Reductive cleavage of tetrahydrofuryl sulfur-substituted oxiranes: Application to the formal synthesis of Kumausyne and Kumausallene

Article abstract of DOI:10.1021/jo051459k

Readily available sulfnyl and sulfonyl tetrahydrofuran methanol derivatives have been transformed efficiently into a variety of substituted tetrahydrofuryl alcohols by treatment with (PhSe)₂ in the presence of an excess of NaBH₄. Alt

Full text of DOI:10.1021/jo051459k

Reductive Cleavage of Tetrahydrofuryl Sulfur-Substituted
Oxiranes: Application to the Formal Synthesis of Kumausyne and
Kumausallene
Roberto Ferna´ndez de la Pradilla,* Carolina Alhambra, Alejandro Castellanos,
Jorge Ferna´ndez, Pilar Manzano, Carlos Montero, Mercedes Uren˜a, and Alma Viso
Instituto de Quı´mica Orga´nica, CSIC, Juan de la Cierva, 3, 28006 Madrid, Spain
iqofp19@iqog.csic.es
Received July 14, 2005
Readily available sulfinyl and sulfonyl tetrahydrofuran methanol derivatives have been transformed
efficiently into a variety of substituted tetrahydrofuryl alcohols by treatment with (PhSe)2 in the
presence of an excess of NaBH4. Alternatively, oxirane cleavage with MgI2 produces the related
ketones, amenable to stereocontrolled reduction. This reductive cleavage methodology has been
applied to short formal syntheses of trans-Kumausyne and Kumausallene.
Introduction
illustrate the influence of the oxidation state at sulfur,
the nature of the R substituent, and the different stereo-
chemistry at positions 2 and 5 on the outcome of the
desired reductive oxirane cleavage. All substrates were
prepared by our reported methodology.4 Scheme 2 also
shows some substrates used in the second part of this
study (10a,c and 11a,c) that were prepared from di-
Substituted tetrahydrofurans are ubiquitous structural
motifs in bioactive natural products,1 and this has
motivated a sustained interest in the development of
synthetic methods to prepare these heterocycles.2 A few
years ago, we reported an expedient route to 2,5-cis
tetrahydrofuryl sulfur-substituted oxiranes, B, by the
nucleophilic epoxidation of hydroxyl sulfinyl dienes A
(Scheme 1).3 The scope and limitations of this protocol
were later disclosed along with an effective alternative
stepwise methodology to access 2,5-cis and 2,5-trans
sulfonyl derivatives, B.4 In this article, we report in full
our efforts to develop stereoselective routes to ketones C
and alcohols D from precursors B, as well as the
application of these methodologies to the formal synthe-
ses of marine natural products Kumausyne, 1, and
Kumausallene, 2.
(2) For a review on synthetic approaches to tetrahydrofurans, see:
(a) Boivin, T. L. B. Tetrahedron 1987, 43, 3309-3362. (b) Harmange,
J.-C.; Figade`re, B. Tetrahedron: Asymmetry 1993, 4, 1711-1754. (c)
Elliot, M. C. J. Chem. Soc., Perkin Trans. 1 2002, 2301-2323. (d)
Casiraghi, G.; Zanardi, F.; Battistini, G.; Rassu, G.; Appendino, G.
Chemtracts: Org. Chem. 1998, 11, 803-828. (e) Jaramillo, C.; Knapp,
S. Synthesis 1994, 1-20. (f) Du, Y.; Linhardt, R. J.; Vlahov, I. R.
Tetrahedron 1998, 54, 9913-9959. For leading references, see: (g)
Micalizio, G. C.; Roush, W. R. Org. Lett. 2000, 2, 461-464. (h) Pilli, R.
A.; Riatto, V. B.; Vencato, I. Org. Lett. 2000, 2, 53-56. (i) Xiong, Z.;
Corey, E. J. J. Am. Chem. Soc. 2000, 122, 4831-4832. (j) Lei, A.; He,
M.; Wu, S.; Zhang, X. Angew. Chem., Int. Ed. 2002, 41, 3457-3460.
(k) Kim, H. C.; Woo, S. W.; Seo, M. J.; Jeon, D. J.; No, Z.; Kim, H. R.
Synlett 2002, 1691-1693. (l) Smitrovich, J. H.; Woerpel, K. A. Synthesis
2002, 2778-2785. (m) Ajamian, A.; Gleason, J. L. Org. Lett. 2001, 3,
4161-4164. (n) Evans, D. A.; Sweeney, Z. K.; Rovis, T.; Tedrow, J. S.
J. Am. Chem. Soc. 2001, 123, 12095-12096. (o) Loh, T.-P.; Hu, Q.-Y.;
Tan, K.-T.; Cheng, H.-S. Org. Lett. 2001, 3, 2669-2672. (p) Cohen, F.;
MacMillan, D. W. C.; Overman, L. E.; Romero, A. Org. Lett. 2001, 3,
1225-1228. (q) Chakraborty, T. K.; Das, S.; Raju, T. V. J. Org. Chem.
2001, 66, 4091-4093. (r) Wakabayashi, K.; Yorimitsu, H.; Oshima, K.
J. Am. Chem. Soc. 2001, 123, 5374-5375. (s) Angle, S. R.; El-Said, N.
A. J. Am. Chem. Soc. 2002, 124, 3608-3013.
Reductive Cleavage. Scheme 2 gathers the sub-
strates chosen for the initial part of this study; these
substrates (3a-e, 4a-e, and 5a,c,d) were selected to
(1) For reviews, see: (a) Westley, J. W. Polyether Antibiotics:
Naturally Occurring Acid Ionophores; Marcel Dekker: New York, 1982;
Vols. 1 and 2. (b) Faul, M. M.; Huff, B. E. Chem. Rev. 2000, 100, 2407-
2473. (c) Alali, F. Q.; Liu, X.-X.; McLaughlin, J. L. J. Nat. Prod. 1999,
62, 504-540. (d) Bermejo, A.; Figade`re, B.; Zafra-Polo, M. C.; Bar-
rachina, I.; Estornell, E.; Cortes, D. Nat. Prod. Rep. 2005, 22, 269-
303. (e) Hacksell, H.; Daves, G. D. The Chemistry and Biochemistry
of C-Nucleosides and C-Aryl Glycosides. In Progress in Medicinal
Chemistry; Ellis, G. P., West, G. B., Eds.; Elsevier: London, 1985; Vol
22, pp 1-65.
(3) Ferna´ndez de la Pradilla, R.; Montero, C.; Priego, J.; Mart´ınez-
Cruz, L. A. J. Org. Chem. 1998, 63, 9612-9613.
(4) Ferna´ndez de la Pradilla, R.; Manzano, P.; Montero, C.; Priego,
J.; Mart´ınez-Ripoll, M.; Mart´ınez-Cruz, L. A. J. Org. Chem. 2003, 68,
7755-7767.
10.1021/jo051459k CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/18/2005
J. Org. Chem. 2005, 70, 10693-10700
10693

Ferna´ndez de la Pradilla et al.
SCHEME 1
NaBH4 since this species had been used before for the
reductive cleavage of simple sulfinyl oxiranes.5 The
results obtained in this study are shown in Table 1. To
obtain ketones 14 or 15, we used conditions that avoided
an excess of reducing agent, and we found that the
treatment of substrate 3a with a large excess of nucleo-
philic selenium species in EtOH gave ketone 12a as a
single isomer of undetermined stereochemistry, along
with some recovered starting material (Table 1, entry 1).
The use of a smaller excess of reagent on the less-reactive
sulfone 4a gave a complex mixture of products from
which selenyl ketone 13a (5%) and starting material 4a
(35%) were isolated, again in a very slow reaction (Table
1, entry 2). Resubmission of 13a to the reaction conditions
gave ketone 12a. These experiments had been started
under argon but without special precautions to secure
an inert atmosphere until completion, and it was ob-
served that the white suspension initially formed turned
yellow after just a few hours. This observation was
attributed to oxidative destruction of the reactive species
to regenerate (PhSe)2, and thus the remaining experi-
ments were carried out under argon and with slow
bubbling of argon through the reaction medium.
SCHEME 2
Sulfonyl oxiranes 4a and 4b and sulfinyl oxirane 3c,
under these optimized conditions, rendered good to
excellent yields of the expected ketones but as practically
nonselective mixtures of 2,5-trans and 2,5-cis isomers 14
and 15 (Table 1, entries 3-5). These results indicated
that the nucleophilic selenium species was too basic to
allow for the preservation of the stereochemistry of the
starting material upon transformation to the desired
ketone. At this stage, we hypothesized that the use of a
large excess of NaBH4, along with slow bubbling of argon
through the reaction, could allow for the use of less
(PhSe)2 and, more importantly, perhaps reduce the
carbonyl with useful stereocontrol, prior to epimerization.
The results obtained under these conditions are gath-
ered in Table 1, entries 6-12. In these cases, the precise
stoichiometry of the reaction was quite case-dependent,
but a substantial reduction of the amount of (PhSe)2 could
be achieved, especially for the more reactive sulfinyl
oxiranes. With regard to the selectivity of the process,
in all cases the all-cis diols 16 were predominant but the
selectivities ranged from low (Table 1, entry 7) to excel-
lent (Table 1, entry 6). Unfortunately, these conditions
were not effective for 2,5-trans isomers 5a and 5d that
gave nonselective mixtures of the expected carbinols 18
and 19 (Table 1, entries 13 and 14).
The general structure of these products was derived
primarily from their 1H and 13C NMR data, and the
stereochemical assignments rely on detailed correlations
hydrofurans 6a/7a and 6c/7c,4 by smooth silylation to
afford protected sulfonyl dihydrofurans 8a,c and 9a,c
that were easily separated by chromatography (Scheme
2). The stereoselective nucleophilic epoxidation of these
substrates with tBuOOK gave the desired substrates
10a,c and 11a,c as single isomers and with good yields.
It should be pointed out that these epoxidations are
significantly more selective than those carried out on the
related substrates with free primary alcohols.4 Three of
these silylated sulfonyl oxiranes were deprotected to the
known alcohols (+)-4a, (+)-4c, and (-)-5c to allow for a
firm structural assignment.
1
particularly based on H NMR spectra and comparison
with literature values,2s as well as the completion of the
formal syntheses described below. The difference in
chemical shifts measured for both H-4 is particularly
diagnostic, with ∆δ ) 0.4-0.6 ppm for 16 and 18 and ∆δ
) 0-0.2 ppm for 17 and 19. On the other hand, isomers
16 had smaller values for J2-3 (2.4-2.8 Hz) than isomers
17 (4.4 Hz); this is in contrast with the measurements
for 2,5-trans derivatives 18 (J2-3 ) 0-1.8 Hz) and 19 (J2-3
) 2.6-2.7 Hz).
With regard to the oxirane cleavage protocol, we
selected the one-step procedure entailing the use of a
nucleophilic selenium species prepared from (PhSe)2 and
(5) Satoh, T.; Kaneko, Y.; Izawa, T.; Sakata, K. Bull. Chem. Soc.
Jpn. 1985, 58, 1983-1990.
10694 J. Org. Chem., Vol. 70, No. 26, 2005

Reductive Cleavage of Sulfur-Substituted Oxiranes
TABLE 1. Reductive Cleavage of Sulfur-Substituted Oxiranes with (PhSe2)/NaBH4
entry
substrate
conditions
productsa
yieldb %
1
3a
4a
4a
4b
3c
4a
4b
3c
4c
3d
4d
4e
5a
5d
6.25 equiv of (PhSe)2, 11.5 equiv of NaBH4, 24 h
3.0 equiv of (PhSe)2, 5.5 equiv of NaBH4, 48 h
3.0 equiv of (PhSe)2, 5.0 equiv of NaBH4, 7 h, Ar
3.0 equiv of (PhSe)2, 5.0 equiv of NaBH4, 7 h, Ar
5.0 equiv of (PhSe)2, 9.0 equiv of NaBH4, 2 h, Ar
1.5 equiv of (PhSe)2, 11.0 equiv of NaBH4, 44 h, Ar
1.0 equiv of (PhSe)2, 8.0 equiv of NaBH4, 7 h, Ar
0.6 equiv of (PhSe)2, 5.0 equiv of NaBH4, 1.5 h, Ar
1.5 equiv of (PhSe)2, 11.0 equiv of NaBH4, 17 h, Ar
0.6 equiv of (PhSe)2, 5.0 equiv of NaBH4, 1.5 h, Ar
1.5 equiv of (PhSe)2, 11.0 equiv of NaBH4, 2.5 h, Ar
2.0 equiv of (PhSe)2, 14.0 equiv of NaBH4, 7 h, Ar
3.0 equiv of (PhSe)2, 14.0 equiv of NaBH4, 96 h, Ar
1.5 equiv of (PhSe)2, 11.0 equiv of NaBH4, 8 h, Ar
12a
13a
54c
40d
81e
92f
57
2
3
14a (70) 15a (30)
14b (60) 15b (40)
14c (60) 15c (40)
16a
16b (70) 17b (30)
16c (80) 17c (20)
16c (80) 17c (20)
16d (86) 17d (14)
16d (85) 17d (15)
16e (80) 17e (20)
18a (58) 19a (42)
18d (60) 19d (40)
4
5
6
73g
59
7
8
83
9
62
10
11
12
13
14
94
90
51
48
84
a Diastereomeric ratios are shown in parentheses. b Combined yields of pure products after column chromatography. c Reaction carried
out without bubbling argon; starting material (16%) recovered. d Reaction carried out without bubbling argon; starting material (35%)
recovered. e Traces of starting material and 12a also isolated. f Starting material (9%) recovered. g Starting material (17%) recovered.
While this cleavage/reduction protocol could be useful,
at least for 2,5-cis derivatives, we decided to explore
alternative routes to access the ketones. After briefly
evaluating the cleavage with MgBr2,6 which should be
followed by a potentially troublesome reductive dehalo-
genation, we considered the use of an excess of MgI2. It
was envisioned that the iodoketone intermediate could
react with iodide to generate I2 (in analogy to the
selenium protocol described above) and thus produce the
desired dehalogenated ketone in a one-pot procedure and
under acidic conditions. Scheme 3 gathers the results
obtained for the reductive cleavage of sulfonyl oxiranes
with freshly prepared magnesium halides. To test this
idea, sulfonyl oxirane 5c was treated with MgBr2 to
afford bromoketone 20 as a single isomer; then sulfonyl
oxirane 4a was treated with an excess of MgI2 in Et2O
to produce in a rapid reaction 2,5-cis ketone 15a as a
single isomer and in excellent yield. Similarly, from
protected substrates 10a,c and 11a,c, prepared as de-
scribed above, diastereomerically pure ketones (+)-21a,c
and (+)-22a,c were obtained uneventfully in excellent
yields. Conditions to achieve the stereoselective reduction
of these ketones were then explored briefly on (+)-21c
and (+)-22c, and L-selectride gave excellent results to
prepare 2,5-trans alcohol (-)-23c with complete selectiv-
ity, and NaBH4 gave 2,5-cis alcohol (-)-24c with com-
plete stereocontrol.
Formal Syntheses of Kumausyne and Kumaus-
allene. The 2,5-disubstituted-3-oxygenated tetrahydro-
furan unit is present in a variety of natural products.
Among these, we focused our attention on halogenated
metabolites isolated from the genus Laurencia, especially
on (-)-trans-Kumausyne, 1,7 (Scheme 4) and (-)-Kumaus-
allene, 2,8 which appeared as suitable targets for an
application of our methodologies. These products have
been synthesized before,9,10 and upon inspection of the
existing literature we envisioned that a unified approach
to these targets could rely on lactol 25 (Scheme 4),
notwithstanding the fact that since our efforts had begun
with (-)-menthyl sulfinate, the unnatural enantiomer
would be obtained. trans-Kumausyne had been prepared
from lactol ent-25 (P ) TBDPS) in six steps,9b and (()-
(7) cis- and trans-Kumausyne: Suzuki, T.; Koizumi, K.; Suzuki, H.;
Kurosawa, E. Chem. Lett. 1983, 1643-1646.
(8) Kumausallene: Suzuki, T.; Koizumi, K.; Suzuki, H.; Kurosawa,
E. Chem. Lett. 1983, 1639-1642.
(6) Durst, T.; Tin, K.-C.; Reinach-Hirtzbach, F.; Decesare, J. M.;
Ryan, M. D. Can. J. Chem. 1979, 57, 258-266.
J. Org. Chem, Vol. 70, No. 26, 2005 10695

Ferna´ndez de la Pradilla et al.
SCHEME 3
substituted oxiranes 3f and 4f by the reductive oxirane
cleavage followed by selective carbonyl reduction de-
scribed in this article. Furthermore, we considered that
benzylated lactol 25 (P ) Bn) could be prepared expedi-
ently by our methodology and result in a further im-
provement of Evans’ route. Finally, if viable, the use of
unprotected lactol 25 (P ) H), in principle available in
just five steps from (-)-menthyl sulfinate, could result
in further shortening of both Overman’s and Evans’
routes. Therefore, we decided to examine the preparation
of these lactols 25 (P ) TBDPS, Bn, H) from our sulfinyl
and sulfonyl oxiranes.
Scheme 5 gathers the preparation of suitably protected
derivatives from the known oxiranes (+)-3f and (+)-4f,
available in three steps from (-)-menthyl sulfinate.4
Silylation under standard conditions proceeded smoothly
to afford silyl ethers (+)-28a, (+)-28b, and (-)-29b in
good yield. In contrast, formation of benzyl ethers 28c
and (+)-29c required more experimentation and gave
just fair yields of the desired products. Initially, we
focused our attention on the stepwise oxirane cleavage
and carbonyl reduction; however, the treatment of sul-
fonyl oxirane (-)-29b with MgI2 led to complex mixtures
of products under different experimental conditions. This
led us to reexamine the one-pot selenium-based protocol
to prepare the desired target 26.
Table 2 summarizes our results on the reductive
cleavage of these synthetic precursors. Not too unexpect-
edly, when the cleavage was carried out in the absence
of excess NaBH4, R,â-unsaturated ketone (-)-31b was
obtained by a facile double bond migration under these
basic conditions (Table 2, entry 1). Ketone (-)-31b was
produced as a single isomer, but the stereochemistry of
the exocyclic alkene has not been studied in detail.
Entries 2 and 3 show that the use of a large excess of
NaBH4 affords mixtures of the desired alcohols (+)-26b
and (-)-32b in good yields and with reasonable selectivi-
ties. Unprotected precursors (+)-3f and (+)-4f were also
transformed into diols (+)-26d and (-)-32d, albeit with
reduced selectivity (Table 2, entries 4 and 5).
Kumausallene had been prepared from dioxabicyclic
derivative 27 in 13 steps.10a Alternatively, Evans et al.
reported a synthesis of Kumausallene that entailed a 10-
step sequence to access benzylated lactol ent-25 (P ) Bn)
from commercially available cis-4-benzyloxy-2-buten-1-
ol, followed by a three-step sequence to produce bromo-
allene ent-28 that was transformed in four additional
steps to Kumausallene.10c It was envisioned that Over-
man’s intermediate 27 could arise from 25 by a Wittig
reaction followed by intramolecular Michael cyclization.
In turn, 25 would be obtained from monoprotected diol
26 by a Wacker protocol,11 and 26 could arise from sulfur-
SCHEME 4
10696 J. Org. Chem., Vol. 70, No. 26, 2005

Reductive Cleavage of Sulfur-Substituted Oxiranes
TABLE 2. Reductive Cleavage of Precursors for trans-Kumausyne and Kumausallene
entry
substrate
conditions
P
productsa
yieldb %
1
2
3
4
5
6
(+)-28b
(+)-28b
(-)-29b
(+)-3f
(+)-4f
28c
5.0 equiv of (PhSe)2, 9.0 equiv of NaBH4, 0 °C, 1 h
3.0 equiv of (PhSe)2, 26.0 equiv of NaBH4, 0 °C, 3 h
4.0 equiv of (PhSe)2, 28.0 equiv of NaBH4, 28 h
3.0 equiv of (PhSe)2, 26.0 equiv of NaBH4, 0 °C, 4 h
3.0 equiv of (PhSe)2, 26.0 equiv of NaBH4, rt, 19 h
3.0 equiv of (PhSe)2, 16.0 equiv of NaBH4, 1 equiv of AcOH,
0 °C, 2 h
4.5 equiv of (PhSe)2, 30.0 equiv of NaBH4, 3 equiv of AcOH,
rt, 4 days
3.0 equiv of (PhSe)2, 12.0 equiv of NaBH4, 1 equiv of AcOH,
0 °C to rt, 1.5 h
TBDPS
TBDPS
TBDPS
H
(-)-31b
73
77
69
69
60
70
(+)-26b (87) (-)-32b (13)
(+)-26b (87) (-)-32b (13)
(+)-26d (76) (-)-32d (24)
(+)-26d (76) (-)-32d (24)
(+)-26c (76) 32c (24)
H
Bn
7
8
(+)-29c
(+)-28a
Bn
(+)-26c (76) 32c (24)
80c
80
TBDMS
(+)-26a (83) (-)-32a (17)
a Diastereomeric ratios are shown in parentheses. b Combined yields of pure products after column chromatography. c Starting material
(50%) recovered, included in the yield.
SCHEME 5
that had been already examined by us with the old batch
of reagent. After extensive experimentation and examin-
ing the purity of the new reagent, we found that addition
of 1-3 equiv of AcOH had very significant beneficial
effects on the rates and conversions of these reactions.12
We tentatively attribute this profound difference in
reactivity to the presence of very small amounts of acidic
contaminants in the “old” bottle of (PhSe)2, and therefore
we now routinely add AcOH in these cleavages.
Under these conditions, sulfinyl benzyl ether 28c gave
a good yield of the expected diol derivatives (+)-26c and
32c with fair selectivity. In contrast, the related sulfone
(+)-29c was much less reactive and 50% starting mate-
rial was recovered (Table 2, entries 6 and 7). Finally
TBDMS ether (+)-28a was also examined with results
that largely paralleled the TBDPS analogue (Table 2,
entries 2 and 8).
Our formal synthesis of trans-Kumausyne was com-
pleted by a regiocontrolled Wacker protocol, well-
documented by Mereyala for related systems, on alcohol
(+)-26b that afforded a good yield of lactol (-)-25b as a
74:26 mixture of anomers with spectral features identical
to those in the literature.9e,13 These conditions were also
shown to be viable for unprotected diol (+)-26d, but in
this case, the use of Cu(OAc)2 gave slightly better results
of unprotected lactol (-)-25d (53%), along with ketone
(-)-33 (21%) (Scheme 6).
At this stage of the project, we ran out of (PhSe)2, and
a new bottle of reagent was purchased. Surprisingly, this
batch of reagent provided inferior results in related
transformations examined concurrently in the group,
with low conversions being observed even for reactions
(9) For a review, see: Ferna´ndez de la Pradilla, R.; Viso, A. Recent
Res. Dev. Org. Bioorg. Chem. 2001, 4, 123-132. For syntheses of
Kumausyne, see: (a) Brown, M. J.; Harrison, T.; Overman, L. E. J.
Am. Chem. Soc. 1991, 113, 5378-5384. (b) Osumi, K.; Sugimura, H.
Tetrahedron Lett. 1995, 36, 5789-5792. (c) Mart´ın, T.; Soler, M. A.;
Betancort, J. M.; Mart´ın, V. S. J. Org. Chem. 1997, 62, 1570-1571.
(d) Lee, E.; Yoo, S.-K.; Cho, Y.-S.; Cheon, H.-S.; Chong, Y. H.
Tetrahedron Lett. 1997, 38, 7757-7758. (e) Boukouvalas, J.; Fortier,
G.; Radu, I.-I. J. Org. Chem. 1998, 63, 916-917. (f) Mereyala, H. B.;
Gadikota, R. R. Tetrahedron: Asymmetry 2000, 11, 743-751. (g)
Garc´ıa, C.; Mart´ın, T.; Mart´ın, V. S. J. Org. Chem. 2001, 66, 1420-
1428. (h) Gadikota, R. R.; Callam, C. S.; Lowary, T. L. J. Org. Chem.
2001, 66, 9046-9051. (i) Chandler, C. L.; Phillips, A. J. Org. Lett. 2005,
7, 3493-3495.
Our efforts toward a formal synthesis of Kumausallene
are shown in Scheme 7. The transformation of (-)-25d
(10) For syntheses of Kumausallene, see: (a) Grese, T. A.; Hutch-
inson, K. D.; Overman, L. E. J. Org. Chem. 1993, 58, 2468-2477. (b)
Lee, E.; Yoo, S.-K.; Choo, H.; Song, H. Y. Tetrahedron Lett. 1998, 39,
317-318. (c) Evans, P. A.; Murthy, V. S.; Roseman, J. D.; Rheingold,
A. L. Angew. Chem., Int. Ed. 1999, 38, 3175-3177. (d) For a review,
see: Hoffmann-Ro¨der, A.; Krause, N. Angew. Chem., Int. Ed. 2004,
43, 1196-1216.
(11) (a) Mereyala, H. B.; Gadikota, R. R.; Krishnan, R. J. Chem.
Soc., Perkin Trans. 1 1997, 3567-3571. (b) Krishnudu, K.; Krishna,
P. R.; Mereyala, H. B. Tetrahedron Lett. 1996, 37, 6007-6010. See
also: (c) Mereyala, H. B.; Gadikota, R. R.; Sunder, K. S.; Shailaja, S.
Tetrahedron 2000, 56, 3021-3026.
(12) The beneficial effect of weak acids on related selenide cleavages
has been noted before. See: Miyashita, M.; Suzuki, T.; Hoshino, M.;
Yoshikoshi, A. Tetrahedron 1997, 53, 12469-12486.
(13) This reaction required substantial optimization. Early experi-
ments with less PdCl2 gave important amounts of byproducts of
undetermined structure, tentatively attributed to transient buildup
of HCl through the catalytic cycle. Unfortunately, the use of Pd(OAc)2
and Cu(OAc)2 (Smith, A. B.; Cho, Y. S.; Friestad, G. K. Tetrahedron
Lett. 1998, 39, 8765-8768.) did not give better results. Finally the
conditions shown in Scheme 7, with continuous slow bubbling of oxygen
through the reaction medium, gave good results.
J. Org. Chem, Vol. 70, No. 26, 2005 10697

Ferna´ndez de la Pradilla et al.
SCHEME 6
catalytic amount of NaOMe in MeOH gave an 86:14
mixture of dioxabicyclic derivatives that were deprotected
with nBu4NF buffered with AcOH to produce Kumaus-
allene precursor (-)-27 (64% two steps, 86:14 mixture)
with spectral features identical to those in the literature.10a
To shorten the sequence, (+)-35b was treated with
nBu4NF that smoothly effected desilylation and cycliza-
tion in a single step with comparable yield but slightly
lower selectivity (72:28). This mixture was treated with
catalytic NaOMe/MeOH without a significant change in
the isomeric ratio. Finally, to further shorten our syn-
thetic sequence, we examined briefly the Wittig reaction
on unprotected lactol (-)-25d that cleanly led to unsat-
urated ester 35d (67:33, E/Z mixture) and the smooth
base-catalyzed cyclization of 35d that led to a 75:25
mixture of epimers of 27.14
SCHEME 7
Conclusions
An expedient route to a variety of substituted tetra-
hydrofurans from readily available sulfur-substituted
oxiranes has been outlined. This reductive cleavage
methodology has been applied to short formal syntheses
of trans-Kumausyne and Kumausallene.
Experimental Section
General Procedure for the Preparation of Silyl Ethers.
To a solution of the substrate in anhydrous CH2Cl2 (0.4 mL/
mmol) at room temperature was added imidazole (1.5 equiv)
and DMAP (0.05 equiv). The solution was cooled to 0 °C and
TBDMSCl or TBDPSCl (1.5 equiv) was added. The mixture
was allowed to warm to room temperature and monitored by
TLC. Upon completion, the reaction was quenched with H2O
(5 mL/mmol) and saturated NH4Cl solution (5 mL/mmol). The
layers were separated, the aqueous phase was extracted with
CH2Cl2 (3 × 5 mL/mmol), and the combined organic extracts
were washed with a saturated solution of NaCl (5 mL/mmol),
dried over MgSO4, filtered, and concentrated under reduced
pressure to give a crude product that was purified by column
chromatography on silica gel using a gradient of the appropri-
ate solvents.
Synthesis of the tert-Butyldimethylsilyl Ethers of
(2R,5R)-5-Isopropyl-4-(p-tolylsulfonyl)-2,5-dihydrofuran-
2-methanol, (-)-8a, and (2S,5R)-5-Isopropyl-4-(p-tolyl-
sulfonyl)-2,5-dihydrofuran-2-methanol, (+)-9a. From a
mixture of sulfonyl dihydrofurans 6a and 7a4 (189 mg, 0.64
mmol, 1.0 equiv) with TBDMSCl (152 mg, 0.96 mmol, 1.5
equiv), according to the general procedure (8 h), a 43:57
mixture of silyl ethers (-)-8a and (+)-9a was obtained.
Purification by chromatography (30-70% CH2Cl2/hexane) gave
(-)-8a (76 mg, 29%) and (+)-9a (137 mg, 61%) as colorless
oils. Data for (-)-8a: Rf 0.25 (CH2Cl2). [R]20D -105.1 (c 1.39).
1H NMR (300 MHz) δ -0.04 (s, 3 H), -0.01 (s, 3 H), 0.69 (d, 3
H, J ) 6.8 Hz), 0.81 (s, 9 H), 0.98 (d, 3 H, J ) 6.9 Hz), 2.06
(hept d, 1 H, J ) 6.9, 1.5 Hz), 2.43 (s, 3 H), 3.56 (dd, 1 H, J )
10.4, 5.6 Hz), 3.71 (dd, 1 H, J ) 10.4, 3.8 Hz), 4.80 (ap tt, 1 H,
J ) 6.8, 6.1, 1.6 Hz), 4.84 (dd, 1 H, J ) 6.1,1.4 Hz), 6.72 (ap t,
1 H, J ) 1.5 Hz), 7.32 (d, 2 H, J ) 8.5 Hz), 7.76 (d, 2 H, J )
8.6 Hz). 13C NMR (75 MHz) δ -5.5, -5.5, 14.3, 18.2, 19.9, 21.6,
25.8 (3 C), 31.8, 64.9, 86.1, 89.4, 128.1 (2 C), 129.9 (2 C), 136.7,
140.6, 144.1, 144.9. IR (film): 3355, 2956, 2929, 2855, 1597,
1463, 1322, 1158, 836, 666 cm-1. MS (ES): 843 [2M + Na]+,
433 [M + Na]+ (100%), 411 [M + 1]+. Anal. Calcd for
C21H34O4SSi: C, 61.42; H, 8.35; S, 7.81. Found: C, 61.75; H,
into 28 in two steps following the route disclosed by
Evans et al.10c would result in an expedient route to
Kumausallene (11 steps). Thus, the one-pot Wittig de-
silylation protocol was tested on (-)-25d to afford an
excellent yield of enyne diol (+)-34 as an 86:14 E/Z
mixture. To our dismay, the treatment of (+)-34 with an
excess of TBCD in a very slow reaction gave a complex
mixture of allenes in which 28 and its bromoallene
epimer were just minor constituents. In view of these
disappointing results, and considering that the initial
steps of our route to the benzylated lactol, originally used
by Evans, occurred with fair yields and moderate selec-
tivities (Scheme 5 and Table 2, entries 6 and 7), we
decided to shift our efforts to Overman’s intermediate 27.
Thus, a Wittig reaction on lactol (-)-25b gave unsat-
urated ester (+)-35b as a separable 83:17 mixture of E/Z
isomers in excellent yield. Treatment of (+)-35b with a
(14) It should be mentioned that these last experiments were carried
out on a small scale and the yields were not determined.
10698 J. Org. Chem., Vol. 70, No. 26, 2005

Reductive Cleavage of Sulfur-Substituted Oxiranes
8.04; S, 7.60. Data for (+)-9a: Rf 0.16 (CH2Cl2). [R]20 +63.0
up in EtOAc, filtered through a cotton plug, and washed
carefully with additional EtOAc. The solution was concen-
trated under reduced pressure to produce a crude product that
was purified by chromatography on silica gel using the
appropriate mixture of eluents, to produce substantial amounts
(70-80%) of the nonpolar (PhSe)2 and the reaction products.
D
(c 1.55). 1H NMR (300 MHz) δ 0.02 (s, 3 H), 0.03 (s, 3 H), 0.63
(d, 3 H, J ) 6.7 Hz), 0.85 (s, 9 H), 0.96 (d, 3 H, J ) 7.0 Hz),
2.08 (hept d, 1H, J ) 6.9, 1.4 Hz), 2.43 (s, 3 H), 3.55 (dd, 1 H,
J ) 10.0, 6.9 Hz), 3.76 (dd, 1 H, J ) 10.1, 5.2 Hz), 4.78-4.83
(m, 2 H), 6.74 (s, 1 H), 7.33 (d, 2 H, J ) 9.0 Hz), 7.76 (d, 2 H,
J ) 8.3 Hz). 13C NMR (75 MHz) δ -5.4, 14.9, 18.3, 19.8, 21.7,
25.8 (3 C), 30.4, 64.9, 85.1, 89.1, 128.1 (2 C), 129.9 (2 C), 136.6,
140.9, 144.2, 144.9. IR (film): 3500, 2956, 2932, 2876, 1786,
1757, 1597, 1558, 1464, 1323, 1154, 667 cm-1. MS (ES): 843
[2M + Na]+, 433 [M + Na]+, 411 [M + 1]+ (100%). Anal. Calcd
for C21H34O4SSi: C, 61.42; H, 8.35; S, 7.81. Found: C, 61.57;
H, 8.42; S, 7.89.
Synthesis of (()-(2S,3S,5S)-2-(p-Fluorophenyl)-5-hy-
droxymethyltetrahydrofuran-3-ol, 16d, and (()-(2S,3R,5S)-
2-(p-Fluorophenyl)-5-hydroxymethyltetrahydrofuran-3-
ol, 17d. From a solution of epoxy sulfoxide 3d2 (52 mg, 0.15
mmol, 1.0 equiv) in 0.6 mL of EtOH and a suspension
generated from (PhSe)2 (28 mg, 0.09 mmol, 0.6 equiv) in 0.75
mL of EtOH and NaBH4 (29 mg, 0.75 mmol, 5 equiv), according
to the general procedure (1.5 h), an 86:14 mixture of diols 16d
and 17d was obtained. Purification by chromatography (20-
80% EtOAc/hexane) gave 16d (26 mg, 81%) and 17d (4 mg,
13%) as colorless oils. In a related experiment from a solution
of epoxy sulfone 4d4 (1.57 g, 4.3 mmol, 1.0 equiv) in 8.0 mL of
EtOH and a suspension generated from (PhSe)2 (2.01 g, 6.45
mmol, 1.5 equiv) in 26 mL of EtOH and NaBH4 (1.80 g, 47.3
mmol, 11 equiv), according to the general procedure (2.5 h),
an 85:15 mixture of diols 16d and 17d was obtained. Purifica-
tion by chromatography (40-80% EtOAc/hexane) gave 16d
(698 mg, 77%) and 17d (123 mg, 13%) as colorless oils. Data
General Procedure for Nucleophilic Epoxidation of
Vinyl Sulfoxides and Sulfones with tBuOOK. A two-
necked round-bottomed flask fitted with a tube in T for
entrance and exit of argon and a polyethylene stopper was
charged with anhydrous THF (5 mL/mmol) and 2-4 equiv of
oil-free KH (washed with hexane and dried). The mixture was
cooled to 0 °C, and then 2-4.5 equiv of tBuOOH (80% in
tBuOOtBu) was added. After being stirred at room tempera-
ture for 10 min, the resulting solution was cooled to 0 °C and
a solution of 1 equiv of the corresponding vinyl sulfoxide in
THF (7 mL/mmol), previously dried over 4 Å sieves, was added
dropwise. The reaction mixture was stirred at 0 °C until the
starting material disappearance, monitored by TLC. The
reaction was then quenched with a 10% solution of Na2S2O4
(4 mL/mmol), diluted with EtOAc (10 mL/mmol), and the
layers were separated. The aqueous layer was extracted with
EtOAc (3 times, 10 mL/mmol), and the combined organic
extracts were washed with a saturated solution of NaCl (4 mL/
mmol), dried over anhydrous MgSO4, filtered, and concentrated
under reduced pressure to give a crude product, which was
purified by column chromatography on silica gel, using a
gradient of mixtures of EtOAc/hexane or EtOAc/CH2Cl2.
Product ratios were determined by integration of well-resolved
1
for 16d: Rf 0.13 (50% EtOAc/CH2Cl2). H NMR (300 MHz) δ
2.04 (ddd, 1 H, J ) 14.0, 3.4, 0.9 Hz), 2.47 (ddd, 1 H, J ) 14.0,
9.4, 5.2 Hz), 2.90 (br s, 2 H), 3.60 (dd, 1 H, J ) 11.5, 3.0 Hz),
3.91 (dd, 1 H, J ) 11.5, 2.4 Hz), 4.21 (dd, 1 H, J ) 4.9, 2.9
Hz), 4.33 (ddd, 1 H, J ) 9.9, 5.7, 3.1 Hz), 4.80 (d, 1 H, J ) 2.7
Hz), 7.00-7.07 (m, 2 H), 7.32-7.37 (m, 2 H). 13C NMR (75
MHz) δ 36.7, 64.5, 73.0, 77.8, 85.1, 115.2 (2 C, JC-F ) 21.2
Hz), 128.5 (2 C, JC-F ) 8.1 Hz), 132.4, 162.3 (JC-F ) 245.8 Hz).
IR (film): 3350, 2938, 1606, 1512, 1445, 1294, 1222, 1157,
1118, 1059, 927, 877, 850, 808, 777 cm-1. MS (ES): 235 [M +
Na]+ (100%), 230, 213 [M + 1]+. Anal. Calcd for C11H13FO3:
C, 62.26; H, 6.17. Found: C, 62.45; H, 6.46. Data for 17d: Rf
1
signals in the H NMR of the crude reaction mixtures.
1
0.08 (50% EtOAc/CH2Cl2). H NMR (200 MHz) δ 1.92 (ddd, 1
Synthesis of (+)-(2S,3S,4S,5R)-2(tert-Butyldimethyl-
silyloxymethyl)-3,4-epoxy-5-isopropyl-4-(p-tolylsulfonyl)-
tetrahydrofuran, (+)-11a. From a solution of sulfonyl dihy-
drofuran (+)-9a (62 mg, 0.151 mmol, 1.0 equiv) in 1.1 mL of
THF, with tBuOOK (4.0 equiv, in THF), according to the
general procedure (0 °C, 30 min), sulfonyl oxirane (+)-11a was
obtained as a single isomer. Purification by chromatography
(2-10% EtOAc/hexane) gave (+)-11a (52 mg, 81%) as a
H, J ) 13.0, 6.8, 3.8 Hz), 1.94 (br s, 2 H), 2.09 (ddd, 1 H, J )
13.2, 8.4, 6.8 Hz), 3.64 (dd, 1 H, J ) 11.9, 5.3 Hz), 3.86 (dd, 1
H, J ) 11.9, 2.9 Hz), 4.20 (ap dt, 1 H, J ) 7.0, 4.1 Hz), 4.37
(m, 1 H), 4.69 (d, 1 H, J ) 4.4 Hz), 7.02 (ap t, 2 H, J ) 8.7 Hz),
7.32 (ap dd, 2 H, J ) 8.4, 5.3 Hz). 13C NMR (50 MHz) δ 36.3,
64.7, 78.6, 77.8, 87.5, 115.4 (2 C, JC-F ) 21.5 Hz), 127.6 (2 C,
JC-F ) 8.0 Hz), 136.0, 162.5 (JC-F ) 246.0 Hz). IR (film): 3368,
2928, 1606, 1511, 1225, 1045, 832, 757 cm-1. MS (ES): 235
[M + Na]+ (100%), 213 [M + 1]+. Anal. Calcd for C11H13FO3:
C, 62.26; H, 6.17. Found: C, 62.54; H, 6.33.
colorless oil. Data for (+)-11a: Rf 0.32 (100% CH2Cl2). [R]20
D
1
+14.5 (c 2.57). H NMR (300 MHz) δ 0.05 (s, 3 H), 0.06 (s, 3
H), 0.87 (s, 9 H), 1.01 (t, 6 H, J ) 6.9 Hz), 2.37 (ap hept d, 1
H, J ) 6.7, 3.2 Hz), 2.45 (s, 3 H), 3.61-3.73 (m, 2 H), 3.96 (ap
t, 1 H, J ) 5.1, 3.8 Hz), 4.01 (d, 1 H, J ) 3.4 Hz), 4.02 (s, 1 H),
7.35 (d, 2 H, J ) 8.0 Hz), 7.80 (d, 2 H, J ) 8.3 Hz). 13C NMR
(75 MHz) δ -5.5, -5.4, 17.8, 18.2, 21.3, 21.7, 25.8 (3 C), 29.5,
62.4, 66.9, 77.7, 78.0, 84.5, 129.3 (2 C), 129.8 (2 C), 133.6, 145.9.
IR (film): 3384, 2956, 2930, 2876, 2855, 1597, 1471, 1332,
1257, 1150, 1086, 838 cm-1. MS (ES): 427 [M + 1]+, 449 [M +
Na]+(100%). Anal. Calcd for C21H34O5SSi: C, 59.12; H, 8.03;
S, 7.52. Found: C, 59.36; H, 8.29; S, 7.73.
General Procedure for Reductive Cleavage of Sulfinyl
and Sulfonyl Oxiranes with (PhSe)2 and NaBH4. A two-
necked round-bottomed flask was charged with a solution
of (PhSe)2 (0.6-5.0 equiv) in EtOH (3.5-6 mL/mmol of di-
selenide), and argon was slowly bubbled through the solution
with a pipet. To this solution was added powdered NaBH4 (5-
14 equiv) portionwise at room temperature, and the mixture
was stirred until evolution of H2 subsided (ca. 1 h). In many
cases, AcOH (1 equiv) was added to ensure reproducibility and
faster reaction rate. Then, a solution of 1 equiv of sulfinyl or
sulfonyl oxirane in EtOH (2-5 mL/mmol) was added, and the
reaction was monitored by TLC until disappearance of the
starting material. The reaction was then quenched with a
saturated NH4Cl solution (3 drops/mmol), the solvent was
removed under reduced pressure, and the residue was taken
General Procedure for Cleavage of Sulfonyl Oxiranes
with MgBr2 and MgI2. A two-necked round-bottomed flask
fitted with a condenser was charged with a suspension of dry
Mg turnings (stored in an oven overnight) in anhydrous Et2O
(5 mL/mmol of Mg). Then, 1,2-diiodoethane or 1,2-dibromo-
ethane (1.2 equiv relative to Mg) was added at room temper-
ature, and the mixture was stirred for about 30 min or until
disappearance of Mg was observed, to produce a ca. 0.2 M
solution of MgX2 in Et2O that could be stored in a refrigerator
under argon for a few days. To a cold (0 °C) solution of epoxy
sulfone, previously azeotroped with anhydrous cyclohexane,
in anhydrous Et2O (5 mL/mmol of sulfone) was added MgX2
(5-9 equiv of the above solution), and the reaction mixture
was allowed to warm to room temperature and monitored by
TLC until disappearance of starting material was observed.
The reaction was then quenched with 5% NaHCO3 solution
(10 mL/mmol) and 1 M Na2S2O4 solution (10 mL/mmol), the
layers were separated, the aqueous layer was extracted with
EtOAc (3 × 10 mL/mmol), and the combined organic layers
were washed with a saturated solution of NaCl and dried over
anhydrous MgSO4. Removal of the drying agent and concen-
tration of the solution under reduced pressure gave a crude
product that was purified by column chromatography on silica
gel using the appropriate mixture of eluents.
J. Org. Chem, Vol. 70, No. 26, 2005 10699

Ferna´ndez de la Pradilla et al.
Synthesis of (+)-(2R,5R)-5-(tert-Butyldimethylsilyl-
oxymethyl)-2-isopropyl-3-oxotetrahydrofuran, (+)-22a.
From sulfonyl oxirane (+)-11a (31 mg, 0.073 mmol, 1.0 equiv)
in 0.4 mL of Et2O, with MgI2 (2.9 mL from a ca. 0.2 M solution
in Et2O, 8.0 equiv) according to the general procedure (0 °C to
room temperature, 46 h), and after chromatography (10-20%
1]+. Anal. Calcd for C14H28O3Si: C, 61.72; H, 10.36. Found:
C, 61.37; H, 10.53.
Acknowledgment. This research was supported by
DGICYT (BQU2003-02921) and CAM (GR/SAL/0823/
2004). We thank JANSSEN-CILAG for generous ad-
ditional support. We thank CAM, CSIC, and MEC for
doctoral fellowships to C.M. and A.C. and M.U.
CH2Cl2/hexane), ketone (+)-22a (16.3 mg, 82%) was obtained.
1
Data for (+)-22a: Rf 0.31 (CH2Cl2). [R]20 +64.4 (c 0.64). H
D
NMR (300 MHz) δ 0.07 (s, 6 H), 0.88 (s, 9 H), 0.89 (d, 3 H, J
) 6.1 Hz), 1.01 (d, 3 H, J ) 7.0 Hz), 2.03 (hept d, 1H, J ) 6.9,
3.9 Hz), 2.39 (s, 1 H), 2.41 (d, 1 H, J ) 2.3 Hz), 3.64 (d, 1 H,
J ) 3.9 Hz), 3.77 (dd, 1 H, J ) 11.1, 4.0 Hz), 3.85 (dd, 1 H, J
) 11.1, 3.8 Hz), 4.14-4.22 (m, 1 H, J ) 7.4, 3.8 Hz). 13C NMR
(75 MHz) δ -5.4, -5.3, 16.9, 18.4, 18.8, 25.9 (3 C), 30.3, 40.0,
64.4, 75.8, 85.6, 216.4. IR (film): 2926, 2855, 1741, 1710, 1436,
1269, 1162 cm-1. MS (ES): 295 [M + Na]+ (100%), 273 [M +
Supporting Information Available: Experimental
procedures and characterization for new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JO051459K
10700 J. Org. Chem., Vol. 70, No. 26, 2005