Synthesis of a ring F building block for the ciguatoxins

Article abstract of DOI:10.1016/S0040-4020(02)00037-6

The preparation of a new ring F building block for the marine polyether toxins, the ciguatoxins, employing ascorbic acid as a chiral pool starting material, is reported.

Full text of DOI:10.1016/S0040-4020(02)00037-6

TETRAHEDRON
Pergamon
Tetrahedron 58 -2002) 1779±1787
Synthesis of a ring F building block for the ciguatoxins
²
Silas Bond and Patrick Perlmutterp
School of Chemistry, Monash University, P.O. Box 23, 3800 Melbourne, Vic., Australia.
Received 16 August 2001; accepted 14 December 2001
AbstractÐThe preparation of a new ring F building block for the marine polyether toxins, the ciguatoxins, employing ascorbic acid as a
chiral pool starting material, is reported. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
prepared by nucleophilic allylation and ring closing
metathesis -RCM) of 4.
The consumption of ®sh containing accumulated
ciguatoxins -CTXs, e.g. P-CTX-3C 1; Fig. 1) can lead to
disease in humans. This disease, known as ciguatera, is
endemic throughout the tropics. As the CTXs are present
in very low concentrations -,0.1 ppb) in toxic ®sh, a very
sensitive assay, such as an immunoassay, is required for
their detection.1,2 As part of our program3 directed towards
the synthesis of immunogenic domains of the CTXs, we
report herein the details of our preparation of a CTX ring
F building block.4
2. Results and discussion
2.1. Preparation of alkene 4
3-Allyl ascorbic acid derivatives have been reported
previously. For example, introduction of an allyl group at
C3 of ascorbic acid acetonide using a two-step process has
been reported by Wimalasena and Mahindaratne.16 Their
method involved the sequence of -i) O-allylation -using
K2CO3 in a DMSO/THF solution) and -ii) thermal
rearrangement. Although we found this procedure works
quite well, the acetonide group in the product proved to
be insuf®ciently robust to survive subsequent reactions.
Thus we employed 6, the cyclohexylidene acetal of ascorbic
acid, easily prepared on large scale by adaptation of
Tanaka's isoascorbic acid cyclohexylidene acetal prepara-
tion. Unfortunately, the aforementioned allylation condi-
tions proved ineffective for 6. Allylation under neutral
conditions proved more satisfactory. Thus, neutralisation
of a THF solution of 6 with 1N NaOH-aq) followed by treat-
ment with allyl bromide and stirring for 10 h yielded a
mixture of O-allylated 7 and C-allylated 4a and b
-Scheme 2). Claisen rearrangement then gave complete
conversion to a 4:1 mixture of the epimeric C-allylated
furandiones 4a and b.
The overall strategy at the outset of this work was to develop
methods to prepare medium-sized lactones from readily
available, enantiomerically pure lactones, such as ascorbic
acid 5. The structure of the ring F lactone building block 2,
the subject of this report, and its retrosynthetic analysis are
shown in Scheme 1. We envisaged that this could be
achieved from bicyclic 3 after sequential ring opening5,6
and selective reduction.7±15 Bicyclic 3, in turn, could be
2.2. Conversion of 4 into bicyclic 3 %R1ˆTMS)
As the mixture of 4a and b could not be puri®ed without
decomposition, nucleophilic allylation of the ketone was
attempted directly on this crude material. Thus allylation
using either allyl zinc or Grignard reagents were examined
under a variety of conditions. Irrespective of the conditions
employed these reactions failed to consume all of the start-
ing material, even with a large excess of allylating reagent.
Addition to the corresponding tetramethylsilane -TMS)
ether gave far better results. Thus, treatment of 4a and b
Figure 1.
Keywords: ciguatoxin; ring expansions; ascorbic acid; oxidation; ring clos-
ing metathesis.
p
Corresponding author. Tel.: 161-3-9905-4522; fax: 161-3-9905-4597;
e-mail: p.perlmutter@sci.monash.edu.au
New address: Biota Holdings Limited, Biota Chemistry Laboratories,
School of Chemistry, Monash University, P.O. Box 23, 3800 Melbourne,
Vic., Australia.
²
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020-02)00037-6

1780
S. Bond, P. Perlmutter / Tetrahedron 58 (2002) 1779±1787
Scheme 1.
Scheme 2.
Scheme 3.
with trimethylsilyl chloride and imidazole in THF at rt for
only 2 h gave an excellent yield of TMS ethers 8a and
b-Scheme 3) which were easily separated by chromato-
graphy.
-Scheme 3). -Allylation of 8a using allylmagnesium
bromide or NH4Cl-aq)-activated allyl zinc reagents failed to
drive the reaction to completion with no selectivity being
observed.)
Allylation under anhydrous conditions using an allyl zinc
reagent, however, provided diene 9 in excellent yield
The relative stereochemistry of 9 could not be determined
using spectroscopic methods and was ultimately established
Figure 2.

S. Bond, P. Perlmutter / Tetrahedron 58 (2002) 1779±1787
1781
Scheme 4.
by X-ray crystallographic analysis of the crystalline bicyclic
product 3 -R1ˆTMS) obtained from RCM of 9 -vide infra).
Examination of the crystal structure established that allyl-
ation had occurred exclusively from the Si face of the
ketone. Pre-complexation of ketone 8a with anhydrous
ZnBr2, prior to addition of the allylating reagent, yielded
identical results. This suggests it is unlikely that side
chain or O-TMS coordination of the allyl zinc reagent is
controlling the selectivity. Variation of the reaction
temperature had little effect on the diastereoselectivity
with a small amount of starting material being returned at
lower temperatures. Conformational analysis of 8a provides
a possible explanation for the diastereoselectivity -Fig. 2).
Conformation A is most likely favoured as conformation B
suffers from a serious pseudo-1,3-diaxial interaction and
should therefore be signi®cantly higher in energy. Confor-
mation A also offers the appropriate -Si) face of the ketone
with relatively unhindered access.
crystallography of the product 3 -R1ˆTMS, Scheme 1)
con®rmed the relative stereochemistry at the ring junctions
as trans -Scheme 4).
RCM studies on the diastereomeric mixtures of dienes 10a±
c obtained from additions to the mixture of ketones 4a and b
yielded mixtures from which isobenzofuranones 11a±c
were isolated. These reactions always yielded a small
amount of starting material irrespective of reaction times
up to 24 h. Extending the reaction time beyond 24 h and/
or raising the reaction temperature led to a reduction in yield
-Scheme 5).
Puri®cation of the reaction products 11a±d allowed for their
individual stereochemical assignment by direct correlation
each of their corresponding to diene precursors 10a±d. The
3aS, 7aS isomer 11b was assigned cis relative stereo-
chemistry based on its very rapid -,1 min) cleavage
under oxidative conditions -Pb-OAc)4) and based on its
proportional representation in the crude mixture. The
trans assignments of the 3aR, 7aS isomer 11a and 3aS,
7aR isomer 11c were suggested based on their lack of
RCM of the diene 9 using Grubbs' catalyst in toluene at
608C proceeded readily under the described conditions
with all of the starting material being consumed. X-Ray
Scheme 5.
Scheme 6.

1782
S. Bond, P. Perlmutter / Tetrahedron 58 (2002) 1779±1787
proceed to an appreciable extent. Cleavage using
PhI-OAc)2/I2 was successful, however, it was found to
21
be less ef®cient compared with the HgO/I2 method
-Scheme 6).
Tricarbonyl 12 proved to be very susceptible to hydration
producing a variable mixture of 12, its monohydrate 13 and
the bridged hydrate 14 -Fig. 3).
6
Direct reduction of 12 with NaBH4 or triethylborohydride
Figure 3.
yielded complex mixtures which could not be characterised.
Reduction with l-selectridew at approximately 21008C
yielded a mixture of the four diastereomers with the desired
dihydroxyoxoninone 15 being favoured, albeit in only a 3:2
ratio, over the other isomers. Stepwise reduction of the
mixture yielded a more satisfying result with both chemo-
and stereoselective reduction of the a-dicarbonyl system
being effected in cooled glacial acetic acid with NaCNBH3
as the reducing agent. The selectivity in favour of the
desired hydroxyoxonindione 15 over 16 was 95:5 with no
concomitant reduction of the isolated ketone carbonyl under
these conditions -Scheme 7). Conversion of 15 into benzo-
ate 18 yielded a crystalline derivative whose structure
con®rmed the relative stereochemistry of 15 as that shown.
reactivity towards Pb-OAc)4 and their proportional repre-
sentation in the crude mixtures. The assignment of isomer
11c was con®rmed directly by X-ray crystallography and
isomer 11a indirectly via chemical transformation of 9.
Negligible amounts of 11d were present and hence could
not be isolated.
2.3. Completion of the synthesis of 2
Cleavage of 11b proceeded at an exceedingly rapid rate with
no starting material being detected ,1 min after the
addition of Pb-OAc)4 at 08C providing 12 in excellent
yield. Cleavage of the trans fused bicycles proved some-
what more dif®cult. Neither isobenzofuranones 11a,c nor
the O-TMS isobenzofuranone 3 could be cleaved using
Pb-OAc)4. Addition of an acid catalyst with the Pb-OAc)4,
which has been reported to effect trans bicyclic diol
cleavages through a non-cyclic mechanism, also failed to
Reduction of the isolated ketone using diol 15, monosilyl
ether 17 and monobenzoate 18 was investigated employing
l-selectridew as a hydride source -Scheme 8).
yield results.17 Allowing a mixture of the isobenzofuranone
As shown the desired diastereomer was favoured in all cases
with the O-TMS oxonindione 17 yielding a 14:1 ratio in
favour of 21 -which is the ®nal target 2, R2ˆTMS, see
Scheme 1). Assignment of relative stereochemistry was
made based on J8,9.5
18±20
and HgO/I2
to re¯ux whilst being irradiated with a
tungsten lamp did induce b-®ssion of isobenzofuranones
11a,c and O-TMS isobenzofuranone 3. Interestingly both
irradiation and heat were required for the reaction to
Scheme 7.
Scheme 8.

S. Bond, P. Perlmutter / Tetrahedron 58 (2002) 1779±1787
1783
3. Conclusions
-SGP) was 0±10%A 0±15mins then hold 10%A 15±
30mins. Retention times are in minutes.
Synthetic equivalents for the target oxoninone 2 have been
successfully prepared in nine steps from ascorbic acid
cyclohexylidene 6. Although the initial steps proceed in
modest yield this synthesis represents the ®rst preparation
of enantiomerically pure oxoninones from ascorbic acid. A
key, highly selective oxidative cleavage±reduction
sequence from the O-TMS isobenzofuranone 3 provided
hydroxyoxonindione 15 in 43% overall yield for the two
steps.
4.2. Compound data
4.2.1. C-Allylated furandiones 4a,b. Suf®cient aqueous 1N
NaOH -,200 mL) was added to a solution of O-cyclo-
hexylidene furanone 6 -51.2 g, 0.2 mol) in THF -460 mL)
such that a pH of between 7 and 8 was obtained. Allyl
bromide was added dropwise -18.9 mL, 0.22 mol) to this
solution and the mixture allowed to stir for 10 h. The
solution was extracted with EtOAc -3£300 mL) and the
combined extracts washed with brine -200 mL), dried
-MgSO4) and the solvent evaporated in vacuo to yield an
orange oil -23.5 g). The oil was dissolved in toluene
-650 mL) and heated to re¯ux for 6 h after which time the
solvent was removed in vacuo to yield a mixture of the C3
allyl epimers -4±5:1, 4a/4b) as a dark orange oil -22.5 g).
The purity was estimated to be .80% by NMR spectro-
scopy. This material was used in subsequent reactions with-
out puri®cation. The data presented here are for the mixture.
4a 1H NMR -300 MHz, CDCl3) d 1.33±1.70 -m, 10H), 2.65
-bd, Jˆ,7.5 Hz, 2H), 4.05 -dd, Jˆ8.7, 7.0 Hz, 1H), 4.17
-dd, Jˆ8.7, 6.9 Hz, 1H), 4.51 -dt, Jˆ6.9, 2.0 Hz, 1H), 4.65
-d, Jˆ2.0 Hz, 1H), 5.20±5.28 -m, 2H), 5.60±5.74 -m, 1H);
13C NMR -75 MHz, CDCl3) d 23.8, 23.9, 25.1, 34.8, 34.9,
39.6, 64.4, 71.8, 74.2, 81.5, 111.8, 123.1, 127.4, 172.5,
205.4; HRMS Calcd for C15H20O6Na -M1Na1): m/z
319.116, Found 319.117. 4b. Resonance used for diastereo-
The methodology described here forms a solid basis on
which to build syntheses directed at the preparation of
more extended fragments of the central portion of the
CTX marine polyether toxins. Investigations towards
improving and extending this process are currently under-
way in our laboratories and will be reported in due course.
4. Experimental
4.1. General methods and materials
Melting points were determined on a Ko¯er hotstage and are
uncorrected. Elemental microanalyses were performed by
the Australian Microanalytical Service, National Analytical
Laboratories, Melbourne or the University of Otago,
Dunedin, New Zealand. Optical rotations were recorded
on a Perkin±Elmer Model 141 Polarimeter. Infrared -IR)
spectra were recorded on a Perkin±Elmer 1600 Series
Fourier Transform spectrophotometer -cm21 scale) and
refer to thin ®lms of liquids -neat) or paraf®n -Nujol)
mulls of solids between NaCl plates. High resolution
hydrogen-1 nuclear magnetic resonance -1H NMR) spectra
were recorded at 300 MHz on a Bruker DPX-300 spectro-
meter or 400 MHz on a Bruker Avance DRX 400
1
meric ratio determination: H NMR -300 MHz, CDCl3) d
2.72 -bd, Jˆ,7 Hz, 2H, H8).
4.2.2. Allylfurandione TMS ether 8a. Imidazole -1.38 g,
20 mmol) and chlorotrimethylsilane -2.5 mL, 20 mmol)
were added to crude furandiones 4a,b -2.96 g, 10 mmol)
in THF -50 mL) under N2. The solution was allowed to
stir for 2 h after which time it was diluted with Et2O
-50 mL) and poured into water -100 mL). The etherial
layer was separated and the aqueous layer extracted with
two further portions of Et2O -25 mL). The combined
organic extracts were dried -MgSO4) and the solvent
removed in vacuo to yield an oil which was puri®ed by
¯ash chromatography -10% Et2O/hexanes) to yield 8a as a
single isomer -1.71 g, 46%). FTIR -neat) 2938s, 1816s,
1
spectrometer. The H NMR spectral data refer to deuterio-
chloroform solutions -CDCl3) using residual solvent as
internal reference -d 0.00 ppm). Carbon-13 nuclear
magnetic resonance -13C NMR) spectra were recorded at
75 MHz on a Bruker APX-300 spectrometer or 100 MHz
on a Bruker Avance DRX 400 spectrometer. Mass
spectrometry -ESI) was performed using samples in
methanol on a Micromass Platform QMS Electrospray
mass spectrometer. High-resolution mass spectra -HRMS)
for accurate mass determinations were recorded on a Bruker
BioApex 47e FTMS ®tted with an analytical electrospray
source using NaI for accurate mass calibration -accuracy
^3 ppm). Low-resolution mass spectra were recorded on
a VG micromass 70/70F or a VG TRIO-1 mass spectrometer
with an ion source temperature of 2008C and electron
impact energy of 70 eV. X-Ray crystallography was
performed on a Nonius Kappa CCD. HPLCs were run on
a Waters Alliance 2690 HPLC with a 996 Photodiode array
detector using Zorbax RX-SIL normal phase silica
-5 micron) columns with dimensions of 4.6£150 mm for
analytical work and 9.4£259 mm for semi-preparative
with ¯ow rates of 0.3 mL/min and 4.0 mL/min for analytical
and semi-preparative respectively. An analytical gradient
pro®le -AGP) of Isopropanol-A)/hexanes-B) was used
with gradient of 0±10%A during 0±10mins then hold
10%A 10±29mins. The semi-preparative gradient pro®le
1
1771s cm21; H NMR -300 MHz, CDCl3) d 0.14 -s, 9H),
1.17±1.87 -m, 10H), 2.56 -d, Jˆ7.6 Hz, 2H), 3.95 -dd,
Jˆ8.5, 7.3 Hz, 1H), 4.09 -dd, Jˆ8.5, 6.7 Hz, 1H), 4.46±
4.52 -m, 1H), 4.53 -d, Jˆ2.3 Hz, 1H), 5.08±5.18 -m, 2H),
5.57±5.71 -m, 1H); 13C NMR -75 MHz, CDCl3) d 1.6, 23.6,
24.9, 35.0, 35.1, 42.9, 64.5, 73.1, 76.2, 81.0, 111.2, 121.6,
128.1, 171.9, 205.1; HRMS Calcd for C18H28O6Si -M1): m/z
368.166, Found 368.166.
4.2.3. Diallylfuranones 10a±d. Method 1. Freshly titrated
allyl magnesium chloride -23 mL, 1.1 M, 25.3 mmol) was
added to the crude furandione 4a,b -2.96 g, 10 mmol) in
THF -100 mL) at 2788C. The mixture was allowed to stir
for 4 h after which time the reaction was quenched by the
addition of sat. aqueous NH4Cl -100 mL) and allowed to
warm to rt. The product was extracted with EtOAc
-3£50 mL), dried -MgSO4) and the solvent evaporated in
vacuo to yield an orange coloured oil. Puri®cation by ¯ash
chromatography -25% EtOAc/hexanes) yielded the cis

1784
S. Bond, P. Perlmutter / Tetrahedron 58 (2002) 1779±1787
diallylated furanone 10b -430 mg, 13%, Rf 0.6, TLC 40%
EtOAc/hexanes) as a pure compound along with a fraction
consisting of the trans isomers 10a and c along with 4a -4:1
10a/4a) and other minor unidenti®ed impurities -360 mg, Rf
0.5, TLC 40% EtOAc/hexanes). This sample was not
puri®ed further. Method 2. Allyl bromide -8.8 mL,
0.1 mol), sat. aqueous NH4Cl -100 mL) and Zn -6.57 g,
0.1 mol) were added sequentially to a solution of the
crude furandione 4a,b -5.96 g, 20 mmol) in THF -20 mL)
which was being vigorously stirred at rt. After a short
initiation time a temperature rise from 25 to 508C was
observed with a signi®cant gaseous evolution. The reaction
was allowed to stir vigorously for 1 h then diluted with Et2O
-40 mL). After separation of the organic layer the aqueous
phase was extracted further with Et2O -3£40 mL). The
combined extracts were dried -MgSO4) and the solvent
evaporated in vacuo to yield a pale yellow oil. The crude
1H NMR spectrum indicated a 2:3:1 ratio of the cis diallyl-
ated furanone 10b, trans diallylated furanone 10a and start-
ing material 4a as determined by integration of the H5
doublets at 4.18, 4.24 and 4.65 ppm, respectively. Two
¯ash chromatography puri®cations on silica gel -25%
EtOAc/hexanes) yielded one fraction consisting of a
mixture of cis diallylated furanone 10b and starting material
4a in a 1:1 ratio and another fraction consisting of the trans
diallylated furandione 10a and starting material 4a -1.31 g,
11:1, 10a/4a). Products from this reaction were not puri®ed
further 1 h sat. aqueous NH4Cl -25 mL) was added. The
organic phase was separated and the aqueous phase
extracted with Et2O -3£20 mL). The combined organic
extracts were dried -MgSO4) and the solvent evaporated in
vacuo to yield a pale yellow oil. Puri®cation by ¯ash chro-
matography -30% Et2O/hexanes) yielded the 9 -1.05 g,
26
94%) as a colourless oil. [a]D ˆ145.4 -c 1.0, CHCl3);
1
FTIR -neat) 3452b, 3079s, 2948s, 1770s, 1639s cm21; H
NMR -300 MHz, CDCl3) d 0.22 -s, 9H), 1.24±1.70 -m,
10H), 2.29 -dd, Jˆ14.0, 8.3 Hz, 1H), 2.35 -s, 1H), 2.52±
2.60 -m, 2H), 2.61±2.69 -m, 1H), 3.93 -dd, Jˆ8.5, 7.2 Hz,
1H), 4.10 -dd, Jˆ8.4, 6.5 Hz, 1H), 4.15 -d, Jˆ6.0 Hz, 1H),
4.34 -ddd, Jˆ7.2, 6.5, 6.0 Hz, 1H), 5.13 -m, 4H), 5.82±6.08
-m, 2H); 13C NMR -75 MHz, CDCl3) d 2.3, 23.9, 24.0, 25.2,
35.2, 35.9, 37.3, 39.3, 65.4, 72.9, 80.5, 82.6, 83.0, 110.6,
119.4, 121.8, 131.9, 132.6, 174.7; HRMS Calcd for
C21H34O6SiNa -M1Na1): m/z 433.202, Found 433.201;
Anal. Calcd for C21H34O6Si: C, 61.4; H, 8.4. Found: C,
61.4; H, 8.4%.
4.2.5. cis-Isobenzofuranone 11b. A solution of Grubbs'
catalyst -80 mg, 10 mol%) in degassed toluene -0.5 mL)
was added at 608C under Ar to a solution of cis diallyl
furanone 10b -136 mg, 0.40 mmol) in degassed toluene
-7 mL). The solution was stirred at this temperature for
20 h after which time the solvent was removed in vacuo.
The residue was dissolved in ether and passed through a
silica plug to remove some of the black tar like material.
Collection of the ®ltrate and evaporation of the solvent
yielded a clear, black coloured oil. Flash chromatography
of the oil -75% Et2O/hexanes) yielded starting material 10b
-14 mg) and 11b -75 mg, 59%) as a white foam. 11b.
29
further. 10b. [a]D ˆ161.8 -c 1.17, CHCl3); FTIR -CHCl3)
1
3426b, 2935s, 1785s cm21; H NMR -300 MHz, CDCl3) d
1.24±1.67 -m 10H), 2.29 -dd, Jˆ14.3, 8.5 Hz, 1H), 2.42±
2.61 -m, 3H), 3.99 -dd, Jˆ8.4, 7.7 Hz, 1H), 4.14 -dd, Jˆ8.3,
6.8 Hz, 1H), 4.18 -d, Jˆ2.0 Hz, 1H), 4.48 -ddd, Jˆ7.7, 6.8,
2.0 Hz, 1H), 4.53 -s, 1H), 5.16±5.22 -m, 4H), 5.83±6.03 -m,
2H); 13C NMR -75 MHz, CDCl3) d 23.9, 24.0, 24.9, 34.9,
35.3, 37.0, 38.1, 65.5, 73.2, 78.0, 78.3, 79.0, 112.0, 119.7,
119.8, 130.6, 131.4, 175.1; HRMS Calcd for C18H26O6Na
-M1Na1): m/z 361.163, Found 361.161.
26
[a]D ˆ247.0 -c 2.2, CHCl3); FTIR -CHCl3) 3444b,
3036w, 2937s, 2861m, 1784s, 1638m cm21 1H NMR
;
-400 MHz, CDCl3) d 1.34±1.69 -m, 10H), 2.38±2.42 -m,
2H), 2.40±2.48 -m, 1H), 2.53±2.62 -m, 1H), 3.74 -bs, 1H),
4.03 -dd, Jˆ8.5, 7.4 Hz, 1H), 4.17 -dd, Jˆ8.5, 6.8 Hz, 1H),
4.21 -d, Jˆ2.3 Hz, 1H), 4.49 -ddd, Jˆ7.3, 6.8, 2.3 Hz, 1H),
5.61±5.68 -m, 2H); 13C NMR -100 MHz, CDCl3) d 23.8,
24.8, 30.6, 32.9, 34.8, 35.2, 65.4, 72.4, 74.7, 76.1, 78.0,
112.0, 122.5, 123.0, 177.1; HRMS Calcd for C16H22O6Na
-M1Na1): m/z 333.131, Found 333.129.
4.2.4. Diallylfuranone TMS ether 9a. Method 1. Allyl
bromide -0.45 mL, 5.2 mmol), sat. aqueous NH4Cl -5 mL)
and Zn -330 mg, 5.0 mmol) were sequentially added to a
vigorously stirred solution of the TMS ether 8a -368 mg,
1.0 mmol) in THF -1 mL) at rt. The heterogenous mixture
was allowed to stir for 1 h then diluted with Et2O -5 mL) and
the organic layer separated. The aqueous layer was
extracted further with Et2O -2£5 mL) and the combined
extracts dried -MgSO4). Evaporation of the solvent in
vacuo yielded a pale yellow oil. NMR analysis indicated a
1:1 mixture of the starting TMS ether 8a and 9 whose NMR
spectra were identical to those obtained from Method 2.
Method 2. A crystal of I2 followed by allyl bromide
-2.4 mL, 28 mmol) was added dropwise, with stirring, to
zinc powder -3.6 g, 56 mmol) in THF -15 mL) under N2.
The temperature during the addition was maintained
below 208C with the aid of an ice bath once the reaction
had been initiated as indicated by the discolouration of the I2
and accompanied temperature rise. Upon completing the
addition the mixture was stirred for a further 10 min and
the excess zinc ®ltered off under N2. A portion of the allyl
zinc reagent -6.1 mL, 3.6 equiv. based on 100% conversion)
was subsequently added in one portion to a solution of the
O-TMS furandione 8a -1.0 g, 2.7 mmol) in THF -25 mL). A
slight temperature rise was observed and after stirring for a
4.2.6. trans-Isobenzofuranones 11a and c. Grubbs'
catalyst -340 mg, 10 mol%) in degassed toluene -5 mL)
was added under Ar to a puri®ed mixture of diallylated
furanones 10a±c -1.34 g, 4.0 mmol) in degassed toluene
-85 mL) at 708C. The solution was stirred for 24 h at
between 60 and 708C after which time the solvent was
evaporated and the black residue puri®ed by ¯ash chromato-
graphy using 75% Et2O/hexanes as the eluant. Two
fractions were collected the ®rst of which was a white
foam consisted of a 2:1 mixture of the two trans isomers
11a and c -182 mg, 15%), respectively, and a second
fraction containing a 10:1 mixture of 11b and c -524 mg,
42%), respectively, as a white foam. Pure trans isobenzo-
furanone 11c was crystallised -Et2O/hexanes) from the ®rst
mixture and its absolute con®guration determined by X-ray
crystallography. Repuri®cation of the ®ltrate yielded the
trans-isobenzofuranone 11a as a pure compound. 11a.
25
[a]D ˆ234.5 -c 1.0, CHCl3); FTIR -CHCl3) 3330b,
1
3020s, 2939s, 1790s cm21; H NMR -300 MHz, CDCl3) d
1.35±1.73 -m, 10H), 2.33 -ddd, Jˆ17.0, 5.2, 1.0 Hz, 1H),

S. Bond, P. Perlmutter / Tetrahedron 58 (2002) 1779±1787
1785
2.42±2.55 -m, 3H), 2.75±2.83 -m, 1H), 4.11 -t, Jˆ8.1 Hz,
1H), 4.24 -dd, Jˆ8.2, 6.6 Hz, 1H), 4.30 -ddd, Jˆ8.0, 6.6,
1.0 Hz, 1H), 4.52 -d, Jˆ1.0 Hz, 1H), 5.71±5.77 -m, 1H),
5.85±5.91 -m, 2H); 13C NMR -75 MHz, CDCl3) d 23.5,
23.7, 24.6, 29.2, 29.7, 34.5, 35.0, 65.6, 71.8, 71.9, 76.3,
85.2, 112.3, 122.3, 125.0, 175.3; HRMS Calcd for
C16H22O6Na -M1Na1): m/z 333.131, Found 333.132. 11c.
resulted in partial decomposition with the recovered
material having a different and variable ratio of 12 and the
proposed products 13 and 14 along with other unidenti®ed
impurities. Method 2. Yellow HgO -2.83 g, 13 mmol) was
added under Ar to a stirred solution of isobenzofuranone
TMS ether 3 -2.0 g, 5.2 mmol) in rigorously deoxygenated
benzene -250 mL). The mixture was heated to re¯ux by
irradiation with a white light -275 W). Whilst at re¯ux a
solution of I2 -3.17 g, 12.5 mmol) in deoxygenated benzene
-300 mL) was added over a 1 h period. The reaction was
irradiated and re¯ux maintained until all of the starting
material had been consumed as determined by TLC -75%
Et2O/hexanes). Once complete -,8±9 h) the reaction
mixture was cooled and ®ltered through a Celitew plug
which was eluted with Et2O. The ®ltrate was transferred
to a separating funnel and washed sequentially with sat.
aqueous Na2S2O5 -2£200 mL) and brine -100 mL). The
organic layer was dried -MgSO4) and the solvent evaporated
26
Mp 177±1788C; [a]D ˆ181.3 -c 1.5, acetone); FTIR
-CHCl3) 3444b, 2924s, 1787s, 1771s cm21 1H NMR
;
-300 MHz, CDCl3) d 1.37±1.78 -m, 10H), 2.25±2.36 -m,
2H), 2.51±2.60 -m, 1H), 2.66 -s, 1H), 2.72±2.80 -m, 1H),
4.08 -dd, Jˆ8.4, 7.4 Hz, 1H), 4.21 -dd, Jˆ8.4, 6.7 Hz, 1H),
4.50 -ddd, Jˆ7.4, 6.7, 2.8 Hz, 1H), 4.55 -s, 1H), 4.69 -d,
Jˆ2.8 Hz, 1H), 5.69±5.79 -m, 2H); 13C NMR -75 MHz,
CDCl3) d 23.9, 23.9, 24.9, 29.0, 30.4, 35.0, 35.3, 65.5,
73.2, 75.2, 76.7, 81.2, 111.7, 122.9, 123.0, 174.9; HRMS
Calcd for C16H22O6 -M1): m/z 310.142, Found 310.141;
Anal. Calcd for C16H22O6: C, 61.9; H, 7.2; Found: C,
61.9; H, 7.2%.
1
in vacuo to yield a pale yellow coloured foam -1.74 g). H
NMR analysis of this material indicated some residual
TMS-I by-product along with three signi®cant compounds
in a ratio of 5:1:1 with 12 as the major component as deter-
mined by NMR. The other two compounds were assigned as
13 and 14 as described in the lead tetraacetate cleavage
above. NMR data for the crude trione 12: 1H NMR
-400 MHz, CDCl3) d 1.31±1.67 -m, 10H), 2.99 -ddd,
Jˆ10.7, 6.3, 1.4 Hz, 1H), 3.18 -dd, Jˆ13.2, 6.2 Hz, 1H),
3.85 -dd, Jˆ8.9, 5.6 Hz, 1H), 4.06±4.14 -m, 2H), 4.12
-dd, Jˆ8.9, 7.0 Hz, 1H), 4.64 -ddd, Jˆ7.0, 5.6, 2.8 Hz,
1H), 5.22 -d, Jˆ2.8 Hz, 1H), 5.63 -dt, Jˆ10.5, 6.2 Hz,
1H), 5.75±5.83 -m, 1H); 13C NMR -100 MHz, CDCl3) d
23.8, 23.9, 25.0, 34.4, 35.7, 41.7, 43.0, 64.8, 74.0, 79.1,
111.0, 123.2, 129.0, 160.2, 189.5, 202.5.
4.2.7. Isobenzofuranone TMS ether 3. Grubbs' catalyst
-440 mg, 20 mol%) in degassed toluene -10 mL) was
added at 608C to diallylated TMS ether 9 -1.1 g,
2.7 mmol) in degassed toluene -270 mL) under Ar. The
solution was heated at between 60 and 708C for 20 h after
which time the solvent was evaporated in vacuo. The
residue was dissolved in CH2Cl2 and ®ltered through a silica
gel plug. After washing the product was eluted with Et2O
and the solvent evaporated. Trituration of the resultant black
solid with portions of hot hexanes, containing a hint of
CH2Cl2, removed the black colour. Crystallisation
-hexanes/CH2Cl2) of the pale black solid yielded 3
-630 mg, 61%) as white needles. A second crop of 3
-91 mg, 9%) was obtained by recrystallisation of the mother
liquor. Slow recrystallisation -Et2O/hexanes) yielded
crystals suitable for X-ray crystallography. Mp 1438C;
4.2.9. Hydroxyoxonindione 15. Sodium cyanoborohydride
-190 mg, 2.0 mmol) was added in four portions to a frozen
-,138C) solution of the crude oxonintrione 12 -460 mg,
1.5 mmol, derived via HgO/I2 method from 675 mg of 3)
in glacial acetic acid -15 mL). After the additions had been
made the reaction was allowed to warm to rt over a 20 min
period then slowly quenched by addition in portions to a
mixture of EtOAc -20 mL) and ice cold, sat. aqueous
NaHCO3 -100 mL) in a separating funnel. The NaHCO3
solution was kept cold and saturated during the quenching
by the addition of ice and solid NaHCO3. Once all of the
acetic acid had been neutralised the organic layer was
separated and the aqueous layer extracted with EtOAc
-2£20 mL). The combined organic extracts were dried
-MgSO4) and the solvent removed in vacuo to yield an
27
[a]D ˆ127.7 -c 0.47, CHCl3); FTIR -CHCl3) 3461b,
1
2943s, 1787s cm21; H NMR -300 MHz, CDCl3) d 0.15
-s, 9H), 1.37±1.72 -m, 10H), 2.00 -bs, 1H), 2.07±2.14 -m,
1H), 2.40±2.60 -m, 3H), 3.67 -dd, Jˆ8.3, 6.5 Hz, 1H), 4.08
-dd, Jˆ8.3, 6.5 Hz, 1H), 4.39 -d, Jˆ10.0 Hz, 1H), 4.58 -dt,
Jˆ10.0, 6.5 Hz, 1H), 5.65±5.69 -m, 1H), 5.75±5.79, -m,
1H); 13C NMR -75 MHz, CDCl3) d 1.3, 23.8, 24.0, 25.1,
29.1, 30.0, 34.7, 36.4, 65.5, 74.6, 74.7, 75.8, 90.5, 110.1,
122.9, 124.3, 174.3; HRMS Calcd for C19H30O6SiNa
-M1Na1): m/z 405.171, Found 405.170; Anal. Calcd for
C19H30O6Si: C, 59.7; H, 7.9; Found: C, 59.7; H, 8.0%.
4.2.8. Oxonintrione 12. Method 1. Lead tetraacetate
-63 mg, 0.14 mmol) was added at 08C to cis isobenzo-
furanone 11b -40 mg, 0.13 mmol) in CH2Cl2 -1.3 mL).
The brown solution was allowed to stir at this temperature
for 10 min after which time the reaction was quenched
with H2O -2 mL) and ®ltered through a plug of Celitew,
eluting with CH2Cl2. The organic layer was separated and
the aqueous layer extracted with CH2Cl2 -2£2 mL). The
combined organic extracts were dried -MgSO4) and the
solvent evaporated in vacuo to yield a pale brown oil
-36 mg). 1H NMR analysis of the oil indicated three signi®-
cant compounds in varying ratios with 12 as the major
component with the proposed gem-diol 13 and bis-hemia-
cetals 14 as the minor components. Attempted puri®cation,
by ¯ash chromatography, of this product on silica gel
1
oil. A H NMR spectrum of the crude material indicated a
95:5 ratio of hydroxyoxonindione 15 and its C3 epimer 16.
Puri®cation by ¯ash chromatography -25% Et2O/CH2Cl2)
yielded a white solid which was crystallised -Et2O) provid-
ing 15 -232 mg, 43% over two steps) as ®ne needles. Mp
26
115±1178C; [a]D ˆ1311.9 -c 1.0, CHCl3); FTIR -CHCl3)
;
3500b, 3021s, 2937s, 1760s, 1726s cm21 1H NMR
-300 MHz, [D6] acetone) d 1.30±1.81 -m, 10H), 2.42
-ddd, Jˆ13.3, 6.3 Hz, 3.2 1H), 2.74 -dd, Jˆ10.8, 6.4 Hz,
1H), 2.99±3.09 -m, 1H), 3.83 -dd, Jˆ8.7, 6.1 Hz, 1H),
3.81±3.90 -m, 1H), 4.12 -dd, Jˆ8.5, 6.9 Hz, 1H), 4.42 -d,
Jˆ5.5 Hz, 1H), 4.55 -ddd, Jˆ6.9, 6.1, 3.1 Hz, 1H), 4.70±
4.77 -m, 1H), 5.15 -d, Jˆ2.9 Hz, 1H), 5.42 -dt, Jˆ10.7,
6.6 Hz, 1H), 5.83 -dtt, Jˆ10.8, 6.1, 1.3 Hz, 1H); 13C NMR

1786
S. Bond, P. Perlmutter / Tetrahedron 58 (2002) 1779±1787
-75 MHz, [D6] acetone) d 24.5, 24.7, 25.8, 33.6, 35.6, 36.3,
42.2, 65.4, 71.2, 76.1, 78.3, 111.0, 124.7, 130.2, 173.9,
205.9; HRMS Calcd for C16H22O6K -M1K1): m/z
349.105, Found 349.105. 16. Resonances used for diastereo-
meric ratio determination: 1H NMR -300 MHz, [D6]
acetone) d 4.90 -d, Jˆ2.3, 1H, H9).
conditions were maintained for 30 min after which time the
reaction was quenched with sat. aqueous NH4Cl -5 mL) and
allowed to warm to rt. The resultant mixture was diluted
with EtOAc -5 mL) and the organic phase separated. The
aqueous phase was extracted further with EtOAc -2£5 mL)
and the combined extracts dried -MgSO4). Evaporation of
the solvent in vacuo yielded an oil which was subsequently
®lter through a silica plug -Et2O) to removed the bulk of the
borane. 1H NMR analysis indicated a 4:1 mixture of the two
C8 epimers -19/20). This sample was not puri®ed further.
However, the major adduct 19 was obtained diastereo-
merically pure during the puri®cation of the hydroxyoxo-
ninone TMS ether 21 by preparative HPLC -see below).
Desilylation of hydroxyoxoninone TMS ether 21 provided
dihydroxyoxoninone 19 when using an hexane/isopropanol
HPLC gradient. HPLC -AGP) 20.3; -SGP) 18.6;
4.2.10. Oxonindione TMS ether 17. Chlorotrimethylsilane
-122 mL, 0.96 mmol) and imidazole -82 mg, 1.2 mmol)
were added to a stirred solution of hydroxyoxonindione 15
-100 mg, 0.32 mmol) in THF -1.5 mL) at rt under N2. The
reaction was stirred for 12 h after which time it was
quenched with sat. aqueous NH4Cl -5 mL) and extracted
with EtOAc -3£5 mL). The combined extracts were dried
-MgSO4) and the solvent removed in vacuo to yield a
colourless oil. Puri®cation by ¯ash chromatography -15%
Et2O/hexanes) yielded a white solid which was crystallised
-Et2O) to yield 17 -122 mg, 99%) as feathery white needles.
28
[a]D ˆ291.0 -c 1.6, CHCl3); FTIR -CHCl3) 3460b,
1
3019s, 2941s, 1741s, 1656b cm21; H NMR -300 MHz,
26
Mp 63±648C; [a]D ˆ1251.9 -c 0.94, CHCl3); FTIR
[D6] acetone) d 1.18±1.55 -m, 10H), 2.16±2.23 -m, 1H),
2.25±2.36 -m, 2H), 2.47±2.67 -m, 1H), 3.81 -dd, Jˆ8.4,
6.3 Hz, 1H), 4.02 -dd, Jˆ8.4, 6.7 Hz, 1H), 4.05±4.09 -m,
1H), 4.10±4.19 -m, 1H), 4.20 -d, Jˆ6.2 Hz, 1H), 4.41 -ddd,
Jˆ6.8, 6.3, 3.7 Hz, 1H), 4.43 -d, Jˆ5.7 Hz, 1H), 4.67 -bdd,
Jˆ,8.5, 3.4 Hz, 1H,), 5.62±5.66 -m, 1H), 5.67±5.70 -m,
1H); 13C NMR -75 MHz, [D6] acetone) d 24.5, 24.6, 25.8,
34.6, 35.1, 35.7, 36.4, 65.7, 70.3, 71.2, 75.5, 80.0, 109.9,
126.1, 131.1, 176.3; HRMS Calcd for C16H24O6Na
-M1Na1): m/z 335.147, Found 335.146. 20. Resonances
used for diastereomeric ratio determination: 1H NMR
-300 MHz, [D6] acetone) d 4.88 -bdd, J9±10ˆ7.7 Hz,
J9±8ˆ3.5 Hz, 1H, H9).
1
-CHCl3) 2944m, 1771s, 1727s cm21; H NMR -400 MHz,
[D6] acetone) d 0.14 -s, 9H), 1.28±1.75 -m, 10H), 2.36
-ddd, Jˆ13.2, 6.1, 3.3 Hz, 1H), 2.71±2.77 -m, 1H), 3.06±
3.12 -m, 1H), 3.82 -dd, Jˆ8.6, 6.1 Hz, 1H), 3.80±3.87 -m,
1H), 4.11 -dd, Jˆ8.6, 6.8 Hz, 1H), 4.53 -ddd, Jˆ6.8, 6.1,
3.2 Hz, 1H), 4.85±4.91 -m, 1H), 5.14 -d, Jˆ3.1 Hz, 1H),
5.42 -dt, Jˆ10.6, 6.6 Hz, 1H), 5.76±5.84 -m, 1H); 13C NMR
-100 MHz, [D6] acetone) d 0.0, 24.5, 24.7, 25.8, 34.6, 35.6,
36.4, 42.3, 65.5, 72.1, 76.1, 78.1, 111.0, 124.6, 130.3, 172.5,
205.8; HRMS Calcd for C19H30O6SiNa -M1Na1): m/z
405.171, Found 405.171.
4.2.11. Oxonindione benzoate 18. Benzoic anhydride
-66 mg, 0.29 mmol), triethylamine -40 mL, 0.29 mmol)
and a catalytic amount of DMAP were added to hydroxy-
oxonindione 15 -60 mg, 0.19 mmol) in THF -1 mL) at rt.
The solution was stirred at rt for 2 h after which time the
reaction was quenched with sat. aqueous NH4Cl -5 mL) and
extracted with Et2O -3£5 mL). The combined organic
extracts were washed with sat. aqueous NaHCO3
-2£5 mL) and the organic layer dried -MgSO4). Evaporation
of the solvent in vacuo and recrystallisation -Et2O/hexanes)
yielded 18 -80 mg, 99%) as colourless crystals which were
suitable for X-ray crystallography. Mp 1688C;
4.2.13. Hydroxyoxoninone TMS ether 21. A 1 M solution
of l-selectridew -0.36 mL, 0.36 mmol) in THF was added
under N2 to a solution of oxonindione TMS ether 17 -92 mg,
0.24 mmol) in THF -3 mL) whilst cooling using a pentane/
liquid N2 cooling bath -,21338C). The solution was
allowed to react for 30 min and was then quenched with
sat. aqueous NH4Cl -5 mL) and allowed to warm to rt.
The resultant mixture was diluted with EtOAc -5 mL) and
the organic layer separated. The aqueous layer was
extracted further with EtOAc -2£5 mL) and the combined
organic extracts dried -MgSO4) and the solvent evaporated
in vacuo to yield an oil. 1H NMR analysis of the oil
indicated the borane by-product was still present along
with a 14:1 mixture of 21 and its C8 epimer 22, respectively.
The borane was removed using a silica plug eluting with
20% Et2O/CH2Cl2 which co-eluted the epimers -58 mg,
25
[a]D ˆ147.3 -c 0.9, CHCl3); FTIR -CHCl3) 3036s,
2936s, 1775s, 1727s cm21 1H NMR -300 MHz, [D6]
;
acetone) d 1.10±1.77 -m, 10H), 2.68±2.76 -m, 1H), 2.76±
2.86 -m, 1H), 3.28±3.38 -m, 1H), 3.85±4.01 -m, 1H), 3.89
-dd, Jˆ8.7, 6.0 Hz, 1H), 4.16 -dd, Jˆ8.7, 6.9 Hz, 1H), 4.59
-ddd, Jˆ6.9, 6.0, 3.0 Hz, 1H), 5.21 -d, Jˆ3.0 Hz, 1H), 5.56
-dt, Jˆ10.7, 6.6 Hz, 1H), 5.78±5.86 -m, 1H), 5.99 -dtt,
Jˆ10.8, 6.2, 1.3 Hz, 1H), 7.51±7.56 -m, 2H), 7.65±7.68
-m, 1H), 8.07±8.10 -m, 2H); 13C NMR -75 MHz, [D6]
acetone) d 24.5, 24.7, 25.8, 30.2, 35.6, 36.4, 42.2, 65.5,
73.1, 76.1, 78.8, 111.2, 126.0, 129.2, 129.5, 130.4, 130.5,
134.4, 165.9, 169.2, 205.3; HRMS Calcd for C23H26O7Na
-M1Na1): m/z 437.158, Found 437.158; Anal. Calcd for
C23H26O7: C, 66.7; H, 6.3; Found: C, 66.6; H, 6.2%.
25
63%) as a colourless oil. 21. [a]D ˆ252.0 -c 1.5,
CHCl3); FTIR -CHCl3) 3469b, 2936s, 1755s cm21 1H
;
NMR -300 MHz, [D6] acetone) d 0.11 -s, 9H), 1.28±1.56
-m, 10H), 2.17±2.30 -m, 2H), 2.30±2.43 -m, 1H), 2.43±
2.62 -m, 1H), 3.80 -dd, Jˆ8.4, 6.2 Hz, 1H), 4.00±4.10 -m,
1H), 4.03 -dd, Jˆ8.4, 6.7 Hz, 1H), 4.15±4.27 -m, 1H), 4.41
-ddd, Jˆ6.7, 6.1, 3.6 Hz, 1H), 4.69 -bdd, Jˆ,8.7, 3.4 Hz,
1H), 5.56±5.67 -m, 1H), 5.63±5.73 -m, 1H); 13C NMR
-75 MHz, [D6] acetone) d 0.0, 25.6, 25.9, 35.4, 35.5,
35.9, 36.5, 66.0, 71.2, 71.2, 75.5, 79.9, 110.0, 125.7,
131.6, 174.8; HRMS Calcd for C19H32O6SiNa -M1Na1):
m/z 407.187, Found 407.187. 22. Resonances used for
4.2.12. Dihydroxyoxoninone 19. A 1 M solution of
l-selectridew -0.12 mL, 0.12 mmol) in THF was added
under N2 to 3-hydroxyoxonindione 15 -15 mg, 48 mmol)
in THF -0.5 mL) at 21008C and the temperature was main-
tained after the addition between 2100 and 2788C. These
1
diastereomeric ratio determination: H NMR -300 MHz,
[D6] acetone) d 4.90 -bdd, J9±10ˆ7.0 Hz, J9±8ˆ5.1, 1H,
H9).

S. Bond, P. Perlmutter / Tetrahedron 58 (2002) 1779±1787
1787
4.2.14. Hydroxyoxoninone benzoate 23. A solution of
l-selectridew -53 mL, 53 mmol) in THF was added under
N2 to oxonindione benzoate 18 -20 mg, 48 mmol) in THF
-2 mL) at between 2100 and 2908C. After the addition the
temperature was maintained at 2908C for 30 min and the
reaction was then quenched with sat. aqueous NH4Cl -5 mL)
and allowed to warm to rt. The mixture was diluted with
EtOAc -5 mL) and the organic phase separated. The
aqueous phase was extracted with EtOAc -2£5 mL) and
the combined organic phases were dried -MgSO4). Evapora-
tion of the solvent in vacuo yielded an oil. 1H NMR analysis
of the oil indicated a mixture consisting of the residual
borane by-product and a 10:1 mixture of 23 and its C8
epimer 24, respectively. The mixture was puri®ed using
preparative TLC -75% Et2O/hexanes) collecting the bands
at Rf 0.61 and 0.71 which, respectively, were 23 -14.6 mg,
73%) and the C8 epimer 24 contaminated with other
crystal structures. P. P. is grateful to the Australian Research
Council for ®nancial support and S. B. is grateful to the
Australian Government for an Australian Postgraduate
Award scholarship.
References
1. Lewis, R. J. Memoirs of the Queensland Museum 1994, 34,
549±553.
2. Lewis, R. J. Toxicon 2001, 39, 97±106.
3. Campi, E. M.; Eriksson, L.; Guy, S. T.; Jackson, W. R.;
Perlmutter, P. J. Mol. Catal. 1998, 143, 243±252.
4. Bond, S.; Perlmutter, P. J. Chem. Soc., Chem. Commun. 2000,
567±568.
5. Oishi, T.; Maruyama, M.; Shoji, M.; Maeda, K.; Kumahara,
N.; Tanaka, S.; Hirama, M. Tetrahedron 1999, 55, 7471±
7498.
24
unidenti®ed impurities -1.8 mg). 23. [a]D ˆ295.9 -c 1.0,
CHCl3); FTIR -CHCl3) 3486bs, 2938s, 1759m, 1725s cm21
;
6. Oishi, T.; Shoji, M.; Maeda, K.; Kumahara, N.; Hirama, M.
Synlett 1996, 1165±1167.
1H NMR -300 MHz, [D6] acetone) d 1.36±1.42 -m, 2H),
1.52±1.64 -m, 8H), 2.20±2.47 -m, 2H), 2.58±2.68 -m, 1H),
2.79±2.94 -m, 1H), 3.80 -dd, Jˆ8.4, 6.9 Hz, 1H), 4.04 -dd,
Jˆ8.4, 6.7 Hz, 1H), 4.15±4.35 -m, 1H), 4.46 -dt, Jˆ6.8,
3.3 Hz, 1H), 4.51 -d, Jˆ6.0 Hz, 1H), 4.73 -bdd, Jˆ,8.6,
3.4 Hz, 1H), 5.02±5.21 -m, 1H), 5.69±5.85 -m, 1H), 5.76±
5.89 -m, 1H), 7.51±7.57 -m, 2H), 7.57±7.70 -m, 1H), 8.06±
8.10 -m, 2H); 13C NMR -75 MHz, [D6] acetone) d 24.6,
25.9, 30.3, 35.3, 36.0, 36.4, 65.7, 71.2, 72.4, 75.4, 79.9,
110.1, 124.9, 129.5, 130.4, 130.6, 132.9, 134.3, 166.1,
171.6; HRMS Calcd for C23H28O7Na -M1Na): m/z
439.173, Found 439.173. 24. Resonances used for diastereo-
meric ratio determination: 1H NMR -300 MHz, [D6]
acetone) d 4.88±4.93 -m, 1H, H9).
7. Curtis, N. R.; Holmes, A. B.; Looney, M. G. Tetrahedron Lett.
1992, 33, 671±674.
8. Curtis, N. R.; Holmes, A. B. Tetrahedron Lett. 1992, 33, 675±
678.
9. Bendall, J. G.; Payne, A. N.; Screen, T. E. O.; Holmes, A. B. J.
Chem. Soc., Chem. Commun. 1997, 1067±1068.
10. Carling, R. W.; Clark, J. S.; Holmes, A. B. J. Chem. Soc.,
Perkin Trans. 1 1992, 83±94.
11. Fujiwara, K.; Awakura, D.; Tsunashima, M.; Nakamura, A.;
Honma, T.; Murai, A. J. Org. Chem. 1999, 64, 2616±2617.
12. Nicolaou, K. C.; Wallace, P. A.; Shi, S.; Ouellette, M. A.;
Bunnage, M. E.; Gunzner, J. L.; Agrios, K. A.; Shi, G.;
È
Gartner, P.; Yang, Z. Eur. J. Chem. 1999, 5, 618±627.
13. Nicolaou, K. C.; Prasad, C. V. C.; Ogilvie, W. W. J. Am.
Chem. Soc. 1990, 112, 4988±4989.
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 144205, 144206 and 166204.
Copies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge CB2
1EZ, UK -fax: 144-0)-1223-336033 or e-mail: deposit@
ccdc.cam.ac.uk). Deposited data may be accessed by the
journal and checked as part of the refereeing process. If
data are revised prior to publication, a replacement ®le
should be sent to CCDC.
14. Nicolaou, K. C.; McGarry, D. G.; Somers, P. K.; Kim, B. H.;
Ogilvie, W. W.; Yiannikouros, G.; Prasad, C. V. C.; Veale,
C. A.; Hark, R. R. J. Am. Chem. Soc. 1990, 112, 6263±6276.
15. Nicolaou, K. C.; McGarry, D. G.; Somers, P. K.; Veale, C. A.;
Furst, G. F. J. Am. Chem. Soc. 1987, 109, 2504±2506.
16. Wimalasena, K.; Mahindaratne, M. P. D. J. Org. Chem. 1994,
59, 3427±3432.
17. Bell, R. P.; Rivlin, V. G.; Waters, W. A. J. Chem. Soc. 1958,
1696±1697.
18. Goosen, A.; Laue, H. A. H. J. Chem. Soc. C 1969, 995±997.
19. Goosen, A.; Laue, H. A. H. J. Chem. Soc. B 1969, 383±385.
20. O'Dell, D. E.; Loper, J. T.; Macdonald, T. L. J. Org. Chem.
1988, 53, 5225±5229.
Acknowledgements
 Â
21. Freire, R.; Marrero, J. J.; Rodrõguez, M. S.; Suarez, E.
Tetrahedron Lett. 1986, 27, 383±386.
We are grateful to Dr G. D. Fallon for determining X-ray