Synthesis of (-)-xialenon A by enantioselective α-deprotonation- rearrangement of a meso-epoxide

Article abstract of DOI:10.1016/j.tet.2003.09.023

The first total synthesis of (-)-xialenon A was achieved via enantioselective transannular desymmetrisation of a substituted cycloctene oxide using an organolithium/(-)-α-isosparteine combination. Oxidation of the resulting bicyclo[3.3.0]octanol gave an enone which underwent stereoselective conjugate allylation; subsequent α′-hydroxylation on the more hindered face of a derived enone using hypervalent iodine chemistry led to the natural product.

Full text of DOI:10.1016/j.tet.2003.09.023

Tetrahedron 59 (2003) 9719–9728
Synthesis of (2)-xialenon A by enantioselective a-deprotonation-
rearrangement of a meso-epoxide
*
David M. Hodgson, Jean-Marie Galano and Martin Christlieb
Dyson Perrins Laboratory, Department of Chemistry, University of Oxford, South Parks Road, Oxford OX1 3QY, UK
Received 2 July 2003; revised 13 August 2003; accepted 5 September 2003
Abstract—The first total synthesis of (2)-xialenon A was achieved via enantioselective transannular desymmetrisation of a substituted
cycloctene oxide using an organolithium/(2)-a-isosparteine combination. Oxidation of the resulting bicyclo[3.3.0]octanol gave an enone
which underwent stereoselective conjugate allylation; subsequent a0-hydroxylation on the more hindered face of a derived enone using
hypervalent iodine chemistry led to the natural product.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
stereocentres in such bicyclo[3.3.0]octane systems.
Arising from our studies on enantioselective transannular
desymmetrisation of substituted cycloctene oxide 6 ([Si]¼
ButMe2Si) using organolithium/(2)-sparteine [or (2)-a-
isosparteine] combinations,2,3 we considered that alcohol
5 could be used to address the above issues and, if suc-
cessful, provide an attractive entry point to the xialenons,
especially (2)-xialenon A (1), and potentially (2)-xialenon
E (2) (Scheme 1). Thus, we planned to oxidise alcohol 5 to
enone 4. It was anticipated that the envelope shape of enone
4 should bias both allyl conjugate addition and, hopefully,
oxidation of the intermediate enolate from the convex face
to give hydroxyketone 3 with the desired stereochemistry;
xialenon A would then be achieved via elimination of the
tertiary silyloxy group. In the present paper we report full
details of our studies culminating in the synthesis of
xialenon A (1) by the above strategy,4 together with
attempts to use a common intermediate to synthesise
xialenon E (2).
Recently isolated by Thiericke and co-workers, (2)-xia-
lenon A 1 and its congeners are naturally occurring
secondary metabolites from culture broths of the
Streptomyces genus (Fig. 1).1 The xialenons possess a
bicyclo[3.3.0]octane skeleton, which is rarely found
without additional ring appendages in nature. The structural
study by spectroscopy and derivatisation established that
(2)-xialenon A (1) is (2R,3R,4S)-3-allyl-3,4,5,6-tetrahydro-
2,4-dihydroxy-2H-pentalen-1-one.
Figure 1. The family of xialenon natural products.
The principal issues confronting a total synthesis of
xialenon A (1) are introduction, with the correct stereo-
chemistry, of the lone hydroxyl group at C-4 and of the cis
vicinal hydroxyl and allyl groups. Also, the presence of
unsaturation at the ring fusion, aside from being part of a
potentially sensitive enone, serves to flatten the bicycle and,
when compared with the corresponding saturated bicycle,
this removes the important facial biasing (concave, convex)
effect typically relied on to control the introduction of
Keywords: desymmetrization; epoxide; lithiation; sparteine; total synthesis.
*
Corresponding author; e-mail: david.hodgson@chem.ox.ac.uk
Scheme 1.
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.09.023

9720
D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
Scheme 2.
2. Results and discussion
to examine conversion of this intermediate to the xialenons.
Formation of bicyclic alcohol 5 proceeds by enantio-
selective a-deprotonation of the epoxide 6 using the
organolithium mediated by the chiral diamine;3 in related
work we have recently established that the resultant
transient oxiranyl anion can be trapped with electrophiles
to provide a conceptually new route to enantioenriched
epoxides,6 however in the present case warming allows
transannular C–H insertion to occur generating the bicyclic
alcohol 5.
Alcohol 5 was prepared essentially following our previously
described 4-step route from 1,5-cyclooctadiene (COD, 7)
(Scheme 2).2 Some observations made during the scale-up
of this sequence are worthy of note, however. First, mono-
epoxidation of COD 7 was best affected using peracetic acid
(92% yield, compared to 54% yield previously obtained
with m-CPBA). Second, dihydroxylation of the resulting
mono-epoxide 8 was found to give the corresponding
epoxydiol 10 as previously reported. But, interestingly, the
mass balance of this reaction has now been identified as an
approximately equal mixture of 9-oxabicyclo[4.2.1]nonane-
2,5-diol 11 and 9-oxabicyclo[3.3.1]nonane-2,6-diol 12, for
which the former isomer was unambiguously assigned by
X-ray crystallographic analysis of the exo-mono-2,4-
dinitrobenzoate derivative 13 (Section 4.1.3). This work
indicates that the selectivity in the dihydroxylation of mono-
epoxide 8 is 3:1, in favour of the desired epoxydiol 10,
and that the minor epoxydiol 9 isomer undergoes (non-
regioselective)5 transannular ring opening of the epoxide.
Disilylation of epoxydiol 10 gave desymmetrisation sub-
strate 6 in near quantitative yield; whilst we have also
prepared the latter from COD via initial dihydroxylation,
then disilylation and a completely stereoselective epoxida-
tion,2 the low yield in the initial dihydroxylation leads us to
prefer the route via initial monoepoxidation of COD. Scale-
up of the key transannular desymmetrisation step led to a
slightly lower yield (and ee) of alcohol (þ)-5 than we had
previously reported2 (55% yield, 80% ee, vs 72% yield and
89% ee), but this provided sufficient material for us to begin
Recent reports by Nicolaou et al.7 led us to consider that
alcohol 5 could be directly oxidised to enone 4 in one step
by treatment with o-iodoxydobenzoic acid (IBX).8 Treat-
ment of alcohol 5 with IBX (2.2 equiv.) in a 2:1 mixture of
PhF/DMSO (0.1 M) at 558C for 4 h gave ketone 14 in 90%
yield. Extending the reaction time did produce some enone,
but led to lower mass recovery overall (52% of a 1.5:1
mixture of enone 4/ketone 14). Raising the temperature
gave progressively greater mass loss. Attempted reaction of
ketone 14 with IBX (1.4 equiv.) in a 2:1 mixture of PhF/
DMSO (0.1 M) at 558C for 16 h returned the ketone
˚
unchanged, whereas addition of TsOH and 4 A sieves
(16 h, 558C) gave an 89% yield of a 1:1 mixture of
recovered ketone and enone 4. We also examined the more
recently reported IBX-4-methylpyridine-N-oxide com-
plex7c,d with ketone 14, but none of the desired enone 4
was obtained. We then decided to use a more robust
oxidation sequence. Indeed, conversion of alcohol 5 to the
enone 4 was best effected in a stepwise manner via TPAP-
NMO oxidation9 to the ketone 14 (97% yield, Scheme 3),
Scheme 3.

D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
9721
followed by further palladium-induced oxidation10 of the
derived silyl enol ether using Pd(OAc)2 (80% yield from
ketone 14).
for the minor epimer endo-3 NOEs were observed between
C(2)H and C(6a)H and no NOEs were detected between
C(2)H and C(3)H). The modest exo-selectivity in the
oxidation compared with the completely exo-selective
cuprate allylation may originate from two factors: first, the
a-carbon to the ketone is less subject to the concave
shielding effect of the bicycle; and second the newly
introduced allyl side-chain now reduces accessibility on the
convex face.
At this stage of the synthesis two strategies were
investigated to introduce the allyl side-chain from the
convex face of 4. Consideration that the tertiary silyloxy
substitutent might make conjugate addition difficult led us
to first examine an oxy-Cope rearrangement following
1,2-addition of an allyl group to the enone functionality. In
the event, addition of allyllithium11 to enone 4 gave two
products arising from exo-face addition of the allyl group
directly to the ketone carbon (Scheme 3); these products
were allylic alcohol 15 (17% yield) and allylic ether 16
(36% yield)—the later resulting from a ButMe2Si migration.
Pleasingly, allylic alcohol 15 underwent [3,3] sigmatropic
rearrangement when treated with KH/18-crown-6, to give
the desired allylated ketone 17 in quantitative yield. An
analysis of the product’s stereochemistry by 1H NMR
showed NOEs between C(6a)Hexo and one of the allylic
protons, thus supporting the intial exo-attack and pericyclic
step. Unfortunately, allylic ether 16 could not be similarly
coaxed to undergo silyloxy-Cope rearrangement even with
heating up to 4508C (heating 16 to 4508C in a sand bath
caused considerable decomposition; 20% recovered starting
material 16 with detectable (1H NMR) expulsion of propene
to return enone 4). 1,2-Addition of allylmagnesium bromide
to enone 4 was even less satisfactory, since it produced only
compound 16 (80% yield).
Scheme 4.
In model studies on the key elimination (c.f. 3 to 1) using
ketone 14 we found that basic conditions (e.g. 2 equiv. of
DBU in MeCN at 258C for 48 h) caused elimination of the
tertiary silyloxy group yielding the corresponding fused
enone 19 in 90% yield (Scheme 5). Unfortunately,
application of this method to hydroxyketone 3 only resulted
in its complete decomposition.
The second strategy studied was to introduce the allyl side-
chain in a 1,4-fashion. We started by investigating the use of
allyltrimethylsilane. Whilst Sakurai reaction using a
standard protocol12 or an allylbarium reagent13 gave none
of the desired product 17, the recent allylation procedure
developed by Robertson et al.14 using allyltrimethylsilane
and catalytic bistrifluoromethanesulfonimide gave the
required allylated ketone 17 in 54% yield. Encouraged by
this result, copper-mediated allylation was studied15 with
the best result being obtained using the procedure of
Lipshutz and co-workers (Scheme 4).15a Furthermore,
dependent on the work-up conditions (depending on
whether the reaction was worked-up with a saturated
solution of NH4Cl, or with water first for 30 min then
addition of a saturated solution of NH4Cl) either silyl enol
ether 18 or allylated ketone 17 were recovered (both in 60%
yields) and with complete stereoselectivity once again.
Having secured the allyl side-chain introduction, we then
focused on a-hydroxylation via Rubottom oxidation16 of
silyl enol ether 18. This procedure gave hydroxyketone 3 as
an easily separable mixture of epimers (85% yield in total,
3:1 in favour of the desired diastereoisomer as determined
by NOE experiments; for the major epimer exo-3 NOEs
were observed between C(2)H and C(5)Hendo/C(6)Hendo;
Scheme 5.
Suspecting that the presence of the hydroxyl group in
hydroxyketone 3 was the origin of the problem, we
investigated reversing the sequence of elimination and
hydroxylation steps. In order to examine hydroxylation by
this latter strategy, enone 20 was prepared in 94% yield
from allylated ketone 17 via a smooth (E1cB) elimination
using DBU in MeCN for 48 h (Scheme 6).
Many procedures are known for the oxidation of an
a0,b-unsaturated ketone to an a0-acetoxyenone or
a -hydroxyenone,17 but most of them would likely oxidise
enone 20 from the undesired, less-hindered face (trans to
the allyl side-chain, with the convex facial bias also not
now present). However, hypervalent iodine oxidation
of cyclic, enolisable carbonyl compounds leading to
a-hydroxydimethylacetals is known to proceed via an
oxidation–inversion process, with the hydroxyl group
being introduced on the same same face as bulky
substituents present in b- or g-positions.18 In the event,
when enone 20 was exposed to a solution of KOH in dry
Scheme 6.

9722
D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
Figure 2. Comparison of chiral HPLC data for xialenon A (1) samples.
MeOH at 2158C for 15 min, followed by addition of
iodobenzene diacetate (IBD), a mixture of a0-hydroxydi-
methylacetals 21 was obtained (7:1, as determined by H
Desilylation of hydroxyenone 22 using TBAF in THF gave
(2)-xialenon A (1) {88% yield, [a]2D5¼283.3 (c 0.54,
MeOH) [lit.1 [a]2D5¼2108.0 (c 0.66, MeOH)]}, with data
(1H, 13C, IR, MS, TLC) consistent with the natural
material.1 As the absolute configuration of alcohol (þ)-5
is known,2 the consistency in the sign of the specific rotation
values of synthetic and natural xialenon A (1) confirm the
2R,3R,4S absolute stereochemistry of xialenon A (1)
originally determined by Thiericke et al.1 by bis-esterifica-
tion of xialenon A (1) with the individual enantiomers of
1
NMR analysis of the C(2)H signals). The crude mixture of
a0-hydroxydimethylacetals 21, diluted in CH2Cl2, was
briefly treated with 10% sulfuric acid at 258C to yield,
after column chromatography, 70% of the desired hydroxy-
enone 22 and 10% of the isomer epimeric at the hydroxyl
group, epi-22 (the cis relative stereochemistry between
C(2)H and C(3)H in hydroxyenone 22 was provisionally
assigned from the larger 3J value between C(2)H and C(3)H
(6.1 Hz) relative to that observed for epi-22 (2.5 Hz), and
comparison with 3J between C(2)H and C(3)H (6.1 Hz)
reported for natural xialenon A (1)1 and, ultimately, by
conversion of 22 to 1). This transformation is a rare, efficient
a0-hydroxylation of an a,b-unsaturated ketone using
hypervalent iodine chemistry; usually such a reaction is
complicated by partial g-oxidation (or double bond
dioxygenation).18,19 In the present case, the presence of
both the g-silyloxy group and the g0-allyl side-chain on
opposite faces of enone 20 may be shielding19 the enone
system from such unwanted side reactions.
1
2-phenylbutyric acid and H NMR analysis using the rules
of Helmchen.20 Comparison by chiral HPLC of racemic
xialenon 0 A00 (prepared via desymmetrisation of 6 using
N,N,N0,N ,N -pentamethyldiethylenetriamine as ligand with
BusLi in 70% yield)2 with an authentic sample of the natural
product, and our enantioenriched material indicated that
the latter was of 80% ee (Fig. 2), indicating that no loss of
enantiointegrity had occurred in the transformations from
enantioenriched 5 to 1.
In order to probe the generality of our strategy to other
xialenons we also examined the potential of both epimeric
Scheme 7.

D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
9723
alcohols 3 in an approach towards xialenon E (2).
Introduction of the characteristic 2-hydroxyenone unit of
xialenon E was achieved using Cu(OAc)2·H2O21 in
methanol for 24 h to give hydroxyenone 23 in 80% yield
(Scheme 7). Hydroxyenone 23 was then examined in three
different strategies to try to access xialenon E (2). The first
method consisted in applying the conditions of Cossio et al.
(pyridinium dichromate in combination with chlorotri-
methylsilane)22 for (double) desilylation followed by in
situ oxidation of the liberated secondary alcohol to hope-
fully give xialenon E (2) in one-pot. Unfortunately, these
conditions appeared too drastic and gave complete
decomposition of 23. The second approach consisted of
two separate steps. Deprotection was best accomplished
using HF in MeCN to give diol 24 in 70% yield.
Unfortunately, oxidation once again gave complete
decomposition of the starting diol after only 5 min reaction
time, even using the mild oxidants Dess–Martin period-
inane or TPAP-NMO. Thinking that the free tertiary alcohol
could be a problem in the stability of the substrate, we
decided to attempt selective deprotection to the secondary
alcohol of 23 and perform its oxidation. Fluorosilicic acid is
probably the best reagent when combined with appropriate
solvent to selectively deprotect the secondary ButMe2Si
group.23 Using an improved protocol (cat. fluorosilicic acid
in MeCN/ButOH 9:1)23b no reaction took place. However,
by using 1 equiv. of fluorosilicic acid, a mixture of starting
material, monoprotected alcohol 25 (35% yield) and diol 24
were recovered in a 3:3:1 ratio (1H NMR ratio). This low
selectivity could be due to the comparatively accessible exo-
tertiary ButMe2Si group compared with the more hindered
endo-secondary ButMe2Si group. Unfortunately, further
attempts to oxidise 25 into 26 resulted in complete
decomposition of this material. It is likely that xialenon E
(2) and related structure such as 26 are not stable under the
above conditions. Indeed, it has been observed that, unlike
xialenon A (1), xialenons B, C, D, and E are ‘rapidly
decomposed by the addition of bases or acid’.1
radical, hydrocarbons from CaH2, and alcohols from their
magnesium alkoxides. All reactions were monitored by
TLC using commercially available (Merck or Camlab)
plates, pre-coated with a 0.25 mm layer of silica containing
a fluorescent indicator. Visualisation of reaction com-
ponents was achieved with 254 nm light, and KMnO4 dip.
Organic layers were dried (MgSO4), evaporated with a
Buchi rotary evaporator, followed by drying on a high
vacuum oil pump (,1 mm Hg). Column chromatography
was carried out on Kieselgel 60 (40–63 mm). IR spectra
were recorded as either KBr discs or thin films, using a
Perkin–Elmer 1750 FTIR spectrophotometer. Peak inten-
sities are specified as strong (s), medium (m) or weak (w).
Only selected absorbancies are reported. [a]D values are
given in 1021 deg cm2 g21. 1H NMR spectra of compounds
were recorded in CDCl3 unless otherwise stated, using
a Varian Gemini 200 (200 MHz) or Bruker DPX400
(400 MHz). Chemical shifts (d) are reported relative to
CHCl3 (d 7.27). Coupling constants (J) are given in Hz,
multiplicities are given as multiplet (m), doublet (d), triplet
(t) and quartet (q). 13C NMR spectra were recorded on
the Bruker DPX400 (100 MHz) or a DPX200 (50 MHz).
Chemical shifts are reported relative to CDCl3 (central line
of triplet d 77.0) unless stated otherwise. Mass spectra were
obtained from the EPSRC Mass Spectrometry Service
Centre, Swansea with a Micromass, ZAB-E instrument
or 900 XLT high resolution double focusing mass
spectrometer with tandem ion trap. Alternatively they
were recorded in-house using a VG mass Lab. TRIO1
(GCMS) or Micromass platform APCI spectrometer. All
organolithiums were titrated against N-(2-methylphenyl)di-
methylethyl amide before use. A known amount of the
amide was dissolved in THF (2 mL) and the organolithium
was then added to this stirring solution until a colour change
is observed (clear colourless to bright orange).‘Petroleum
ether’ refers to the fraction boiling in the range 30–408C.
Chiral stationary phase HPLC was performed using a Daicel
Chiralpak AD column (4.6 mm£250 mm) on a Gilson
system with 712 Controller Software and a 118 UV–VIS
dectector set at 254 nm. Chiral gas chromatography was
carried out using a Ce instruments Trace GC (thermoquest)
machine with a CP Chirasil Dex-CD column. Retention
times for major (tRmj) and minor (tRmn) enantiomers are
given in minutes.
3. Conclusion
We have accomplished the first total synthesis of xialenon A
(1) in 12 steps from 1,5-cyclooctadiene via the previously
known alcohol 5. The allyl side-chain has been introduced
with complete stereoselectivity, and a highly stereoselective
a0-hydroxylation of the hindered face of an enone using
hypervalent iodine chemistry has been achieved. This
chemistry reported herein constitutes the first application
of our recently reported enantioselective a-deprotonation
transannular C–H insertion of epoxides in natural product
synthesis.
4.1.1. 5,6-Epoxycyclooctene 8. A solution of peracetic acid
(40% in acetic acid; 21.1 g, 111 mmol) in CH2Cl2 (250 mL)
was added dropwise to a stirred mixture of 1,5-cyclo-
octadiene 7 (10.0 g, 92.6 mmol) and Na2CO3 (58.8 g,
555 mmol) in CH2Cl2 (280 mL) at 08C. The mixture was
stirred at 08C for 6 h. Water (100 mL) was then added and
the mixture filtered. The filtrate was extracted with CH2Cl2
(3£100 mL), washed with brine (100 mL), dried and the
solvent removed at reduced pressure. Rapid filtration
through a plug of silica gave monoepoxide 8 (10.5 g,
92%) as a colourless oil; data as previously reported.2
4. Experimental
4.1. General
4.1.2. (1R p,4S p,5R p,8S p)-9-Oxabicyclo[6.1.0]nonane-4,5-
diol 10, (1R p,2S p,5S p,6S p)-9-oxabicyclo[4.2.1]nonane-
2,5-diol 11 and (1R p,2R p,5R p,6S p)-9-oxabicyclo[3.3.1]-
nonane-2,6-diol 12. OsO4 (4% w/w in H2O; 2 mL,
0.3 mmol) was added to a stirred solution of monoepoxide
8 (10.0 g, 80.6 mmol) and NMO (24.0 g, 205 mmol) in 1:1
All reactions requiring anhydrous conditions were con-
ducted in flame-dried apparatus under an atmosphere of
argon. Syringes and needles for the transfer of reagents were
oven-dried and allowed to cool in a desiccator over P2O5
before use. Ethers were distilled from benzophenone ketyl

9724
D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
THF/H2O (200 mL) at 08C. The solution was allowed to
warm to 258C and stirred for 15 h before being cooled to
08C. Excess Na2S2O4 was added. After filtration the solution
was concentrated under reduced pressure and transferred to
a liquid–liquid extractor and continuously extracted with
EtOAc for 24 h, then with CH2Cl2 for 72 h. The combined
organic phases were dried and evaporated under reduced
pressure. Purification of the residue by column chromato-
graphy (5% MeOH in CH2Cl2) gave the epoxydiol 10 (9.2 g,
73%) as a white solid (data as previously reported)2 and an
inseparable mixture of 9-oxabicyclo[4.2.1]nonane-2,5-diol
11 and 9-oxabicyclo[3.3.1]nonane-2,6-diol 12 (3.2 g, 25%)
as a colourless oil; data for 11 and 12: Rf 0.27 (10% MeOH
in CH2Cl2); nmax(KBr)/cm21 3337s, 2936s, 1037m, 1000m;
dH (400 MHz, CD3OD) 4.32–4.26 (2H, m), 3.92–3.87
(1H, m), 3.80–3.72 (3H, m), 3.64–3.57 (2H, m), 2.25–2.15
(1H, m), 2.07–1.94 (4H, m), 1.92–1.66 (8H, m), 1.58–1.13
(3H, m); dC (100 MHz, CD3OD) 85.0 (CH 11), 81.7
(CH 11), 77.2 (CH 11), 72.6 (CH 11), 72.5 (CH), 70.5
(CH), 68.4 (CH), 68.2 (CH), 30.3 (CH2 11), 28.5 (CH2 11),
27.4 (CH2 11), 26.6 (CH2), 26.5 (CH2), 25.8 (CH2),
22.8 (CH2 11), 17.7 (CH2); m/z (ESþ), 181 (MþNaþ,
100%), (Found MþNaþ 181.0841, C8H14O3Na requires
181.0841).
4.1.5. (1S,3aS,4S,6aS)-3a,4-Bis(tert-butyldimethylsilyl-
oxy)octahydropentalen-1-ol 5. PriLi25 (2.2 M in petroleum
ether, 11.6 mL, 25.5 mmol) was added over 15 min to
a stirred solution of (2)-a-isosparteine·H2O (3.28 g,
13.0 mmol) in 30 mL of ether at 2908C. After stirring for
1 h at this temperature a solution of epoxide 6 (2.0 g,
5.2 mmol) in ether (10 mL) was added over 20 min. The
solution was then stirred at 2908C for a further 6 h, before
being allowed to slowly warm to 258C overnight. The
reaction was cooled to 08C and H3PO4 (0.5 mol dm23
,
50 mL) was added dropwise. The organic layers were
washed with saturated aq. NaHCO3 (2£20 mL) and brine,
dried and the solvent evaporated under reduced pressure.
Column chromatography of the residue (10% Et2O in light
petroleum) gave 5 (1.10 g, 55, 80% ee); [a]2D5¼þ28.2 (c 1.0,
CHCl3), data as previously reported.2
4.1.6. (3aS,4S,6aR)-3a,4-Bis(tert-butyldimethylsilyloxy)-
hexahydro-2H-pentalen-1-one 14. Tetrapropylammonium
perruthenate (5 mol%, 55 mg, 0.15 mmol) was added in one
portion to a stirred mixture of alcohol 5 (1.2 g, 3.1 mmol),
4-methylmorpholine N-oxide (546 mg, 4.67 mmol) and
˚
powdered 4 A molecular sieves (1.6 g) in CH2Cl2 (15 mL)
at 258C. After 2 h the reaction mixture was filtered through
a short pad of silica gel, eluting with CH2Cl2. The filtrate
was evaporated and the residue was purified by column
chromatography (98:2 petroleum ether/Et2O) to afford 14
(1.17 g, 97%); [a]D25¼220.6 (c 1.0, CHCl3); Rf (10% Et2O/
petroleum ether) 0.35; nmax(thin film)/cm21 2955s, 2930s,
1754s, 1472m, 1258s, 1166s, 1107m, 1061s, 835s, 776s; dH
(400 MHz) 3.93–3.91 (1H, m, C(4)H), 2.53–2.34 (4H, m),
2.02–1.92 (4H, m), 1.60–1.57 (1H, m), 0.86 (9H, s, CMe3),
0.84 (9H, s, CMe3), 0.11 (6H, s, SiMe), 0.07 (3H, s, SiMe),
0.05 (3H, s, SiMe); dC (100 MHz) 218.8 (CvO), 91.4
(COSi), 81.8 (CHOSi), 58.7 (CHCO), 39.6, 34.0, 29.9, 25.7
(CMe3), 25.6 (CMe3), 23.8, 17.8 (CMe3), 22.5 (SiMe),
22.6 (SiMe), 24.6 (SiMe), 24.7 (SiMe); m/z (CIþ), 402
(MþNHþ4 , 100%), 327 (10), 270 (20), 253 (25), 138 (65),
and 121 (30) (Found MþNHþ4 402.2855, C20H44NO3Si2
requires 402.2860).
4.1.3. (1R p,2S p,5S p,6S p)-5-Hydroxy-9-oxabicyclo[4.2.1]-
non-2-yl 3,5-dinitrobenzoate 13.24 3,5-Dinitrobenzoyl
chloride (6.4 g, 27.8 mmol) was added to a stirred solution
of 11 and 12 (2.0 g, 12.7 mmol), Et3N (8.8 mL, 63.3 mmol)
and imidazole (185 mg, 2.7 mmol) in CH2Cl2 (50 mL) at
258C. After 24 h the reaction mixture was washed with
water (50 mL) and extracted with CH2Cl2. The combined
organic layers were washed with saturated aq. Na2CO3
(50 mL) and brine (30 mL). The organic layer was dried and
evaporated to leave a yellow residue. Purification of the
residue by column chromatography (1% MeOH in CH2Cl2)
gave mixture of mono-benzylated alcohols (2.45 g, 55%) as
a yellow solid which was recrystallised (DCM:hexane) to
give 13 as white crystals; mp 148–1508C; Rf 0.57 (10%
MeOH:CH2Cl2); nmax(KBr)/cm21 3247w, 2938w, 1725m,
1544s, 1345s, 1280s; dH (400 MHz,) 9.23 (1H, t, J¼2 Hz),
9.17 (2H, d, J¼2 Hz), 5.06 (1H, dd, J¼10, 6 Hz, C(2)H,),
4.57–450 (1H, m), 4.10–4.00 (3H, m), 2.45–2.35 (1H, m),
2.33–2.21 (1H, m), 2.19–2.05. (2H, m), 2.00–1.85 (2H, m),
1.75–1.47 (2H, m), dC (100 MHz) 162.0 (CvO), 148.6
(Ar), 134.0 (Ar), 129.5 (Ar), 122.4 (Ar), 82.4 (CHO), 81.8
(CHO), 81.4 (CHO), 72.2 (CHO), 30.6 (CH2), 27.5 (CH2),
24.8 (CH2), 22.7 (CH2); m/z (CIþ), 370 (MþNH4þ, 100%),
356 (60) (Found MþNHþ4 370.1245, C15H20O8N3 requires
370.1235).
4.1.7. (3aS,4S,6aR)-3a,4-Bis(tert-butyldimethylsilyloxy)-
4,5,6,6a-tetrahydro-3aH-pentalen-1-one 4. BunLi (1.5 M
in hexanes, 2.10 mL, 3.15 mmol) was added dropwise to a
solution of N,N-diisopropylamine (0.45 mL, 3.12 mmol) in
THF (10 mL) at 08C. After 30 min, the solution was cooled
to 2788C and a solution of ketone 14 (600 mg, 1.56 mmol)
in THF (3 mL) was added dropwise. After 30 min Me3SiCl
(0.39 mL, 3.12 mmol) was added dropwise. After 1 h at
2788C, the reaction was allowed to warm to 2258C, then
diluted with Et2O before addition of water and extraction
with Et2O (3£50 mL). The combined organic phases were
dried and concentrated to leave a residue, which was passed
through a short pad of Florisil (eluting with petroleum
ether). The resulting crude silyl enol ether was rapidly used
in the next step. Pd(OAc)2 (330 mg, 1.56 mmol) was added
to a solution of the crude silyl enol ether in dry MeCN
(6 mL) and the mixture was stirred overnight at room
temperature. The solvent was evaporated and the black
residue was filtered through a pad of silica gel (eluting with
CH2Cl2). The filtrate was washed with saturated aq.
NaHCO3 to remove traces of acetic acid. The organic
phase was washed with brine, dried and evaporated under
4.1.4. (1R p,4S p,5R p,8S p)-4,5-Bis(tert-butyldimethylsilyl-
oxy)-9-oxabicyclo[6.1.0]nonane 6.24 Epoxydiol 10 (4.1 g,
26 mmol), ButMe2SiCl (9.9 g, 65 mmol) and imidazole
(33.5 g, 492 mmol) were stirred at 258C in DMF (15 mL)
for 18 h. The solution was diluted with water (200 mL) and
CH2Cl2 (200 mL). The aqueous layer was extracted with
CH2Cl2 (4£CH2Cl2), dried, and evaporated under reduced
pressure to give a clear oil. Purification of the residue by
column chromatography (10% Et2O in light petroleum)
gave the epoxide 6 (10 g, 98%) as a clear oil which
solidified on standing in the refrigerator over several days;
data as previously reported.2

D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
9725
reduced pressure. The residue was purified by column
chromatography (98:2 petroleum ether/Et2O) to afford 4
(480 mg, 80%) as a colourless oil; [a]2D5¼227.8 (c 1.0,
CHCl3); Rf (10% Et2O/petroleum ether) 0.35; nmax(thin
film)/cm21 2955s, 2930s, 1706s, 1471w, 1245s, 912w,
738w; dH (400 MHz) 7.48 (1H, d, J¼6 Hz, CHvCHCO),
6.24 (1H, d, J¼6 Hz, CHvCHCO), 4.17 (1H, dd, J¼10,
6 Hz, C(4)H), 2.44 (1H, dd, J¼19, 9 Hz, C(6a)H), 1.90–
1.70 (3H, m), 1.28–1.18 (1H, m), 0.86 (9H, s, CMe3), 0.87
(9H, s, CMe3), 0.10 (3H, s, SiMe), 0.04 (3H, s, SiMe), 0.04
(3H, s, SiMe) and 20.01 (3H, s, SiMe); dC (100 MHz) 210.1
(CvO), 164.0, (CHvCHCO), 134.8 (CHvCHCO), 90.8
(COSi), 81.9 (CHOSi), 55.1 (CHCO), 31.7, 25.7 (CMe3),
25.6 (CMe3), 22.8, 18.0 (CMe3), 17.8 (CMe3), 22.4 (SiMe),
23.0 (SiMe), 24.7 (SiMe) and 24.8 (SiMe); m/z (CIþ)
400.3 (40%, MþNH4þ), 383 (60, MHþ), 270 (20, MþNHþ4 –
TBSO) and 121 (100) (Found (ESþ) MþHþ 383.2443,
C20H39O3Si2 requires 383.2438).
2.77 (1H, dm, J¼14 Hz, allylic CH2), 2.57 (1H, dd, J¼11,
3 Hz, C(6a)H), 2.46 (1H, dd, J¼16, 7 Hz, C(2)Hendo), 2.32
(1H, m, C(3)H), 2.22 (1H, ddd, J¼16, 11, 1 Hz, C(2)Hexo),
2.06 (1H, dm, J¼14 Hz, allylic CH2), 1.84 (3H, m), 1.44
(1H, m), 0.91 (9H, s, CMe3), 0.90 (9H, s, CMe3), 0.14 (3H, s,
SiMe), 0.10 (6H, s, 2£SiMe) and 0.06 (3H, s, SiMe); dC
(100 MHz) 218.9 (CvO), 138.0 (CvC), 115.3 (CvC),
90.3 (COSi), 82.3 (CHOSi), 57.9 (CHCO), 44.4 (homo-
allylic CH), 40.0, 35.8, 33.0, 25.9 (CMe3), 25.8 (CMe3),
22.4, 18.4 (CMe3), 17.9 (CMe3), 22.5 (SiMe), 22.5 (SiMe),
24.2 (SiMe) and 24.8 (SiMe); m/z (CIþ) 442 (100%,
MNHþ4 ), 293 (80) (Found (ESþ) MþNHþ4 442.3177,
C23H48NO3Si2 requires 442.3172).
4.1.9. (3S,3aS,4S,6aR)-3-Allyl-3a,4-bis(tert-butyldi-
methylsilyloxy)hexahydro-2H-pentalen-1-one 17. CuI
(776 mg, 4.10 mmol) and anhydrous LiBr (358 mg,
4.10 mmol) were placed in a 3-neck 25 mL round-bottom
flask. While stirring the inorganic reagents, the flask was
evacuated and heated for five minutes then flushed with
argon (3 times). THF (10 mL) was injected, and the mixture
was stirred for 5 min to yield a yellow homogeneous
solution, which was cooled to 2788C. Allylmagnesium
bromide (1.0 M in Et2O, 3.58 mL, 3.58 mmol) was added
dropwise to the yellow solution at 2788C to yield a tan
solution. TMSCl (500 mL, 3.96 mmol) was added followed
immediately by the addition of enone 4 (500 mg,
1.27 mmol) in THF (3 mL). After 30 min at 2788C the
reaction mixture was hydrolysed with water (15 mL) then
allowed to quickly warm to 258C over 30 min, followed by
the addition of saturated aq. NH4Cl solution (30 mL).
Extraction with ether was followed by combining the
organic layers and drying. The solvent was removed in
vacuo, and the resulting residue was subjected to flash
chromatography (96:4 petroleum ether/Et2O) to yield
allylated ketone 17 (333 mg, 60%); data as above.
4.1.8. (1S,3aS,4S,6aS)-4-Allyl-4,6a-bis(tert-butyldi-
methylsilyloxy)-1,2,3,3a,4,6a-hexahydropentalen-1-ol 16
and (3S,3aS,4S,6aR)-3-allyl-3a,4-bis(tert-butyldimethyl-
silyloxy)hexahydro-2H-pentalen-1-one 17. BunLi (1.3 M
solution in hexanes, 261 mL, 0.34 mmol) was added to a
solution of allyltributyltin (113 mg, 0.34 mmol) in THF
(2 mL) at 2788C. The mixture was stirred at 2788C for 1 h
before addition of enone 4 (108 mg, 0.28 mmol) in THF
(1 mL). After a further 3 h saturated aq. NH4Cl (5 mL) was
added. The mixture was diluted with water (5 mL) and ether
(10 mL) and the layers separated. The aqueous layer was
extracted with Et2O (5 mL) and the combined organic
fractions were dried and evaporated under reduced pressure.
Purification of the residue by column chromatography (39:1
petroleum ether/Et2O) gave allylic alcohol 15 (19 mg, 17%)
and allylic ether 16 (41 mg, 36%) as colourless oils. Allylic
alcohol 15 (19 mg, 0.045 mmol) was then added to a solu-
tion of KH (20 mg, 0.50 mmol) and 18-crown-6 (150 mg,
0.50 mmol) in THF (5 mL). After 16 h at 258C the reaction
mixture was hydrolysed with water (15 mL) and extracted
with ether (3£15 mL). The combined organic layers were
dried and evaporated under reduced pressure. Purification
of the residue by column chromatography (39:1 petroleum
ether/Et2O) gave allylated ketone 17 (19 mg, 100%); data
for allylic ether 16: Rf (3% Et2O/petroleum ether) 0.29;
4.1.10. (2R,3S,3aS,4S,6aR)-3-Allyl-3a,4-bis(tert-butyldi-
methylsilyloxy)hexahydro-2-hydroxy-2H-pentalen-1-
one exo-3 and (2S,3S,3aS,4S,6aR)-3-allyl-3a,4-bis(tert-
butyldimethylsilyloxy)hexahydro-2-hydroxy-2H-penta-
len-1-one endo-3. The exact procedure (and scale) of
Section 4.1.9 above was followed until work-up. After
30 min at 2788C the reaction was quenched with saturated
aq. NH4Cl solution (20 mL). Extraction with Et2O
(3£20 mL) was followed by combining the organic layers
and drying. The solvent was removed in vacuo, and the
resulting residue was passed through a short pad of Florisil
(eluting with petroleum ether) to yield silyl enol ether 18
(390 mg, 60%). The silyl enol ether was then rapidly used in
the next step. To a mixture of KHCO3 (84 mg, 0.84 mmol)
and m-CPBA (50 mg, 0.20 mmol) in hexanes (3 mL) at
2158C was added silyl enol ether 18 (84 mg, 0.17 mmol)
in hexanes (2 mL) at 2158C. The mixture was allowed to
warm to 258C over 2 h, then filtered, extracted with Et2O
(30 mL) and washed with a saturated aq. Na2SO3 (10 mL),
saturated aq. NaHCO3 (10 mL) and dried. Column chroma-
tography of the residue gave the a-hydroxy ketones exo-3
(48 mg, 64%) and endo-3 (16 mg, 21%) as colourless oils;
data for exo-3: Rf 0.55 (50% Et2O/petroleum ether);
n
max(thin film)/cm21 3464w, 2956s, 2930s, 1641s, 1258s,
1086w, 837s, 776s; dH (200 MHz) 5.90–5.80 (1H, m,
CHvCH2), 5.85 (1H, d, J¼6 Hz, CHvCH), 5.69 (1H, d,
J¼6 Hz, CHvCH), 5.18–5.08 (2H, m, CHvCH2), 3.75
(1H, d, J¼3 Hz, C(1)H), 3.35 (1H, brs, OH), 2.50–2.15 (4H,
m), 2.05–1.95 (1H, m), 1.80 (1H, m), 1.75–1.65 (1H, m),
0.89 (9H, s, CMe3), 0.87 (9H, s, CMe3), 0.09 (3H, s, SiMe),
0.07 (9H, s, 3£SiMe); dC (50 MHz) 139.9 (CvC), 134.4
(CvC), 133.7 (CvC), 117.4 (CvC), 99.4 (CO), 80.9 (CO),
79.1 (CO), 57.7, 44.3, 36.8, 25.8 (CMe3), 25.6 (CMe3), 20.0,
18.1 (CMe3), 17.9 (CMe3), 22.5 (SiMe), 22.8 (SiMe) and
24.8 (2£SiMe), m/z (CIþ) m/z 425 (100%, MHþ) and 407
(MHþ218); (Found MþHþ 425.2906, C23H45O3Si2
requires 425.2907); data for allylated ketone 17: [a]2D5¼
þ22.0 (c 1.0, CHCl3), Rf (10% Et2O/petroleum ether) 0.35;
n
max(thin film)/cm21 2957m, 2254m, 1734m, 1472m, 911s;
742s; dH (400 MHz) 5.79 (1H, dddd, J¼17, 10, 8, 7 Hz,
CHvCH2), 5.03 (1H, dm, J¼17 Hz, CHvCH2), 4.97 (1H,
dm, J¼10 Hz, CHvCH2), 4.16 (1H, dd, J¼9, 6 Hz, C(4)H),
n
max(thin film)/cm21 3460s, 2969s, 2933s, 1746s, 1641m,
1361m, 1257w; dH (400 MHz, D2O in CDCl3) 5.98–5.83
(1H, m, CHvCH2), 5.13 (1H, dd, J¼17, 2 Hz, CHvCH2),

9726
D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
5.02 (1H, dd, J¼10, 2 Hz, CHvCH2), 4.22 (1H, dd, J¼8,
5 Hz, C(4)H), 4.13 (1H, d, J¼5 Hz, C(2)H), 2.74 (1H, dd,
J¼10, 4 Hz, C(6a)H), 2.63–2.53 (1H, m, allylic CH2),
2.42–.33 (2H, m, allylic CH2 and C(3)H), 2.00–1.90 (2H,
m, C(5)H exo, C(6)H exo), 1.89–1.80 (1H, m, C(6)H endo),
1.45–1.33 (1H, m, C(5)H endo), 0.91 (9H, s, CMe3), 0.90
(9H, s, CMe3), 0.18 (3H, s, SiMe), 0.16 (3H, s, SiMe), 0.12
(3H, s, SiMe), 0.07 (3H, s, SiMe); dC (100 MHz) 216.6
(CvO), 137.4 (CHvCH2), 116.3 (CHvCH2), 90.0 (COSi),
82.4 (CHOSi), 76.6 (CHOH), 55.8 (CHCO), 44.5 (homo-
allylic CH), 33.7 (CH2CHOSi), 28.3 (allylic CH2), 25.8
(CMe3), 25.8 (CMe3), 23.0 (CH2), 18.3 (CMe3), 17.9
(CMe3), 22.3 (SiMe), 22.4 (SiMe), 24.1 (SiMe), 24.8
(SiMe); m/z (CIþ) 458 (100%, M þNHþ4 ), (Found (ESþ)
MþNHþ4 458.3126, C23H48NO4Si2 requires 458.3122); data
for endo-3: Rf 0.39 (50% Et2O/petroleum ether); nmax(thin
film)/cm21 3480w, 2956s, 2938s, 1749s, 1362w, 1096s,
836s, 775s; dH (400 MHz, D2O in CDCl3) 6.09 (1H, dddd,
J¼17, 10, 8, 6 Hz, CHvCH2), 5.17 (1H, dm, J¼17 Hz,
CHvCH2), 5.06 (1H, dm, J¼8 Hz, CHvCH2), 4.37 (1H,
dd, J¼13, 3 Hz, C(2)H), 4.13 (1H, dd, J¼11, 6 Hz, C(4)H),
2.90–2.85 (1H, m, allylic CH2), 2.68–2.63 (1H, m,
C(6a)H), 2.47–2.35 (1H, m, allylic CH2), 2.09–1.80 (4H,
m, C(3)H, C(6)H exo and endo, C(5)H exo), 1.36–1.26 (1H,
m, C(H)5 endo), 0.94 (9H, s, CMe3), 0.91 (9H, s, CMe3),
0.14 (3H, s, SiMe), 0.12 (3H, s, SiMe), 0.10 (3H, s, SiMe),
0.08 (3H, s, SiMe); dC (100 MHz) 216.0 (CvO), 138.5
(CHvCH2), 115.4 (CHvCH2), 86.1 (COSi), 82.8
(CHOSi), 80.5 (CHOH), 54.1 (CHCO), 46.9 (homoallylic
CH), 35.1 (allylic CH2), 32.9 (CH2CHOSi), 25.8 (2£CMe3),
21.6 (CH2), 18.3 (CMe3), 17.9 (CMe3), 22.5 (SiMe), 22.6
(SiMe), 24.1 (SiMe), 24.8 (SiMe); m/z (CIþ) 458
(100%, MþNHþ4 ), (Found (ESþ) MþNHþ4 458.3127,
C23H48NO4Si2 requires 458.3122).
0.40 (50% Et2O/petroleum ether); nmax(thin film)/cm21
2956s, 2936s, 1704s, 1253m, 1098m, 836m, 775m; dH
(400 MHz) 5.71 (1H, ddt, J¼17, 10, 7 Hz, CHvCH2),
5.06 (1H, dm, J¼17 Hz, CHvCH2), 5.06–5.04 (1H, m,
CHvCH2), 5.02 (1H, br t, J¼6 Hz, C(4)H), 3.08–3.00
(1H, m, C(2)H), 2.85 (1H, dd, J¼19, 6 Hz, allylic CH2),
2.68–2.60 (1H, m, C(3)H), 2.50–2.40 (2H, m), 2.41 (1H,
dd, J¼19, 2 Hz, allylic CH2), 2.28–2.19 (1H, m), 2.15–2.05
(2H, m), 0.91 (9H, s, CMe3), 0.12 (3H, s, SiMe), 0.11 (3H, s,
SiMe); dC (100 MHz) 203.5 (CvO), 186.8 (CvCCHOSi),
148.9
(CvCCHOSi),
134.9
(CHvCH2),
117.4
(CHvCH2), 73.0 (CHOSi), 46.3, 38.9, 36.6, 34.7 (homo-
allylic CH), 25.7 (CMe3), 22.8, 18.0 (CMe3), 24.4 (SiMe),
24.9 (SiMe); m/z (CIþ) 161 (100%, M231), 293 (80,
MHþ); (Found (ESþ) MþHþ 293.1934, C17H29O2Si
requires 293.1937).
4.1.13. (2R,3R,4S)-3-Allyl-4-(tert-butyldimethylsilyloxy)-
3,4,5,6-tetrahydro-2-hydroxy-2H-pentalen-1-one 22 and
(2S,3R,4S)-3-allyl-4-(tert-butyldimethylsilyloxy)-3,4,5,6-
tetrahydro-2-hydroxy-2H-pentalen-1-one epi-22. KOH
(432 mg, 7.68 mmol) in dry MeOH (4 mL) was stirred
until dissolved, then the solution was cooled to 2158C and a
solution of 20 (150 mg, 0.51 mmol) in MeOH (2 mL) was
added. After 15 min, PhI(OAc)2 (249 mg, 0.78 mmol) was
added in one portion. The reaction mixture was stirred at
2158C for 1 h and then at 258C for 2 h and concentrated
under reduced pressure to give the crude a0-hydroxy-
dimethylacetals 21 [7:1 mixture as determined by 1H NMR
of C(2)H, for 21 dH (400 MHz) 4.40 (d, J¼7 Hz) and for
epi-21 dH 4.01 (d, J¼5 Hz)]. [In a separate experiment
column chromatography (8:2 petroleum ether/Et2O) gave
separable 21 and epi-21, as well as compounds 22. While 21
was stable enough to be characterised by NMR, epi-21 was
almost immediately hydrolysed in CDCl3. Data for 21: dH
(250 MHz, CDCl3 in D2O) 5.97–5.80 (1H, m, CHvCH2),
5.07 (1H, dm, J¼17 Hz, CHvCH2), 5.02–4.96 (1H, brd,
J¼10 Hz, CHvCH2), 4.89–485 (1H, m, C(4)H), 4.42 (1H,
d, J¼7 Hz, C(2)H), 3.40 (3H, s, OMe), 3.35 (3H, s, OMe),
2.86–2.76 (1H, m), 2.76–2.23 (4H, m), 2.05–1.87 (2H, m),
0.90 (9H, s, CMe3), 0.09 (3H, s, SiMe), 0.08 (3H, s,
SiMe); dC (63 MHz) 155.0 (CvC), 143.9 (CvC), 138.0
(CHvCH2), 115.4 (CHvCH2), 80.3 (CHO), 74.1 (CHO),
50.6 (OMe), 50.2 (OMe), 41.2 (homoallylic CH), 38.0, 33.8,
27.3, 25.8 (CMe3), 18.1 (CMe3), 24.3 (SiMe), 4.7 (SiMe)].
Crude 21 was dissolved in CH2Cl2 and treated with 10%
sulphuric acid at 258C for 10 min. After neutralisation with
saturated aq. NaHCO3 and extraction with ether, the
combined organic layers were dried and removed under
vacuum. Flash chromatography (7:3 petroleum ether/Et2O)
gave 22 (110 mg, 70%) and epi-22 (16 mg, 10%); data for
22: [a]2D5¼249.6 (c 1.0, CHCl3), Rf 0.24 (50% Et2O/
petroleum ether); nmax(thin film)/cm21 3424m, 2954s,
2930s, 1713s, 1638m, 1471m, 1255s, 837s, 776m; dH
(400 MHz) 5.75 (1H, ddt, J¼17, 10, 7 Hz, CHvCH2), 5.06
(1H, dm, J¼17 Hz, CHvCH2), 5.00 (1H, dm, J¼10 Hz,
CHvCH2), 4.96 (1H, m, C(4)H), 4.51 (1H, dd, J¼6, 3 Hz,
C(2)H), 3.28–3.21 (1H, m, C(3)H), 2.95 (1H, d, J¼3 Hz,
OH), 2.64–2.56 (1H, m), 2.54–2.42 (2H, m), 2.35–2.23
(2H, m), 2.11–2.02 (1H, m), 0.92 (9H, s, CMe3), 0.14
(3H, s, SiMe), 0.12 (3H, s, SiMe); dC (100 MHz) 204.0
(CvO), 184.0 (CvCCHOSi), 144.7 (CvCCHOSi), 136.1
(CHvCH2), 116.9 (CHvCH2), 78.9 (CHOH), 73.5
4.1.11. (4S)-4-(tert-Butyldimethylsilyloxy)-3,4,5,6-tetra-
hydro-2H-pentalen-1-one 19. DBU (117 mL, 0.78 mmol)
was added to a stirred solution of ketone 14 (150 mg,
0.39 mmol) in anhydrous MeCN (3.0 mL), and stirred at
258C for 2 days. The solvent was evaporated and the crude
mixture was subjected to flash chromatography (7:3
petroleum ether/Et2O) to give the enone 19 (88 mg, 90%)
as a colourless oil; Rf 0.19 (50% Et2O/petroleum ether);
n
max(thin film)/cm21 2954s, 2938s, 1705s, 1257m, 1096m,
838m, 775m; dH (200 MHz) 5.00–4.93 (1H, m, C(4)H),
2.77–2.60 (4H, m), 2.57–2.31 (2H, m), 2.30–2.08 (2H, m),
0.92 (9H, s, CMe3), 0.12 (3H, s, SiMe), 0.11 (3H, s, SiMe);
dC (50 MHz) 204.6 (CvO), 184.8 (CvCCHOSi), 148.9
(CvCCHOSi), 74.7 (CHOSi), 40.8, 38.9, 25.8 (CMe3),
23.3, 22.8, 18.3 (CMe3), 24.6 (SiMe), 24.8 (SiMe); m/z
(CIþ) 140 (100%), 123 (70), 270 (40), 253 (30); (Found
(ESþ) MþHþ 253.1622, C14H25O2Si requires 253.1624);
the ee (80%) was determined by chiral GC (100–1408C,
18C min21, 1 mL min21) tRmj, 116.0; tRmn, 121.4.
4.1.12. (3S,4S)-3-Allyl-4-(tert-butyldimethylsilyloxy)-
3,4,5,6-tetrahydro-2H-pentalen-1-one 20. DBU (152 mL,
1.02 mmol) was added to a stirred solution of ketone 17
(288 mg, 0.69 mmol) in anhydrous MeCN (3.0 mL), and
stirred at 258C for 2 days. The solvent was evaporated and
the crude mixture was subjected to flash chromatography
(8:2 petroleum ether/Et2O) to give the enone 20 (190 mg,
94%) as a colourless oil. [a]2D5¼235.3 (c 1.0, CHCl3); Rf

D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
9727
(CHOSi), 40.0 (homoallylic CH), 37.9, 33.4, 25.7 (CMe3),
23.1, 18.1 (CMe3), 24.4 (SiMe), 24.8 (SiMe); m/z (CIþ)
149 (100%), 309 (80); 131 (65), (Found MþHþ 309.1883,
C17H29SiO3 requires 309.1886); data for epi-22: [a]2D5¼
264.0 (c 1.0, CHCl3), Rf 0.18 (50% Et2O/petroleum ether);
(9H, s, CMe3), 0.85 (9H, s, CMe3), 0.08 (3H, s, SiMe), 0.05
(3H, s, 2£SiMe), 0.03 (3H, s, SiMe); dC (100 MHz)
202.0 (CvO), 149.9 (COHvC), 143.5 (COHvC), 134.0
(CHvCH2), 116.4 (CHvCH2), 88.2 (COSi), 81.4 (CHOSi),
53.8 (CHCO), 32.5 (CH2), 31.5 (CH2), 25.9 (CMe3), 25.8
(CMe3), 22.1 (CH2), 18.2 (CMe3), 18.0 (CMe3), 23.1
(SiMe), 23.2 (SiMe), 24.5 (SiMe), 24.9 (SiMe).
n
max(thin film)/cm21 3428w, 2930s, 2857s, 1713s, 1472w,
1340w, 1257s, 838s, 776s; dH (400 MHz) 5.82 (1H, ddt,
J¼17, 10, 7 Hz, CHvCH2), 5.19 (1H, dm, J¼17 Hz,
CHvCH2), 5.13 (1H, dm, J¼10 Hz, CHvCH2), 5.07
(1H, dm, J¼6 Hz, C(4)H), 4.15 (1H, d, J¼3 Hz, C(2)H),
2.97–2.90 (1H, m, C(3)H), 2.95 (1H, br s, OH), 2.66–2.59
(2H, m), 2.57–2.51 (1H, m), 2.34–2.25 (2H, m), 2.12–2.06
(1H, m), 0.91 (9H, s, CMe3), 0.13 (3H, s, SiMe), 0.11 (3H, s,
SiMe). dC (100 MHz) 203.2 (CvO), 184.2 (CvCCHOSi),
145.8 (CvCCHOSi), 135.1 (CHvCH2), 118.3 (CHvCH2),
82.4 (CHOH), 74.1 (CHOSi), 44.7 (CH), 38.3, 34.7, 26.1
(CMe3), 23.8, 18.4 (CMe3), 23.8 (SiMe), 24.5 (SiMe); m/z
(CIþ) 148 (100%), 70 (35), 131 (20); (Found MþHþ
309.1882, C17H29SiO3 requires 309.1886).
4.1.16. (3aR,4S,6aR)-3-Allyl-4,5,6,6a-tetrahydro-2,3a,4-
trihydroxy-3aH-pentalen-1-one 24. HF (40 % w/w H2O;
1 mL, 22.0 mmol) was added to a stirred solution of 23
(35 mg, 0.08 mmol) in MeCN (3 mL) at 258C. After 1 h the
reaction was neutralised with a saturated aq. NaHCO3, and
after filtration the solvent was evaporated under reduced
pressure. Flash column chromatography (1:20 MeOH/
CH2Cl2) gave 24 (12 mg, 70%); Rf 0.54 (10% Et2O/
methanol); dH (400 MHz, CD3OD) 6.07 (1H, ddt, J¼17, 10,
7 Hz, CHvCH2), 5.16 (1H, dm, J¼17 Hz, CHvCH2), 5.01
(1H, br dq, J¼10 Hz, CHvCH2), 4.10 (1H, dd, J¼7, 5 Hz,
C(4)H), 3.30–3.20 (2H, m, allylic CH2), 2.43 (1H, dd, J¼9,
4 Hz, C(6a)H), 2.02–1.94 (1H, m), 1.92–1.83 (1H, m),
1.77–1.70 (1H, m), 1.55–1.47 (1H, m); dC (100 MHz,
CD3OD) 203.0 (CvO), 151.1(COHvC), 143.5 (COHvC),
135.0 (CHvCH2), 115.1 (CHvCH2), 86.3 (COH), 78.9
(CHOH), 56.4 (CHCO), 33.1 (CH2), 30.8 (CH2), 22.1(CH2).
4.1.14. (2R,3R,4S)-3-Allyl-3,4,5,6-tetrahydro-2,4-dihy-
droxy-2H-pentalen-1-one, xialenon A (1). To a solution
of 22 (85 mg, 0.28 mmol) in THF (10 mL) was added TBAF
(1.0 M solution in THF, 0.50 mL, 0.50 mmol) at 08C. After
2 h, the reaction mixture was quenched with water (10 mL)
and extracted with ether (3£25 mL). The combined organic
phases were dried and the solvent was removed under
vacuum. The residue was subjected to flash column
chromatography (1:3 petroleum ether/Et2O) to give xial-
enon A (1) (49 mg, 88%) as a colourless oil; [a]2D5¼283.3
(c 0.54, MeOH) [lit.1 [a]D25¼2108.0 (c 0.66, MeOH)], Rf
0.33 (10% MeOH/CH2Cl2); nmax(thin film)/cm21 3368s,
2934w, 2860w, 1699s, 1630m, 1438w, 1050s, 920s; dH
(400 MHz) 5.83 (1H, dddd, J¼17, 10, 8, 7 Hz, CHvCH2),
5.12 (1H, dq, J¼17, 2 Hz, CHvCH2), 5.06 (1H, dm, J¼
10 Hz, CHvCH2), 5.03 (1H, br dd, J¼6 Hz, C(4)H), 4.50
(1H, d, J¼6 Hz, C(2)H), 3.34–3.29 (1H, m, C(3)H), 2.79
(1H, br s, OH), 2.74–2.66 (1H, m, C(5)H), 2.63–2.55 (1H,
m), 2.54–2.49 (1H, m), 2.36 (1H, dddd, J¼16, 6, 3, 2 Hz,
C(6)H), 2.20–2.07 (2H, m); dC (100 MHz) 204.0
(CvO), 183.7 (CvCCHOH), 145.5 (CvCCHOH), 136.2
(CHvCH2), 117.2 (CHvCH2), 78.6 ((CCO), 73.3 (CHOH),
40.0 (CH), 37.2, 33.4, 23.1; m/z (CIþ) 195 (100%, MþHþ),
(Found MþHþ 195.1027, C11H15O3 requires 195.1021); the
ee (80%) was determined by chiral HPLC (50:50 EtOH/
heptane, 0.1 mL min21) tRmj, 52.1; tRmn, 43.7.
4.1.17. (3aR,4S,6aR)-3-Allyl-3a-(tert-butyldimethylsilyl-
oxy)-4,5,6,6a-tetrahydro-2,4-dihydroxy-3aH-pentalen-1-
one 25. H2SiF6 (20–25% w/w in H2O; 100 mL, 0.22 mmol)
was added to a stirred solution of 23 (39 mg, 0.09 mmol)
in MeCN/tert-butanol (9:1, 3 mL). After 24 h the reaction
was neutralised with saturated aq. NaHCO3 (10 mL) and
extracted with ether. The organic layer was dried and
concentrated to leave a residue. Flash column chroma-
tography (1:1 petroleum ether/Et2O) gave 24 (2 mg, 12%),
25 (10 mg, 35%) and starting material 23 (14 mg, 35%);
data for 25: Rf 0.25 (1:1 petrol/ether); dH (400 MHz, CDCl3)
6.11 (1H, dddd, J¼17, 10, 8, 7 Hz, CHvCH2), 5.27 (1H,
dq, J¼17, 2 Hz, CHvCH2), 5.18 (1H, dq, J¼10, 2 Hz,
CHvCH2), 4.05 (1H, br dd, J¼5 Hz, C(4)H), 3.33 (1H,
ddm, J¼15, 6 Hz, allylic CH2), 3.15 (1H, ddm, J¼15, 8 Hz,
allylic CH2), 2.70 (1H, dd, J¼10, 5 Hz, C(6a)H), 2.10–2.02
(1H, m), 1.98–1.82 (2H, m), 1.72–1.65 (1H, m), 0.91 (9H,
s, CMe3), 0.09 (3H, s, SiMe), 0.06 (3H, s, SiMe).
Acknowledgements
4.1.15. (3aR,4S,6aR)-3-Allyl-3a,4-bis(tert-butyldimethyl-
silyloxy)-4,5,6,6a-tetrahydro-2-hydroxy-3aH-pentalen-
1-one 23. Cu(OAc)2·H2O (37 mg, 0.19 mmol) was added in
one portion to a solution of 3 (41 mg, 0.09 mmol) in MeOH
(3 mL) at 258C. After being stirred for 24 h, the reaction was
hydrolysed with water (10 mL) and extract with ether
(3£15 mL). The combined organic layers were dried and
concentrated to leave a residue. Flash column chroma-
tography (7:3 petroleum ether/Et2O) gave hydroxyenone 23
(33 mg, 80%) as an oil; Rf 0.51 (50% Et2O/petroleum
ether); dH (400 MHz) 6.08 (1H, dddd, J¼17, 10, 7, 6 Hz,
CHvCH2), 6.00 (1H, s, OH), 5.18 (1H, dm, J¼17 Hz,
CHvCH2), 5.09 (1H, dq, J¼10, 2 Hz, CHvCH2), 4.08–
4.05 (1H, m, C(4)H), 3.26 (1H, ddt, J¼17, 6, 2 Hz, allylic
CH2), 3.18 (1H, ddt, J¼17, 7, 1 Hz, allylic CH2), 2.62 (1H,
br dd, C(6a)H), 1.87–1.76 (3H, m), 1.49–1.36 (1H, m), 0.89
We thank the EPSRC for Research Grants (GR/L39940 and
GR/N26320), and the European Community for a Marie
Curie Fellowship (HPMF-CT-2001-01274 to J.-M. G.).
We also thank E. Gras for X-ray analyses of 6 and 13, the
EPSRC National Mass Spectrometry Service Centre for
¨
mass spectra and I. Sattler and R. Thiericke (Hans-Knoll-
Institute for Natural Products Research) for a gift of natural
xialenon A (1) and spectral data.
References
1. Tang, Y.-Q.; Sattler, I.; Thiericke, R.; Grabley, S.; Feng, X.-Z.
Eur. J. Org. Chem. 2000, 2401–2406.

9728
D. M. Hodgson et al. / Tetrahedron 59 (2003) 9719–9728
2. Hodgson, D. M.; Cameron, I. D.; Christlieb, M.; Green, R.;
Lee, G. P.; Robinson, L. A. J. Chem. Soc., Perkin Trans. 1
2001, 2161–2174.
13. Yanagisawa, A.; Habaue, S.; Yasue, K.; Yamamoto, H. J. Am.
Chem. Soc. 1994, 116, 6130–6141.
14. Kuhnert, N.; Peverley, J.; Robertson, J. Tetrahedron Lett.
1998, 39, 3215–3216.
3. For recent reviews, see: (a) Hodgson, D. M.; Gras, E. Synthesis
2002, 1625–1642. Hodgson, D. M.; Tomooka, K.; Gras, E.
Enantioselective synthesis by lithiation adjacent to oxygen and
subsequent rearrangement. In Topics in Organometallic
Chemistry: Organolithiums in Enantioselective Synthesis;
Hodgson, D. M., Ed.; Springer: Berlin, 2003; pp 217–250.
4. Hodgson, D. M.; Galano, J.-M.; Christlieb, M. Chem.
Commun. 2002, 2436–2437.
15. Lipshutz, B. H.; Ellsworth, E. L.; Dimock, S. H.; Smith, R. A.
J. J. Am. Chem. Soc. 1990, 112, 4404–4410. Stack, D. E.;
Klein, W. R.; Rieke, R. D. Tetrahedron Lett. 1993, 34,
3063–3066. Lipshutz, B. H.; Hackmann, C. J. Org. Chem.
1994, 59, 7437–7444.
16. Rubottom, G. M.; Gruber, J. M. J. Org. Chem. 1978, 43,
1599–1602.
5. Cope, A. C.; Fisher, B. S.; Funke, W.; McIntosh, J. M.;
McKervey, M. A. J. Org. Chem. 1969, 34, 2231–2234. Eaton,
P. E.; Millikan, R. Synthesis 1990, 483–484.
17. Demir, A. S.; Jeganathan, A. Synthesis 1992, 235–247.
18. Moriarty, R. M.; Prakash, O. Org. React. (NY) 1999, 54,
273–418.
6. Hodgson, D. M.; Gras, E. Angew. Chem., Int. Ed. 2002, 41,
2376–2378.
19. Tamura, Y.; Yakura, T.; Terashi, H.; Haruta, J.; Kita, Y. Chem.
Pharm. Bull. 1987, 35, 570–577.
7. Nicolaou, K. C.; Zhong, Y.-L.; Baran, P. S. J. Am. Chem. Soc.
2000, 122, 7596–7597. Nicolaou, K. C.; Montagnon, T.;
Baran, P. S.; Zhong, Y.-L. J. Am. Chem. Soc. 2002, 124,
2245–2258. Nicolaou, K. C.; Montagnon, T.; Baran, P. S.
Angew. Chem., Int. Ed. 2002, 41, 993–996. Nicolaou, K. C.;
Gray, D. L. F.; Montagnon, T.; Harrison, S. T. Angew. Chem.,
Int. Ed. 2002, 41, 996–1000.
20. Helmchen, G. Tetrahedron Lett. 1974, 15, 1527–1530.
21. Feigenbaum, A.; Pete, J.-P.; Scholler, D. J. Org. Chem. 1984,
49, 2355–2360.
22. Cossio, F. P.; Aizpura, J. M.; Palomo, C. Can. J. Chem. 1986,
64, 224–231.
23. Pilcher, A. S.; Hill, D. K.; Shimshock, S. J.; Waltermire, R. E.;
DeShong, P. J. Org. Chem. 1992, 57, 2492–2495. Pilcher,
A. S.; DeShong, P. J. Org. Chem. 1993, 58, 5130–5134.
24. Crystallographic data (excluding structure factors) for the
structures in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication
numbers CCDC 221381 (13) and CCDC 221382 (6). Copies of
the data can be obtained, free of charge, on application to CCDC,
12 Union Road, Cambridge CB2 1EZ, UK (fax: þ44-1223-
336033 or e-mail: deposit@ccdc.cam.ac.uk).
8. Frigerio, M.; Santagostino, M.; Sputore, S. J. Org. Chem.
1999, 64, 4537–4538.
9. Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P.
Synthesis 1994, 639–666.
10. Ito, Y.; Hirao, T.; Saegusa, T. J. Org. Chem. 1978, 43,
1011–1013.
11. Desponds, O.; Schlosser, M. J. Organomet. Chem. 1991, 409,
93–101.
12. Hosomo, A.; Sakurai, H. J. Am. Chem. Soc. 1977, 99,
1673–1675.
25. For the preparation of PriLi, see: http://www.syntheticpages.
com/search.php.