) cmꢀ1
;
1H NMR
(400 MHz, CDCl3):
(m, 6H),1.13 (s, 3H),1.14 (m,1H),1.18e1.51 (m, 7H),1.55e1.72 (m, 3H),
1.87 (m, 1H); 13C NMR (100 MHz, CDCl3):
d¼0.79 (apparent br s, 6H), 0.87 (s, 3H), 0.90e1.06
[
a
]
22 þ37.0 (c 1, CHCl3), lit. [
a
]
22 þ39.2; IR (thin film): 3365 (m br),
d
¼15.4,18.1, 18.3,18.5, 20.6,
D
D
2930 (s), 1717 (m), 1700 (m), 1461 (s), 1378 (s), 1242 (s), 1191 (s)
21.5, 23.8, 33.2, 33.5, 39.2, 39.7, 42.0, 44.5, 56.1, 64.4, 74.5; HRMS
ꢃ
cmꢀ1; 1H NMR (400 MHz, CDCl3):
d
¼0.82 (s, 3H), 0.88 (s, 3H), 0.96
calculated for C16H30O[MD ] 238.2297, found 238.2296.
(dd, J¼12.0, 2.5 Hz, 1H), 1.11 (s, 3H), 1.16e1.54 (m, 6H), 1.37 (s, 3H),
1.61e1.74 (m, 2H), 1.81 (dt, J¼12.5, 3.0 Hz, 1H), 1.95 (m, 1H), 2.07 (s,
4.1.5. (1S,3aS,5aS,9aS,9bS)-1-Hydroxy-3a,6,6,9a-tetra-methyl-deca-
hydro-naphtho[2,1-b]furan-2(3aH)-one (24)9. To a stirred solution
of diisopropylamine (20.2 mL, 144.1 mmol) in anhydrous THF
1H), 10.01 (d, J¼1.5 Hz, 1H); 13C NMR (100 MHz, CDCl3)
¼17.6, 18.2,
d
19.9, 21.4, 25.3, 33.2, 33.3, 37.4, 39.8, 41.6, 42.7, 55.1, 71.3, 72.8,

6328
J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
(100 mL) was added dropwise a 1.0 M solution of Bu2Mg in hexane
(72.0 mL, 72.0 mmol). The resulting solution was heated at reflux
for 1 h, after which time it was allowed to cool to room tempera-
ture. A solution of 8-epi-sclareolide 239 (5.17 g, 20.7 mmol) in an-
hydrous THF (25 mL) was then added dropwise. The reaction
mixture was stirred at room temperature for 1 h, then cooled to
ꢀ78 ꢁC for 2 h before the addition of solid oxodi-peroxy-
molybdenum(pyridine)-(hexamethylphosphoric triamide) MoOPH
(20.95 g, 48.0 mmol) in one portion. The mixture was allowed to
warm up to ꢀ25 ꢁC overnight. Addition of water (7 mL) to the
resulting brown solution gave a white precipitate, which was
filtered, and the filtrate was extracted with Et2O (4ꢄ50 mL). The
combined extracts were washed with 10% aqueous HCl solution
and brine, then dried over MgSO4, filtered and evaporated to give
a beige solid. This crude product mixture was purified by flash
chromatography (silica gel, hexanes/EtOAc,10:1) to yield 24 (2.65 g,
9.95 mmol, 48%) as a white solid. Data for 24: Rf 0.20 (silica gel,
hexanes/EtOAc, 4:1); mp 83e84 ꢁC; 1H NMR (400 MHz, CDCl3):
19.2, 22.2, 32.7, 34.0, 34.1, 40.1, 41.2, 42.5, 45.1, 57.2, 62.0, 67.7, 73.5,
ꢃ
74.3; HRMS calculated for
293.2084.
C
16H30NaO3[MD
]
293.2087, found
4.1.7. (1R,2S,4aS,8aS)-1-Formyl-2,5,5,8a-tetramethyl-decahydro-
naphthalen-2-yl formate (29). To an ice-cold, stirred solution of 27
(497 mg, 1.85 mmol) in THF (25 mL) was added dropwise a solution
of NaIO4 (463 mg, 2.17 mmol) in water (25 mL). After stirring for
20 min, the resulting precipitate was filtered and the filtrate was
extracted with Et2O (4ꢄ40 mL). The combined organic extracts
were washed with saturated aqueous Na2S2O3 solution (100 mL)
and brine (100 mL), dried over MgSO4 and filtered. Evaporation of
the solvent under reduced pressure gave the aldehyde 29 as a white
solid (326 mg, 1.22 mmol, 66%). Data for 29: Rf 0.75 (silica gel,
20
hexanes/EtOAc, 1:1); mp 69e72 ꢁC; [
a
]
þ28.8 (c 0.56, CHCl3); IR
D
(KBr disc) 3409 (w), 2963 (s), 2752 (w), 1711 (s), 1460 (m), 1391 (m),
1166 (s) cmꢀ1; 1H NMR (400 MHz, CDCl3):
d¼0.83 (d, J¼3.0 Hz, 1H),
0.85e0.87 (m, 3H), 0.88 (s, 3H), 1.03e1.21 (m, 2H), 1.29 (s, 3H),
1.31e1.47 (m, 3H), 1.51 (s, 3H), 1.52e1.67 (m, 5H), 2.85 (dt, J¼15.0,
3.0 Hz, 1H), 8.12 (s, 1H), 9.92 (d, J¼5.0 Hz, 1H); 13C NMR (100 MHz,
d
¼0.88 (s, 3H), 0.90 (s, 3H), 0.94 (s, 3H), 1.02e1.23 (m, 3H),
1.39e1.67 (m, 5H), 1.53 (s, 3H), 1.71e1.85 (m, 3H), 2.06 (m, 1H), 3.64
(br s 1H), 4.39 (d, J¼4.0 Hz, 1H); 13C NMR (100 MHz, CDCl3):
d¼16.6,
CDCl3):
d
¼16.8, 17.8, 18.0, 21.7, 26.1, 33.2, 33.4, 36.9, 38.1, 39.8, 41.5,
ꢃ
18.0,18.1, 21.8, 31.6, 32.8, 33.2 (two peaks superimposed), 35.8, 41.8,
42.1, 48.7, 61.3, 71.3, 86.4, 177; HRMS calculated for
C16H25O3[MꢀH]ꢀ 265.1809, found 265.1812.
55.0, 70.7, 83.2, 159.8, 205.2; HRMS calculated for C16H26O3[Mþ
266.1882, found 266.1893.
]
Second product 25 also obtained as a white solid (689 mg,
1.33 mmol,13%). Data for 25: Rf 0.65 (silica gel, hexanes/EtOAc, 4:1);
4.1.8. (1R,2S,4aS,8aS)-2-Hydroxy-2,5,5,8a-tetramethyl-decahydro-
naphthalene-1-carbaldehyde (28). To an ice-cold stirred solution of
26 (303 mg, 1.12 mmol) in THF (20 mL) was added dropwise a so-
lution of NaIO4 (280 mg, 1.31 mmol) in water (20 mL). After stirring
for 20 min at 0 ꢁC, the resulting precipitate was filtered and the
filtrate was extracted with Et2O (4ꢄ35 mL). The combined extracts
were then washed with aqueous Na2S2O3 solution and brine, dried
over Mg2SO4 and filtered. Evaporation of the solvent under reduced
pressure gave the aldehyde 28 as a white solid (250 mg, 1.05 mmol,
94%). Data for 28: Rf 0.50 (silica gel, hexanes/EtOAc, 5:1); mp
mp 176e178 ꢁC; 1H NMR (400 MHz, CDCl3):
d
¼0.85 (s, 6H), 0.90 (s,
3H), 0.92 (s, 3H), 0.96 (s, 3H), 1.00e1.05 (m, 1H), 1.07 (s, 3H),
1.09e1.27 (m, 5H), 1.39 (s, 3H), 1.41e1.47 (m, 4H), 1.49 (s, 3H),
1.52e1.77 (m, 11H), 1.86e1.99 (m, 2H), 2.74 (d, J¼4.0 Hz, 1H), 2.80
(d, J¼8.5 Hz, 1H), 4.18 (t, J¼9.0 Hz, 1H); 13C NMR (100 MHz, CDCl3):
d¼17.2, 17.7, 17.9, 18.1, 18.2, 18.6, 21.1, 21.8, 29.1, 31.7, 31.9, 32.2, 32.2,
33.3, 33.4, 34.0, 36.6, 36.9, 41.9, 42.3, 42.6, 44.4, 46.2, 47.8, 50.2, 57.3,
62.5, 75.9, 81.4, 88.0, 100.6, 178.9 ppm; HRMS calculated for
C32H51O5[MꢀH]ꢀ 515.3742, found 515.3740.
64e68 ꢁC; [
a]
21 þ122.4 (c 0.37, CHCl3); IR (KBr disc) 3512 (s), 2945
D
(s), 2849 (m), 2741 (w), 1698 (s), 1459 (m) cmꢀ1
;
1H NMR (CDCl3,
4.1.6. (1S,3aS,5aS,9aS,9bS)-3a,6,6,9a-Tetramethyldodeca-hydro-
naphtho-[2,1-b]furan-1,2-diol (27) and (S)-1-((1S,2S,4aS,8aS)-2-hy-
droxy-2,5,5,8a-tetramethyl-decahydro-naphthalen-1-yl)ethane-1,2-
diol (26). To a stirred ice-cold solution of 24 (1.99 g, 7.47 mmol) in
dry Et2O (150 mL) was added dropwise a 1 M solution of DIBAL-H
in THF (57.4 mL, 57.4 mmol). The reaction mixture was allowed to
warm up to room temperature and stirring was continued for
40 h, after which time it was quenched with a 10% aqueous so-
lution of HCl (100 mL). The organic layers were separated, and
the aqueous layer was extracted with Et2O (4ꢄ50 mL). The
combined organic extracts were washed with a saturated aque-
ous solution of K2CO3 (100 mL) and brine (100 mL), and then
dried over Mg2SO4. After filtration, evaporation under reduced
pressure gave a white solid, which was purified by flash chro-
matography (silica gel, hexanes/EtOAc, 8:1) to yield 27 (1.10 g,
400 MHz):
d
¼0.84 (s, 3H), 0.88 (s, 3H), 0.89e0.96 (m, 1H), 1.15 (s,
3H), 1.18 (s, 3H), 1.15e1.80 (m, 11H), 2.12 (d, J¼3.0 Hz, 1H), 10.04 (d,
J¼2.5 Hz, 1H); 13C NMR (100 MHz, CDCl3):
¼16.9, 17.9, 18.2, 21.5,
d
31.0, 33.3, 33.5, 39.8, 40.2, 41.4, 41.6, 55.1, 69.3, 71.3, 209.7; HRMS
ꢃ
calculated for C15H26NaO2[MD ] 261.1825, found 261.1823.
4.1.9. (1R,2S,4aS,8aS)-1-(2,2-Dibromo-vinyl)-2,5,5,8a-tetrame-
thyldeca-hydronaphthalen-2-ol (30). Ph3P (1.83 g, 6.96 mmol) was
added to a solution of CBr4 (1.15 g, 3.48 mmol) in CH2Cl2 (25 mL) at
0 ꢁC and the resulting mixture was stirred at 0 ꢁC for 30 min.
A solution of aldehyde 28 (415 mg, 1.74 mmol) in CH2Cl2 (15 mL)
was then added dropwise and the reaction mixture was stirred at
0 ꢁC for 2 h. The mixture was then diluted with Et2O (100 mL),
washed with brine and dried over MgSO4. Evaporation of the or-
ganic solvents under reduced pressure gave a residue, which was
purified by flash chromatography (silica gel, hexanes/EtOAc, 8:1) to
4.10 mmol, 55%) as a white solid. Data for 27: Rf 0.50 (silica gel,
25
hexanes/EtOAc, 1:1); mp 122e125 ꢁC; [
a
]
ꢀ63.4 (c 0.74, CHCl3);
yield dibromide 30 as a white solid (616 mg, 1.56 mmol, 90%). Data
D
23
IR (KBr disc) 3458 (br s), 3330 (br s), 2921 (s), 2863 (s), 2664 (w),
for 30: Rf 0.70 (silica gel, hexanes/EtOAc, 5:1); mp 66e68 ꢁC; [
a]
D
1461 (m) cmꢀ1; 1H NMR (400 MHz, CDCl3):
d
¼0.86 (s, 3H), 0.91 (s,
þ39.5 (c 1.32, CHCl3); IR (KBr disc) 3614 (w br), 2909 (s), 2846 (s),
1603 (m), 1456 (m), 1388 (m) cmꢀ1 1H NMR (400 MHz, CDCl3):
3H), 0.95 (s, 3H), 0.96e1.22 (m, 5H), 1.34e1.39 (m, 1H), 1.41 (s,
3H), 1.42e1.65 (m, 6H), 1.80e1.92 (m, 2H), 4.20 (t, J¼4.5 Hz, 1H),
;
d
¼0.86 (s, 3H), 0.90 (s, 3H), 0.91e0.94 (m, 1H), 1.08 (s, 3H), 1.14 (s,
5.26 (d, J¼5.0 Hz, 1H); 13C NMR (100 MHz, CDCl3):
d
17.0, 18.1,
3H),1.16e1.23 (m, 1H),1.36e1.45 (m, 2H), 1.47e1.66 (m, 6H), 1.80 (d,
J¼7.5 Hz, 1H), 1.99 (d, J¼11.0 Hz, 1H), 6.62 (d, J¼11.0 Hz, 1H); 13C
18.4, 21.9, 32.5, 33.2, 33.4, 34.0, 35.6, 42.1, 42.9, 48.8, 64.3, 74.5,
ꢃ
82.6, 96.1; HRMS calculated for C16H28NaO3[Mþ ] 291.1931, found
NMR (100 MHz, CDCl3):
d
¼16.0, 18.1, 18.2, 21.8, 30.8, 33.3, 33.5, 39.7,
291.1923.
41.9, 42.2, 55.3, 62.9, 72.4, 77.2, 89.4, 137.6; HRMS calculated for
ꢃ
Further elution gave 26 as a white solid (568 mg, 2.10 mmol,
28%). Data for 26: Rf 0.25 (silica gel, hexanes/EtOAc, 1:1); mp
C16H26Br2O [MD ] 394.0331, found 394.0326.
143e146 ꢁC; 1H NMR (400 MHz, acetone-d6):
d
¼0.86 (s, 3H), 0.87
(s, 3H), 1.22 (s, 3H), 1.31 (s, 3H), 3.50 (s, 1H), 3.75e3.87 (m, 3H),
4.21e4.25 (m, 1H); 13C NMR (100 MHz, acetone-d6):
¼17.2, 19.0,
4.1.10. (1R,2S,4aS,8aS)-1-Ethynyl-2,5,5,8a-tetramethyl-decahydro-
naphthalen-2-ol (31). A solution of n-BuLi in hexanes (1.6 M,
7.45 mL, 11.9 mmol) was added to a solution of dibromide 30
d

J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
6329
(1.57 g, 3.98 mmol) in anhydrous THF (25 mL) at ꢀ78 ꢁC and the
resulting mixture was stirred at this temperature for 1 h. The re-
action mixture was then quenched by the addition of saturated
aqueous NH4Cl solution (25 mL). The mixture was then extracted
with Et2O (2ꢄ50 mL), washed with water (50 mL) and brine
(50 mL) and then dried over MgSO4. Filtration and evaporation of
the organic solvents in vacuo gave crude alkyne 31, which was used
without further purification. Data for 31: Rf 0.60 (silica gel, hexanes/
olefinic peaks seen; 13C NMR (100 MHz, CDCl3):
d
¼142.3, 125.0;
ꢃ
HRMS calculated for C16H28[MD ] 220.2191, found 220.2192 (of
mixture of 33 and 3). Data for 3: Rf 0.80 (silica gel, hexanes/
EtOAc, 20:1); 1H NMR (CDCl3, 400 MHz):
d
¼5.39 (s, 1H); 13C NMR
(100 MHz, CDCl3):
d
¼13.6, 18.9, 21.9, 22.1, 23.8, 33.0, 33.3, 33.7,
42.2, 57.2, 122.0, 135.6.
4.1.14. Preparation of (1R,2S)-1-ethyl-1,2,5,5-tetra-methyl-1,2,3,4,5,-
6,7,8-octahydro-naphthalene oxide (40). To a solution of 34
EtOAc, 4:1); [
a]
21 þ18.4 (c 0.49, CHCl3); IR (KBr disc) 3519 (s), 3440
D
(m br), 3289 (s), 2924 (s), 2108 (w), 1460 (m), 1385 (m) cmꢀ1
;
1H
(185 mg, 0.84 mmol) in anhydrous CH2Cl2 (10 mL) and MeOH
(1 mL), O3 was bubbled through for 10 min at ꢀ78 ꢁC. After this
time, O2 was passed through the reaction mixture for 2 min until
residual ozone was removed. Me2S (0.50 mL, 6.81 mmol) was then
added at ꢀ78 ꢁC and stirring continued until the reaction mixture
warmed up to room temperature over 3 h. Evaporation of the or-
ganic solvents in vacuo gave a residue, which was purified by flash
chromatography (silica gel, hexanes/EtOAc, 50:1) to yield the ep-
oxide 40 as a colourless liquid (126 mg, 0.53 mmol, 63%). Data for
40: Rf 0.50 (silica gel, hexanes/EtOAc, 20:1); IR (film) 2962 (s), 2920
(s), 2878 (s), 1465 (m), 1383 (m), 1261 (s) cmꢀ1; 1H NMR (500 MHz,
NMR (400 MHz, CDCl3):
d
¼0.80 (dd, J¼12.0, 2.5 Hz,1H), 0.85 (s, 3H),
0.88 (s, 3H), 0.91e0.99 (m, 1H), 1.16 (s, 3H), 1.28 (s, 3H), 1.30e1.44
(m, 3H), 1.45e1.59 (m, 3H), 1.59e1.69 (m, 1H), 1.87 (dq, 13.0, 3.0 Hz,
1H), 1.94 (dt, J¼14.0, 3.0 Hz, 1H), 2.05 (d, J¼2.5 Hz, 1H), 2.26 (d,
J¼2.5 Hz, 1H); 13C NMR (100 MHz, CDCl3)
d 15.7, 18.1, 18.3, 21.6, 31.6,
33.2, 33.3, 37.8, 39.6, 40.5, 42.0, 54.7, 55.1, 71.0, 74.1, 82.7; HRMS
ꢃ
calculated for C16H26O [MD ] 234.1984, found 234.1975.
4.1.11. (1R,2S,4aS,8aS)-1-Ethyl-2,5,5,8a-tetramethyl-decahydro-
naphthalen-2-ol (19). A catalytic amount of 10% palladium on car-
bon (90 mg) was added to a solution of alkyne 31 (895 mg,
3.82 mmol) in anhydrous MeOH (20 mL) at room temperature
under a nitrogen atmosphere. The system was then purged with
hydrogen and left stirring for 24 h under a hydrogen atmosphere.
Careful TLC analysis showed that the starting material had been
consumed and a less-polar product had been formed. The reaction
mixture was then filtered through Celite (washed with MeOH) and
the filtrate was evaporated to give alkane 19 as a white solid
CDCl3):
d
¼0.74 (d, J¼6.5 Hz, 3H), 0.81 (s, 3H), 0.86 (t, J¼7.5 Hz, 3H),
0.89e0.93 (m, 1H), 0.96 (s, 3H), 0.98 (s, 3H), 1.00e1.11 (m, 2H),
1.26e1.37 (m, 2H), 1.37e1.45 (m, 4H), 1.64 (ddd, J¼14.0, 12.0, 4.5 Hz,
1H), 1.68e1.76 (m, 1H), 1.80e1.86 (m, 1H), 1.89 (dt, J¼14.0, 3.5 Hz,
1H); 13C NMR (125 MHz, CDCl3):
d
¼8.9, 16.4, 17.4, 17.8, 24.3, 24.9,
26.4, 26.4, 26.9, 29.5, 32.8, 33.2, 38.0, 39.3, 68.8, 69.9; HRMS cal-
ꢃ
culated for C16H28ONa [MD ] 259.2032, found 259.2035.
(910 mg, 3.82 mmol) in 97% yield for two steps. Data for 19: Rf 0.65
4.1.15. Preparation of (1R,2S)-1-ethyl-1,2,5,5-tetra-methyl-1,2,3,5,6,7-
hexahydro-naphthalene (47) and 5-(2-ethyl-6,6-dimethylcyclohex-1-
en-1-yl)-3-methylpentan-2-one (46). To a stirred solution of 40
(62 mg, 0.26 mmol) in anhydrous CH2Cl2 (3 mL) was added BF3$OEt2
(0.1 mL, 0.81 mmol). After stirring at ꢀ78 ꢁC for 1 h, the reaction
mixture was quenched with H2O and the organic layer was extracted
with CH2Cl2 (30 mL), dried over MgSO4 and filtered. Evaporation of
the solvent gave a pale yellow oil, which was purified by flash
chromatography (silica gel, hexanes/EtOAc, 50:1) to yield 47 (12 mg,
0.06 mmol, 21%). Data for 47: Rf 0.70 (silica gel, hexanes/EtOAc,16:1);
IR (film) 3041 (w), 2964 (s), 2921 (s), 1456 (m), 1379 (m); 1H NMR
25
(silica gel, hexanes/EtOAc, 4:1); mp 49e51 ꢁC; [
a
]
þ6.4 (c 0.56,
D
CHCl3); IR (KBr disc) 3520 (m br), 2926 (s), 2866 (s), 1711 (w), 1458
(m) cmꢀ1; 1H NMR (400 MHz, CDCl3):
d
¼0.71e0.81 (m, 2H), 0.83 (s,
3H), 0.84e0.86 (m, 1H), 0.87 (s, 3H), 0.88e0.92 (m, 1H), 0.94 (s, 3H),
0.98 (t, J¼15.0, 8.0 Hz, 3H), 1.14 (s, 3H), 1.16e1.23 (m, 1H), 1.28e1.46
(m, 4H), 1.47e1.52 (m, 2H), 1.53e1.56 (m, 1H), 1.56e1.64 (m, J¼13.0,
3.5, 3.5 Hz, 1H), 1.68e1.78 (m, 2H); 13C NMR (100 MHz, CDCl3):
d
¼15.0, 17.9, 18.0, 18.2, 18.3, 21.6, 30.5, 33.2, 33.4, 38.9, 39.0, 42.0,
ꢃ
42.1, 55.9, 61.3, 73.2; HRMS calculated for C16H30O[MD ] 238.2297,
found 238.2298.
(CDCl3, 500 MHz):
d
¼0.77 (t, J¼7.5 Hz, 3H), 0.81 (d, J¼7.0 Hz, 3H),
4.1.12. Preparation of (1R,2S)-1-ethyl-1,2,5,5-tetra-methyl-1,2,3,4,5,6,7,8-
octahydro-naphthalene (34). To an ice-cold solution of 14 (53 mg,
0.22 mmol) in CH2Cl2 (3 mL), was added dropwise BF3$OEt2 (0.14 mL,
1.11 mmol). After stirring at 0 ꢁC for 5 h, the organic layer was
extracted with CH2Cl2 (40 mL), dried over MgSO4 and filtered. Evap-
oration of the solvent gave 34 as a pale yellow oil (47 mg, 0.21 mmol,
0.91 (s, 3H), 0.99 (s, 3H), 1.04 (s, 3H),1.21e1.33 (m, 2H),1.36e1.50 (m,
2H), 1.55e1.66 (m, 1H), 1.79 (dt, J¼18.5, 4.5 Hz, 1H), 2.10e2.24 (m,
1H), 2.32e2.40 (m, 2H), 5.39e5.42 (m, 1H), 5.44e5.48 (m, 1H); 13C
NMR (100 MHz, CDCl3):
33.4, 34.7, 36.9, 40.0, 115.8, 120.6, 138.1, 141.0.
d
¼8.1, 16.1, 21.5, 23.3, 27.2, 28.2, 30.7, 31.1,
Further elution gave 46 (34 mg, 0.14 mmol 55%) as a colourless
oil. Data for 46: Rf 0.25 (silica gel, hexanes/EtOAc, 16:1); 1H NMR
(CDCl3, 500 MHz): 0.95 (t, J¼7.5 Hz, 3H), 0.96 (s, 3H), 0.98 (s, 3H),
1.11 (d, J¼8.0 Hz, 3H), 1.37e1.44 (m, 4H), 1.53e1.58 (m, 2H),
1.65e1.76 (m, 2H), 1.85e1.96 (m, 4H), 2.15 (s, 3H), 2.50 (sextet,
J¼8.0 Hz,1H); 13C NMR (100 MHz, CDCl3): 13.0,16.0,19.5, 26.0, 26.4,
28.0, 28.71, 28.73, 29.4, 34.3, 34.9, 39.9, 48.1, 133.1, 136.1, 212.6;
95%). Data for 34: Rf 0.80 (silica gel, hexanes/EtOAc, 20:1); [
a
]
24 ꢀ101.6
D
(c 0.38, CHCl3); IR (film) 2964 (s), 2926 (s), 2877 (s), 1731 (w), 1459 (m),
1
1380 (m) cmꢀ1; H NMR (CDCl3, 400 MHz):
d
¼0.67 (t, J¼7.5 Hz, 3H),
0.80 (s, 3H), 0.83 (d, J¼7.0 Hz, 3H), 0.97 (s, 3H), 0.99 (s, 3H), 1.27e1.66
(m, 9H), 1.72e1.82 (m, 1H), 1.90e2.02 (m, 3H); 13C NMR (100 MHz,
CDCl3):
d
¼8.4, 16.1, 20.0, 21.0, 25.2, 25.7, 27.2, 27.7, 28.3, 29.3, 32.7, 34.5,
ꢃ
ꢃ
40.0, 40.7, 132.5, 136.8; HRMS calculated for C16H28[MD ] 220.2191,
found 220.2195.
HRMS calculated for C16H28ONa [MD ] 259.2032, found 259.2036.
4.1.16. (4aS,5S,8aS)-5-Ethyl-1,1,4a-trimethyl-6-methylene-decahy-
dro-naphthalene (32). To an ice-cold stirred solution of 14 (52 mg,
0.22 mmol) in 2,6-lutidine (2.5 mL) was added dropwise a solution
of MeSO2Cl (0.1 mL, 1.29 mmol). After stirring for 24 h at 0 ꢁC the
reaction mixture was poured onto ice, neutralised with 10% HCl
solution and extracted with Et2O (4ꢄ10 mL). The combined extracts
were then washed with water and brine, dried over MgSO4 and
filtered. Evaporation of the solvent under reduced pressure gave
the olefin 32 as a pale yellow oil (47 mg, 0.21 mmol, 98%): Data for
32: Rf 0.80 (silica gel, hexanes/EtOAc, 20:1); 1H NMR (400 MHz,
4.1.13. Preparation of (4aS,8aS)-8-ethyl-4,4,7,8a-tetra-methyl-1,2,3,-
4,4a,5,6,8a-octahydro-naphthalene (33) and (4aS,5S,8aS)-5-ethyl-
1,1,4a,6-tetramethyl-1,2,3,4,4a,5,8,8a-octahydro-naphthalene (3). To
a solution of 14 (50 mg, 0.21 mmol) in CH2Cl2 (3 mL), was added
dropwise BF3$OEt2 (0.13 mL, 1.05 mmol). After stirring at ꢀ78 ꢁC
for 20 min the reaction was quenched with H2O and the organic
layer was extracted with CH2Cl2 (40 mL), dried over MgSO4 and
filtered. Evaporation of the solvent gave an inseparable mixture
of 33 and 3 as a pale yellow oil, which could not be fully assigned
due to signal overlap in the NMR spectrum. Data for 33: Rf 0.80
(silica gel, hexanes/EtOAc, 20:1); 1H NMR (CDCl3, 400 MHz): no
CDCl3):
d
¼0.68 (s, 3H), 0.81 (s, 3H), 0.88 (s, 6H), 0.93e1.80 (m, 12H),
1.91e2.10 (m, 1H), 2.40 (ddd, J¼13.0, 4.5, 2.5 Hz, 1H), 4.51 (d,

6330
J.H. George et al. / Tetrahedron 66 (2010) 6321e6330
J¼1.5 Hz, 1H), 4.83 (d, J¼1.5 Hz, 1H); 13C NMR (100 MHz, CDCl3):
5. Sato, A.; Kurabayashi, M.; Nagahori, H.; Ogiso, A.; Mishima, H. Tetrahedron Lett.
1970, 11, 1095.
6. Teresa, J. P.; Urones, J. G.; Marcos, I. S.; Bermejo, F.; Basabe, P. Phytochemistry
d
¼13.5, 14.4, 19.4, 21.7, 24.5, 33.6, 33.6, 38.4, 39.1, 39.7, 42.2, 42.3,
ꢃ
55.5, 59.1, 106.1, 148.4 ppm; HRMS calculated for C16H28[MD
220.2191, found 220.2194.
]
1983, 22, 2273.
7. Hadley, M. S.; Halsall, T. G. J. Chem. Soc., Perkin Trans. 1 1974, 1334.
8. Kuchkova, K. I.; Chumakov, Y. M.; Simonov, Y. A.; Bocelli, G.; Panasenko, A. A.;
Vlad, P. F. Synthesis 1997, 1045.
9. Quideau, S.; Lebon, M.; Lamidey, A.-M. Org. Lett. 2002, 4, 3975.
10. (a) Corey, E. J.; Usprong, J. J. J. Am. Chem. Soc. 1956, 78, 5041; (b) Johns, W. F. J.
Org. Chem. 1961, 26, 4583.
11. Fujiwara, N.; Kinoshita, M.; Akita, H. Tetrahedron: Asymmetry 2006, 17, 3037.
12. Matsubara, S.; Sugihara, M.; Utimoto, K. Synlett 1998, 313.
13. (a) Takai, K.; Hotta, Y.; Oshima, K.; Utimoto, K. Tetrahedron Lett. 1978, 19, 2417;
(b) Lombardo, L. Tetrahedron Lett. 1982, 23, 4293.
14. Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 36, 3769.
15. Vedejs, E.; Larsen, S. Org. Synth. Coll. Vol. 1990, 7, 277; Org. Synth. 1986, 64,
127.
Acknowledgements
We thank Dr. Barbara Odell for assistance with NMR studies.
References and notes
1. Merritt, A. T.; Ley, S. V. Nat. Prod. Rep. 1992, 9, 243.
2. Capon, R. J.; Rochfort, S. J.; Ovenden, S. P. B. J. Nat. Prod. 1997, 60, 1261.
3. Comprehensive Natural Products Chemistry; Cane, D. E., Ed.; Pergamon: 1999;
Vol. 2, pp 225e227; and references therein.
4. Akhila, A.; Rani, K.; Thakur, R. S. Phytochemistry 1991, 30, 2573.
16. Hochstetler, A. R. J. Org. Chem. 1975, 40, 1536.
17. Eguchi, T.; Dekishima, Y.; Hamano, Y.; Dairi, T.; Seto, H.; Kakinuma, K. J. Org.
Chem. 2003, 68, 5433.