Enantioselective total synthesis of (+)-eutypoxide B

Article abstract of DOI:10.1021/jo9617326

The enantioselective synthesis of (+)-eutynoxide B, a metabolite isolated from the culture medium of the fungus Eutypa lata, is described. Two highly stereoselective and efficient sequential 1,4-additions to enone systems, derived from D-(-)-quinic acid, are the key to the synthesis. The synthesis of a diastereoisomer of (+)-eutypoxide B, was also accomplished.

Full text of DOI:10.1021/jo9617326

3984
J . Org. Chem. 1997, 62, 3984-3988
En a n tioselective Tota l Syn th esis of (+)-Eu typ oxid e B
M. Teresa Barros,†,‡ Christopher D. Maycock,*,‡,§ and M. Rita Ventura‡
Faculdade de Cieˆncias e Tecnologia da Universidade Nova de Lisboa, Departamento de Qu´ımica,
2825 Monte da Caparica, Instituto de Tecnologia Qu´ımica e Biolo´gica, Rua da Quinta Grande 6,
Apartado 127, 2780 Oeiras, and Faculdade de Cieˆncias da Universidade de Lisboa, Departamento de
Quı´mica, Rua Ernesto de Vasconcelos, 1700 Lisboa, Portugal
Received September 10, 1996X
The enantioselective synthesis of (+)-eutypoxide B, a metabolite isolated from the culture medium
of the fungus Eutypa lata, is described. Two highly stereoselective and efficient sequencial 1,4-
additions to enone systems, derived from D-(-)-quinic acid, are the key to the synthesis. The
synthesis of a diastereoisomer of (+)-eutypoxide B, was also accomplished.
In tr od u ction
conditions and by employing relatively inexpensive re-
agents and equipment.
Recently, several highly oxygenated cyclohexane-based
metabolites, in many cases epoxides, have been isolated
from bacteria, fungi, higher plants, and even molluscs.
These compounds have diverse biological activities rang-
ing from antifungal, antibacterial, and antitumor to
phytotoxic and enzyme inhibitory.1-3
The fungus, Eutypa lata, is one of the pathogenic
agents responsible for the vineyard die-back disease.1,2
This disease is known as “dying arm” or “eutypiosis” and
is linked to toxic secondary fungal metabolites. E. lata
is widely distributed in several countries and is known
to attack many other woody species.2
Eutypoxide B, 1, is one of the secondary metabolites,
isolated from the culture medium of E. lata, but has not
yet been identified as the cause of the above mentioned
disease. For further biological studies, it was necessary
to obtain larger quantities by an efficient synthesis. Its
enantioselective synthesis, however, poses a challenge in
that the core cyclohexane ring is highly substituted and
with groups requiring mild reaction conditions. Its
racemic synthesis has been reported2 and also an asym-
metric4 one by a route which is dependent upon a retro-
Diels-Alder deprotection of a carbon-carbon double
bond.
Resu lts a n d Discu ssion
The starting material chosen was (-)-quinic acid, 2, a
cyclohexane-based natural substance isolated from the
plant species Cinchona which has been used successfully
in several other syntheses.5-11 It has the advantage of
being readily available and suitably functionalized for
conversion to a wide range of interesting structures,
particularly polysubstituted cyclohexanes. Our approach
capitalized upon the high stereoselectivity of 1,4-addi-
tions to R,â-unsaturated ketones derived from quinic
acid.12 We demonstrate here the use of two highly
stereoselective sequential 1,4-additions to introduce the
required configuration at key positions in the cyclohexane
ring. Our initial target was a simple and efficient
preparation of the intermediate 4 which was to be
the starting point of our synthesis. The first three
steps leading to â-hydroxy ketone 3 (Scheme 1) have
already been described elsewhere.13 The enone 4, [R]20
D
+137.4 (c 1.12, CH2Cl2), mp 28-29 °C, was obtained in
an excellent yield, 99%, by acetylation of the â-hydroxy
ketone 3, mp 79-80 °C, followed by a very facile14 in situ
elimination with diisopropylethylamine. Interestingly,
the corresponding benzoate9 is more stable than the
acetate and can be readily isolated. Compound 4 has also
been prepared from the same â-hydroxy ketone by
Danishefsky and others12 using sulfonation followed by
elimination in a one-pot process.
(5) Eberle, M.; Arigoni, D. Helv. Chem. Acta 1960, 43, 1508.
(6) Barton, D. H. R.; Gero, S. D.; Maycock, C. D. J . Chem. Soc.,
Chem. Commun. 1980, 1089.
(7) Elliot, J . D.; Kelson, A. B.; Purcell, N.; Stoodley, R. J . J . Chem.
Soc., Perkin Trans. 1 1983, 2441.
(8) Trost, B. M.; Romero, A. G. J . Org. Chem. 1986, 51, 2332.
(9) Barros, M. T.; Godinho, L. S.; Maycock, C. D.; Santos, A. G.
Tetrahedron Lett. 1992, 33, 4633.
(10) Kamenecka, T. M.; Overman, L. E. Tetrahedron Lett. 1994, 35,
4279.
(11) Toyota, M.; Nishikawa, Y.; Fukumoto, K. Tetrahedron 1996,
52, 10347.
(12) (a) Audia, J . E.; Boisvert, L.; Patten, A. D.; Villalobos, A.;
Danishefsky, S. J . J . Org. Chem. 1989, 54, 3738. (b) J eroncic, L. O.;
Cabal, M.-P.; Danishefsky, S. J . J . Org. Chem. 1991, 56, 387. The
cyclohexylidene analogue of compound 4 had been observed previously
in work related to ref 6.
(13) Grewe, R.; Nolte, E. Ann. Chem. 1952, 575, 1. Haslam, E.
Shikimic acid Metabolism and Metabolites; Wiley-Interscience: New
York, 1993.
(14) The ease with which this elimination occurs is probably due to
the boat conformation of the cyclohexane ring which is imposed by the
isopropylidene acetal ring. In unconstrained systems this elimination
occurs less readily. See ref 13.
We report here an efficient enantioselective synthesis
of (+)-eutypoxide B, 1, the enantiomer corresponding to
the natural product,2 without recourse to drastic reaction
† Faculdade de Cieˆncias e Tecnologia Universidade Nova de Lisboa.
‡ Instituto de Tecnologia Qu´ımica e Biolo´gica.
§ Faculdade de Cieˆncias da Universidade de Lisboa.
X Abstract published in Advance ACS Abstracts, May, 1, 1997.
(1) Renaud, J . M.; Tsoupras, G.; Stoeckli-Evans, H.; Tabacchi, R.
Helv. Chem. Acta 1989, 72, 1262.
(2) Defrancq, E.; Gordon, J .; Brodard, A.; Tabacchi, R. Helv. Chem.
Acta 1992, 75, 276.
(3) Kamikubo, T.; Ogasawara, K. Tetrahedron Lett. 1995, 36, 1685.
Nakamori, K.; Natsuura, H.; Ioshihara, A.; Koda, I. Phytochemistry
1994, 35, 835. Ichihara, A.; Kobayashi, M.; Oda, K.; Sakamura, S.
Tetrahedron 1979, 35, 2861. Ichihara, A.; Kobayashi, M.; Oda, K.;
Sakamura, S. Tetrahedron 1980, 36, 183. Muhlenfeld, A.; Achenbach,
H. Phytochemistry 1988, 27, 3853. Ichihara, A. Synthesis 1987, 207.
(4) Takano; S.; Moriya, M.; Ogasawara, K. J . Chem. Soc., Chem.
Commun. 1993, 614.
S0022-3263(96)01732-X CCC: $14.00 © 1997 American Chemical Society

Enantioselective Total Synthesis of (+)-Eutypoxide B
J . Org. Chem., Vol. 62, No. 12, 1997 3985
Sch em e 1a
complex is soluble in mixtures of Me2S and ethereal
solvents, and the sulfide ligand, Me2S (bp 37 °C), is easily
separated from reaction products. We used a stoichio-
metric quantity of Me2S(CuBr) to try to prevent the
occurrence of uncomplexed vinylmagnesium bromide that
led to compound 17. The temperature was kept below
-20 °C, because of the thermal instability of the vinyl
cuprate reagent, and above -40 °C to achieve a reason-
able reaction rate between the vinylmagnesium bromide
and Cu(I). We obtained the 1,4-addition product as the
sole isolable material, but the yield was only 39%, since
partial decomposition of the starting material had oc-
curred.
To improve the yield of this step, we took advantage
of the greater stability and reactivity of the higher-order
cyanocuprate17 (CHdCH2)2Cu(CN)Li2. To prepare this
reagent in situ, we also had to prepare the vinyllithium
necessary for this reaction18 by transmetalation on tet-
ravinyltin. The conjugate addition of (CHdCH2)2Cu-
(CN)Li2 to enone 4 afforded the ketone 5 [R]20 +26.3 (c
D
2.05, CH2Cl2) and its base-induced elimination product
6. No addition to this double bond was observed even
though an excess of organocuprate was used so we
assume that elimination occurred after the reagent had
been destroyed. Since 6 was the next required product
it was unnecessary to optimize this reaction or to purify
this mixture. Thus, it was treated with a catalytic
quantity of aqueous 0.5 N NaOH in THF, to convert all
a
(a) Acetone, dry HCl, 89%; (b) Ac2O, pyridine, 92%; (c) (i)
LiAlH4, Et2O; (ii) NaIO4, H2O, 5 < pH < 6, 91% (two steps); (d)
Ac2O, (i-Pr)2NEt, DMAP, CH2Cl2, 0 °C, 99%; (e) (CHdCH2)2-
CuCNLi2, Et2O, -30 °C; (f) 0.5 N NaOH, THF, 0 °C; (g) TBDMSCl,
(i-Pr)2NEt, CH2Cl2, DMAP, 0 °C/rt 60%; (e-h) t-BuOOH, Triton
B, THF, 0 °C, 94%; (i) CeCl3‚7H2O, NaBH4, methanol, -78 °C/0
°C, 98% (1:1, 9:10).
the remaining ketone 5 into the enone 6, [R]20 -109.3
D
(c 1.85, CH2Cl2), in virtually quantitative yield. Without
purification, we proceeded to the protection of the hy-
droxyl group (Scheme 1) to obtain the silyl ether 7, [R]20
D
-142.9 (c 0.96, CH2Cl2), in 82% yield. The overall yield
of these three steps, without purification of the interme-
diates, was 60%, the lowest individual yield being from
the 1,4-addition reaction. In all previous uses of these
compounds as intermediates,10-12 one of the double bonds
has been saturated such that the elimination served as
a deoxygenation method. This synthesis used the func-
tionality to the fullest, and both double bonds were used
to introduce chirality and to functionalize the ring.
Our first challenge was to introduce a vinyl group by
a 1,4-addition reaction at the R,â-unsaturated ketone. It
was thought that a vinyl group would survive the
conditions necessary to arrive at a point where an
aldehyde group could be generated from it and onto which
the five-carbon side chain could be constructed. We
envisioned that this aldehyde would be generated by
ozonolysis of the double bond, a much simpler and cleaner
method than dihydroxylation and oxidation. Our initial
efforts to obtain the vinyl ketone 5, via 1,4-addition, with
vinylmagnesium bromide and catalytic or stoichiometric
amounts of CuI or CuCN15 were unsuccessful, due to the
insolubility of these Cu(I) salts in ethereal solvents and
the partial decomposition of the thermally unstable vinyl
cuprate formed. Thus, instead of a regiospecific 1,4-
addition of the vinyl group, we obtained a mixture of
On exposure to t-BuOOH and a catalytic amount of
Triton B, in THF,19 7 afforded exclusively the epoxide 8,
[R]20 -9.24 (c 1.12, CH2Cl2), in 94% isolated yield,
D
revealing the influence of the adjacent bulky TBS ether
group on the selectivity of the epoxidation which arises
from a second 1,4-addition. The use of the less bulky
epoxidizing agent, hydrogen peroxide, in place of the
above, resulted in the loss of diastereofacial selectivity,
producing a mixture of epoxides which proved difficult
to separate.
ketone 5 and the 1,2-addition product 17, [R]20 +29.0
D
(c 0.65, CH2Cl2) which was difficult to separate.
Reduction of the ketonic carbonyl group with sodium
borohydride in methanol was readily achieved, and we
obtained the two diastereoisomers 9, [R]20 +5.9 (c 1.74,
D
CH2Cl2), and 10, [R]20 -48.7 (c 0.875, CH2Cl2), in high
D
yield with a diastereoselectivity of 40:60 (ratio 9/10, 0.7).
This ratio demonstrates the somewhat unexpected low
influence of the t-BuMe2Si group on the diastereoselec-
tivity of the reduction.
In an attempt to avoid this side reaction, we used the
readily available, crystalline complex, Me2S(CuBr).16 This
(17) Lipshutz, B. H.; Wilhelm, R. S.; Kozlowski, J . A. Tetrahedron
1984, 40, 5005. Lipshutz, B. H. Synthesis 1987, 325.
(18) Nugent, W. A.; Hobbs, F. W., J r. Organic Syntheses; Wiley: New
York, 1993; Coll. Vol. VIII, p 112.
(19) Grieco, P. A.; Nishizawa, M.; Oguri, T.; Burke, S. D.; Marinovic,
N. J . Am. Chem. Soc. 1977, 99, 5773. Yang, N. S.; Finnegan, R. A. J .
Am. Chem. Soc. 1958, 80, 5845.
(15) House, H. O.; Respess, W. L.; Whitesides, G. M. J . Org. Chem.
1966, 31, 3128. Kretchemer, R. B.; Mihelich, E. D.; Waldron, J . J . J .
Org. Chem. 1972, 37, 4483.
(16) House, H. O.; Chu, C.-Y.; Wilkins, J . M.; Umen, M. J . J . Org.
Chem. 1975, 40, 1460. Shea, K. J .; Pham, P. Q. Tetrahedron Lett. 1983,
24, 1003. House, H. O.; Umen, M. J . J . Org. Chem. 1973, 38, 3893.
(20) Barros, M. T.; Alves, C. M.; Santos, A. G.; Godinho, L. S.;
Maycock, C. D. Tetrahedron Lett. 1995, 36, 2321.

3986 J . Org. Chem., Vol. 62, No. 12, 1997
Barros et al.
Sch em e 2a
Having achieved this biprotected cyclic nucleus, we
then proceeded to build up the side chain. Ozonolysis of
11 produced the expected intermediate ozonide 18 which
proved to be surprisingly stable and could be isolated
readily. Both diastereomers were evident from the NMR
spectrum and in fact the ozonolysis was slightly diaste-
reoselective. This ozonide was resistant to extended
periods in the presence of excess dimethyl sulfide and
zinc/acetic acid but was reduced rapidly to the aldehyde
12, [R]20 -3.1 (c 1.03, CH2Cl2) by triphenylphosphine
D
in 92% overall yield for the two steps. No simultaneous
attack at the epoxide oxygen was observed.
Owing to the insolubility of the inexpensive Grignard
reagent, (2-methyl-1-propen-3-yl)magnesium chloride in
ethereal solvents, the exact concentration of the reagent
was difficult to determine. An excess of the organomag-
nesium reagent produced undesirable products formed
by epoxide ring opening. Thus, the addition to the
aldehydic carbonyl of 12 had to be carried out with
precaution in order to assure the exclusive formation of
the two diastereoisomeric alcohols 13 and 14. By careful
addition of the Grignard with monitoring, we were able
to obtain a good yield of the two separable diastereomers
13 [R]20 -14.3 (c 1.295, CH2Cl2) and 14 [R]20 -7.2 (c
a
( j) TBDMSCl, CH2Cl2, (i-Pr)2NEt, DMAP, 0 °C/rt, 99%; (k) (i)
O3, CH2Cl2, -78 °C; (ii) PPh3, Et2O, rt, 92% (two steps); (l)
CH3C(CH2)CH2MgCl, Et2O, 0 °C, 96%; (m) periodinane, pyridine,
CH2Cl2, 0 °C/rt, 99%; (n) DBN, CH2Cl2, rt, 99%; (o) Bu4NF, THF,
cat. H2O, rt, 93%.
We attempted to improve the diastereoselectivity of
this reduction, by adding 1 equiv of cerium(III) chloride,20
but we only succeeded in increasing the ratio 9/10 of the
two alcohols to unity. By performing the reduction with
zinc borohydride, the diastereomeric ratio was raised to
about 60:40, 9:10, but the yield was unsatisfactory. This
improvement was probably due to complexation of the
zinc21 to the epoxide oxygen thus delivering the hydride
to the upper face. However, since the secondary products
appeared to be derived from epoxide ring opening we
suspect that an interaction between the basic epoxide
oxygen atom with the acidic zinc is beneficial for the
reduction but also promotes epoxide ring opening and is
thus detrimental to the overall yield of the required
product. We have not studied this reaction exhaustively,
but we believe that it is possible to increase the diaste-
reoselectivity of this reduction by varying the reaction
conditions. Although not attempted Mitsunobu technol-
ogy should provide a means of converting 10 to 9 in two
steps.
D
D
0.57, CH2Cl2), mp 51-52 °C. At this stage in our
asymmetric synthesis, we had obtained, with good ste-
reocontrol up to the reduction step (8 to 9), compounds
with six chiral centers.
Oxidation of the mixture of 13 and 14 was accom-
plished with periodinane, a mild and selective reagent
for the oxidation of primary and secondary alcohols to
aldehydes and ketones.22 By performing the reaction in
the presence of pyridine and using the thiosulfate work-
up, with sodium bicarbonate buffer,22 it was possible to
maintain nearly neutral conditions throughout the entire
reaction and isolation sequence. In this way, we obtained
compound 15, [R]20 -7.4 (c 0.81, CH2Cl2), mp 59-60°,
D
in almost quantitative yield.
For the quantitative isomerization of 15 to give the R,â-
unsaturated ketone 16, [R]20 -6.1 (c 1.12, CH2Cl2), we
D
used a catalytic amount of DBN, instead of 1 equiv of
the related base DBU, as was reported.4 These last three
steps could be carried out successfully without purifica-
tions which reduced significantly the time consumed for
this synthesis.
The reduced compound also provided us with an
opportunity to assign the signals for H-2 and H-3 in the
NMR spectrum. In compound 8 the corresponding H
NMR signals indicated vicinal proton coupling only
between themselves (H-2 and H-3). Upon reduction to
alcohols 9 and 10 additional coupling was observed for
one of the proton signals (H-2) along with an expected
upfield shift compared to 8.
1
Finally, the removal of the two TBS groups was
achieved under nonacidic conditions and in excellent yield
using Bu4NF in THF containing a small amount of water.
This afforded (+)-eutypoxide B, 1, [R]23 +57.3 (c 0.84,
D
anhydrous CHCl3) (lit.4 (-)-eutypoxide (ent-1), [R]23D -56.6
(c 0.68, CHCl3)) identified by spectroscopic comparison2
as the natural product isolated from the fungus E. lata.
This deprotection afforded several byproducts when
carried out under anhydrous conditions with the same
reagent. On the other hand no reaction occurred when
The two diastereoisomers were separated, and the
alcohol 9 was protected with TBDMSCl, in the presence
of diisopropylethylamine and DMAP, to afford fully
protected compound 11, [R]20 +10.3 (c 1.23, CH2Cl2), in
D
99% yield (Scheme 2).
(22) Dess, D. B.; Martin, J . C. J . Org. Chem. 1987, 48, 4155. Dess,
D. B.; Martin, J . C. J . Am. Chem. Soc. 1991, 113, 7277.
(21) Cerium could also function in a similar way to zinc but without
the consequent ring opening.
(23) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of
Laboratory Chemicals, 2nd ed.; Pergamon Press: New York, 1980.

Enantioselective Total Synthesis of (+)-Eutypoxide B
J . Org. Chem., Vol. 62, No. 12, 1997 3987
tion in the next step. In order to characterize compounds 5
and 6, a sample of the previous mixture was purified by
preparative TLC (AcOEt/hexane 4:6) to afford 0.039 g (34%)
of 5 and 0.24 g (29%) of 6, both as colorless oils. Compound 5:
this transformation was attempted using more than a
trace amount of water.
The epimeric alcohol 10, was converted, following the
same reaction sequence, into compound 19, [R]20 -13.7
D
[R]20 +26.3 (c 2.05, CH2Cl2). 1H NMR (CDCl3): δ 5.89-5.80
D
(c. 0.52, CH2Cl2), a previously unreported diastereoisomer
(1H, m); 5.22 (1H, d, J ) 10.0 Hz); 5.14 (1H, d, J ) 17.8 Hz);
4.61-4.58 (1H, q,); 4.38 (1H, dd, J ) 7.0 Hz, J ) 3.8 Hz); 2.88
(1H, m); 2.68-2.59 (3H, m); 2.37 (1H, dd, J ) 17.7 Hz, J ) 5.6
Hz); 1.48 (3H, s); 1.37 (3H, s). 13C NMR (CDCl3): δ 208.3;
137.1; 116.7; 108.1; 75.3; 72.3; 41.4; 40.4; 38.4; 26.5; 23.9. FT-
IR (film): 2987,2935, 2912 (CsH); 1718 (CdO). Compound
of 1, in good overall yield.
Con clu sion
The fungal metabolite eutypoxide B has been synthe-
sized in a stereospecific manner using an efficient
transformation of quinic acid. This synthesis demon-
strates that similar technology may be useful for the
synthesis of analogous metabolites. We are attempting
such syntheses as well as studying the 1,4-additions to
other, quinic acid derived, cyclohexenone systems in order
to develop routes to the syntheses of the enantiomers of
some of the intermediates described here.
6: [R]20 -109.3 (c 1.85, CH2Cl2). 1H NMR (CDCl3): 6.96 (1H,
D
d, J ) 10.3 Hz); 6.01 (1H, d, J ) 10.0 Hz); 5.80 (1H, m); 5.31
(2H, d, J ) 5.5 Hz);4.35 (1H, d, J ) 8.6 Hz); 2.72 (1H, m); 2.58
(1H, dd, J ) 16.7 Hz, J ) 4.0 Hz); 2.35 (1H, dd, J ) 16.9 Hz,
J ) 13.3 Hz). FT-IR (Film): 3433 (OdH); 2876 (CdH); 1685
(CdO, R,â-unsat ketone).
(4S,5R)-4-Hyd r oxy-5-vin yl-2-cycloh exen -1-on e (6). To
a solution of 5 (1.476 g, 7.5 mmol) in THF, at 0 °C, was added
a catalytic amount of aqueous NaOH (0.5 N). The mixture
was stirred at 0 °C until all the starting material had been
consumed. Saturated aqueous NH4Cl was added, and the
mixture was extracted with ethyl ether. The organic extracts
were dried (MgSO4) and concentrated. The residue obtained
(0.990 g) was used in the next reaction without further
purification. The NMR spectrum of 6 is described in the
previous experiment.
Exp er im en ta l Section
Gen er a l. Melting points were determined with a capillary
apparatus and are uncorrected. 1H NMR spectra were ob-
tained at 300 MHz in CDCl3 with chemical shift values (δ) in
ppm downfield from tetramethylsilane, and 13C NMR spectra
were obtained at 100.61 MHz in CDCl3. Microanalyses were
performed by the ITQB analytical services using a combustion
apparatus. FT-IR (ν, cm-1): liquid samples were measured
as thin films on NaCl windows. Medium pressure preparative
column chromatography: silica gel Merck 60 H. Preparative
TLC: silica gel Merck 60 GF254. Analytical TLC: Aluminum-
backed silica gel Merck 60 F254. Specific rotations ([R]20D) were
measured on an automatic polarimeter. Reagents and solvents
were purified and dried according to ref 23. All the reactions
were carried out in an inert atmosphere (argon), unless
otherwise indicated.
(4S,5R)-4-[(ter t-Bu t yld im et h ylsilyl)oxy]-5-vin yl-2-cy-
cloh exen -1-on e (7). To a solution of 6 (0.831 g, 6.0 mmol) in
CH2Cl2 (4.16 mL), at 0 °C, were added diisopropylethylamine
(2.6 mL, 15.5 mmol, 2.5 equiv), a catalytic amount of DMAP,
and tert-butyldimethylsilyl chloride (TBDMSCl) (1.812 g, 12
mmol, 2 equiv) in CH2Cl2 (4.16 mL). The solution was stirred
at rt for 2 days. Water was then added, and the mixture was
vigorously stirred for 15 min and extracted with CH2Cl2. The
organic layer was dried (MgSO4) and evaporated. The residue
was purified by column chromatography (AcOEt/hexane 2/98)
to afford 1.24 g (82%) of 7 as a colorless oil. [R]20 -142.9 (c
D
0.96, CH2Cl2). 1H NMR (CDCl3): δ 6,78 (1H, dd, J ) 10.2 Hz,
J ) 2.1 Hz); 5.95 (1H, d, H-2); 5.821 (1H, m); 5.16 (1H, s); 5.11
(1H, d, J ) 9.2 Hz); 4.27 (1H, tt); 2.75 (1H, m); 2.58 (1H, dd,
J ) 16.4 Hz, J ) 4.0 Hz); 2.32 (1H, dd, J ) 16.5 Hz, J ) 12.6
Hz); 0,92 (9H, s); 0.12, 0.11 (6H, 2s). FT-IR (Film): 1693
(CdO, R,â-unsat ketone).
(4S ,5R )-4,5-(Isop r op ylid e n e d ioxy)cycloh e x-2-e n -1-
on e (4). To a solution of 3 (1.5 g, 8 mmol) in CH2Cl2 (6 mL),
at 0 °C, was added a catalytic amount of 4-(dimethylamino)py-
ridine (DMAP), diisopropylethylamine (2.8 mL, 16 mmol, 2
equiv), and acetic anhydride (0.91 mL, 9.6 mmol, 1.2 equiv).
After stirring for 3 h at 0 °C all of the starting material had
been consumed. The reaction mixture was washed with
saturated aqueous NaHCO3 solution and extracted with
CH2Cl2. The organic layer was dried (MgSO4) and concen-
trated. Kugelrohr distillation gave 1.34 g (99%) of 4, as a
(2S,3R,4S,5R)-4-[(ter t-Bu tyldim eth ylsilyl)oxy]-2,3-epoxy-
5-vin ylcycloh exa n -1-on e (8). To a solution of 7 (0.723 g,
2.9 mmol) in THF (6 mL) at 0 °C was added Triton B
(N-benzyltrimethylammonium hydroxide, 40 wt % solution in
methanol, 0.069 mL, 0.023 mmol, 0.058 equiv) and t-BuOOH
(tert-butyl hydroperoxide anhydrous, 3.0 M solution in isooc-
tane, 4.77 mL, 14.3 mmol, 5 equiv). After 2 h at 0 °C, water
and saturated aqueous NH4Cl were added, and the mixture
was extracted with ethyl ether, dried (MgSO4), and evaporated
to give a colorless residue, which was purified by column
chromatography. Elution with AcOEt/hexane 2/98 afforded 8
(0.723 g, 94%) as a colorless oil. [R]20D -9.24 (c 1.12, CH2Cl2).
1H NMR (CDCl3): δ 5.76 (1H, m); 5.11 (1H, d, J ) 7.2 Hz);
5.02 (1H, s); 3.99 (1H, d, J ) 7.2 Hz); 3.46 (1H, d, J ) 3.7 Hz);
3.27 (1H, d, J ) 3.6 Hz); 2.57 (2H, m); 2.19 (1H, d, J ) 10.3
Hz); 0.91 (9H, s); 0.15, 0.10 (6H, 2s). FT-IR (film): 1728 (CdO,
ketone).
colorless oil that slowly crystallized as white needles. [R]20
D
+137.4 (c 1.12, CH2Cl2) (lit. [R]D +147.5 (c 0.48, CHCl3)). mp
28-29 °C (lit. 39-40 °C). 1H NMR (CDCl3): δ 6.83 (1H, d, J
) 10.4 Hz); 6.20 (1H, d, J ) 10.4 Hz); 4.90 (1H, s); 4.86 (1H,
s); 3.10 (1H, d, J ) 17.6 Hz); 2.97 (1H, dd, J ) 17.6 Hz, J )
4.1 Hz); 1.55 (3H, s); 1.54 (3H, s). 13C NMR (CDCl3): δ 195.3;
145.8; 128.8; 109.9; 73.3; 71.0; 38.7; 27.7; 26.5. FT-IR (film):
2989, 2935, 2910 (CsH); 1682 (CdO, R,â-unsat ketone). Anal.
Calcd for C9H12O3 (168.19419): C 64.27, H 7.19. Found: C
64.03, H 7.23.
(3R,4S,5R)-3,4-(Isop r op ylid en ed ioxy)-5-vin ylcycloh ex-
a n -1-on e (5). A three-necked, round-bottomed flask was
flame-dried and fitted with an internal thermometer. Under
a positive pressure argon atmosphere, the flask was charged
with tetravinyltin (1.937 mL, 0.011 mol) and anhydrous ether
(48.1 mL). The solution was cooled to 0 °C, and methyllithium
in ether (24.18 mL, 1.6 M, 0.039 mol) was slowly added to the
stirred solution. After 15 min, the vinyllithium mixture was
cooled to -78 °C for 20 min. The septum on one neck was
briefly removed, and copper(I) cyanide (1.846 g, 0.021 mol) was
added all at once. The bath and reaction were allowed to warm
slowly over about 1 h, until the internal temperature attained
-30 °C. A solution of 4 (1.3 g, 7.7 mmol) in 3.12 mL in ether
was added dropwise to the cuprate at -30 °C. Saturated
aqueous NH4Cl was slowly added while the temperature of
the system was allowed to rise. The mixture was extracted
with ether, dried (MgSO4), and evaporated to yield a mixture
(0.825 g) of 5 and 6, which was used without further purifica-
(1R,2S,3R,4S,5R)-4-[(ter t-Bu tyld im eth ylsilyl)oxy]-2,3-
ep oxy-1-h yd r oxy-5-vin ylcycloh exa n e (9) a n d (1S,2S,-
3R,4S,5R)-4-[(ter t-Bu t yld im et h ylsilyl)oxy]-2,3-ep oxy-1-
h yd r oxy-5-vin ylcycloh exa n e (10). To a solution of 8 (0.544
g, 2.0 mmol) in methanol (6 mL) was added CeCl3‚7H2O (0.775
g, 2.08 mmol, 1 equiv). At -78 °C was added NaBH4 (0.308
g, 8.14 mmol, 4 equiv). The reaction mixture was stirred at
this temperature for 30 min and then at 0 °C for 30 min.
Saturated NH4Cl was added, followed by extractions with ethyl
ether. The organic extracts were dried (MgSO4), and the
solvent was evaporated to yield a residue. Purification by
column chromatography (AcOEt/hexane 1:9) afforded 0.268 g
of 9 and 0.262 g of 10 (98% yield, dr 1:1), both as colorless
oils. Compound 9: [R]20 +5.9 (c 1.74, CH2Cl2). 1H NMR
D
(CDCl3): δ 5.70 (1H, m); 5.08 (1H, d, J ) 7.0 Hz); 5.02 (1H, s);

3988 J . Org. Chem., Vol. 62, No. 12, 1997
Barros et al.
4.32 (1H, t); 3.60 (1H, d, J ) 9.0 Hz); 3.20 (1H, s); 3.14 (1H, d,
J ) 3.4 Hz); 2.28 (1H, m); 1.70 (1H, broad s); 1.54 (2H, m);
0.91 (9H, s); 0.12, 0.07 (6H, 2s). FT-IR (film): 3412 (O-H).
Anal. Calcd for C14H26O3Si (270.44702): C 62.18, H 9.69.
Found: C 62.20, H 9.73. Compound 10: [R]20D -48.7 (c 0.875,
CH2Cl2). 1H NMR (CDCl3): δ 5.70 (1H, m); 5.06 (1H, s); 5.02
(1H, d, J ) 2.9 Hz); 4.14-4.08 (1H, m); 3.59 (1H, d, J ) 9.1
Hz); 3.37 (1H, broad s); 3.19 (1H, d, J ) 3.8 Hz); 2.00-1.93
(1H, m); 1.87 (1H, broad s); 1.70 (1H, qq); 1.29 (1H, m); 0.90
(9H, s); 0.12, 0.06 (6H, 2s). 13C NMR (CDCl3): δ 139.5; 115.6;
70.4, 68.4; 60.0; 56.5; 46.0; 29.7; 25.7; 18.0; -4.5; -4.7. FT-
IR (film): 3410 (OsH).
IR (KBr): 3547 (O-H). Anal. Calcd for C23H46O4Si2
(442.79167): C 62.39, H 10.47. Found: C 62.43, H 10.26.
1-[(1R,2S,3R,4S,5R)-2,5-Bis[(ter t-bu tyldim eth ylsilyl)oxy]-
3,4-ep oxycycloh exyl]-3-m eth yl-3-bu ten -1-on e (15). To a
solution of a mixture of 13 and 14 (0.090 g, 0.2 mmol) in
CH2Cl2 (6 mL) at 0 °C were added periodinane (0.109 g, 0.26
mmol, 1.2 equiv) and pyridine (0.6 mL). The reaction mixture
was stirred at rt until all the starting material had been
consumed (30 min). The mixture was then diluted with ethyl
ether and poured into saturated aqueous NaHCO3 containing
a sevenfold excess of Na2S2O3. The mixture was stirred to
dissolve the solid, and the layers were separated. The ether
layer was washed with saturated NaHCO3 and with water.
After extractions with ethyl ether, the organic layer was dried
(MgSO4) and concentrated. Purification by preparative TLC
(AcOEt/hexane 1:9) gave 0.089 g (99%) of 15 that crystallized
(1R,2S,3R,4S,5R)-1,4-Bis[(ter t-bu tyld im eth ylsilyl)oxy]-
2,3-ep oxy-5-vin ylcycloh exa n e (11). To a solution of 9 (0.335
g, 1.2 mmol) in 3.1 mL of CH2Cl2 at 0 °C were added
diisopropylethylamine (0.54 mL, 3.2 mmol, 2.5 equiv), a
catalytic amount of DMAP, and TBDMSCl (0.367 g, 2.4 mmol,
2 equiv) in 3.1 mL of CH2Cl2. The solution was stirred at rt
for one day. Water was added, and the mixture was vigorously
stirred for 15 min and then extracted with CH2Cl2, dried
(MgSO4), and concentrated. The residue obtained was purified
by column chromatography (AcOEt/hexane 2.5/97.5) to yield
0.428 g (99%) of 11, as a white solid below 4 °C (at rt it is a
as white needles. [R]20 -7.4 (c 0.81, CH2Cl2). Mp 59-60 °C.
D
1H NMR (CDCl3): δ 4.93 (1H, s); 4.81 (1H, s); 4.24 (1H, broad
s); 4.13 (1H, d, J ) 9.5 Hz); 3.20-2.94 (5H, m); 1.74 (3H, s);
1.47-1.43 (2H, m); 0.94, 0.89 (18H, 2s); 0.15, 0.13, 0.12, 0.04
(12H, 4s). FT-IR (KBr): 1709 CdO, ketone); 1651 (CdC).
Anal. Calcd for C23H44O4Si2 (440.77573): C 62.68, H 10.06.
Found: C 62.63, H 10.33.
colorless oil). [R]20 +10.3 (c 1.23, CH2Cl2). 1H NMR
D
(CDCl3): δ 5.71 (1H, m); 5.04 (1H, d, J ) 5.7 Hz); 4.91 (1H, s);
4.24 (1H, d); 3.56 (1H, d, J ) 9.6 Hz); 3.11 (1H, d, J ) 2.9 Hz);
3.06 (1H, broad s); 2.36 (1H, m); 1.43-1.39 (2H, m); 0.92 (18H,
2s); 0.12, 0.11, 0.09, 0.07 (12H, 4s). 13C NMR (CDCl3): δ 140.8;
115.1; 71.4, 65.2; 58.0, 55.3; 39.4; 30.5; 25.8, 25.7; 18.1; -4.4,
-4.6, -4.8, -4.9. FT-IR (film): 3080, 2993, 2953, 2930, 2887,
2868 (C-H). Anal. Calcd for C20H40O3Si2 (384.711): C 62.44,
H 10.48. Found: C 62.54, H 10.36.
1-[(1R,2S,3R,4S,5R)-2,5-Bis[(ter t-bu tyldim eth ylsilyl)oxy]-
3,4-epoxycycloh exyl]-3-m eth yl-2-bu ten -1-on e (16). A cata-
lytic amount of DBN was added to a solution of 15 (0.061 g,
0.14 mmol) in CH2Cl2 (0.6 mL). After stirring at rt for about
6 h, water was added, the mixture was extracted with CH2Cl2,
the organic layer was dried (MgSO4), and the solvent was
evaporated. The residue was purified by preparative TLC
(AcOEt/hexane 1:9) to afford 0.060 g (99%) of 16 as a colorless
oil. [R]20D -6.1 (c 1.22, CH2Cl2). 1H NMR (CDCl3): δ 6.06 (1H,
s); 4.25 (1H, d, J ) 2.2 Hz); 4.14 (1H, d, J ) 9.5 Hz); 3.09 (1H,
d, J ) 3.5 Hz); 3.06 (1H, broad s); 2.85-2.76 (1H, m); 2.14
(3H, s); 1.89 (3H, s); 1.48-1.42 (2H, m); 0.92, 0.86 (18H, 2s);
0.13, 0.12, 0.11, 0.01 (12H, 4s). FT-IR (film): 1691 (CdO, R,â-
unsat ketone); 1620 (CdC). Anal. Calcd for C23H44O4Si2
(440.77573): C 62.68, H 10.06. Found: C 62.81, H 10.16.
(1R,2S,3R,4S,5R)-[2,5-Bis(ter t-bu tyld im eth ylsilyl)oxy]-
3,4-ep oxycycloh exyl]m eth a n a l (12). Into a solution of 11
(0.290 g, 0.83 mmol) in 30 mL of CH2Cl2 at -78 °C was bubbled
ozone till the solution turned blue. The temperature was
allowed to rise, and the solvent was evaporated. The residue
was dissolved in ethyl ether (6 mL), and triphenylphosphine
(0.408 g, 1.6 mmol, 2 equiv) was added. The solution was
stirred at rt until all the ozonide had been consumed. The
solvent was evaporated, and the residue was purified by
column chromatography (AcOEt/hexane 0.5/99.5) to afford
1-[(1R,2S,3R,4S,5R)-3,4-Epoxy-2,5-dih ydr oxycycloh exyl]-
3-m eth yl-2-bu ten -1-on e (1). To a solution of 16 (0.083 g, 0.2
mmol) in THF (2.1 mL) at rt, was added a catalytic quantity
of water followed by Bu4NF (0.122 g, 0.47 mmol, 2.5 equiv).
The mixture was stirred at rt till all the starting material had
been consumed. The solution was diluted with ethyl acetate,
and water was added. After stirring for 5 min, the mixture
was washed with saturated NaCl and extracted with ethyl
acetate. The evaporation of the solvent gave a residue that
was purified by preparative TLC (AcOEt) to yield 1 (0.037 g,
93%) as a very viscous colorless oil that foamed under vacuum;
its 1H NMR data were identical with those of the natural
0.295 g (92%) of 12 as a colorless oil. [R]20 -3.1 (c 1.03,
D
CH2Cl2). 1H NMR (CDCl3): δ 9.70 (1H, s); 4.27 (1H, broad s);
4.19 (1H, d, J ) 8.8 Hz); 3.12-3.09 (2H, m); 2.67 (1H, m); 1.64
(1H, m); 1.48 (1H, m); 0.90, 0.87 (18H, 2s); 0.13, 0.11 (12H,
2s). FT-IR (film): 1728 (CdO, aldehyde). Anal. Calcd for
C19H38O4Si2 (386.68331): C 59.02, H.9.91. Found: C 59.11,
H 10.28.
(1R)-1-[(1S,2S,3R,4S,5R)-2,5-Bis[(ter t-bu tyld im eth ylsi-
lyl)oxy]-3,4-ep oxycycloh exyl]-3-m eth yl-3-bu ten -1-ol (13)
a n d (1S)-1-[(1S,2S,3R,4S,5R)-2,5-b is[(ter t-Bu t yld im et h -
ylsilyl)oxy]-3,4-ep oxycycloh exyl]-3-m et h yl-3-b u t en -1-ol
(14). To a solution of 12 (0.250 g, 0.65 mmol) in ethyl ether
(3 mL) at 0 °C was added, from a syringe, an ethereal
suspension of the Grignard reagent, (2-methyl-1-propen-3-
yl)magnesium chloride (0.964 mL, 0.64 mmol, 1.1 equiv), in
small portions. When the reaction was complete (TLC),
saturated aqueous NH4Cl was added, and the mixture was
extracted with ethyl ether, dried (MgSO4), and concentrated.
Purification by column chromatography gave 0.275 g (96%) of
product.2 [R]23 +57.3 (c 0.84, anhydrous CHCl3). 1H NMR
D
(CDCl3): δ 6.15 (1H, s); 4.38 (1H, broad s); 4.26 (1H, d, J )
9.2 Hz); 3.26 (1H, d, J ) 3.1 Hz); 3.22 (1H, broad s); 2.95 (1H,
broad s); 2.67 (1H, m); 2.33 (1H, broad s); 2.16 (3H, s); 1.92
(3H, s); 1.77 (1H, m); 1.51 (1H, m). FT-IR (film): 3412 (O-H);
1683 (CdO); 1616 (CdC). EIMS m/ z (relative intensity) 213
(50), 212 (9), 127 (40), 94 (27), 83 (100). HRMS Calcd for
C
11H16O4: 212.104859. Found: 212.10486.
13 and 14. First diastereoisomer (colorless oil): [R]20 -14.3
D
(c 1.295, CH2Cl2). 1H NMR (CDCl3): δ 4.87, 4.81 (2H, 2s); 4.42
(1H, d, J ) 1.8 Hz); 3.90 (1H, m); 3.80 (1H, d, J ) 9.4 Hz);
3.08 (2H, m,); 2.13 (1H, d, J ) 13.7 Hz); 1.99 (1H, dd, J ) 13.8
Hz, J ) 10.5 Hz; 2.0 (1H, m); 1.76 (3H, s); 1.59 (1H, tt, J )
13.9 Hz, J ) 2.8 Hz); 1.23 (1H, m); 0.93, 0.92 (18H, 2s); 0.19,
0.14, 0.13, 0.11 (12H, 4s). FT-IR (film): 3545 (O-H). Second
diastereoisomer (white solid): [R]20D -7.2 (c 0.57, CH2Cl2). Mp
51-52 °C. 1H NMR (CDCl3): δ 4.81, 4.78 (2H, 2s); 4.29 (1H,
s); 4.01 (1H, dd); 3.91 (1H, d, J ) 9.3 Hz); 3.12 (1H, d, J ) 3.2
Hz); 3.03 (1H, t); 2.20 (1H, dd, J ) 13.6 Hz, J ) 10.0 Hz); 1.99
(1H, dd, J ) 13.5 Hz, J ) 3.7 Hz); 1.75 (3H, s); 1.69 (1H, m);
1.39 (2H, m); 0.94, 0.92 (18H, 2s); 0.18, 0.15, 0.11, 0.10 (12H,
4s). 13C NMR (CDCl3): δ 142.6; 113.4; 67.7; 66.2; 65.5; 58.6;
55.1; 43.4; 39.6; 25.8; 23.9, 22.1; 18.0; -4.4, -4.8, -5.0. FT-
Ack n ow led gm en t. We thank the J unta Nacional
de Investigac¸a˜o Cient´ıfica e Tecnolo´gica for a grant
(Praxis XXI/BD/4527/94) conceded to M.R.V.
Su p p or tin g In for m a tion Ava ila ble: Specific rotations,
1H and 13C NMR data, FTIR, and in some cases combustion
analyses for the compounds 19-25 (41 pages). This material
is contained in libraries on microfiche, immediately follows this
article in the microfilm version of this journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J O9617326