Divergent approach to gabosines and anhydrogabosines: Enantioselective syntheses of (+)-epiepoformin, (+)-EpoKformin, (+)-gabosine A, and gabosines B and F

Article abstract of DOI:10.1002/ejoc.201001527

A divergent approach to polyoxygenated methylcyclohexanes has been applied to synthesize several gabosines and anhydrogabosines. The starting hydroxycyclohexenone, which is readily available in any antipodal form, provides access to both enantiomers of the target compounds. The syntheses of anhydrogabosines involve three main transformations: α-methylation, epoxidation, and sulfur removal, whereas the syntheses of gabosines require an additional epoxide hydrolysis step. The strategy has been applied to the synthesis of (+)-epiepoformin, (+)-epoformin, (+)-gabosine A, and gabosines B and F, through straightforward sequences in good overall yields. A cyclohexenone that is easily available in any antipodal form has been used as the starting material to synthesize various gabosines and anhydrogabosines through a divergent strategy. The syntheses of anhydrogabosines involve three main transformations: α-methylation, epoxidation, and sulfur removal, whereas the syntheses of gabosines require an additional epoxide hydrolysis step. Copyright

Full text of DOI:10.1002/ejoc.201001527

FULL PAPER
DOI: 10.1002/ejoc.201001527
Divergent Approach to Gabosines and Anhydrogabosines: Enantioselective
Syntheses of (+)-Epiepoformin, (+)-Epoformin, (+)-Gabosine A, and
Gabosines B and F
Gladis Toribio,[a] Georgina Marjanet,[a] Ramon Alibés,[a] Pedro de March,[a] Josep Font,[a]
Pau Bayón,*[a] and Marta Figueredo*[a]
Dedicated to Professor José Barluenga on the occasion of his 70th birthday
Keywords: Natural products / Synthetic methods / Epoxidation / Enantioselectivity / Enzymatic resolution / Divergent
synthesis
A divergent approach to polyoxygenated methylcyclohex- tions: α-methylation, epoxidation, and sulfur removal,
anes has been applied to synthesize several gabosines and
anhydrogabosines. The starting hydroxycyclohexenone,
which is readily available in any antipodal form, provides ac-
cess to both enantiomers of the target compounds. The syn-
theses of anhydrogabosines involve three main transforma-
whereas the syntheses of gabosines require an additional ep-
oxide hydrolysis step. The strategy has been applied to the
synthesis of (+)-epiepoformin, (+)-epoformin, (+)-gabosine A,
and gabosines B and F, through straightforward sequences
in good overall yields.
Introduction
The gabosine family comprises a group of secondary me-
tabolites that have been isolated from various Streptomyces
strains with a closely related carba-sugar structure. In 1993,
Thiericke, Zeeck, and co-workers isolated eleven com-
pounds that were named gabosines A–K (Figure 1), and
classified into four structural types.[1] In 2000, the same au-
thors disclosed the isolation of three additional members of
the family: gabosines L, N, and O.[2] The structural diversity
of the gabosines originates from differences in the substitu-
ent positions, degree of unsaturation, and/or the relative
and absolute configuration of their stereogenic centers.
Prior to being named gabosines, some of these compounds
were already known. Thus, gabosine B had formerly been
isolated from Actinomycetes strains,[3] gabosine C was iden-
tical to the previously known antibiotic KD16-U1,[4] and
its crotonyl ester (COTC), also extracted from various
Streptomyces strains,[5] was a recognized antitumor agent.
Concerning their biological activity, it was also found that
gabosines A, B, F, N, and O exhibit DNA-binding proper-
ties.[2,6]
Figure 1. Representative examples of compounds with gabosine
and anhydrogabosine structural pattern.
A number of other compounds, the isolation of which
from natural sources is described in the literature, also pres-
ent structural moieties that are consistent with the gabosine
family. For instance, compound (–)-1 was extracted from
the fungus Phyllostica sp., the “Kurohagare” disease of red
clover,[7] and an undefined stereoisomer of 1 was found in
Ophiosphaerella herpotricha, a cause of spring dead spot of
bermuda grass.[8] Moreover, several known epoxyquinol
[a] Department de Química, Universitat Autònoma de Barcelona,
Edifici Cn, 08193 Bellaterra, Spain
Fax: +34-935811265
E-mail: marta.figueredo@uab.es
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201001527.
1534
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1534–1543

Divergent Approach to Gabosines and Anhydrogabosines
natural products possess anhydrogabosine structures.[9]
Most of these compounds are phytotoxins. Among them,
the first structural assignment corresponded to (+)-epoxy-
don, which presents an oxymethyl substituent at the α-car-
bonyl position (as in gabosines C, D, and E). Other natural
anhydrogabosines display a methyl group at the α-carbonyl
position (as in gabosines A and N), as is the case for (+)-
epoformin and (+)-epiepoformin, among others. (+)-Epo-
formin was first isolated from Penicillium claviforme[10] and
later found in a fungus of Lagerstroemia indica L.,[11] in
Phoma sp.,[12] Ophiosphaerella herpotricha,[8] and Penicil-
lium vulpinum;[13] (+)-epiepoformin was isolated from La-
gerstroemia indica L.[11] It has been described that (+)-epie-
poformin inhibited the germination of lettuce seeds.[11]
Some syntheses of gabosines and anhydrogabosines have
been published, with a variety of strategies being employed,
although several independently developed approaches pres-
ent some common trends. Most synthetic strategies are
either based on Diels–Alder reactions or they start from
chiral pool materials, typically carbohydrates and quinic
acid, although a few examples of alternative approaches
were also described.[14] However, prior to our own work,[15]
there was no suitable synthetic approach to the preparation
of a large number of these compounds from common syn-
thetic intermediates and in any antipodal form. Herein, we
describe the application of our stereodivergent design for
the enantioselective synthesis of gabosines and anhydroga-
bosines to the preparation of (+)-epoformin, (+)-epiepo-
formin, (+)-gabosine A, gabosine B, and gabosine F.
Scheme 1. Preparation and kinetic resolution of the key cyclohex-
enol (Ϯ)-4 and synthesis of ketones (4R)-6 and (4S)-6.
Results and Discussion
Synthetic Strategy: In former investigations, we prepared
a series of chiral equivalents of p-benzoquinone[16] and ex-
plored their utility in synthetic organic chemistry.[17] One
of our goals was to design an easily available and versatile
precursor of polyfunctionalized cyclohexanes in both anti-
podal forms. This objective was accomplished by de-
veloping the protocol depicted in Scheme 1, where, starting
from very accessible raw materials such as p-methoxyphe-
nol, 2, ethylene glycol, and thiophenol, multi-gram quanti-
ties of (Ϯ)-3 can be readily prepared.[18] The reduction of
(Ϯ)-3 with NaBH4 gives the cis alcohol (Ϯ)-4 exclusively.
Treatment of (Ϯ)-4 with vinyl acetate and a catalytic
amount of Novozyme 435 in diisopropyl ether furnishes the
acetate (–)-5 and the unreacted alcohol (+)-4. Methanolysis
of the acetate (–)-5 leads to recovery of the levorotatory
alcohol (–)-4.[15,18] By silylation of alcohols (+)-4 and (–)-4
followed by hydrolysis of the acetal, the α-phenylthio-
ketones (4R)-6 and (4S)-6 are easily synthesized, respec-
tively.
Enone 6 was our material of choice to undertake a sys-
tematic synthesis of gabosines and anhydrogabosines
through the strategy depicted in Scheme 2, which involves
the following transformations: (i) alkylation of the doubly
activated C-6 position of 6 to introduce the methyl or hy-
Scheme 2. Strategy for the synthesis of gabosines and anhydro-
droxymethyl group; (ii) dihydroxylation of the double bond gabosines.
Eur. J. Org. Chem. 2011, 1534–1543
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1535

P. Bayón, M. Figueredo et al.
FULL PAPER
to provide the cis α,β-glycol unit or epoxidation that can
eventually give access to the trans α,β-glycol; and (iii) re-
duction of the C6–S bond or oxidation to the sulfoxide fol-
lowed by pyrolysis to generate the conjugated C–C double
bond, depending on the target compound.
In contrast to previous enantioselective syntheses that
started from chiral pool materials, this divergent design
should give access to a large number of gabosine and anhy-
drogabosine type compounds of both enantiomeric series.
The results of the methylation and dihydroxylation steps
were reported in a previous communication, along with the
syntheses of gabonines O and N, and epigabosines O and
N.[15] We have now explored the methylation–epoxidation
pathway and developed new syntheses of the target com-
pounds. Most reactions were initially performed with race-
mic substrates and, after establishing the best conditions,
were repeated starting from enantiomerically pure precur-
sors.
Methylation and Epoxidation Studies: The methylation of
ketone 6 (Scheme 3) was accomplished by reaction of its
enolate with methyl iodide in tetrahydrofuran (THF). This
transformation was attempted using three different bases
as a means of modulating the diastereofacial selectivity. A
chromatographically separable mixture of the two epimers
7 and 8 was obtained in the three cases in high overall
yields. Within the enantioselective pathway, methylation of
(4S)-6 using sodium hydride as the base furnished an ap-
proximate 1:1 mixture of the two epimers (4S,6S)-7 and
(4S,6R)-8, which were separated for subsequent studies.
The epoxidation of enones 7 and 8 was assayed by reac-
tion with hydrogen peroxide (Method A)[19] or potassium
tert-butylhydroperoxide (Method B)[20] and Triton B in
THF. Using Method A, the epoxidation of enone 7 deliv-
ered a mixture of oxiranes 9 and 10 in a 1:5 ratio and 78%
overall yield, whereas epoxidation of enone 8 furnished a
5:1 mixture of the oxiranes 11 and 12 in 91% total yield.
The observed stereoselectivity was in good agreement with
that found in the dihydroxylation of the same enones with
Scheme 3. Preparation and epoxidation of methylcyclohexanones 7
and 8.
OsO4 and N-methylmorpholine N-oxide (NMO), where we chromatography; alcohol 14 was then converted back into
argued that the oxidant reagent approached the enone the corresponding silylether in 98% yield. Within the op-
mainly on the face opposite to the bulky phenylsulfenyl tically active series, the epoxides (1R,3S,5S,6S)-9 and
group.[15] However, when the epoxidation of 7 and 8 was (1S,3S,5S,6R)-10 were readily prepared from the enone
performed with the more sterically demanding oxidant (4S,6S)-7, and the isomer (1R,3R,5S,6S)-11 from the enone
(Method B), epoxides 9 and 11 were exclusively formed and (4S,6R)-8.
isolated in 96 and 98% yield, respectively. The exclusive for-
Generation of the conjugated double bond. Total syntheses
mation of epoxide 9, with complete inversion of the facial of (+)-epiepoformin and (+)-epoformin, and the formal syn-
selectivity when going from hydrogen peroxide to tert-butyl- theses of (–)-theobroxide and (+)-4-epigabosine A: In pre-
hydroperoxide, provides evidence that the stereoselectivity vious work by other authors, racemic epoformin and epie-
of the former processes (epoxidation with H2O2/Triton B poformin were synthesized from p-benzoquinone through a
and dihydroxylation with OsO4/NMO) were not merely strategy based on Diels–Alder/retro-Diels–Alder reac-
governed by steric factors. For synthetic purposes, we tions.[21] Some years later, an asymmetric version of the
judged it convenient to isolate epoxide 10, which was the same strategy led to the synthesis of (+)-epiepoformin.[22]
major product formed in the oxidation of 7 with hydrogen Two additional enantioselective syntheses of (+)-epiepofor-
peroxide and, relative to C-5, presents a configuration of min were accomplished by using chiral auxiliaries,[23] and
the oxirane stereocenters opposite to that of 9 and 11. Be- another approach made use of the enzymatic reduction of
cause compounds 9 and 10 exhibited identical polarities, a cyclohexanone.[24] (–)-Epiepoformin has recently been
the mixture of epoxides was desilylated to the the corre- synthesized from the enzymatic dihydroxylation product of
sponding alcohols 13 and 14, which were separated by iodobenzene.[25] Both (+)-epoformin and (+)-epiepoformin
1536
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1534–1543

Divergent Approach to Gabosines and Anhydrogabosines
Scheme 4. Oxidation/pyrolysis of 9 and 11 and synthesis of epiepoformin.
were also synthesized from (–)-quinic acid.[26] The overall value of the synthetic material, [α]D = +109 (c = 0.20,
yields of these syntheses range from 4 to 50%. In our ap- EtOH), was in agreement with the previously reported
proach, epoxides 9 and 11 were suitable intermediates for data.[10,26a]
epiepoformin, whereas epoxide 10 was an appropriate pre-
cursor for the synthesis of epoformin. Considering that 9
and 11 are epimers at C3, the oxidation/pyrolysis protocol
providing the conjugated double bond should, in principle,
give rise to the same olefin. In practice, when 9 and 11 were
separately treated with m-chloroperoxybenzoic acid (m-
CPBA) in chloroform at 0 °C and then heated, a mixture
of the endo- and exocyclic olefins 15 and 16 was obtained
in the same ratio (5.5:1) and good overall yield (Scheme 4).
It was described that isomerization of 16 to 15 can be ac-
complished by treatment with Pd/C preactivated under a
hydrogen atmosphere, with variable yields of between 31
Scheme 6. Oxidation/pyrolysis of 10 and synthesis of epoformin.
and 71% because of the sensitivity to small changes in the
reaction conditions.[27] However, this transformation proved
Cleavage of the epoxide. Total synthesis of gabosine A:
Starting from chiral pool materials, two syntheses of (–)-
to be unnecessary because treatment of the mixture of epox-
ides 15 and 16 with the complex Et3N·3HF furnished epie-
gabosine A,[29] the natural enantiomer, and one of its levo-
poformin as the sole product in 86% isolated yield.
rotatory antipode[30] have been described, with overall
In view of these results, a straightforward and highly ef-
yields below 15%. A third synthesis of (–)-gabosine A was
ficient synthesis of epiepoformin was completed in four
completed starting from the enzymatic dihydroxyation
steps starting from 6: (i) methylation to 7 and 8, (ii) epoxid-
product of iodobenzene in 58% total yield.[31] Formally, re-
ation to 9 and 11, (iii) oxidation/pyrolysis to 15 and 16, and
gioselective hydrolysis of the epoxide in (+)-epiepoformin
(iv) deprotection, without performing any chromatographic
or (–)-epoformin could deliver (+)-gabosine A, whereas the
corresponding antipode, which is equally available through
our approach, would deliver the naturally occurring enantio-
separation in 74% total yield (Scheme 5). When this se-
quence of reactions was applied to (4S)-6, the dextrorota-
tory enantiomer of epiepoformin was furnished, [α]D
=
mer. However, it was described that hydrolysis of the epox-
ide in epiepoformin was a troublesome transformation that
could only be achieved by treatment with aqueous sodium
acetate in 45% yield;[23b] furthermore, this reaction deliv-
ered 4-epigabosine A, an epimer of the natural product. In
view of this precedent, for the synthesis of gabosine A, we
decided to assay the hydrolysis of the epoxide with a former
intermediate lacking the conjugated C–C double bond; it
was thought that the higher flexibility could benefit the de-
sired transformation. Accordingly, ketone 11 was treated
with various hydrolytic reagents, including NaOAc/
H2O,[23b] trifluoroacetic acid (TFA)/THF/H2O,[32] and
Sc(OTf)3/AcOH,[33] however, in all cases, this led to full re-
covery of the starting material. In contrast, the reaction of
+315 (c = 1.1, EtOH).[28] The conversion of (+)-epiepofor-
min into (–)-theobroxide and (+)-4-epigabosine A was de-
scribed previously.[23b]
Scheme 5. Synthesis of (+)-epiepoformin.
A parallel sequence of reactions was then applied to ep- 11 with BF3·Et2O in toluene delivered a unique product,
oxyketone 10 (Scheme 6). Oxidation to the sulfoxide and which was identified as 20, in 80% yield (Scheme 7). The
subsequent pyrolysis delivered the expected mixture of en- regio- and stereochemistry of 20 was determined with the
do- and exocyclic olefins 17 and 18, but the less stable iso- help of NMR analyses, including selective NOE experi-
mer 18 readily isomerized to 17 during the chromatographic ments. Considering the relative configuration of the ste-
purification on silica gel. Deprotection of the hydroxyl reocenters in 20, we speculated that its formation may have
group in 17 furnished the target compound. The parallel occurred through an intramolecular process with migration
enantioselective pathway starting from (–)-10 furnished nat- of the sulfide group to open the oxirane, which was elec-
ural epoformin in 63% overall yield. The specific rotation tronically activated by the Lewis acid (path A). To test this
Eur. J. Org. Chem. 2011, 1534–1543
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1537

P. Bayón, M. Figueredo et al.
FULL PAPER
Scheme 7. Attempted hydrolysis of the oxirane in 11 and formation of 20.
hypothesis, the diasteromeric ketone 9 was submitted to
identical reaction conditions and it was found that com-
pound 20 was also formed in similar yield. This result sug-
gested that elimination of thiophenol probably occurred
prior to epoxide opening (path B). In agreement with this
assumption, enone 15 readily delivered alcohol 20 as the
sole product when treated with BF3·Et2O in toluene in the
presence of 1 equiv. thiophenol. The addition of thiophenol
to related epoxyquinones under neutral conditions was
shown to occur with identical regioselectivity, although at
a much lower rate.[34] Enone 15 was then submitted to iden-
tical reaction conditions, except for the absence of thio-
phenol, in an attempt to form the corresponding trans diol
Scheme 8. Synthesis of (+)-gabosine A from (+)-epiepoformin.
21; unfortunately, the starting material was recovered un-
changed.
In a related system, cleavage of 2,3-epoxycyclohexanone
to the trans diol was described to take place in 67% yield
through a BF3·Et2O mediated and acetate-assisted pro-
cess.[35] Considering this precedent, the acetyl derivative of
(+)-epiepoformin, (+)-22, was prepared and treated with
BF3·Et2O in toluene at 0 °C for 2 h (Scheme 8). This reac-
tion furnished a mixture of acetates 23 and 24 that, without
separation, was submitted to methanolysis in basic medium
affording (+)-gabosine A in 87% yield for the two steps.
than the corresponding reaction of the silylether analogues
7 and 8. Moreover, partial hydrolysis of the epoxyacetates
took place spontaneously, furnishing a complex mixture of
products that could not be converted into the target com-
pound in a reproducible manner. Consequently, in this case,
the straightforward approach from (–)-5 to (+)-gabosine A
proved ineffective.
The specific rotation of the synthetic material was [α]D
=
+128 (c = 0.30, MeOH) {ref.[1a] –132 (c = 1, MeOH) for
the natural antipode}.
Because the presence of the vicinal acetate proved to be
crucial for cleavage of the epoxide, we decided to develop a
more straightforward synthesis of gabosine A starting from
the acetate (–)-5, produced through enzymatic resolution of
(Ϯ)-4. The synthetic plan, shown in Scheme 9, would in-
volve seven chemical transformations in only five experi-
mental operations: (i) hydrolysis of the ketal, (ii) methyl-
ation, (iii) stereoselective epoxidation, (iv) oxidation/pyroly-
sis, and (v) epoxide cleavage/methanolysis. Along this se-
quence, the separation of isomers in each individual step
would be unnecessary. In practice, the two initial transfor-
Scheme 9. Attempted synthesis of (+)-gabosine A from acetate (–)-
5.
Reduction of the C–S bond; Total syntheses of gabosines
mations took place as expected, however, unfortunately, the B and F: Shinada and Ohfune described an enantioselective
epoxidation of the acetate (1S)-26 was less stereoselective synthesis of (–)-gabosine B starting from (–)-quinic acid,[29a]
1538
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1534–1543

Divergent Approach to Gabosines and Anhydrogabosines
and a synthesis of the racemate had been previously report- epoxide hydrolysis step that works well only on vicinal ace-
ed.[1b] Recently, an enantioselective synthesis of the dextro- toxy derivatives. Although the stereoselectivities of the α-
rotatory antipode, which is also the natural gabosine F, has methylation and epoxidation steps can be modulated by an
been accomplished from l-arabinose.[36] Formally, gabos- appropriate choice of base or oxidant, respectively, for some
ines B and F could be obtained from dihydroepiepoformin synthetic purposes the separation of diastereomers is un-
and/or dihydroepoformin by stereoselective hydrolysis of necessary. Thus, the total synthesis of (+)-epiepoformin was
the epoxide function. According to our approach, the syn- completed in only four steps from (4S)-6 with 74% overall
thesis of these natural products and their analogues re- yield. The synthesis of (+)-epoformin was accomplished in
quired the reduction of the C(sp3)–S bond in compounds 9 two steps and 63% yield starting from the epoxide
and/or 11 (Scheme 10). This transformation was performed (1R,3R,5R,6S)-10, which can, in turn, be prepared from
by treating each epimeric sulfide with HBu3Sn in the pres- (4R)-6. (+)-Gabosine A was synthesized from (+)-epiepo-
ence of AIBN in toluene as the solvent, at reflux tempera- formin in three steps and 76% yield. Gabosines B and F
ture. Both reactions led to the formation of a mixture of were respectively prepared from (–)-9 or (–)-11 and (+)-9 or
the expected reduction products 29 and 30 in similar yield (+)-11. It is remarkable that the syntheses of these gabos-
(80–90%) and identical ratio (6.6:1), which can presumably ines could also be accomplished starting from (4R)-6 or
be linked to the relative stability of the isomers, which can (4S)-6, respectively, in seven steps without any chromato-
equilibrate under the reaction conditions. Desilylation of graphic separation of the isomers, in 50% overall yield.
the mixture furnished the corresponding alcohols 31 and 32
in 80% overall yield, which were acetylated to give the start-
ing material for the epoxide cleavage. This transformation
Experimental Section
was accomplished through the same two-step protocol used
for gabosine A, without isolation of the intermediate diols,
in 88% total yield and with concomitant epimerization at
C-6 to furnish exclusively one diastereomer; the physical
data of this diastereoisomer were identical to those of natu-
ral gabosines B/F. Starting from (+)-9, (+)-11, or any mix-
ture of both epimers, the synthetic sequence resulted in the
formation of the dextrorotatory product, gabosine F. Anal-
ogously, the levorotatory antipode, gabosine B, was pre-
pared from (–)-9 and/or (–)-11.
General Procedure for the Epoxidation Reaction. Method A: To an
ice-cooled solution of (Ϯ)-7 (500 mg, 1.43 mmol) in THF (8.3 mL),
H2O2 (30% in water, 950 μL, 9.29 mmol) and Triton B (40% in
water, 65 μL, 0.14 mmol) were added. After stirring at 0 °C for
15 min, the reaction mixture was warmed to room temperature and
the progress of the reaction was monitored by GC analysis. After
disappearance of the starting olefin, the reaction mixture was di-
luted with EtOAc (10 mL) and washed with saturated NH4Cl solu-
tion. The organic layer was separated, the aqueous layer was ex-
tracted with EtOAc (3ϫ10 mL), and the combined organic extracts
were dried with anhydrous MgSO4. The solvent was evaporated
under reduced pressure and the residue was purified by flash
chromatography (hexanes/EtOAc, 20:1 to 5:1) to furnish a white
solid that was identified as a 1:5 mixture of epoxides (Ϯ)-9 and
(Ϯ)-10 (397 mg, 1.09 mmol, 76%). The same reaction starting from
(4S,6S)-7 (1.0 g, 2.87 mmol) furnished
a
1:5 mixture of
(1R,3S,5S,6S)-9 and (1S,3S,5S,6R)-10 (820 mg, 2.25 mmol, 78%).
The same reaction starting from (4R,6R)-7 (122 mg, 0.35 mmol)
furnished a 1:5 mixture of (1S,3R,5R,6R)-9 and (1R,3R,5R,6S)-10
(102 mg, 0.28 mmol, 80%).
Method B: The same protocol starting from a solution of (Ϯ)-7
(780 mg, 2.24 mmol) in THF (13 mL), tBuOOH (70% in water,
1.08 mL, 7.86 mmol) and Triton B (40% in water, 102 μL,
0.22 mmol) afforded a solid residue, which was purified by flash
chromatography (hexanes/EtOAc, 5:1) to provide exclusively epox-
ide (Ϯ)-9 (765 mg, 2.10 mmol, 94%) as a white solid. The same
reaction starting from (4S,6S)-7 (200 mg, 0.57 mmol) furnished
(1R,3S,5S,6S)-9 (200 mg, 0.55 mmol, 96%).
Scheme 10. Reduction of the C–S bond in 9 and 11 and synthesis
of gabosine B/F.
(؎)-(1RS,3SR,5SR,6SR)-5-(tert-Butyldimethylsilyloxy)-3-methyl-
3-(phenylthio)-7-oxabicyclo[4.1.0]heptan-2-one [(؎)-9]: Rf = 0.48
(hexanes/EtOAc, 5:1); m.p. 71–74 °C (hexanes/EtOAc); Rt (GC) =
11.545 min. 1H NMR (360 MHz, CDCl3, 25 °C): δ = 7.41 (m, 3 H,
ArH), 7.33 (m, 2 H, ArH), 4.53 [q, 3JH,H ≈ 2.8 Hz, 1 H, 5-H], 3.55
Conclusions
Our design for the synthesis of polyoxygenated cyclohex-
anes has been successfully applied to the enantioselective
syntheses of several gabosine and anhydrogabosine type
compounds. The common starting material for these syn-
theses are the easily available enones (4R)-6 or (4S)-6, de-
pending on the required configuration of the target com-
pound. The syntheses of anhydrogabosines involve three
main transformations: α-methylation, epoxidation, and sul-
fur removal, whereas the syntheses of gabosines with trans
2
3
(m, 2 H, 1-H, 6-H), 2.30 (dd, JH,H = 14.6, JH,H = 2.8 Hz, 1 H,
2
3
4
4-H), 2.01 (ddd, JH,H = 14.6, JH,H = 4.0, JH,H = 1.0 Hz, 1 H, 4-
H), 1.35 (s, 3 H, Me), 0.86 (s, 9 H, tBu), 0.11 (s, 3 H, SiMe), 0.10
(s, 3 H, SiMe) ppm. 13C NMR (90 MHz, CDCl3): δ = 201.3, 137.7,
130.4, 129.8, 128.8, 67.1, 59.3, 55.2, 50.5, 40.1, 27.3, 25.8, 18.1,
–4.7 ppm. IR (ATR): ν = 2927, 2855, 1704, 1132, 1105 cm–1
.
˜
HRMS (CI+): calcd. for C19H28O3SSiNa: 387.1421; found:
configuration of the α,β-glycol unit requires an additional 387.1431 [M + Na+].
Eur. J. Org. Chem. 2011, 1534–1543
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1539

P. Bayón, M. Figueredo et al.
FULL PAPER
(+)-(1R,3S,5S,6S)-9: M.p. 126–128 °C (hexanes/EtOAc). [α]2D0
+6.7 (c = 0.79, CHCl3).
=
tion of anhydrous m-CPBA (224 mg, 1.09 mmol) in CHCl3 (3 mL)
was continuously added during 1 h at the same temperature. The
reaction mixture was monitored by GC analysis until completion.
The solution was washed with saturated NaHCO3 solution and the
organic layer was separated, dried with anhydrous MgSO4 and fil-
tered. The filtrate was concentrated to around 15 mL and heated to
reflux temperature for 1 h. The solvent was removed under reduced
pressure and the residue was purified by flash chromatography
(hexanes/EtOAc, 15:1 to 5:1) to afford a 5.5:1 mixture of epoxides
(Ϯ)-15 and (Ϯ)-16 (215 mg, 0.84 mmol, 69%). The same reaction
starting from a 1:7.6 mixture of (Ϯ)-9 and (Ϯ)-11 (396 mg,
1.09 mmol) furnished a 5.5:1 mixture of (Ϯ)-15 and (Ϯ)-16
(207 mg, 0.81 mmol, 75 %). The same reaction starting from
(1R,3R,5S,6S)-11 (614 mg, 1.68 mmol) furnished a 5.5:1 mixture of
(1R,5S,6S)-15 and (1R,5S,6S)-16 (373 mg, 1.47 mmol, 87%). The
same reaction starting from a 7.1:1 mixture of (1R,3S,5S,6S)-9 and
(1R,3R,5S,6S)-11 (610 mg, 1.68 mmol) furnished a 5.5:1 mixture of
(1R,5S,6S)-15 and (1R,5S,6S)-16 (376 mg, 1.48 mmol, 88%). The
spectroscopic data of 15 and 16 were in accordance with those de-
scribed in the literature.[27]
(؎)-(1RS,3RS,5RS,6SR)-5-(tert-Butyldimethylsilyloxy)-3-methyl-3-
(phenylthio)-7-oxabicyclo[4.1.0]heptan-2-one [(؎)-10]: Rt (GC) =
10.755 min. Epoxide 10 was isolated and characterized after flash
chromatographic separation of the unprotected analogues 13 and
14 and subsequent TBS protection of 14 (see the Supporting Infor-
mation for details).
Mixture of (1RS,3RS,5SR,6SR)- and (1RS,3SR,5RS,6SR)-5-(tert-
butyldimethylsilyloxy)-3-methyl-3-(phenylthio)-7-oxabicyclo[4.1.0]-
heptan-2-one [(؎)-11 and (؎)-12]: Following Method A, starting
from (Ϯ)-8 (250 mg, 0.72 mmol), an inseparable 5:1 mixture of ep-
oxides (Ϯ)-11 and (Ϯ)-12 (239 mg, 0.65 mmol, 91%) was isolated
as a white solid. MS (ESI+): m/z (%) = 387 (100) [M + Na]+.
C19H28O3SSi (364): calcd. C 62.59, H 7.74, S 8.80; found C 62.36,
H 7.90, S 8.62.
(1RS,3RS,5SR,6SR)-5-(tert-Butyldimethylsilyloxy)-3-methyl-3-
(phenylthio)-7-oxabicyclo[4.1.0]heptan-2-one [(؎)-11]: Following
Method B, starting from (Ϯ)-8 (512 mg, 1.47 mmol), epoxide (Ϯ)-
11 (525 mg, 1.44 mmol, 98%) was isolated as a white solid. Rf =
0.44 (hexanes/EtOAc, 5:1); m.p. 100–102 °C (hexanes/EtOAc); Rt
(1RS,5RS,6SR)-5-(tert-butyldimethylsilyloxy)-3-methyl-7-oxabi-
cyclo[4.1.0]hept-3-en-2-one and (1RS,5RS,6SR)-5-(tert-butyldi-
methylsilyloxy)-3-methylene-7-oxabicyclo[4.1.0]heptan-2-one [(؎)-
17]: Following the general procedure and starting from (Ϯ)-10
(119 mg, 0.33 mmol), olefin (Ϯ)-17 (57 mg, 0.22 mmol, 69%) was
isolated. Rt (GC) = 5.145 min. 1H NMR (250 MHz, CDCl3, 25 °C):
1
(GC) = 10.878 min. H NMR (360 MHz, CDCl3, 25 °C): δ = 7.35
3
4
(m, 5 H, ArH), 4.45 (m, 1 H, 5-H), 3.52 (dd, JH,H = 3.4, JH,H
=
=
3
3
4
0.9 Hz, 1 H, 1-H), 3.49 (ddd, JH,H = 3.4, JH,H = 2.2, JH,H
1.2 Hz, 1 H, 6-H), 2.25 (dd, JH,H = 15.3, JH,H = 4.1 Hz, 1 H, 4-
2
3
2
3
4
3
H), 2.09 (ddd, JH,H = 15.3, JH,H = 2.0, JH,H = 1.2 Hz, 1 H, 4-
H), 1.27 (s, 3 H, Me), 0.98 (s, 9 H, tBu), 0.18 (s, 6 H, SiMe2) ppm.
13C NMR (90 MHz, CDCl3, 25 °C): δ = 194.0, 137.9, 130.9, 129.7,
128.7, 64.3, 56.7, 52.9, 52.1, 38.7, 26.0, 22.9, 18.3, –4.6 ppm. IR
δ = 6.13 (m, 1 H, 4-H), 4.74 (m, 1 H, 5-H), 3.67 (dt, JH,H = 4.0,
4
3
3JH,H ≈ JH,H ≈ 2.7 Hz, 1 H, 6-H), 3.43 (d, JH,H = 4.0 Hz, 1 H, 1-
H), 1.80 (m, 3 H, Me), 0.96 (s, 9 H, tBu), 0.18 (s, 6 H, SiMe2) ppm.
13C NMR (100 MHz, CDCl3, 25 °C): δ = 194.4, 141.5, 131.8, 66.3,
54.6, 53.2, 25.9, 15.9, 14.2, –4.4, –4.5 ppm. HRMS (CI+): calcd. for
C13H22O3SiNa 277.1230; found 277.1232 [M + Na+].
(ATR): ν = 2957, 2926, 2855, 1703, 1248, 1107 cm–1.
˜
(+)-(1R,3R,5S,6S)-11: The same reaction starting from (4S,6R)-8
(1.00 g, 2.87 mmol) furnished (1R,3R,5S,6S)-11 (1.02 g, 2.8 mmol,
98%). M.p. 126–128 °C (hexanes/EtOAc); [α]2D0 = +5.5 (c = 0.84,
CHCl3).
(+)-(1R,5R, 6S)-17: T he sa me re ac ti on st ar ti ng from
(1R,3R,5R,6S)-10 (116 mg, 0.32 mmol) furnished (1R,5R,6S)-17
(58 mg, 0.23 mmol, 72%). [α]2D0 = +86 (c = 0.85, CHCl3).
(؎)-Epiepoformin: A solution of epoxides (Ϯ)-15 and (Ϯ)-16
(280 mg, 1.10 mmol) in THF (14.5 mL) at room temperature was
treated with Et3N·3HF (1.0 mL, 6.13 mmol). The reaction mixture
was stirred for 12 h at the same temperature, then CH2Cl2 (16 mL)
and a saturated aqueous solution of NaHCO3 (16 mL) were added.
The organic layer was separated and the aqueous layer was ex-
tracted with CH2Cl2 (3ϫ8 mL). The combined organic extracts
were dried with anhydrous MgSO4 and concentrated under vac-
uum. The oily residue was purified by flash chromatography (hex-
anes/EtOAc, 5:1) to give (Ϯ)-epiepoformin (132 mg, 0.94 mmol,
86%) as a white solid. Rf = 0.56 (EtOAc); m.p. 59–61 °C (hexanes/
Mixture of (؎)-9 and (؎)-11: Following Method B, starting from a
mixture of (Ϯ)-7 and (Ϯ)-8 (230 mg, 0.66 mmol), a mixture of ep-
oxides (Ϯ)-9 and (Ϯ)-11 (233 mg, 0.64 mmol, 97%) was isolated as
a white solid.
(1RS,3SR,5RS,6SR)-5-(tert-Butyldimethylsilyloxy)-3-methyl-3-
(phenylthio)-7-oxabicyclo[4.1.0]heptan-2-one [(؎)-12]: Repeated
flash chromatography of a mixture of (Ϯ)-11 and (Ϯ)-12 allowed
the isolation of an analytical sample of (Ϯ)-12: Rt (GC) =
11.185 min. 1H NMR (360 MHz, CDCl3, 25 °C): δ = 7.38 (m, 5 H,
3
3
ArH), 4.22 (dd, JH,H = 11.3, 4.8 Hz, 1 H, 5-H), 3.53 (dd, JH,H
=
4
3
4.3, JH,H = 1.4 Hz, 1 H, 6-H), 3.46 (d, JH,H = 4.3 Hz, 1 H, 1-H),
3
EtOAc). 1H NMR (250 MHz, CDCl3, 25 °C): δ = 6.45 (ddq, JH,H
2.36 (dd, 2JH,H = 13.1, 3JH,H = 11.3 Hz, 1 H, 4-H), 1.83 (ddd, 2JH,H
4
4
= 5.5, JH,H = 2.6, JH,Me = 1.3 Hz, 1 H, 4-H), 4.65 (br. s, 1 H, 5-
3
4
= 13.1, JH,H = 4.8, JH,H = 1.4 Hz, 1 H, 4-H), 1.27 (s, 3 H, Me),
0.91 (s, 9 H, tBu), 0.11 (s, 3 H, SiMe), 0.10 (s, 3 H, SiMe) ppm.
13C NMR (90 MHz, CDCl3, 25 °C): δ = 202.2, 137.8, 137.7, 130.3,
129.8, 66.1, 60.2, 56.0, 50.7, 38.7, 25.8, 25.1, 18.2, –4.5, –4.6 ppm.
3
4
H), 3.77 (ddd, JH,H = 3.7, 1.3, JH,H = 2.6 Hz, 1 H, 6-H), 3.48
3
4
(dd, JH,H = 3.7, JH,H = 1.1 Hz, 1 H, 1-H), 2.38 (br. s, 1 H, OH),
1.83 (dd, JH,H = 1.3, JH,H = 1.2 Hz, 3 H, Me) ppm. 13C NMR
(100 MHz, CDCl3, 25 °C): δ = 194.3, 138.9, 134.8, 63.5, 57.8, 53.5,
16.1 ppm.
4
5
Mixture of (1R,3S,5S,6S)-9 and (1R,3R,5S,6S)-11: The same reac-
tion starting from a mixture of (4S,6S)-7 and (4S,6R)-8 (600 mg,
1.72 mmol) furnished a mixture of (1R,3S,5S,6S)-9 and
(1R,3R,5S,6S)-11 (627 mg, 1.72 mmol, 100%).
(+)-Epiepoformin: The same reaction starting from (1R,5S,6S)-15
and (1R,5S,6S)-16 (376 mg, 1.48 mmol) furnished (1R,5S,6R)-epi-
epoformin (178 mg, 1.27 mmol, 86 %). M.p. 88–90 °C (hexanes/
EtOAc). [α]2D0 = +315 (c = 1.1, EtOH).
Mixture of (1S,3R,5R,6R)-9 and (1S,3S,5R,6R)-11: The same reac-
tion starting from a mixture of (4R,6R)-7 and (4R,6S)-8 (1.24 g,
3.56 mmol) furnished a mixture of (1S,3R,5R,6R)-9 and
(1S,3S,5R,6R)-11 (1.28 g, 3.51 mmol, 99%).
(؎)-Epoformin: A solution of epoxide (Ϯ)-17 (45 mg, 0.18 mmol)
in THF (2.5 mL) at room temperature, was treated with Et3N·3HF
(167 μL, 1.03 mmol). The reaction mixture was stirred at the same
temperature for 48 h then CH2Cl2 (2.5 mL) and a saturated aque-
ous solution of NaHCO3 (2.5 mL) were added. The organic layer
General Procedure for the Oxidation/Pyrolysis Step: To a solution
of (Ϯ)-11 (400 mg, 1.09 mmol) in CHCl3 (13 mL) at 0 °C, a solu-
1540
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1534–1543

Divergent Approach to Gabosines and Anhydrogabosines
was separated and the aqueous layer was extracted with CH2Cl2
(3ϫ2 mL). The combined organic extracts were dried with anhy-
drous MgSO4 and concentrated under vacuum. The oily residue
was purified by flash chromatography (hexanes/EtOAc, 5:1) to give
epoformin (21 mg, 0.15 mmol, 85%). Rf = 0.40 (hexanes/EtOAc,
0.99 mmol) in anhydrous toluene (9.0 mL), a solution of AIBN
(17.3 mg, 0.10 mmol) and Bu3SnH (315 μL, 1.19 mmol) in anhy-
drous toluene (5.1 mL) was continuously added over 73 min. After
1 min, the reaction mixture was cooled and the solvent was evapo-
rated under vacuum to give an oily residue, which was purified by
flash chromatography (hexanes to hexanes/EtOAc, 5:1) to provide
1
1:1). H NMR (400 MHz, CDCl3, 25 °C): δ = 6.27 (s, 1 H, 4-H),
3
4.63 (br. s, 1 H, 5-H), 3.83 (m, 1 H, 6-H), 3.51 (d, JH,H = 4.0 Hz, a 6.6:1 mixture of epoxides (Ϯ)-29 and (Ϯ)-30 (204 mg, 0.79 mmol,
3
1 H, 1-H), 2.19 (d, JH,H = 9.9 Hz, 1 H, OH), 1.82 (s, 3 H,
Me) ppm. 13C NMR (100 MHz CDCl3, 25 °C): δ = 193.9, 140.3,
132.6, 65.2, 54.6, 53.9, 15.9 ppm.
80%) as an oil. The same protocol starting from (Ϯ)-11 (52 mg,
0.14 mmol) furnished a mixture of (Ϯ)-29 and (Ϯ)-30 (33 mg,
0.13 mmol, 90%) in an identical ratio (see the Supporting Infor-
mation for analytical data).
(+)-Epoformin: The same reaction starting from (1R,5R,6S)-17
(34 mg, 0.13 mmol) furnished (1R,5R,6R)-epoformin (16 mg,
0.12 mmol, 87%) as a colorless oil. [α]2D0 = +109 (c = 0.20, EtOH).
Mixture of (1R,3R,5S,6S)-29 and (1R,3S,5S,6S)-30: The same reac-
tion starting from a mixture of (1R,3S,5S,6S)-9 and (1R,3R,5S,6S)-
11 (500 mg, 1.37 mmol) furnished a mixture of (1R,3R,5S,6S)-29
and (1R,3S,5S,6S)-30 (278 mg, 1.1 mmol, 79%).
(1R,2S,6R)-4-Methyl-5-oxo-7-oxabicyclo[4.1.0]hept-3-en-2-yl Acet-
ate [(+)-22]: To a solution of (+)-epiepoformin (131 mg, 0.93 mmol)
in anhydrous acetonitrile (4.5 mL) at 0 °C under a nitrogen atmo-
sphere, DMAP (126 mg, 1.03 mmol) and acetic anhydride (350 μL,
3.74 mmol) were added. The solution was warmed to room tem-
perature and stirred for 5 min, then the reaction mixture was
poured into ice water and extracted with CH2Cl2. The combined
organic extracts were washed with cold water and a saturated aque-
ous solution of NaHCO3. The solution was dried with MgSO4 and
the solvent was removed under reduced pressure to furnish an oily
residue. Purification of this material by flash chromatography
(EtOAc) yielded (+)-22 as an oil (151 mg, 0.83 mmol, 89%). Rf =
0.68 (EtOAc). [α]2D0 = +94 (c = 0.17, CHCl3). 1H NMR (400 MHz,
CDCl3, 25 °C): δ = 6.36 (dqn, 3JH,H = 6.3, 4JH,H ≈, 4JH,Me ≈ 1.4 Hz,
Mixture of (1S,3S,5R,6R)-29 and (1S,3R,5R,6R)-30: The same re-
action starting from
a
mixture of (1S,3R,5R,6R)-9 and
(1S,3S,5R,6R)-11 (800 mg, 2.19 mmol) furnished a mixture of
(1S,3S,5R,6R)-29 and (1S,3R,5R,6R)-30 (456 mg, 1.78 mmol,
81%).
Mixture of (1RS,3RS,5SR,6RS)- and (1RS,3SR,5SR,6RS)-5-Hy-
droxy-3-methyl-7-oxabicyclo[4.1.0]heptan-2-one [(؎)-31 and (؎)-
32]: A solution of a 6.6:1 mixture of epoxides (Ϯ)-29 and (Ϯ)-30
(336 mg, 1.31 mmol) in THF (17.4 mL) at room temperature, was
treated with Et3N·3HF (1.24 mL, 7.61 mmol). The reaction mix-
ture was stirred for 20 h at the same temperature, then CH2Cl2
(23 mL) and a saturated aqueous solution of NaHCO3 (23 mL)
were added. The organic layer was separated and the aqueous layer
was extracted with CH2Cl2 (3ϫ 15 mL). The combined organic
extracts were dried with anhydrous MgSO4 and concentrated under
vacuum to give a 1:1 mixture of alcohols (Ϯ)-31 and (Ϯ)-32
(149 mg, 1.05 mmol, 80%) as an oily residue. After repeated flash
chromatography (hexanes/EtOAc, 10:1 to 1:1) analytical samples
of (Ϯ)-31 and (Ϯ)-32 were isolated.
3
4
1 H, 3-H), 5.72 (ddd, JH,H = 6.3, 2.5, JH,H = 1.1 Hz, 1 H, 2-H),
3
4
3.72 (ddd, JH,H = 3.5, 2.5, JH,H = 1.4 Hz, 1 H, 1-H), 3.52 (dd,
3JH,H = 3.5, 4JH,H = 1.1 Hz, 1 H, 6-H), 2.12 (s, 3 H, OAc), 1.86 (d,
4JH,H = 1.1 Hz, 3 H, Me) ppm. 13C NMR (100 MHz, CDCl3,
25 °C): δ = 193.7, 170.0, 136.7, 134.4, 64.7, 55.2, 53.1, 20.9,
16.1 ppm. IR (ATR): ν = 2928, 1737, 1684, 1369, 1219 cm–1
.
˜
HRMS (ESI+): calcd. for C9H10O4Na 205.0471; found 205.0472
[M + Na+].
1
(؎)-31: Rf = 0.20 (EtOAc). H NMR (250 MHz, CDCl3, 25 °C): δ
3
(+)-Gabosine A: To an ice-cooled solution of acetate (+)-22 (17 mg,
0.09 mmol) in toluene (1 mL), BF3·Et2O (12 μL, 0.09 mmol) was
added. The reaction mixture was stirred at 0 °C for 2 h, then neu-
tralized with NaHCO3 and extracted with EtOAc (3ϫ 0.5 mL).
The combined organic extracts were dried with anhydrous MgSO4
and the solvent was evaporated under vacuum. The residue was
purified by flash chromatography (EtOAc), affording a 3.1:1 mix-
ture of acetates 23 and 24 (18 mg, 0.09 mmol, 97%) as an oil (see
the Supporting Information for analytical data). The mixture of
acetates was dissolved in MeOH (0.2 mL), NaOMe (4.8 mg,
0.09 mmol) was added, and the mixture was stirred at room tem-
perature for 1 h. The solvent was removed under reduced pressure
and the residue was diluted in water and made slightly acidic with
2% HCl. The aqueous solution was extracted with CH2Cl2 (3ϫ
0.1 mL), the combined organic extracts were dried with anhydrous
MgSO4 and the solvent was evaporated under vacuum. Purification
of the residue by flash chromatography (EtOAc) furnished (+)-ga-
bosine A (12.6 mg, 0.08 mmol, 90%) as a white crystalline solid. Rf
= 0.20 (EtOAc). [α]2D0 = +128 (c = 0.3, MeOH). 1H NMR
= 4.58 (br. s, 1 H, 5-H), 3.56 (t, JH,H = 3.5 Hz, 1 H, 6-H), 3.29 (d,
3
3JH,H = 3.5 Hz, 1 H, 1-H), 2.48 (tq, JH,H = 9.3, 7.2 Hz, 1 H, 3-
3
H), 1.86 (m, 3 H, 2ϫ 4-H, OH), 1.14 (d, JH,H = 7.2 Hz, 3 H,
Me) ppm. 13C NMR (90 MHz, CDCl3, 25 °C): δ = 205.8, 64.9,
56.2, 54.4, 36.3, 32.3, 15.5 ppm. IR (ATR): ν = 3401, 2932, 1704,
˜
1058 cm–1. HRMS (ESI+): calcd. for C7H10O3Na 165.0528; found
165.0527 [M + Na+].
1
(؎)-32: Rf = 0.18 (EtOAc). H NMR (360 MHz, CDCl3, 25 °C): δ
= 4.44 (dd, 3JH,H = 8.6, 5.9 Hz, 1 H, 5-H), 3.57 (m, 1 H, 6-H), 3.34
3
(d, JH,H = 3.8 Hz, 1 H, 1-H), 2.75 (dqd, 3JH,H = 12.1, 6.7, 5.4 Hz,
1 H, 3-H), 2.32 (m, 1 H, 4-H), 1.70 (br. s, 1 H, OH), 1.59 (ddd,
3
3
2JH,H = 13.5, JH,H = 12.1, 8.6 Hz, 1 H, 4-H), 1.03 (d, JH,H
=
6.7 Hz, 3 H, Me) ppm. 13C NMR (90 MHz, CDCl3, 25 °C): δ =
208.3, 65.3, 63.3, 55.5, 34.4, 31.2, 14.8 ppm. IR (ATR): ν = 3430,
˜
2926, 1714, 1082 cm–1. HRMS (ESI+): calcd. for C7H10O3Na
165.0528; found 165.0525 [M + Na+].
Mixture of (1R,3R,5S,6R)-31 and (1R,3S,5S,6R)-32: The same re-
action starting from a 6.6:1 mixture of epoxides (1R,3R,5S,6S)-29
and (1R,3S,5S,6S)-30 (114 mg, 0.44 mmol) furnished a 1:1 mixture
of alcohols (1R,3R,5S,6R)-31 and (1R,3S,5S,6R)-32 (50 mg,
0.35 mmol, 79%).
3
4
(400 MHz, CD3OD, 25 °C): δ = 6.78 (dq, JH,H = 5.6, JH,Me
=
3
1.4 Hz, 1 H, 3-H), 4.42 (m, 1 H, 4-H), 4.36 (d, JH,H = 10.0 Hz, 1
3
4
H, 6-H), 3.76 (dd, JH,H = 10.0, 3.9 Hz, 1 H, 5-H), 1.85 (dd, JH,H
= 1.4, JH,Me = 1.2 Hz, 3 H, Me) ppm. 13C NMR (100 MHz,
5
Mixture of (1S,3S,5R,6S)-31 and (1S,3R,5R,6R)-32: The same reac-
tion starting from a 6.6:1 mixture of epoxides (1S,3S,5R,6R)-29
and (1S,3R,5R,6R)-30 (180 mg, 0.70 mmol) furnished a 1:1 mixture
of alcohols (1S,3S,5R,6S)-31 and (1S,3R,5R,6S)-32 (80 mg,
CD3OD, 25 °C): δ = 201.3, 143.9, 137.7, 75.9, 74.8, 68.3, 16.4 ppm.
Mixture of (1RS,3RS,5SR,6SR)- and (1RS,3SR,5SR,6SR)-5-(tert-
Butyldimethylsilyloxy)-3-methyl-7-oxabicyclo[4.1.0]heptan-2-one
[(؎)-29 and (؎)-30]: To a boiling solution of (Ϯ)-9 (363 mg, 0.56 mmol, 80%). After repeated flash chromatography (hexanes/
Eur. J. Org. Chem. 2011, 1534–1543
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1541

P. Bayón, M. Figueredo et al.
FULL PAPER
EtOAc, 10:1 to 1:1), analytical samples of (1S,3S,5R,6S)-31 and (14 mg, 0.07 mmol) was converted into gabosine B (10 mg,
(1S,3R,5R,6S)-32 were isolated.
0.06 mmol, 90%). [α]2D0 = –88 (c = 0.51, MeOH).
(1S,3S,5R,6S)-31: oil; [α]2D0 = –28 (c = 0.14, CHCl3).
(1S,3R,5R,6S)-32: oil; [α]2D0 = –36 (c = 0.11, CHCl3).
Supporting Information (see footnote on the first page of this arti-
cle): General methods and materials; procedures for the prepara-
tion and characterization data of compounds 7, 8, 10, 13–15, 20,
23–26, 29, 30, 33, and 34; NMR spectra of compounds 7–17, 20,
22–26, 29–34, epiepoformin, epoformin, gabosine A, and gabosine
B/F.
Mixture of (1RS,2SR,4RS,6RS)- and (1RS,2SR,4SR,6RS)-4-
Methyl-5-oxo-7-oxabicyclo[4.1.0]heptan-2-yl Acetate [(؎)-33 and
(؎)-34]: To a solution of a 1:1 mixture of epoxides (Ϯ)-31 and (Ϯ)-
32 (64 mg, 0.45 mmol) in anhydrous acetonitrile (4.2 mL) at 0 °C
under N2, DMAP (61 mg, 0.50 mmol) and acetic anhydride
(171 μL, 1.81 mmol) were added. The reaction mixture was
warmed to room temperature, stirred for 5 min and then poured
into ice-water (2 mL) and extracted with CH2Cl2 (3ϫ 2 mL). The
combined organic layers were washed with cold water and a satu-
rated aqueous solution of NaHCO3, then dried with MgSO4. Re-
moval of the solvent under reduced pressure furnished an oily resi-
due, which was purified by flash chromatography (EtOAc) to yield
a 1:1 mixture of (Ϯ)-33 and (؎)-34 (68 mg, 0.37 mmol, 82%) as an
oil (see the Supporting Information for analytical data).
Acknowledgments
We acknowledge financial support from Ministerio de Educación
y Ciencia (MEC), grant number CTQ2007-60613, and fellowships
from the Direcció General de Recerca (to G. M.) and MEC (to
G. T.).
[1] a) G. Bach, S. Breiding-Mack, S. Grabley, P. Hammann, K.
Hütter, R. Thiericke, H. Uhr, J. Wink, A. Zeeck, Liebigs Ann.
Chem. 1993, 241–250. Synthetic studies led to the structural
reassignment of gabosine K, see: b) G. Mehta, S. Lakshminath,
Tetrahedron Lett. 2000, 41, 3509–3512; c) T. K. M. Shing,
H. M. Cheng, Synlett 2010, 142–144.
[2] Y.-Q. Tang, C. Maul, R. Höfs, I. Sattler, S. Grabley, X.-Z.
Feng, A. Zeeck, R. Thiericke, Eur. J. Org. Chem. 2000, 149–
153.
[3] A. Müller, W. Keller-Schierlein, J. Bielecki, G. Rak, J. Stümpfel,
H. Zähner, Helv. Chim. Acta 1986, 69, 1829–1832.
[4] K. Tatsuta, T. Tsuchiya, N. Mikami, S. Umezawa, H. Umez-
awa, H. Naganawa, J. Antibiot. 1974, 27, 579–586.
[5] a) T. Takeuchi, H. Chimura, M. Hamada, H. Umezawa, O.
Yoshioka, N. Oguchi, Y. Takahashi, A. Matsuda, J. Antibiot.
1975, 28, 737–742; b) H. Chimura, H. Nakamura, T. Takita,
T. Takeuchi, H. Umezawa, K. Kato, S. Saito, T. Tomisawa, Y.
Iitaka, J. Antibiot. 1975, 28, 743–748; c) Y. Sugimoto, H. Su-
zuki, H. Yamaki, T. Nishimura, N. Tanaka, J. Antibiot. 1982,
35, 1222–1230; d) C. F. M. Huntley, D. S. Hamilton, D. J.
Creighton, B. Ganem, Org. Lett. 2000, 2, 3143–3144.
[6] A. Maier, C. Maul, M. Zerlin, S. Grabley, R. Thiericke, J. Anti-
biot. 1999, 52, 952–959.
Mixture of (1R,2S,4R,6R)-33 and (1R,2S,4S,6R)-34: The same re-
action starting from a 1:1 mixture of alcohols (1R,3R,5S,6R)-31
and (1R,3S,5S,6R)-32 (46 mg, 0.32 mmol) furnished a 1:1 mixture
of acetates (1R,2S,4R,6R)-33 and (1R,2S,4S,6R)-34 (48 mg,
0.25 mmol, 80%).
Mixture of (1S,2R,4S,6S)-33 and (1S,2R,4R,6S)-34: The same reac-
tion starting from a 1:1 mixture of alcohols (1S,3S,5R,6S)-31 and
(1S,3R,5R,6S)-32 (36 mg, 0.26 mmol) furnished a 1:1 mixture of
acetates (1S,2R,4S,6S)-33 and (1S,2R,4R,6S)-34 (40 mg,
0.22 mmol, 84%).
(؎)-Gabosine B/F: To an ice-cooled solution of acetates (Ϯ)-33 and
(Ϯ)-34 (76 mg, 0.41 mmol) in toluene (4.5 mL), BF3·Et2O (52 μL,
0.41 mmol) was added and the mixture was stirred at 0 °C for 2 h.
The reaction mixture was neutralized with saturated aqueous
NaHCO3 solution and the aqueous phase was extracted with
EtOAc (3ϫ 2 mL). The combined organic extracts were dried with
anhydrous MgSO4 and the solvent was evaporated under vacuum.
The residue was purified by flash chromatography (EtOAc) to af-
ford a mixture of the expected diols (79 mg, 0.39 mmol, 96%) as a
yellow oil. To a solution of the mixture (48 mg, 0.24 mmol) in
MeOH (0.5 mL), was added NaOMe (12.8 mg, 0.24 mmol) and the
mixture was stirred at room temperature for 1 h. The solvent was
removed under vacuum and the residue was diluted in water and
neutralized with TFA (to ca. pH 7). Purification of the residue by
flash chromatography (CHCl3/MeOH, 20:1) furnished gabosine B/
F (35 mg, 0.22 mmol, 92%) as a white crystalline material. Rf =
[7] a) S. Sakamura, K. Nabeta, S. Yamada, A. Ichihara, Agric.
Biol. Chem. 1971, 35, 1639–1640; b) S. Sakamura, K. Nabeta,
S. Yamada, A. Ichihara, Agric. Biol. Chem. 1975, 39, 403–407.
[8] P. Venkatasubbaiah, N. A. Tisserat, W. S. Chilton, Mycopathol-
ogia 1994, 128, 155–159.
[9] J. Marco-Contelles, M. T. Molina, S. Anjum, Chem. Rev. 2004,
104, 2857–2899.
[10] I. Yamamoto, E. Mizuta, H. Teruji, Y. Toko, Y. Saburo, Takeda
Kenkyushoho 1973, 32, 532–538.
1
[11] H. Nagasawa, A. Suzuki, S. Tamura, Agric. Biol. Chem. 1978,
0.24 (CHCl3/MeOH, 9:1). H NMR (400 MHz, CD3OD, 25 °C): δ
42, 1303–1304.
3
3
= 4.43 (d, JH,H = 10.0 Hz, 1 H, 2-H), 4.13 (br. q, JH,H ≈ 3.0 Hz,
[12] a) P. Venkatasubbaiah, W. S. Chilton, J. Nat. Prod. 1992, 55,
639–643; b) P. Venkatasubbaiah, C. G. Van Dyke, W. S. Chil-
ton, Mycologia 1992, 84, 715–723.
[13] M. Makino, T. Endoh, Y. Ogawa, K. Watanabe, Y. Fujimoto,
Heterocycles 1998, 48, 1931–1934.
[14] For pioneer references on different strategies, see: a) Diels–Al-
der: A. Ichihara, R. Kimura, K. Oda, S. Sakamura, Tetrahe-
dron Lett. 1976, 17, 4741–4744; b) Carbohydrates: S. Mirza,
L.-P. Molleyres, A. Vasella, Helv. Chim. Acta 1985, 68, 988–
996; c) Quinic acid: T. K. M. Shing, Y. Tang, J. Chem. Soc.,
Chem. Commun. 1990, 312.
3
1 H, 4-H), 3.49 (dd, JH,H = 10.0, 3.0 Hz, 1 H, 3-H), 2.96 (dqd,
3JH,H = 13.0, 6.6, 5.9 Hz, 1 H, 6-H), 2.15 (ddd, JH,H
≈
3JH,H
≈
2
3
2
3
14.0, JH,H = 5.9, 3.2 Hz, 1 H, 5-H), 1.45 (td, JH,H = 14.0, JH,H
= 2.6 Hz, 1 H, 5-H), 1.06 (d, JH,H = 6.6 Hz, 3 H, Me) ppm. 13C
3
NMR (100 MHz, CD3OD, 25 °C): δ = 212.3, 80.0, 79.0, 70.6, 39.0,
38.7, 14.6 ppm.
(+)-Gabosine F: The same protocol starting from acetates
(1R,2S,4R,6R)-33 and (1R,2S,4S,6R)-34 (30 mg, 0.16 mmol) fur-
nished a mixture of diols (29 mg, 0.14 mmol, 86%), part of which
(20 mg, 0.10 mmol) was converted into gabosine F (15 mg,
0.10 mmol, 92%). [α]2D0 = +89 (c = 0.32, MeOH).
[15] R. Alibés, P. Bayón, P. de March, M. Figueredo, J. Font, G.
Marjanet, Org. Lett. 2006, 8, 1617–1620.
[16] a) P. de March, M. Escoda, M. Figueredo, J. Font, A. Álvarez-
Larena, J. F. Piniella, J. Org. Chem. 1995, 60, 3895–3897; b) P.
de March, M. Escoda, M. Figueredo, J. Font, Tetrahedron Lett.
1995, 36, 8665–8668; c) P. de March, M. Escoda, M. Fig-
(–)-Gabosine B: The same protocol starting from acetates
(1S,2R,4S,6S)-33 and (1S,2R,4R,6S)-34 (22 mg, 0.12 mmol) fur-
nished a mixture of diols (22 mg, 0.11 mmol, 89%), part of which
1542
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1534–1543

Divergent Approach to Gabosines and Anhydrogabosines
ueredo, J. Font, J. Medrano, An. Quim. 1997, 93, 81–87; d) P.
de March, M. Figueredo, J. Font, J. Medrano, Tetrahedron
1999, 55, 7907–7914.
Carreño, E. Merino, M. Ribagorda, A. Somoza, A. Urbano,
Chem. Eur. J. 2007, 13, 1064–1077.
[24] T. Tachihara, T. Kitahara, Tetrahedron 2003, 59, 1773–1780.
[25] D. M. Pinkerton, M. G. Banwell, A. C. Willis, Org. Lett. 2009,
11, 4290–4293.
[26] a) M. T. Barros, C. D. Maycock, M. R. Ventura, Tetrahedron
1999, 55, 3233–3244; b) M. T. Barros, C. D. Maycock, M. R.
Ventura, Chem. Eur. J. 2000, 6, 3991–3996.
[27] H. Okamura, H. Shimizu, N. Yamashita, T. Iwagawa, M. Nak-
atani, Tetrahedron 2003, 59, 10159–10164.
[28] The specific rotation values described in the literature for epie-
poformin range between [α]D = +221 (c = 0.83, EtOH)[11] and
[α]D = +320 (c = 0.06, EtOH).[24]
[17]
a) P. de March, M. Escoda, M. Figueredo, J. Font, A. Alvarez-
Larena, J. F. Piniella, J. Org. Chem. 1997, 62, 7781–7787; b) P.
de March, M. Figueredo, J. Font, S. Rodríguez, Tetrahedron
2000, 56, 3603–3609; c) P. de March, M. Escoda, M. Fig-
ueredo, J. Font, E. García-García, S. Rodríguez, Tetrahedron:
Asymmetry 2000, 11, 4473–4483; d) F. Busqué, P. de March,
M. Escoda, M. Figueredo, J. Font, S. Rodríguez, Tetrahedron:
Asymmetry 2001, 12, 3077–3080; e) M. Cantó, P. de March,
M. Figueredo, J. Font, S. Rodríguez, A. Álvarez-Larena, J. F.
Piniella, Tetrahedron: Asymmetry 2002, 13, 455–459; f) F.
Busqué, M. Cantó, P. de March, M. Figueredo, J. Font, S.
Rodríguez, Tetrahedron: Asymmetry 2003, 14, 2021–2032; g)
R. Alibés, P. de March, M. Figueredo, J. Font, G. Marjanet,
Tetrahedron: Asymmetry 2004, 15, 1151–1155; h) G. G. Bardají,
M. Cantó, R. Alibés, P. Bayón, P. de March, M. Figueredo, J.
Font, J. Org. Chem. 2008, 73, 7657–7662; i) E. Ferrer, R. Al-
ibés, F. Busqué, M. Figueredo, J. Font, P. de March, J. Org.
Chem. 2009, 74, 2425–2423.
[29] a) T. Shinada, T. Fuji, Y. Ohtani, Y. Yoshida, Y. Ohfune, Syn-
lett 2002, 1341–1343; b) R. N. Monrad, M. Fanefjord, F. G.
Hansen, N. M. E. Jensen, R. Madsen, Eur. J. Org. Chem. 2009,
396–402.
[30] T. K. M. Shing, H. M. Cheng, Org. Biomol. Chem. 2009, 7,
5098–5102.
[31] M. G. Banwell, A. M. Bray, D. J. Wong, New J. Chem. 2001,
25, 1351–1354.
[32] D. A. Henderson, P. N. Collier, G. Pavé, P. Rzepa, A. J. P.
White, J. N. Burrows, A. G. M. Barrett, J. Org. Chem. 2006, 71,
2434–2444.
[18]
[19]
P. Bayón, G. Marjanet, G. Toribio, R. Alibés, P. de March, M.
Figueredo, J. Font, J. Org. Chem. 2008, 73, 3486–3491.
M. T. Barros, P. M. Matias, C. D. Maycock, M. R. Ventura,
Org. Lett. 2003, 5, 4321–4323.
[33] M. Shan, Y. Xing, G. A. O’Doherty, J. Org. Chem. 2009, 74,
5961–5966.
[20] M. T. Barros, C. D. Maycock, M. R. Ventura, J. Org. Chem.
1997, 62, 3984–3988.
[34] P. Wipf, P. Jeger, Y. Kim, Bioorg. Med. Chem. Lett. 1998, 8,
351–356.
[35] G. Mehta, S. R. Pujar, S. S. Ramesh, K. Islam, Tetrahedron
Lett. 2005, 46, 3373–3376.
[36] T. K. M. Shing, K. H. So, W. S. Kwok, Org. Lett. 2009, 11,
[21] A. Ichihara, R. Kimura, K. Oda, K. Moriyasu, S. Sakamura,
Agric. Biol. Chem. 1982, 46, 1879–1883.
[22] T. Kamikubo, K. Ogasawara, Tetrahedron Lett. 1995, 36, 1685–
1688.
5070–5073.
[23] a) H. Shimizu, H. Okamura, N. Yamashita, T. Iwagawa, M.
Nakatani, Tetrahedron Lett. 2001, 42, 8649–8651; b) M. C.
Received: November 12, 2010
Published Online: January 31, 2011
Eur. J. Org. Chem. 2011, 1534–1543
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1543