Synthetic studies towards garsubellin A: Synthesis of model systems and potential mimics by regioselective lithiation of bicyclo[3.3.1]nonane-2,4,9- trione derivatives from catechinic acid

Article abstract of DOI:10.1039/b704311b

Bridgehead lithiations have successfully been carried out on substrates derived from catechinic acid, which possess the core bicyclo[3.3.1]nonane-1,3,5- trione structure present in garsubellin A. Using an external quench method, various electrophiles have been incorporated at the C-5 bridgehead position in a one-step process that appears to be sensitive to the substitution pattern on the bicyclic system. Regioselective lithiation at the C-3 sp² centre was achieved by changing the base used from LDA to LTMP. Following the introduction of a prenyl substituent by bridgehead substitution, annulation of a THF ring, analogous to that in garsubellin A, was possible via an epoxidation-ring opening sequence. Oxidative modification of the catechol substituent of the catechinic acid core was possible to give systems with muconic acid, ortho-quinone or furan 2-carboxylic acid side chains. The Royal Society of Chemistry.

Full text of DOI:10.1039/b704311b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthetic studies towards garsubellin A: synthesis of model systems and
potential mimics by regioselective lithiation of bicyclo[3.3.1]nonane-
2,4,9-trione derivatives from catechinic acid†
Nadia M. Ahmad,a Vincent Rodeschini,a Nigel S. Simpkins,*a Simon E. Wardb and Claire Wilsona
Received 21st March 2007, Accepted 25th April 2007
First published as an Advance Article on the web 15th May 2007
DOI: 10.1039/b704311b
Bridgehead lithiations have successfully been carried out on substrates derived from catechinic acid,
which possess the core bicyclo[3.3.1]nonane-1,3,5-trione structure present in garsubellin A. Using an
external quench method, various electrophiles have been incorporated at the C-5 bridgehead position in
a one-step process that appears to be sensitive to the substitution pattern on the bicyclic system.
Regioselective lithiation at the C-3 sp2 centre was achieved by changing the base used from LDA to
LTMP. Following the introduction of a prenyl substituent by bridgehead substitution, annulation of a
THF ring, analogous to that in garsubellin A, was possible via an epoxidation–ring opening sequence.
Oxidative modification of the catechol substituent of the catechinic acid core was possible to give
systems with muconic acid, ortho-quinone or furan 2-carboxylic acid side chains.
Introduction
The natural products known as polyprenylated acylphloroglu-
cinols (PPAPs) have emerged as an important class of com-
pounds due to their broad range of biological activities.1 These
compounds, isolated from various plants and trees from the
family Clusiaceae, are characterised by the presence of a common
bicyclo[3.3.1]nonane-1,3,5-trione core, decorated with prenyl (or
geranyl) and acyl groups. Important examples include garsubellin
A, hyperforin and clusianone, Fig. 1.
These challenging structures, together with promising thera-
peutic potential have made these molecules attractive targets for
total synthesis.2 As a result, a number of research groups have
described synthetic approaches to these compounds, and recently
garsubellin A (1) has succumbed to total synthesis by the groups
of Shibasaki, and Danishefsky.3,4
We recently described the first total synthesis of clusianone
(3),5 and also a formal synthesis of garsubellin A.6 This work,
inspired by a previous approach of Spessard and Stoltz,7 employed
a malonyl dichloride (5) annulation (Effenberger cyclisation)8 of
a cyclohexanone-derived enol ether 4 to set up the [3.3.1] trione
core structure 6, Scheme 1. Subsequent elaboration to the target
Fig. 1 Representative PPAP natural products.
natural product was then possible by regioselective metallation–
substitution reactions at both the bridgehead C-5 and C-3 sp2
hybridised positions.
at least in our hands, rarely provides more than 25–30% yields.9
Although this approach enabled a rapid access to clusianone,
Secondly, and in line with most other synthetic work in this area,
and has the potential to deliver other PPAPs, it has a number
the method has provided only racemic intermediates to date.10
of deficiencies. First and foremost is the consistently modest
In parallel with our synthetic investigations summarised above,
efficiency of the type of annulation shown in Scheme 1 which,
we also surveyed the literature for alternative promising entries to
enantiopure starting materials for PPAP synthesis. We identified
a little-known, and remarkable, rearrangement of the naturally
occurring, readily available, flavanoid (+)-catechin (7) into the
bicyclo[3.3.1]nonane-1,3,5-trione derivative catechinic acid (8)
under alkaline conditions (Scheme 2).11
aSchool of Chemistry, The University of Nottingham, University Park,
Nottingham, NG7 2RD, UK. E-mail: nigel.simpkins@nottingham.ac.uk
bMedicinal Chemistry, Psychiatry Centre of Excellence for Drug Discovery,
GlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow,
Essex, CM19 5AW, UK
† Electronic supplementary information (ESI) available: Experimental
procedures and/or full spectroscopic data for compounds 19–26, 32–42,
44, 46–50. See DOI: 10.1039/b704311b
Reported over thirty years ago by Sears et al. this highly
stereoselective transformation has not attracted much interest
1924 | Org. Biomol. Chem., 2007, 5, 1924–1934
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Scheme 1 Access to the [3.3.1] core via Effenberger annulation.
Scheme 2 Preparation and protection of catechinic acid.
since. For our purpose, catechinic acid (8) constituted a promising,
Results and discussion
enantiopure entry into the bicyclic core of PPAP type structures,
and might even constitute a chiral pool starting material for
natural products such as garsubellin A.
(i) Regioselective bridgehead substitution
As a prerequisite to a synthetic study towards 1 using 8 as a tem-
plate, two key issues needed to be addressed: (i) the introduction
of appropriate substituents at both bridgehead positions C-1 and
C-5, as well as at position C-3; (ii) the conversion of the catechol
ring at position C-8 of 8 into the gem-dimethyl group present in 1.
The solution to the substitution problems at C-1, C-3 and
C-5 were explored in parallel to our studies using the Effenberger
derived materials (Scheme 1). In this regard the catechinic acid
derived bicyclic systems could function as useful, and readily
available, model systems to probe the scope and regiocontrol of
substitution chemistry. Clearly, if the catechinic acid system was to
provide access to the natural products themselves then degradation
of the C-8 catechol unit was a key issue. In this regard the oxidative
degradation of an aromatic ring into a carboxylic acid derivative is
a well known transformation.12 In turn, a C-8 carboxylic function
would allow introduction of the desired gem-dimethyl through a
methylation–reduction sequence. In order to address these issues,
a systematic study was initiated on suitably protected derivatives
of 8.
In this paper we describe in full detail our results concerning
the bridgehead substitution of catechinic acid derivatives through
the selective formation of bridgehead enolates. Also reported is
the selective introduction of various substituents at the C-3 sp2
position of 8. Additionally, the formation of the THF ring present
in 1 is reported starting with a prenylated derivative, involving
an epoxidation–ring opening sequence. Finally, some interesting
results obtained by the oxidative degradation of the aromatic
moiety of intermediates derived from 8 are described.
Our study began with the protection of the acidic residues present
in catechinic acid (8). Following literature precedent, methylation
under basic conditions afforded regioisomers 9 and 10 in 58%
overall yield from 7 (Scheme 2). The mixture of secondary
alcohols was then treated with TIPSOTf and lutidine to afford
the corresponding silylated vinylogous ester derivatives 11 and
12 in 75% yield (3 : 1 ratio), which were separated by column
chromatography.
With suitable substrates in hand, initial studies of bridgehead
substitution reactions were undertaken. Initial attempts to de-
protonate major regioisomer 11 using excess LDA or LTMP (1
to 5 eq.) in THF at −78 ◦C in the presence of Me3SiCl (in situ
quench) failed to give any silylated product and starting material
was recovered, Scheme 3.
Although previously this in situ quench technique had proved
highly effective for bridgehead silylation of various substrates,13
it proved singularly unsuccessful in this case, starting with either
regioisomer 11 or 12. Similar lack of product formation was also
observed when more conventional external quenching reactions
were attempted, using similar excess of base at low temperature
for up to 3 hours, followed by addition of D2O, or reactive alkyl
halides. After some experimentation under these external quench
conditions, we found that increasing the amount of LDA to 10
equivalents allowed the bridgehead deprotonation of 11 to occur,
and after a quench with prenyl bromide, derivative 13 was isolated
in 42% yield, Scheme 3.
Despite this preliminary success, the modest yield of 13 ob-
tained, together with the fact that the reaction was less effective
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1924–1934 | 1925
©

View Article Online
Scheme 3 Initial metallation attempts.
Fig. 2 X-Ray structure of 16 (ellipsoids are shown at 50% probability).
with a number of other electrophiles, prompted us to modify
the structures of our substrates in order to make bridgehead
lithiation more facile. We were interested to probe the effect
of modifying the bridging ketone, especially since reduction or
protection could generate slightly less rigid structures that might
better accommodate the ‘bridgehead enolate’ character of the
reactive intermediate. The first modification tested was to convert
the bridgehead ketone into the corresponding dimethylketal. Thus,
under acidic conditions, ketones 9 and 10 were converted into
intermediate ketals, and the secondary alcohols were then directly
silylated as previously described to afford the regioisomeric ketals
16 and 17 (Scheme 4).
conditions, various C-5 derivatives 18–26 were furnished in
moderate to good yield (Table 1).
Table 1 Bridgehead alkylation of ketal 16
We were able to clarify any regiochemical ambiguity as X-ray
crystallography confirmed the structure of the major isomer 16
(Fig. 2).
Electrophile
R
Yield
Compound
Prenyl bromide
Me3SiCl
Allyl Br
BnBr
Prenyl
Me3Si
Allyl
Bn
Me
PhCH(OH)
iPrCH(OH)
PhS
75 (89)a
46 (59)a
46
18
19
20
21
22
23
24
25
26
Further studies focused on the lithiation of the major regioiso-
meric ketal 16. As observed with ketone 11, the use of an excess
of LDA or LTMP under in situ quench conditions with various
electrophiles failed to give any bridgehead substituted products
when using ketal 16. In contrast, the use of 5 equivalents of
LDA–LiCl under external quench conditions (Scheme 3), enabled
efficient bridgehead deprotonation–substitution of 16 to afford
prenylated product 18 in a very acceptable 75% yield. The use
of external quench conditions opened up the opportunity to
use various electrophiles, and pleasingly, under our optimised
56
MeI
48 (55)a
51 (1 : 1)b
31 (42)a
67
PhCHO
iPrCHO
Ph2S2
nBu3SnCl
Bu3Sn
63
a Yield in parentheses is based on recovered starting material.
b Diastereomeric ratio.
Scheme 4 Synthesis of [3.3.1] systems with bridging ketal functions.
1926 | Org. Biomol. Chem., 2007, 5, 1924–1934 This journal is The Royal Society of Chemistry 2007
©

View Article Online
The origin of the requirement for a large excess of base is
unclear but lower yields of the products were obtained when
fewer equivalents of base were employed. We saw no evidence
for substitution at the alternative C-1 bridgehead position, and
fortuitously for our planned natural product work quenching
with prenyl bromide proved to be most effective. Heteroatom
substituents such as Si, Sn or S groups could also be efficiently
introduced at the bridgehead position, although attempted iodine
quenching led to intractable mixtures, probably involving C-3
substitution. When benzaldehyde was used as electrophile, the
aldol product 23 was isolated as an equimolar mixture of diastereo-
isomers, whereas the use of isobutyraldehyde resulted in a single
isomer 24, albeit in modest yield.
Surprisingly, reaction of 16 with the acylating agent EtOCOCN
gave product 27, in which substitution had occurred at both the
C-3 and C-5 positions, Scheme 5. This behaviour appears unlikely
to be due to dianion formation (little or no disubstituted product
was observed in most other cases) and could be the result of
activation of the system at C-3 following initial acylation at the
bridgehead.
Scheme 6 Synthesis and lithiation of reduced derivative 29.
sterically encumberedLTMP, ketal16 failed togive any bridgehead
substitution product when the lithiated intermediate was quenched
with prenyl (or allyl) bromide. In contrast, the use of Me3SiCl as
electrophile gave a new product in a modest yield, which proved
to be the vinyl silane 31. Thus by using a bulkier lithium amide,
the site of deprotonation on ketal 16 could be altered. This trend
was further exemplified by reaction of the lithiated species with
different electrophiles (Table 2).
Scheme 5 Double acylation of 16.
Disappointingly, allyl or prenyl substituents could not be
introduced at this position by this method (we later developed
the use of cuprates to effect this type of substitution6). The
use of iodine or diphenyl disulfide as the electrophile led to
complex mixtures. In the case of benzoyl chloride, the benzoylated
derivative 33 was isolated in 50% yield. Increasing the amount
In order to further probe the metallation chemistry of these
systems we also prepared a catechinic acid derivative in which
the bridging ketone was converted into a protected alcohol. Thus,
diastereoselective reduction of ketone 11, using NaBH4, provided
alcohol 28, which was then protected as the corresponding TES
ether (Scheme 6).
In this case, our standard metallation–quenching conditions
furnished the prenylated product 30 in a very modest 26% yield.
This result contradicts our early hypothesis that an sp3-hybridised
bridging atom might prove superior to a more rigid bridging
ketone. The bridgehead substitution is clearly viable in each of
11, 16 and 29, but the overall efficiency of the process is difficult to
predict. It is possible that chelation with the ketal oxygen atoms
plays a part in facilitating lithiation in the case of ketal derivative
16, whereas steric effects may have the opposite effect in the bulky
OTES compound 29.
Table 2 C-3 substitution of ketal 16
(ii) Regioselective sp2 substitution
Electrophile
R
Yield
Compound
Me3SiCl
MeI
Me3Si
Me
PhCO
Bu3Sn
Allyl
45
36
50
61
—
—
31
32
33
34
—
—
So far, our successful results relied on the use of LDA, and
only bridgehead substitution was observed, with no isolable
quantities of side products from substitution at either C-1 or
C-3. It was interesting to study whether the nature of the base
could allow us to modify this selectivity, and based on literature
precedent it appeared that substitution of the vinylogous ester
at C-3 should be possible.14 When changing LDA for the more
PhCOCl
nBu3SnCl
Allyl bromide
Prenyl bromide
a
a
Prenyl
a No reaction, starting material was recovered.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1924–1934 | 1927
©

View Article Online
of LTMP to 10 equivalents resulted in the isolation of the di-
substituted product 35 in 55% yield. As with the previously
described acylation leading to 27 this last result could be due
to sequential substitution or may indicate the intermediacy of a
dianion. However, we were unable to provide further evidence that
the reactive species may be a dianion in this system (for example
quenching with D2O did not give double incorporation).
It is likely that the steric bulk of the aromatic residue at C-8
hinders deprotonation at C-1 in this case, although it should
be noted that Danishefsky accomplished metallation at this
position during the aforementioned garsubellin A synthesis.4
Using TMSCl, MeI or prenyl bromide as electrophiles, derivatives
36–38 were isolated in moderate to good yields. It is worth noting
that changing the base to LTMP did not change the selectivity,
but allowed us to improve the yield of prenylation at the vinylic
position. Having established that the selective introduction of
substituents at the C-5 bridgehead and C-3 sp2 positions was
possible through the use of an appropriate base, we decided
to further advance key intermediates by constructing the five-
membered THF ring present in 1.
(iii) Synthesis of the THF ring
In contrast with the methodologies used previously by the groups
of Danishefsky and Shibasaki during their total synthesis of
1, we wanted to form the THF ring through the epoxidation
of the C-5 bridgehead prenyl residue, followed by a 5-exo-tet
cyclisation involving the C-4 oxygen atom. This cyclisation would
be accompanied by the cleavage of the vinylogous ester. Our first
attempts to form the epoxide of 18 using mCPBA resulted in
very low mass recovery. In contrast, the use of freshly prepared
dimethyldioxirane (DMDO) allowed the epoxidation to proceed
smoothly, affording a diastereomeric mixture of two epoxides.
Due to instability of the epoxides on silica, we explored methods
to directly cyclise the crude diastereomeric mixture, and we
established that reaction with TMSCl effected their conversion
into the separable THF derivatives 39 and 40 (Scheme 7).15
Trimethylsilyl chloride was found to be much more effective
than the corresponding iodide for this cyclisation, which we
assume proceeds via silicon-mediated activation of the epoxide.
Surprisingly, however, we have not been able to directly isolate the
silicon protected alcohol derivatives from this reaction.
Thus far, various substituents had been selectively introduced
at positions C-3 and C-5 by altering the base employed. However,
it was also of interest to introduce substituents at bridgehead
position C-1, as most natural PPAPs have substituents at both
their bridgehead positions. We thought that using the minor
regioisomeric ketal 17 might allow lithiation to occur at position
C-1 by analogy with previous results. Thus, isomer 17 was sub-
jected to our optimised conditions. In this case, bridgehead
substitution was not observed, only vinylic substitution (Table 3).
Table 3 Lithiation-substitution of regioisomeric ketal 17
The relative stereochemistry of the major diastereoisomer 40
was determined through NOE studies and proved to have the
unnatural side chain configuration. The alcohols 39 and 40 were
readily converted into the corresponding silyl ethers 41 and 42. We
decided to use the major silyl ether 42 to probe the feasibility of
the direct installation of a prenyl residue at the C-3 sp2 position, a
result that had eluded us with compound 16. Once again we were
unable to effect C-3 allylation or prenylation using either LDA or
LTMP under the conditions described previously, Scheme 8.
Electrophile
R
Base
Yield
Compound
Me3SiCl
MeI
Prenyl bromide
Prenyl bromide
Me3Si
Me
Prenyl
Prenyl
LDA–LiCl
LDA–LiCl
LDA–LiCl
LTMP
31 (34)a
60
36
37
38
38
25 (33)a
51
a Yields in parentheses based on recovered starting material.
Scheme 7 Side chain epoxidation and cyclisation.
1928 | Org. Biomol. Chem., 2007, 5, 1924–1934 This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Catechol derivatives 46 and 47 were prepared from 8 as depicted
in Scheme 10.
Scheme 8 Attempted C-3 prenylation of 42.
Scheme 10 Synthesis of systems with a free catechol unit.
The facile synthesis of vinyl bromide 44 enabled us to try
the same type of substitution, originating with a halogen–metal
transfer process, but again no product was obtained. The problem
here is certainly one of poor reactivity of the intermediate
organolithium and, as mentioned earlier, we eventually solved this
problem by allylation of a mixed cuprate generated using thienyl
copper cyanide.6 This chemistry was not explored on the catechinic
acid systems.
Initial silylation of the secondary alcohol as well as the two
phenolic positions of 8 led, after methylation of the 1,3 diketone
moiety to an inseparable mixture of tris-silylated derivatives. Selec-
tive deprotection of the phenolic hydroxyls was easily performed
under acidic conditions to provide catechols 46 and 47 which were
then separated by chromatography.
As seen for 11, the use of RuO4 based oxidation led to
the rapid decomposition of the starting material. However, the
presence of a free catechol function opened up the possibility
of a more step-wise strategy, for example via muconic acid or
quinone intermediates. To explore the first of these we followed
the procedure of Gilheany,20 which involved reaction of catechol46
with Pb(OAc)4 in MeOH, and were pleased to isolate the muconic
acid diester derivative 48 in very high yield (Scheme 11).
(iv) Oxidative degradation of the catechol ring
As shown before, in order to set up a synthesis of 1, the aromatic
moiety at position C-8 needed to be converted into the gem-
dimethyl group present in 1. The use of a ruthenium based
oxidation on a suitably protected derivative of catechinic acid 8
was attempted to degrade the aromatic ring to a carboxylic acid
(11 → 45, Scheme 9).
Scheme 11 Oxidative cleavage of 46.
Scheme 9 Attempted oxidative cleavage of 11.
With this new intermediate, we were hoping to be able to reach
an acid derivative similar to 45 by the simultaneous oxidative
cleavage of the two conjugated olefins in 48. However, 48 proved
to be completely inert towards dihydroxylation. In the case of
ozonolysis, the substrate reacted with the initial cleavage of the
disubstituted olefin but no products resulting from the cleavage
of both C–C double bonds of the muconic acid moiety could be
isolated.
In our case, however, a variety of conditions failed to provide any
quantities of acid 45. Use of Sharpless conditions rapidly led to
decomposition of the starting material.16 Alternative conditions
for the in situ formation of the active species RuO4 were tried,
including the use of RuO2 or the complex cis[Ru(bpy)]2Cl2 with
NaIO4 but in each case only decomposition was observed.17,18
Ozonolysis of 11 followed by an acidic work up also failed to
provide derivative 45.19
Another potential route for the degradation of the aromatic
part was explored using the minor regioisomer 47, as depicted in
Scheme 12.
At this stage, a system bearing a free catechol unit was prepared
as a means to make the aromatic moiety more easily oxidisable.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1924–1934 | 1929
©

View Article Online
of these new compounds have been tested for their biological
properties; results will be reported in due course.
Experimental
General methods
All reactions were performed under an atmosphere of nitrogen in
flame dried glassware unless otherwise stated. Organic solvents
and reagents were dried by distillation from the following as
required: THF (sodium–benzophenone), DCM, TMSCl (CaH2).
Allyl bromide, prenyl bromide, benzaldehyde, benzyl bromide,
ethyl cyanoformate, iodomethane, and triethylamine were all
distilled before use. Lead tetraacetate was freshly recrystallised
before use from AcOH–Ac2O. All other solvents and reagents
were used as received from commercial suppliers unless otherwise
stated. RT relates to the temperature range 20–25 ◦C. Reaction
progress was monitored by thin layer chromatography (TLC)
Scheme 12 Formation of side chain quinone 49 and furan 50.
performed on Merck aluminium plates coated with kieselgel F254
.
Visualisation was achieved by a combination of ultraviolet light
(254 nm) and anisaldehyde or acidic potassium permanganate.
Flash chromatography was performed using silica gel (Merck
7734 grade), eluted with the indicated solvent. Melting points
were recorded on a Stuart Scientific SMP3 apparatus and are
uncorrected. Infrared spectra were recorded on a Perkin-Elmer
1600 FT-IR spectrometer as dilute solutions in spectroscopic grade
chloroform within a NaCl cell. NMR spectra were recorded on
a Bruker AV400 or DRX500 machine, using CDCl3 as solvent
at 298 K. Chemical shifts are given in ppm downfield from
tetramethylsilane, using residual protic solvent as an internal
standard. J values are reported in Hz and rounded to the nearest
0.1 Hz. Where required, assignments were confirmed by two-
dimensional homonuclear (1H–1H) and heteronuclear (1H–13C)
correlation spectroscopy on a Brucker AV400 spectrometer. Mass
spectra were obtained using a VG Micromass 70E spectrometer,
using electron impact (EI) at 70 eV or chemical ionisation (CI) or
electrospray ionisation (ESI). Optical rotations were recorded as
dilute solutions in the indicated solvent in a 100 mm cell using a
JASCO DIP370 digital polarimeter at a wavelength of 598 nm.
Oxidation of 47 using Ag2CO3 on Celite (Fetizon’s reagent)21
resulted in an extremely clean conversion to the corresponding
ortho-quinone 49. Then, a one pot dihydroxylation–diol cleavage
sequence was attempted, using a mixture of OsO4 and NaIO4.22
To our surprise, under these conditions, 49 was converted into
a furan carboxylic acid derivative, which was isolated as the
corresponding methyl ester 50 after methylation under basic
conditions. On further checking the literature we established a
reasonable precedent for this transformation in the work of Gierer
and Imsgard.23
Further attempts to cleave the furan ring using ozonolysis failed
to give any usable products.24
Conclusions
In conclusion, the regioselective lithiation of derivatives of cate-
chinic acid (8), possessing the bicyclo[3.3.1]nonane-1,3,5-trione
core has been achieved. In the case of bridging ketal 16, the
use of LDA promoted selective bridgehead substitution, whilst
LTMP enabled metallation–substitution at the C-3 sp2 centre.
With reactive electrophiles, such as acylating agents, we observed
a tendency towards double substitution at both the C-3 and C-5
positions.
With regioisomer 17, the use of either LDA or LTMP led to
lithiation at the sp2 center. These lithiations have been shown to be
dependent on the nature of substitution at the bridging atom with
the bridging dimethylketal proving superior to either a bridging
ketone or a reduced and silicon protected variant.
A THF ring, resembling that present in garsubellin A, was
appended to the catechinic acid system through an epoxidation–
ring opening sequence. Oxidative degradation of the aryl group
of our catechinic acid derivatives to give a carboxylic acid
derivative did not succeed as predicted, although an interesting
transformation of an ortho-quinone intermediate into a furan
derivative was achieved. Although we have not yet been able to
use catechinic acid to achieve an asymmetric synthesis of 1, this
template has provided us with a very rapid access to a wide range
of enantiomerically pure bicyclo[3.3.1]nonane derivatives. Some
(1S,5R,7S,8R)-(+)-8-(3,4-Dimethoxyphenyl)-7-hydroxy-2-
methoxybicyclo[3.3.1]non-2-ene-4,9-one 9 and (1S,5R,7S,8R)-
(+)-8-(3,4-dimethoxyphenyl)-7-hydroxy-4-methoxy-
bicyclo[3.3.1]non-3-ene-2,9-one 10
Solid KOH (600 mg, 6.90 mmol) was added to a solution of
(+)-catechin (7) (2.00 g, 25.0 mmol) in water (80 ml) and the
reaction mixture was then heated under reflux for 105 min. The
reaction mixture was allowed to cool, filtered through a pad
of Amberlite IR-120 and concentrated in vacuo to give a dark
red solid (2.0 g, quantitative). Crude catechinic acid (8) was
used without purification. Dimethyl sulfate (25.0 mL, 36.9 mmol,
7.7 eq.) and K2CO3 (15.7 g, 15.8 mmol, 3.3 eq.) were added
to crude 8 (10.0 g, 4.8 mmol) in acetone (500 mL), and the
solution was heated under reflux for 18 h. The reaction mixture was
allowed to cool, diluted with H2O (200 mL), and extracted with
EtOAc (4 × 100 mL). The organic extract was dried (MgSO4)
and concentrated in vacuo to give the crude product as a fluffy
orange solid. Purification by column chromatography (petroleum
1930 | Org. Biomol. Chem., 2007, 5, 1924–1934
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
ether–EtOAc 3 : 7) gave regioisomers 9 and 10 in a 3 : 1 ratio
(6.4 g, 58% over two steps) data for 9; [a]2D7 +147 (c 1.0 in CHCl3);
mmax (CHCl3) 3463 (br), 2906, 1652, 1603, 1516, 1452, 1379, 1230,
1105, 1068 cm−1; 1H NMR (400 MHz, CDCl3) d 1.89–1.97 (m, 1H,
CHH), 2.23 (br s, 1H, OH), 2.57 (ddd, J 13.0, 8.8, 5.3, 1H, CHH),
3.08 (dd, J 11.0, 3.9, 1H, ArCH), 3.29 (m, 2H, 1-CH, 5-CH),
3.57 (s, 3H, OCH3), 3.85 (s, 3H, OCH3), 3.86 (s, 3H, OCH3), 4.40
4.1, 1H, ArCH), 3.24–3.26 (m, 1H, 5-CH), 3.34–3.37 (m, 1H, 1-
CH), 3.83 (s, 9H, OCH3, 2 × ArOCH3), 4.54 (ddd, J 10.5, 10.5,
=
5.2, 1H, CHOSi(CH(CH3)2)3), 5.80 (s, 1H, C CH), 6.60–6.65 (m,
2H, 2 × ArH), 6.78 (d, J 8.0, 1H, ArH); 13C NMR (125 MHz,
CDCl3) d 12.5 (CH), 17.5 (CH3), 37.1 (CH2), 51.6 (CH), 55.8
(CH3), 55.9 (CH3), 56.2 (CH3), 57.0 (CH) 67.8 (CH), 68.3 (CH),
105.1 (CH), 110.9 (CH), 112.5 (CH), 120.9 (CH), 130.5 (C), 148.5
(C), 148.6 (C), 175.4 (C), 192.5 (C), 204.0 (C); HRMS (CI, NH3)
m/z 488.2585 [M]+, [C27H40O6Si]+ requires 488.2594.
=
(ddd, J 11.0, 11.0, 5.3, 1H, CHOH), 5.70 (s, 1H, C CH), 6.66 (d,
J 2.0, 1H, ArH), 6.69 (dd, J 8.2, 2.0, 1H, ArH), 6.84 (d, J 8.2,
1H, ArH); 13C NMR (100 MHz, CDCl3) d 37.2 (CH2), 53.2 (CH),
56.5 (CH3), 56.5 (CH3), 57.1 (CH3), 58.9 (CH), 59.4 (CH), 66.1
(CH), 105.1 (CH), 111.0 (CH), 111.3 (CH), 119.8 (CH), 129.7 (C),
148.7 (C), 149.2 (C), 174.5 (C), 194.9 (C), 203.9 (C); HRMS (EI)
m/z 333.1362 [M + H]+, [C18H21O6]+ requires 333.1338; 10: [a]D20
+67 (c 0.5 in CHCl3); mmax (CHCl3) 2936, 1736, 1657, 1602, 1462,
(1S,5R,7S,8R)-(+)-8-(3,4-Dimethoxyphenyl)-2-methoxy-1-
(prenyl)-7-triisopropylsilanyloxy-bicyclo[3.3.1]non-2-ene-4,
9-dione 13
A solution of LDA·LiCl was prepared by treatment of a suspen-
sion of DIPA·HCl (256 mg, 2.04 mmol) in THF (4 mL) at −78 ◦C
1
1365, 1142, 908, cm−1; H NMR (400 MHz, CDCl3) d 1.96–2.05
n
(m, 1H, CHH), 2.63 (ddd, J 13.4, 8.6, 5.3, 1H, CHH), 3.09 (dd,
J 10.8, 4.0, 1H, ArCH), 3.33–3.40 (m, 2H, 1-CH, 5-CH), 3.86–
3.89 (m, 9H, OCH3, 2 × ArOCH3), 4.46 (ddd, J 10.8, 10.8, 5.3,
with BuLi (1.6 M solution in hexanes; 2.56 mL, 4.08 mmol).
The solution was allowed to warm to RT and after 10 min re-
◦
cooled to −78 C. This LDA·LiCl solution was added dropwise
=
1H, CHOH), 5.79 (s, 1H, C CH), 6.67–6.72 (m, 2H, 2 × ArH),
via syringe to a solution of bridged ketone 11 (100 mg, 0.20 mmol)
in THF (1 mL) at −78 ◦C resulting in a deep yellow-coloured
solution. The solution was stirred at −78 ◦C for 3 h, followed
by addition of prenyl bromide (0.25 mL, 2.04 mmol, 10 eq.) and
stirring for a further 3 h at −78 ◦C. The reaction mixture was
quenched after this period with H2O (5 mL) followed by extraction
with EtOAc (3 × 5 mL). The organic layers were combined and
washed with saturated aqueous NaCl (5 mL), dried (MgSO4), and
concentrated in vacuo. Purification by column chromatography
(petroleum ether–EtOAc 4 : 1) gave the title compound 13 as a
yellow oil (47 mg, 42%); [a]2D0 +82 (c 1.0 in CHCl3); mmax (CHCl3)
2940, 2258, 1736, 1650, 1095 cm−1; 1H NMR (400 MHz, CDCl3)
d 0.82 (m, 21H, OSi(CH(CH3)2)3), 1.64 (s, 3H, CH3), 1.69 (s, 3H,
OCH3), 1.74–1.78 (m, 1H, 8-CHH), 2.32 (dd, J 12.7, 5.3, 1H, 8-
6.84–6.87 (m, 1H, ArH); 13C NMR (100 MHz, CDCl3) d 35.3
(CH2), 51.6 (CH), 53.3 (CH), 55.8 (CH3), 55.9 (CH3), 56.5 (CH3),
57.1 (CH3), 66.5 (CH), 66.8 (CH), 105.3 (CH), 111.3 (CH), 112.1
(CH), 120.1 (CH), 128.4 (C), 149.0 (C), 149.3 (C), 175.5 (C), 191.9
(C), 203.8 (C); HRMS (ESI) m/z 333.1316 [M + H]+, [C18H21O6]+
requires 333.1338.
(1S,5R,7S,8R)-(+)-8-(3,4-Dimethoxyphenyl)-2-methoxy-7-
triisopropylsilanyloxy-bicyclo[3.3.1]non-2-ene-4,9-dione 11 and
(1S,5R,7S,8R)-(+)-8-(3,4-dimethoxyphenyl)-4-methoxy-7-
triisopropylsilanyloxy-bicyclo[3.3.1]non-3-ene-2,9-dione 12
Triisopropylsilyl trifluoromethanesulfonate (2.06 mL, 7.68 mmol,
1.5 eq.) was added to a solution of bridged ketones 9 and 10 (1.70 g,
5.10 mmol) and 2,6-lutidine (1.79 mL, 15,4 mmol, 3 eq.) in dry
=
CHH), 2.52 (dd, J 15.0, 7.0, 1H, CHHCH C), 2.56 (dd, J 15.0,
=
7.0, 1H, CHHCH C), 3.07 (dd, J 10.3, 4.3, 1H, ArCH), 3.33 (dd,
◦
CH2Cl2 (10.0 mL) at 0 C. The mixture was allowed to warm to
J 4.3, 1H, 5-CH), 3.61 (s, 3H, OCH3), 3.84 (s, 3H, OCH3), 3.86 (s,
3H, OCH3), 4.46 (ddd, J 10.3, 10.3, 5.3, 1H, CHOSi(CH(CH3)2)3),
RT over 3 h, then stirred at RT for 16 h. After this period, the
crude reaction mixture was washed with 2 M HCl (2 × 5 mL)
and extracted with CH2Cl2 (2 × 10 mL). The organic extracts
were combined and dried (Na2CO3), then concentrated in vacuo.
Purification by column chromatography (petroleum ether–EtOAc
4 : 1) gave firstly 11 as a pale yellow solid (1.4 g, 56%); mp 128 ◦C;
[a]2D2 +109 (c 1.0 in CHCl3); mmax (CHCl3) 2939, 1736, 1657, 1604,
1374, 1118, 907 cm−1; 1H NMR (500 MHz, CDCl3) d 0.63–0.75 (m,
21H, OSi(CH(CH3)2)3), 1.83–1.91 (m, 1H, CHH), 2.48–2.52 (m,
1H, CHH), 3.00 (dd, J 10.3, 3.8, 1H, ArCH), 3.17–3.22 (m, 2H, 1-
CH, 5-CH), 3.54 (s, 3H, OCH3), 3.73 (s, 3H, ArOCH3), 3.75 (s, 3H,
ArOCH3), 4.40 (ddd, J 10.3, 10.3, 5.2, 1H, CHOSi(CH(CH3)2)3),
=
=
4.95–5.01 (m, 1H, CH C(CH3)2), 5.75 (s, 1H, C CH), 6.63 (d, J
2.0, 1H, ArH), 6.66 (dd, J 8.2, 2.0, 1H, ArH), 6.80 (d, J 8.2, 1H,
ArH); 13C NMR (100 MHz, CDCl3) d 12.5 (CH), 17.8 (CH3), 18.0
(CH3), 25.9 (CH3), 29.4 (CH2), 45.8 (CH2), 54.9 (CH3), 55.8 (CH3),
56.0 (CH), 56.2, (CH3), 59.3 (CH3), 62.6 (C), 68.2 (CH), 105.2
(CH), 110.0 (CH), 117.0 (CH), 119.1 (CH), 120.5 (CH), 122.2
(CH), 131.6 (C), 134.5 (C), 148.5 (C), 148.7 (C), 173.8 (C), 196.8
(C), 205.9 (C); HRMS (EI) m/z 557.3293 [M + H]+, [C32H49O6Si]+
requires 557.3298.
(1S,5S,7S,8R)-(+)-8-(3,4-Dimethoxyphenyl)-7-hydroxy-2,9,9-
trimethoxy-bicyclo[3.3.1]non-2-en-4-one 14 and (1R,5R,7S,8R)-
(−)-8-(3,4-dimethoxyphenyl)-7-hydroxy-4,9,9-trimethoxy-
bicyclo[3.3.1]non-3-en-2-one 15
=
5.67 (s, 1H, C CH), 6.52–6.53 (m, 1H, ArH), 6.55 (dd, J 8.2, 1.9,
1H, ArH), 6.70 (d, J 8.2, 1H, ArH); 13C NMR (125 MHz, CDCl3)
d 12.8 (CH), 17.5 (CH3), 39.0 (CH2), 54.6 (CH), 55.8 (CH3),
55.9 (CH3), 56.4 (CH3), 59.1 (CH), 59.4 (CH), 67.6 (CH), 105.1
(CH), 111.0 (CH), 111.6 (CH), 120.4 (CH), 131.5 (C), 148.5 (C),
148.7 (C), 174.9 (C), 194.8 (C), 204.1 (C); HRMS (CI, NH3) m/z
489.2633 [M + H]+, [C27H41O6Si]+ requires 489.2671 and secondly
12 as a yellow oil (0.42 g, 19%); [a]2D7 +6.0 (c 1.0 in CHCl3); mmax
(CHCl3) 2942, 1743, 1650, 1594, 1462, 1373, 1123 cm−1; 1H NMR
(500 MHz, CDCl3) d 0.75–0.90 (m, 21H, OSi(CH(CH3)2)3), 1.96–
2.04 (m, 1H, CHH), 2.53–2.59 (m, 1H, CHH), 3.06 (dd, J 10.5,
Trimethyl orthoformate (0.50 mL, 4.50 mmol, 1.5 eq.) and pTsOH
(57 mg, 0.30 mol, 0.1 eq.), were added to a solution of bridged
ketones 9 and 10 (1.00 g, 3.0 mmol) in MeOH (20 mL), and
the mixture was heated under reflux for 24 h. After this period,
the solution was allowed to cool then concentrated in vacuo.
Purification by column chromatography (EtOAc–petroleum ether
7 : 3) gave a mixture of ketals 14 and 15 in a 3 : 1 ratio (898 mg, 79%);
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1924–1934 | 1931
©

View Article Online
data for 14: [a]3D0 +169 (c 0.9 in CHCl3); mmax (CHCl3) 3463, 2906,
1.93–1.99 (m, 1H, CHH), 2.13 (ddd, J 12.8, 5.4, 3.7, 1H, CHH),
2.89–2.92 (m, 2H, CH, CH), 3.06 (dd, J 10.7, 3.7, 1H, ArCH),
3.12 (s, 3H, OCH3), 3.21 (s, 3H, OCH3), 3.75 (s, 3H, OCH3), 3.81
(s, 3H, ArOCH3), 3.82 (s, 3H, ArOCH3), 4.31 (ddd, J 10.7, 10.7,
1
1652, 1603, 1516, 1230, 1105, 1068, cm−1; H NMR (400 MHz,
CDCl3) d 1.84–1.92 (m, 1H, CHH), 2.17 (ddd, J 12.8, 8.8, 5.6, 1H,
CHH), 2.92 (m, 2H, 1-CH, 5-CH), 3.13 (s, 3H, OCH3), 3.21 (dd,
J 10.4, 3.6, 1H, ArCH), 3.31 (s, 3H, OCH3), 3.50 (s, 3H, OCH3),
3.86 (s, 3H, ArOCH3), 3.87 (s, 3H, ArOCH3), 4.17–4.29 (m, 1H,
=
5.4, 1H, CHOSi(CH(CH3)2)3), 5.56 (s, 1H, C CH), 6.57–6.60 (m,
2H, 2 × ArH), 6.75 (d, J 7.9, ArH); 13C NMR (125 MHz, CDCl3)
d 12.6 (CH), 17.9 (CH3), 32.4 (CH2), 42.5 (CH), 47.2 (CH3), 48.6
(CH3), 49.3 (CH), 55.7 (CH3), 55.9 (CH3), 56.4 (CH3), 57.0 (CH),
68.9 (CH), 100.9 (C), 103.0, (CH), 110.8 (CH), 112.9 (CH), 121.0
(CH), 132.3 (C), 147.9 (C), 148.4 (C), 176.6 (C), 197.4 (C); HRMS
(ES) m/z 535.3086 [M + H]+, [C29H47O7Si]+ requires 535.3091.
=
CHOH), 5.51 (s, IH, C CH), 6.67 (d, J 2.0, 1H, ArH), 6.70 (dd, J
8.0, 2.0, 1H, ArH), 6.84 (d, J 8.0, 1H, ArH); 13C NMR (100 MHz,
CDCl3) d 32.0 (CH2), 47.4 (CH), 48.7 (CH3), 48.8 (CH3), 49.3
(CH), 49.8 (CH), 55.7 (CH3), 55.8 (CH3), 55.9 (CH3), 66.4 (CH),
101.3 (C), 103.4 (CH), 111.3 (CH), 111.6 (CH), 120.0 (CH), 131.4
(C), 148.4 (C), 149.2 (C), 175.7 (C), 190.0 (C); HRMS (ES) m/z
379.1768 [M + H]+, [C20H27O7]+ requires 379.1757; data for 15: [a]2D0
−5 (c 0.5 in CHCl3); mmax (CHCl3) 2939, 1650, 1611, 1462, 1375,
1108, 1068, 1027, cm−1; 1H NMR (400 MHz, CDCl3) d 1.94–2.00
(m, 1H, CHH), 2.21 (ddd, J 12.9, 8.8, 5.5, 1H, CHH), 2.94–2.99
(m, 2H, 1-CH, 5-CH), 3.12 (dd, J 10.9, 4.0, 1H, ArCH), 3.15 (s,
3H, OCH3), 3.33 (s, 3H, OCH3), 3.78 (s, 3H, OCH3), 3.85 (s, 3H,
ArOCH3), 3.87 (s, 3H, ArOCH3), 4.24 (ddd, J 10.9, 10.9, 5.5, 1H,
Crystal data for 16. C29H46O7Si, M = 534.75, monoclinic, a =
◦
˚
8.2824(6), b = 13.7389(10), c = 12.7110(10) A, b = 91.2840(10) ,
3
˚
U = 1446.04(19) A , T = 150 K, space group P21 (no. 4), Z = 2,
l(Mo-Ka) = 0.124 mm−1, 12515 reflections measured, 6307 unique
(Rint = 0.031) which were used in all calculations. The final wR(F2)
was 0.100 (all data) and the Flack parameter refined to 0.09(9)
confirming the determination of the absolute configuration.‡
=
CHOH), 5.55 (s, 1H, C CH), 6.65–6.67 (m, 2H, 2 × ArH), 6.81–
Typical procedure A for lithiation substitution using LDA (Table 1
and Table 3): (1S,5S,7S,8R)-(+)-8-(3,4-dimethoxyphenyl)-
2,9,9-trimethoxy-5-(prenyl)-7-triisopropylsilanyloxy-
bicyclo[3.3.1]non-2-en-4-one 18
6.83 (m, 1H, ArH); 13C NMR (100 MHz, CDCl3) d 30.4 (CH2),
42.4 (CH), 47.2 (CH3), 48.7 (CH3), 48.7 (CH), 55.8 (CH3), 55.8
(CH3), 56.4 (CH3), 56.5 (CH), 67.2 (CH), 101.0 (C), 103.1 (CH),
111.3 (CH), 112.5 (CH), 120.2 (CH), 130.3 (C), 148.4 (C), 149.1
(C), 176.9 (C), 196.5 (C); HRMS (ESI) m/z 379.1756 [M + H]+,
[C20H27O7]+ requires 379.1757.
A solution of LDA·LiCl was prepared by treatment of a suspen-
sion of DIPA·HCl (180 mg, 1.31 mmol) in THF (2.5 mL) at −78 ◦C
n
with BuLi (1.64 mL, 2.62 mmol, 1.6 M solution in hexanes).
(1S,5S,7S,8R)-(+)-8-(3,4-Dimethoxyphenyl)-2,9,9-trimethoxy-7-
triisopropylsilanyloxy-bicyclo[3.3.1]non-2-en-4-one 16 and (1R,
5R, 7S, 8R)-(+)-8-(3,4-dimethoxyphenyl)-4,9,9-trimethoxy-7-
triisopropylsilanyloxy-bicyclo[3.3.1]non-3-en-2-one 17
The solution was allowed to warm to RT and after 10 min re-
cooled to −78 C. This LDA·LiCl solution was added dropwise
◦
via syringe to a solution of◦bridged ketone 16 (140 mg, 0.26 mmol)
in THF (1.4 mL) at −78 C resulting in a deep yellow-coloured
solution. The solution was stirred at −78 ◦C for 3 h, followed
by addition of prenyl bromide (0.30 mL, 2.62 mmol, 10 eq.) and
stirring for a further 3 h at −78 ◦C. The reaction mixture was
quenched after this period with H2O (5 mL) followed by extraction
with EtOAc (3 × 5 mL). The organic layers were combined and
washed with saturated aqueous NaCl (5 mL), dried (MgSO4), and
concentrated in vacuo. Purification by column chromatography
(petroleum ether–EtOAc 4 : 1) gave the title compound 18 as a
light brown oil (118 mg, 75%); [a]2D3 +150 (c 0.9 in CHCl3); mmax
(CHCl3) 2940, 2865, 1649, 1615, 1462, 1381, 1132, 1055 cm−1; 1H
NMR (500 MHz, CDCl3) d 0.77–0.88 (m, 21H, OSi(CH(CH3)2)3),
1.62 (s, 3H, CH3), 1.67 (s, 3H, CH3), 1.82–1.84 (m, 2H, 8-CH2),
2,6-Lutidine (0.64 mL, 5.55 mmol, 3 eq.) and triisopropylsilyl
trifluoromethanesulfonate (0.75 mL, 3.70 mmol, 1.5 eq.) were
added to a solution of 14 and 15 (0.7 g, 1.85 mmol) in dry CH2Cl2
(10 mL) at 0 ◦C. The reaction mixture was allowed to warm to RT
over 1 h then stirred at RT overnight. The reaction mixture was
washed with 2 M HCl (3 × 7 mL) and saturated aqueous NaCl (1 ×
10 mL), extracted with CH2Cl2 (2 × 10 mL), dried (MgSO4), then
concentrated in vacuo. Purification by column chromatography
(petroleum ether–EtOAc 85 : 15) gave firstly bridged ketal 16 as
a white solid (480 mg, 56%); [a]2D6 +154 (c 1.0 in CHCl3); mmax
(CHCl3) 2941, 2865, 1650, 1604, 1462, 1379, 1106 cm−1; 1H NMR
(500 MHz, CDCl3) d 0.83–0.93 (m, 21H, OSi(CH(CH3)2)3), 1.90–
1.97 (m, 1H, CHH), 2.19 (ddd, J 12.8, 5.6, 3.7, 1H, CHH), 2.93–
2.94 (m, 2H, 1-CH, 5-CH), 3.15 (s, 3H, OCH3), 3.19 (dd, J 10.6,
3.7, 1H, ArCH), 3.35 (s, 3H, OCH3), 3.60 (s, 3H, OCH3), 3.87
(s, 3H, ArOCH3), 3.39 (s, 3H, ArOCH3), 4.33 (ddd, J 10.7, 10.7,
=
2.47 (dd, J 15.0, 7.0, 1H, CHHCH C(CH3)2), 2.60 (dd, J 15.0,
=
7.0, 1H, CHHCH C(CH3)2), 2.92 (d, J 4.0, 1H, 5-CH), 3.07 (dd,
J 10.4, 4.0, 1H, ArCH), 3.22 (s, 3H, OCH3), 3.47 (s, 3H, OCH3),
3.54 (s, 3H, OCH3), 3.82 (s, 3H, ArOCH3), 3.84 (s, 3H, ArOCH3),
4.22 (ddd, J 10.4, 10.4, 7.0, 1H, CHOSi(CH(CH3)2)3), 5.47–5.52
=
5.6, 1H, CHOSi(CH(CH3)2)3), 5.58 (s, 1H, C CH), 6.66 (d, J 1.8,
=
=
(m, 1H, (CH3)2C CH), 5.53 (s, 1H, C CH), 6.60 (d, J 1.9, 1H,
1H, ArH), 6.68 (dd, J 8.2, 1.8, 1H, ArH), 6.82 (d, J 8.2, 1H,
ArH); 13C NMR (125 MHz, CDCl3) d 12.8 (CH), 17.9 (CH3), 34.0
(CH2), 47.3 (CH3), 48.3 (CH), 48.6 (CH3), 49.5 (CH), 49.9 (CH),
55.6 (CH3), 55.7 (CH3), 55.9 (CH3), 68.2 (CH), 101.3 (C), 103.12
(CH), 110.8 (CH), 111.9 (CH) 120.8 (CH), 133.2 (C), 147.9 (C),
148.5 (C), 176.6 (C), 198.9 (C); HRMS (ES) m/z 535.3069 [M +
H]+, [C29H47O7Si]+ requires 535.3091; and secondly bridged ketal
17 as a yellow oil (250 mg, 22%); [a]3D0 +81 (c 0.9 in CHCl3); mmax
(CHCl3) 2941, 2865, 1650, 1604, 1462, 1379, 1106 cm−1; 1H NMR
(500 MHz, CDCl3) d 0.77–0.87 (m, 21H, CHOSi(CH(CH3)2)3),
ArH), 6.64 (dd, J 8.2, 1.9, 1H, ArH), 6.77 (d, J 8.2, 1H, ArH); 13
C
NMR (125 MHz, CDCl3) d 12.6 (CH), 17.9 (CH3), 25.9 (CH3),
30.2 (CH2), 41.1 (CH2), 48.2 (CH), 49.7 (CH3), 50.6 (CH), 51.1
(CH3), 55.5 (CH3), 55.7 (CH3), 55.9 (CH3), 57.1 (C), 68.7 (CH),
103.6 (CH), 103.9 (C), 110.8 (CH), 111.9 (CH), 121.0 (CH), 122.3
(CH), 131.5 (C), 133.2 (C), 147.9 (C), 148.4 (C), 174.7 (C), 201.2
‡ CCDC reference number 641180. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/b704311b
1932 | Org. Biomol. Chem., 2007, 5, 1924–1934
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
(C); HRMS (EI) m/z 603.3712 [M + H]+, [C34H55O7Si]+ requires
603.3717.
with saturated aqueous NaCl (10 mL), dried (MgSO4), then
concentrated in vacuo. Purification by column chromatography
(petroleum ether–EtOAc 8 : 2) gave the title compound 29 as a pale
yellow oil (500 mg, 86%); [a]3D0 −50 (c 1.0 in CHCl3); mmax (CHCl3)
(1S,5R,7S,8R)-(+)-8-(3,4-Dimethoxyphenyl)-4-oxo-2,9,9-tri-
methoxy-7-triisopropylsilanyloxy-bicyclo[3.3.1]non-2-ene-3,5-dicar-
boxylic acid diethyl ester 27. Isolated as an amorphous yellow
solid (55 mg, 45%); [a]3D0 +173 (c 0.5 in CHCl3); 1H NMR
(500 MHz, CDCl3) d 0.75–0.90 (m, 21H, OSi(CH(CH3)2)3), 1.29
(t, J 7.1, 3H, CH3), 1.33 (t, J 7.1, 3H, CH3), 2.26 (t, J 12.9, 1H,
CHH), 2.39 (dd, J 12.9, 5.3, 1H, CHH), 3.15 (d, J 4.5, 1H, CH),
3.22 (s, 3H, OCH3), 3.28 (dd, J 10.6, 4.5, 1H ArCH), 3.3 (s, 3H,
OCH3), 3.37 (s, 3H, OCH3), 3.86 (s, 3H, ArOCH3), 3.89 (s, 3H,
ArOCH3), 4.21–4.38 (m, 5H, CHOSi(CH(CH3)2)3), OCH2CH3,
OCH2CH3), 6.77–6.83 (m, 3H, ArH, ArH, ArH); 13C NMR
(125 MHz, CDCl3) d 13.9 (CH3), 14.2 (CH3), 12.5 (CH), 18.0
(CH3), 38.9 (CH2), 48.8 (CH), 49.0 (CH3), 49.3 (CH3), 49.4 (CH),
55.8 (CH3), 56.0 (CH3), 56.8 (CH3), 61.6 (CH2), 61.7 (CH2), 62.9
(C), 68.0 (CH), 101.8 (C), 111.0 (CH), 111.1 (CH), 115.1 (C),
121.6 (CH), 132.2 (C), 148.2 (C), 148.8 (C), 165.4 (C), 170.3 (C),
171.9 (C), 192.7 (C). IR mmax 2913, 2847, 1729, 1614, 1463, 1373,
1249, 1089, 1051 cm−1; HRMS (ES) m/z 679.3508 [M + H]+,
[C35H55O11Si]+ requires 679.3513.
1
2918, 1660, 1612, 1462, 1107, 831 cm−1; H NMR (400 MHz,
CDCl3) d 0.64 (q, J 7.9, 6H, OSi(CH2(CH3)2)3), 0.77–0.89 (m,
21H, OSi(CH(CH3)2)3), 1.01 (t, J 7.9, 9H, OSi(CH2(CH3)2)3), 2.04
(ddd, J 12.4, 8.6, 5.2, 1H, CHH), 2.14–2.18 (m, 1H, CHH), 2.61–
2.67 (m, 2H, 1-CH, 5-CH), 3.28 (dd, J 10.5, 3.8, 1H, ArCH),
3.55 (s, 3H, OCH3), 3.82 (s, 3H, ArOCH3), 3.85 (s, 3H, ArOCH3),
4.09 (t, J 3.6, 1H, CHOSi(CH2CH3)3), 4.21 (ddd, J 10.5, 10.5, 5.2,
=
1H, CHOSi(CH(CH3)2)3), 5.53 (s, 1H, C CH), 6.59 (d, J 1.9, 1H,
ArH), 6.63 (dd, J 8.2, 1.9, 1H, ArH), 6.78 (d, J 8.2, 1H, ArH); 13
C
NMR (125 MHz, CDCl3) d 4.8 (CH2), 6.9 (CH3), 12.7 (CH), 18.0
(CH3), 18.1 (CH3), 31.8 (CH2), 45.7 (CH), 50.8 (CH), 51.3 (CH),
55.6 (CH3), 55.7 (CH3), 55.9 (CH3), 68.2 (CH), 68.7 (CH), 104.1
(CH), 110.8 (CH), 111.9 (CH), 121.0 (CH), 134.0 (C), 147.7 (C),
148.4 (C), 178.8 (C), 200.7 (C); HRMS (ESI) m/z 605.3724 [M +
H]+, [C33H57O6Si2]+ requires 605.3694.
(1R,5S,7S,8R,9R)-(+)-8-(3,4-Dimethoxyphenyl)-2-methoxy-5-
(prenyl)-9-triethylsilanyloxy-7-triisopropylsilanyloxy-
bicyclo[3.3.1]non-2-en-4-one 30
(1R,5S,7S,8R,9R)-(+)-8-(3,4-Dimethoxyphenyl)-9-hydroxy-2-
methoxy-7-triisopropylsilanyloxy-bicyclo[3.3.1]non-2-en-4-one 28
A solution of LDA·LiCl was prepared by treatment of a suspen-
sion of DIPA·HCl (141 mg, 1.03 mmol) in THF (2.5 mL) at −78 ◦C
Sodium borohydride (60.0 mg, 1.54 mmol, 1.5 eq.) in CH3OH
(4 mL) was added to a solution of 9 (500 mg, 1.02 mmol) in 2 :
1 CH3OH : CH2Cl2 (15 mL) at −78 ◦C. The reaction mixture
n
with BuLi (1.28 mL, 2.05 mmol, 1.6 M solution in hexanes).
◦
The solution was allowed to warm to RT and after 10 min re-
cooled to −78 ◦C. The LDA·LiCl solution was added dropwise
via syringe to a solution of bicyclic compound 29 (124 mg,
0.20 mmol) in THF (2 mL) at −78 ◦C resulting in a yellow-
coloured solution. The solution was stirred at −78 ◦C for 3 h,
followed by addition of prenyl bromide (0.25 mL, 2.05 mmol,
10 eq.) and stirring for a further 4 h at −78 ◦C. The reaction
mixture was quenched after this period with H2O (5 mL) followed
by extraction with EtOAc (3 × 5 mL). The organic layers were
combined and washed with saturated aqueous NaCl (15 mL),
dried (MgSO4), and concentrated in vacuo. Purification by column
chromatography (petroleum ether–EtOAc 4 : 1) gave the title
compound 30 as a pale yellow oil (35 mg, 26%); [a]2D2 +73 (c
0.15 in CHCl3); mmax (CHCl3) 2256, 1816, 1793, 1380, 1095, 947,
888, 641 cm−1; 1H NMR (400 MHz, CDCl3) d 0.66–0.74 (m, 6H,
OSi(CH2CH3)3), 0.78–0.88 (m, 21H, OSi(CH(CH3)2)3), 1.06 (t, J
was stirred at −78 C for 1 h. After this period the reaction was
quenched with aqueous NaOH (2 mL, 2 M), extracted with EtOAc
(3 × 5 mL). The organic extracts were combined and washed with
saturated aqueous NaCl (aq.) (1 × 10 mL), dried (MgSO4), then
concentrated in vacuo. Purification by column chromatography
(petroleum ether–EtOAc 3 : 7) gave the title compound 28 as a
◦
fluffy white solid (470 mg, 94%); mp 178–180 C; [a]2D4 +163 (c
0.5 in CHCl3); mmax (CHCl3) 3606, 2940, 1646, 1597, 1463, 1378,
1
1116, 1026, 882 cm−1; H NMR (400 MHz, CDCl3) d 0.79–0.88
(m, 21H, OSi(CH(CH3)2)3), 1.99–2.20 (m, 2H, 8-CH2), 2.32 (d, J
3.5, 1H, OH), 2.78 (m, 2H, 1-CH, 5-CH), 3.29 (dd, J 10.2, 4.1,
1H, ArCH), 3.56 (s, 3H, OCH3), 3.83 (s, 3H, ArOCH3), 3.86 (s,
3H, ArOCH3), 4.20–4.27 (m, 2H, CHOSi(CH(CH3)2)3, CHOH),
=
5.55 (s, 1H, C CH), 6.62–6.65 (m, 2H, 2 × ArH), 6.78 (d, J 8.7,
1H, ArH); 13C NMR (100 MHz, CDCl3) d 12.6 (CH), 17.9 (CH3),
18.1 (CH3), 31.7 (CH2), 45.7 (CH), 49.9 (CH), 50.4 (CH), 55.6
(CH3), 55.7 (CH3), 56.0 (CH3), 68.3 (CH), 68.3 (CH), 104.0 (CH),
110.9 (CH), 112.2 (CH), 120.8 (CH), 133.6 (C), 147.8 (C), 148.5
(C), 178.7 (C), 200.2 (C); HRMS (ESI) m/z 491.2811 [M + H]+,
[C27H43O6Si]+ requires 491.2828.
=
8.0, 9H, OSi(CH2CH3)3), 1.64 (s, 3H, C C(CH3)2), 1.65 (s, 3H,
=
C C(CH3)2), 1.74 (dd, J 12.0, 5.2, 1H, CHH), 1.93 (t, J 12.0,
=
10.3, 1H, CHH), 2.20 (dd, J 15.0, 7.0, 1H, CHHCH C(CH3)2),
2.56 (dd, J 15.0, 7.0, 1H, CHHCH C(CH3)2), 2.69 (dd, J 4.0,
4.0, 1H, 5-CH), 3.20 (dd, J 10.3, 4.0, 1H, ArCH), 3.54 (s, 3H,
OCH3), 3.81 (s, 3H, ArOCH3), 3.85 (s, 3H, ArOCH3), 4.07 (d,
J 4.0, 1H, CHOSi(CH2CH3)3), 4.16 (ddd, J 10.3, 10.3, 5.2, 1H,
=
(1R,5S,7S,8R,9R)-(−)-8-(3,4-Dimethoxyphenyl)-2-methoxy-
9-triethylsilanyloxy-7-triisopropylsilanyloxy-bicyclo[3.3.1]non-
2-en-4-one 29
=
CHOSi(CH(CH3)2)3), 4.91 (t, J 7.0, 1H, CH C(CH3)2), 5.54 (s,
=
1H, C CH), 6.60 (d, J 1.9, 1H, ArH), 6.63 (dd, J 8.2, 1.9, 1H,
Triethylsilyl triflate (0.33 mL, 1.3 mmol, 1.5 eq.) was added to a
solution of 28 (470 mg, 9.56 mmol) and 2,6-lutidine (0.33 mL,
2.6 mmol, 3 eq.) in dry CH2Cl2 (10 mL) at −78 ◦C and the
solution stirred at −78 ◦C for 1 h. The reaction was quenched
with aqueous HCl (10 mL, 2 M) and extracted with CH2Cl2
(3 × 8 mL). The organic extracts were combined and washed
ArH), 6.78 (d, J 8.2, 1H, ArH); 13C NMR (100 MHz, CDCl3)
d 5.1 (CH2), 7.0 (CH3), 12.7 (CH), 18.0 (CH3), 18.2 (CH3), 25.8
(CH3), 26.9 (CH3), 31.3 (CH2), 39.4 (CH2), 45.8 (CH), 51.2 (CH),
53.4 (CH), 55.4 (CH3), 55.6 (CH3), 55.9 (CH3), 68.8 (CH), 70.0
(CH), 104.5 (CH), 110.9 (CH), 111.9 (CH), 119.7 (CH), 121.0
(CH), 133.5 (C), 147.7 (C), 148.4 (C), 177.0 (C), 201.4 (C); HRMS
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1924–1934 | 1933
©

View Article Online
(ESI) m/z 557.3293 [M − OSi(CH2CH3)3]+, [C32H49O6Si]+ requires
Notes and references
557.3298.
1 For a recent and comprehensive review of the area, see (a) R. Ciochina
and R. B. Grossman, Chem. Rev., 2006, 106, 3963–3986. See also; (b) L.
Verotta, Phytochem. Rev., 2002, 1, 389–407; (c) R. M. Wilson and S. J.
Danishefsky, Acc. Chem. Res., 2006, 39, 539.
2 For the most recent contribution, see P. Nuhant, M. David, T. Pouplin,
B. Delpech and C. Marazano, Org. Lett., 2007, 9, 287.
3 A. Kuramochi, H. Usuda, K. Yamatsugu, M. Kanai and M. Shibasaki,
J. Am. Chem. Soc., 2005, 127, 14200.
4 D. R. Siegal and S. J. Danishefsky, J. Am. Chem. Soc., 2006, 128, 1048.
5 V. Rodeschini, N. M. Ahmad and N. S. Simpkins, Org. Lett., 2006, 8,
5283.
6 N. M. Ahmad, V. Rodeschini, N. S. Simpkins, S. E. Ward and A. J.
Blake, J. Org. Chem., 2007, DOI: 10.1021/jo070388h.
7 S. J. Spessard and B. M. Stoltz, Org. Lett., 2002, 4, 1943.
8 K.-H. Scho¨nwa¨lder, P. Kollatt, J. J. Stezowski and F. Effenburger, Chem.
Ber., 1984, 117, 3280.
9 In a more recent synthesis of clusianone, using a similar approach,
slightly improved yields of bridged adduct were obtained from the
Effenberger cyclisation by use of BF3–OEt2 as Lewis acid mediator, see
ref. 2.
10 (a) The starting material for the Shibasaki synthesis was demonstrated
to be available in 95% ee, see ref. 3; (b) for an enantiospecific approach,
see G. Mehta and M. K. Bera, Tetrahedron Lett., 2004, 45, 1113.
11 K. D. Sears, R. L. Casebier, H. L. Hegert, G. H. Stout and L. E.
McCandlish, J. Org. Chem., 1974, 39, 3244.
12 L. N. Mander and C. M. Williams, Tetrahedron, 2003, 59, 1105.
13 (a) G. M. P. Giblin, D. T. Kirk, L. Mitchell and N. S. Simpkins, Org.
Lett., 2003, 5, 1673; (b) A. J. Blake, G. M. P. Giblin, D. T. Kirk, N. S.
Simpkins and C. Wilson, Chem. Commun., 2001, 2668.
14 (a) O. Miyata and R. R. Schmidt, Tetrahedron Lett., 1982, 23, 1793;
(b) For a more recent example, see C. W. Zapf, B. A. Harrison, C. Drahl
and E. J. Sorensen, Angew. Chem., Int. Ed., 2005, 44, 6533.
15 TMSCl is a well established reagent for the conversion of epoxides to
chlorohydrins, usually promoted by catalysts, see for example (a) L.-W.
Xu, L. Li, C.-G. Xia and P.-Q. Zhao, Tetrahedron Lett., 2004, 45, 2435;
(b) L.-S. Wang and T. K. Hollis, Org. Lett., 2003, 5, 2543.
16 P. H. J. Carlsen, T. Katsuki, V. S. Martin and K. B. Sharpless, J. Org.
Chem., 1981, 46, 3936.
17 D. M. Jones, B. Nillson and M. Szelke, J. Org. Chem., 1993, 58, 2286.
18 A. K. Chakraborti and U. R. Ghatak, J. Chem. Soc., Perkin Trans. 1,
1985, 2605.
19 See for example: T. Kamo, N. Hirai, K. Iwami, D. Fujioka and H.
Ohigashi, Tetrahedron, 2001, 57, 7649.
20 J. G. Walsh, P. J. Furlong, L. A. Byrne and D. C. Gilheany, Tetrahedron,
1999, 55, 11519.
21 V. Balogh, M. Fetizon and M. Golfier, J. Org. Chem., 1971, 36, 1339.
22 (a) R. Pappo, D. S. Allen, Jr., R. U. Lemieux and W. S. Johnson, J. Org.
Chem., 1956, 21, 478; (b) H. Vorbrueggen and C. Djerassi, J. Am. Chem.
Soc., 1962, 84, 2990.
Typical procedure B for lithiation substitution using LTMP
(Table 2): (1S,5S,7S,8R)-(+)-8-(3,4-dimethoxyphenyl)-2,9,9-
trimethoxy-7-triisopropylsilanyloxy-3-trimethylsilanyl-
bicyclo[3.3.1]non-2-en-4-one 31
n
A solution of LTMP was prepared by adding BuLi (0.59 mL,
0.93 mmol, 1.6 M solution in hexanes) to tetramethylpiperidine
(0.19 mL, 0.93 mmol, 5 eq.) in THF (1 mL) at −78 ◦C. The
solution was allowed to warm to RT and after 10 min re-cooled
to −78 ◦C. This LTMP–THF solution was added dropwise via
syringe to a solution of bridged ketone 16 (100 mg, 0.19 mmol)
in THF (1 mL) at −78 ◦C resulting in a deep yellow-coloured
solution. The solution was stirred at −78 ◦C for 3 h, followed
by addition of trimethylsilyl chloride (0.24 mL, mmol, 10 eq.)
and stirring for a further 3 h at −78 ◦C. The reaction mixture was
quenched after this period with H2O (5 mL) followed by extraction
with EtOAc (3 × 5 mL). The organic layers were combined and
washed with saturated aqueous NaCl (5 mL), dried (MgSO4), and
concentrated in vacuo. Purification by column chromatography
(petroleum ether–EtOAc 4 : 1) gave the title compound 31 as a
clear viscous oil (53 mg, 45%); [a]2D2 +169 (c 1.0 in CHCl3); mmax
(CHCl3) 2941, 2866, 1634, 1565, 1462, 1366, 1141, 1108, 1027,
881 cm−1; 1H NMR (500 MHz, CDCl3) d 0.22 (s, 9H, Si(CH3)3),
0.72–0.89 (m, 21H, OSi(CH(CH3)2)3), 1.81–1.87 (m, 1H, CHH),
2.13 (ddd, J 12.5, 9.2, 5.3, 1H, CHH), 2.80–2.82 (m, 1H, 1-CH),
2.86 (s, 3H, OCH3), 3.11 (s, 3H, OCH3), 3.16 (dd, J 10.9, 3.0,
1H, ArCH), 3.22–3.24 (m, 1H, 5-CH), 3.31 (s, 3H, OCH3), 3.83
(s, 3H, ArOCH3), 3.86 (s, 3H, ArOCH3), 4.25 (ddd, J 10.9, 10.9,
5.3, 1H, CHOSi(CH(CH3)2)3), 6.76–6.81 (m, 3H, 3 × ArH); 13C
NMR (125 MHz, CDCl3) d 0.82 (CH3), 12.6 (CH), 17.9 (CH3),
18.1 (CH3), 34.4 (CH2), 44.3 (CH), 47.3 (CH3), 48.5 (CH3), 48.7
(CH), 49.7 (CH3), 54.4 (CH3), 55.7 (CH3), 56.0 (CH3), 56.1 (CH3),
68.1 (CH), 101.2 (C), 111.1 (CH), 116.8 (C), 132.9 (C), 148.2 (CH),
148.7 (C), 180.9 (C), 201.9 (C); HRMS (ES) m/z 607.3461 [M +
H]+, [C32H55O7Si2]+ requires 607.3486.
Acknowledgements
23 J. Gierer and F. Imsgard, Acta Chem. Scand., Ser. B, 1977, 31, 537; J.
Gierer and F. Imsgard, Acta Chem. Scand., Ser. B, 1977, 31, 546.
24 C. Koshima, S. Hibi, T. Maruyama, K. Harada and Y. Omote,
J. Heterocycl. Chem., 1987, 24, 637.
We acknowledge DEFRA and a Marie-Curie Fellowship for
support of VR, and GSK, Harlow, UK and the University of
Nottingham for support of NMA.
1934 | Org. Biomol. Chem., 2007, 5, 1924–1934
This journal is
The Royal Society of Chemistry 2007
©