Practical synthesis of (±)-chlorovulone II

Article abstract of DOI:10.1021/jo972073f

We describe a total synthesis of (±)-chlorovulone II that is 10 steps shorter than the best alternative currently available (nine vs 19 steps). The key event of the synthesis is an aldol addition of the enolate of ethyl acetate into 4-cyelopentene-1,3-dione, a substance that has received little attention as an educt for prostanoid synthesis and for which little is known about carbonyl 1,2-addition with enolates. In addition, we provide chemical and stereochemical details of a route to a key intermediate toward the title compound that involves a carbonyl-ene reaction and a radical addition to an aldehyde carbonyl.

Full text of DOI:10.1021/jo972073f

1668
J . Org. Chem. 1998, 63, 1668-1675
P r a ctica l Syn th esis of (()-Ch lor ovu lon e II
Marco A. Ciufolini* and Shuren Zhu
Department of Chemistry, MS 60, Rice University, 6100 Main Street, Houston, Texas 77005-1892
Received November 11, 1997
We describe a total synthesis of (()-chlorovulone II that is 10 steps shorter than the best alternative
currently available (nine vs 19 steps). The key event of the synthesis is an aldol addition of the
enolate of ethyl acetate into 4-cyclopentene-1,3-dione, a substance that has received little attention
as an educt for prostanoid synthesis and for which little is known about carbonyl 1,2-addition with
enolates. In addition, we provide chemical and stereochemical details of a route to a key
intermediate toward the title compound that involves a carbonyl-ene reaction and a radical addition
to an aldehyde carbonyl.
In tr od u ction
Chlorovulone II, 1, is a marine prostanoid isolated from
the Okinawan soft coral, Clavularia viridis.1 This organ-
ism has been a rich source of bioactive and structurally
novel prostanoids, which have received much attention
owing to their unusual architecture and physiological
activity.2 Prostanoids from C. viridis were first reported
in 1982, when two groups simultaneously described
clavulone I, 2, and related substances (Figure 1).3 Vari-
ous related compounds were subsequently isolated from
the same source.4
F igu r e 1.
Clavulone I and congeners display antiinflammatory
activity in a fertilized chicken egg assay (30 µg/mL).3a
More significantly, these substances have shown remark-
able cytotoxicity and antiproliferative activity against
several transformed cell lines.5 Antitumor activity is
especially pronounced in halogenated congeners of cla-
vulones, and chlorovulone II is one of the most active
members of the group. The activity of 1 against HL-60
cells corresponds to an IC50 of 30 nM (10 ng/mL), about
1 order of magnitude greater than that of clavulone I on
a molar basis.6 Preliminary studies have also defined
some requirements for maximum activity. In particular,
a C-10-11 olefin or epoxide is essential; a halogen atom
at C-10 increases activity (Cl ) F > Br ) I > H), as does
a C-5,7 diene; the hydroxyl group at C-12, the sole
stereogenic carbon atom in the molecule, is required for
highest activity, but remarkably, the antipodes of 1
possess identical potency.6a
(1) Iguchi, K.; Kaneta, S.; Mori, K.; Yamada, Y.; Honda, A.; Mori,
Y. Tetrahedron Lett. 1985, 26, 5787.
The mode of cytotoxic action of chlorovulone and
related substances remains poorly understood. Of course,
chlorovulone could interfere with the biological functions
of prostaglandins, due to its great similarity to the latter
family of compounds. Alternatively, it may function as
a relatively nonselective alkylating agent (it is a reactive
Michael acceptor) that could damage proteins, glu-
tathione, and other key factors involved in cellular
function, including, perhaps, nucleic acids. The latter
mode of action could explain why the stereochemistry of
the C-12 OH is not important for activity. As shown
below, conjugate addition of a bionucleophile (BioNu,
Scheme 1) could promote expulsion of that OH group,
preparing the molecule for subsequent Michael-type
events and causing it to behave as a multiple alkylating
agent; possibly a cross-linking one. It must be stressed,
however, that at this time the mechanisms of Scheme 1
belong to the realm of conjecture.
(2) Review: Gerwick, W. H. Chem. Rev. 1993, 93, 1807 and
references cited therein. See also: Gribble, G. W. Fort. Chem. Org.
Naturst. 1996, 68, 110.
(3) (a) Kikuchi, H.; Tsukitani, Y,; Iguchi, K.; Yamada, Y. Tetrahedron
Lett. 1982, 23, 5171. (b) Kobayashi, M.; Yasuzawa, T.; Yoshihara, M.;
Akutsu, H.; Kyogoku, Y.; Kitagawa, I. Tetrahedron Lett. 1982, 23, 5331.
The absolute stereochemistry of these compounds was established by
degradative and chiroptical techniques [(c) Kobayashi, M.; Yasuzawa,
T.; Yoshihara, M.; Son, B. W.; Kyogoku, Y.; Kitagawa, I. Chem. Pharm.
Bull. 1983, 31, 1440. (d) Kikuchi, H.; Tsukitani, Y.; Iguchi, K.; Yamada,
Y. Tetrahedron Lett. 1983, 24, 1549] and confirmed by synthesis. (e)
Hashimoto, S.; Arai, Y.; Hamanaka, N. Tetrahedron Lett. 1985, 26,
2679. (f) Corey, E. J .; Mehrotra, M. M. J . Am. Chem. Soc. 1984, 106,
3384. (g) Nagaoka, H.; Miyakoshi, T.; Yamada, Y. Tetrahedron Lett.
1984, 25, 3621. (h) Shibasaki, N.; Ogawa, Y. Tetrahedron Lett. 1985,
26, 3841. (i) Zhu, J .; Yang, J . Y.; Klunder, A. J . H.; Liu, Z. Y.;
Zwanenburg, B. Tetrahedron 1995, 51, 5847. (j) Takeda, K.; Nakajima,
A.; Yoshii, E. Synlett 1997, 255.
(4) (a) Iguchi, K.; Yamada, Y.; Kikuchi, H.; Tsukitani, Y. Tetrahedron
Lett. 1983, 24, 4433. (b) Iguchi, K.; Kaneta, S.; Mori, K.; Yamada, Y.;
Honda, A.; Mori, Y. J . Chem. Soc., Chem. Commun. 1986, 981. (c)
Iguchi, K.; Kaneta, S.; Mori, K.; Yamada, Y. Chem. Pharm. Bull. 1987,
35, 4375.
(5) (a) Honda, A.; Yamamoto, Y.; Mori, Y.; Yamada, Y.; Kikuchi, H.
Biochem. Biophys. Res. Commun. 1985, 130, 515. (b) Fukushima, M.;
Kato, T.; Yamada, Y.; Kitagawa, I.; Kurozumi, S.; Scheuer, P. J . Proc.
Am. Assoc. Cancer Res. 1985, 26, 980. In vivo testing of some
Clavularia metabolites has furnished promising results. Thus, clavu-
lone II gave a T/C ) 151 at 10 mg kg-1 day-1 for 5 days and a T/C of
160 at 20 mg kg-1 day-1. Additional biological effects are described
in: Honda, A.; Hong, S.; Yamada, Y.; Mori, Y. Res. Commun. Chem.
Pathol. Pharmacol. 1991, 72, 363.
Consensus exists that arachidonic acid is the progeni-
tor of Clavularia prostanoids. On the other hand, several
(6) (a) Honda, A.; Mori, Y.; Iguchi, K.; Yamada, Y. Mol. Pharmacol.
1987, 32, 530. (b) Fukushima, M.; Kato, T. Adv. Prostaglandin
Thromboxane Leukotriene Res. 1985, 15, 415. (c) Honda, A.; Mori, Y.;
Iguchi, K.; Yamada, Y. Prostaglandins 1988, 36, 621.
S0022-3263(97)02073-2 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/11/1998

Synthesis of (()-Chlorovulone II
J . Org. Chem., Vol. 63, No. 5, 1998 1669
Sch em e 1
Sch em e 3a
a
Key: (a) t-BuOK, 0 °C, 71%; (b) pyridine, cat. DMAP, rt, 91%.
route that should permit site-specific labeling of various
ring atoms in the molecule. Chemical highlights of the
work reported herein include the use of commercial
4-cyclopentene-1,3-dione as a building block for marine
prostanoids; the use of our carbonyl-ene reaction as a
key step toward intermediate 3, and the investigation of
chemical and stereochemical aspects of radical additions
to aldehyde carbonyls.
Sch em e 2
Ca r bon yl-En e Rou te to In ter m ed ia te 3
Enone 3 is the key intermediate for our synthesis
(Scheme 2). The presence of an ethyl ester, instead of
Yamada’s tert-butyl ester,8 should allow direct DIBAL
reduction to an aldehyde, as a prelude to installation of
the two side chains as described earlier.8 Our strategy
for 3 rests on our carbonyl ene-like reaction,9 as apparent
from Scheme 2. The key enone 3 could result through
free-radical cyclization (cf. 4) of a suitable derivative of
the ene adduct of 2-methoxypropene (“2-MP”) with an
aldehyde of generic structure 5. Plausible substrates for
the planned transformations were aldehydes 6-8 or even
a glyoxal analog, 9. Whereas ene reaction of 6-8 would
be followed by suitable activation of the vinyl ether and
radical cyclization into an olefinic π system, use of glyoxal
equivalent 8 would ultimately entail a radical-carbonyl
cyclization.10,11
The preparation of aldehyde 6 was troublesome. Ethyl
R,R-diethoxyacetate and ethyl acetate condensed easily
in the presence of t-BuOK to give â-keto ester 11,
treatment of which with AcCl and pyridine in the
presence of DMAP produced 12 (Scheme 3). However,
it was not possible to promote selective hydrolysis of the
ketal in 12. Mild acidic conditions released the acetate
to give back 11, while stronger acid damaged the mol-
ecule. Ethyl 4-oxocrotonate (7)12 was advanced easily to
15 by carbonyl ene reaction with 2-MP, methoxybromi-
nation of the resulting 13 (note: MIP ) 2-methoxyiso-
propyl) with NBS/MeOH and radical cyclization (77%
overall, Scheme 4). Unfortunately, introduction of the
questions regarding details of the biosynthesis of these
compounds remain unanswered, despite much activity
in this area.7 Unresolved issues, both in biosynthesis
and, more importantly, in pharmacology, could probably
be explored if a stable supply of materials were to be
secured. It appears that the sole practical solution for
this problem resides in a total synthesis, because the
natural compounds appear to be quite rare. This seems
to be especially the case for the highly bioactive chloro-
vulone II, which constitutes only 0.002% of the freeze-
dried weight of C. viridis.1-4 Specifically labeled material
may also be useful, or even necessary, to deal effectively
with several of the above questions. Again, synthesis
offers the sole practical avenue to such compounds: a
biosynthetic route involving labeled precursors is unat-
tractive, given the paucity of 1 in its natural source.
Interestingly, only one total synthesis of 1, by Yamada
and collaborators,8 has been recorded in the literature,
while several syntheses of clavulones are known.3e-g The
Yamada synthesis leads to (-)-chlorovulone II, the
antipode of the natural product, in 19 linear steps and
in ca. 6% yield from L-tartaric ester, by way of the tert-
butyl ester analog of intermediate 3 (Scheme 2). How-
ever, there may not be a pressing need for an enantiose-
lective synthesis of 1, since the two antipodes of the
molecules have identical potency. In this paper, we
describe a practical, efficient total synthesis of (()-1 that
requires nine linear steps from commercial materials and
that proceeds in 21% overall yield. We also disclose a
(9) Deaton, M. V.; Ciufolini, M. A. Tetrahedron Lett. 1993, 34, 2409.
(10) Flies, F.; Lalande, R.; Maillard, B. Tetrahedron Lett. 1976, 17,
439.
(11) (a) Tsang, R.; Dickson, J .K., J r.; Pak, H.; Walton, R.; Fraser-
Reid, B. J . Am. Chem. Soc. 1987, 109, 3484. (b) Walton, R.; Fraser-
Reid, B. J . Am. Chem. Soc. 1991, 113, 5791. Physical organic studies:
(c) Beckwith, A. L. J .; Raner, K. D. J . Org. Chem. 1992, 57, 4954.
Applications: (d) Grissom, J . W.; Klingberg, D. J . Org. Chem. 1993,
58, 6558. Related processes: (e) Naito, T.; Tajiri, K.; Harimoto, T.;
Ninomiya, I.; Kiguchi, T. Tetrahedron Lett. 1994, 35, 2205. (f) Enholm,
E. J .; Xie, Y.; Abboud, K. A. J . Org. Chem. 1995, 60, 1112. (g) Hays,
D. S.; Fu, G. J . Am. Chem. Soc. 1995, 117, 7283. (h) Yuasa, Y.; Ando,
J .; Shibuya, S. J . Chem. Soc., Perkin Trans. 1 1996, 793.
(12) Available in a low 20% yield by a laborious SeO2 oxidation of
ethyl crotonate: Rambaud, R.; Vessiere, M. Bull. Soc. Chim. Fr. 1961,
1567. A similar oxidation of ethyl 3-acetoxycrotonate (from ethyl
acetoacetate and Ac2O/pyridine) as a route to 6 was unsuccessful.
(7) Cf. (a) Corey, E. J .; Matsuda, S. P. T. Tetrahedron Lett. 1987,
28, 4247. (b) Corey, E. J .; D’Alarcao, M.; Matsuda, S. P. T.; Lansbury,
P. T., J r.; Yamada, Y. J . Am. Chem. Soc. 1987, 109, 289. (c) Brash, A.
R. J . Am. Chem. Soc. 1989, 111. 1891. (d) Brash, A. R.; Baertschi, S.
W.; Ingram, C. D.; Harris, T. M. Proc. Natl. Acad. Sci. U.S.A. 1988,
85, 3382.
(8) Nagaoka, H.; Iguchi, K.; Miyakoshi, T.; Yamada, N.; Yamada,
Y. Tetrahedron Lett. 1986, 27, 223.

1670 J . Org. Chem., Vol. 63, No. 5, 1998
Ciufolini and Zhu
Sch em e 4a
group is attributable to the generation of formic acid
through over oxidation of formaldehyde, coproduced with
20 and arising from the terminal methylene portion of
the olefin. The problem persisted when the ozonolysis
reaction was run with suspended NaHCO3. This forced
upon us an awkward protection-deprotection sequence
prior to ozonolysis (cf. 20 f 22, Scheme 5). Triethylsilyl
ether 24b ultimately emerged as a substrate for carbon-
yl-radical cyclization. Parallel studies defined a more
direct route to an alternative substrate starting with
cinnamaldehyde, 17. Thus, methoxybromination of ene
adduct 19 followed by ozonolysis furnished aldehyde 24a ,
which was readily purified by chromatography without
loss of the MIP group.
a
Key: (a) 10 equiv, 0.5 mol % of 1:1 Yb(fod)3-AcOH complex,
rt, 90%; (b) MeOH, CH2Cl2, K2CO3, -78 °C, 99%; (c) AIBN, PhH,
reflux, 85%.
Radical cyclization of 24a and 24b with n-Bu3SnH and
AIBN in refluxing benzene revealed new and interesting
aspects of this poorly explored reaction. As shown in
Scheme 6, compound 24a produced a mixture of cis
alcohol 26a (56% chromatographed, major product), trans
alcohol 27a (6% chrom), and aldehyde 28a (15% chrom).
By contrast, compound 24b furnished a mixture of cis
alcohol 26b (31% chrom) and aldehyde 28b (27% chrom).
Trans alcohol 27b escaped detection, possibly due to the
small quantities obtained.17 The stereochemistry of 26
and 27 was assigned on the basis of cyclic acetonide
formation (Scheme 7). Thus, in all cases radical cycliza-
tion was appreciably diastereoselective for the cis isomer
of the product.
Sch em e 5a
While a fair amount of stereochemical information is
already available regarding radical additions to olefins,18
the corresponding reactions of aldehydes have not been
studied in detail.11 It is significant that the observed
selectivity is consistent with a Felkin-Ahn-like mode of
addition19 of the radical species to the carbonyl π system
(Scheme 8). This may reflect the fact that the major
frontier orbital interaction in the current process is
probably one between the carbonyl π* (LUMO) and the
radical SOMO. Felkin-Ahn-type stereoelectronic effects
that reduce the energy of the LUMO, thereby narrowing
the SOMO-LUMO energy gap, should accelerate the rate
of radical addition, much as they accelerate nucleophilic
additions. It should be emphasized that the applicability
of the Felkin-Ahn model to radical additions to carbonyls
remains to be researched in greater detail.
Aldehyde 28 is believed to result from fragmentation
of presumed oxy radical intermediates 32/33 (cf. Scheme
8).11a,b This fragmentation seems to occur at a rate
comparable to that of bimolecular H-atom transfer from
the tin hydride to the oxy radicals. Attempts to suppress
fragmentation by varying the concentration of reactants,
or by providing a large amount of n-Bu3SnH in the hope
of favoring bimolecular H-transfer, were unsuccessful.
Whereas this may not be surprising, since intramolecular
processes are often much faster than bimolecular ones,
it is also noteworthy, in that it signifies that the ratio of
cyclization product to aldehyde is an intrinsic function
Key: (a) 10 equiv, 0.5 mol % of 1:1 Yb(fod)3-AcOH complex, rt,
80% (R ) H), 100% (R ) Ph); (b) MeOH, CH2Cl2, K2CO3, -78 °C,
97% (R ) H), 87% (R ) Ph); (c) CSA, MeOH, 84%; (d) Et3SiCl,
Et3N, CH2Cl2, 95%; (e) MeOH, CH2Cl2, NaHCO3, -78 °C; then
Me2S, 78% (R ) H), 80% (R ) Ph).
requisite tertiary OH unit at this stage would have taken
numerous steps. Finally, reactive ynal 8, available by
MnO2 oxidation13 of 1-acetoxy-2-butyn-4-ol,14 was a poor
substrate for our ene reaction.15
We thus turned our attention to a suitable glyoxal
equivalent. Aldehydes containing ketal functionalities
may react poorly in the ene process; furthermore, the
multitude of ketals present in various late intermediates
would have complicated the issue of selective hydrolysis.
This ruled out the use of a true glyoxal monoketal in our
synthesis. Ozonolytic cleavage of an olefin constituted
an appealing way to reveal the requisite aldehyde func-
tion. Accordingly, we focused on the ene reaction of 2-MP
with enals. The ene adduct of acrolein, 18,16 underwent
smooth methoxybromination, but ozonolysis of the result-
ing 20 occurred with concomitant loss of the MIP group
(Scheme 5). The resulting R-hydroxy aldehyde was
isolable, but it was difficult to handle. Release of the MIP
(13) Review: Fatiadi, A. J . Synthesis 1976, 65.
(14) Reaction of 2-butyne-1,4-diol with 1.0 equiv of Ac2O in pyridine
gave the monoacetate in 50% chromatographed yield, plus diacetate
(25%) and unreacted starting diol (25%).
(17) The stereochemistry of 26-27 was assigned on the basis of
cyclic acetonide formation as described in: Ciufolini, M. A.; Zhu, S.
Tetrahedron Lett. 1997, 38, in press.
(15) A catalyst for enantioselective ene-like reactions similar to ours
has been described by: Carreira, E.; Lee, W.; Singer, R. A. J . Am.
Chem. Soc. 1995, 117, 3649. The Carreira catalyst is reported to
produce excellent results with ynals, but not with enals. By contrast,
our catalytic system works well with enals, but ynals are less
successful.
(16) This ene product was obtained in 80% yield, but it was
contaminated with small amounts of byproducts of unknown structure.
Purification to homogeineity was difficult.
(18) Cf., e.g.: (a) Morikawa, T.; Washio, Y.; Harada, S.; Hanai, R.;
Kayashita, T.; Nemoto, H.; Shiro, M.; Taguchi, T. J . Chem. Soc., Perkin
Trans. 1 1995, 271. (b) Lesueur, C.; Nauguier, R.; Bertrand, M. P.;
Hoffmann, P.; DeMesmaecker, A. D. Tetrahedron 1994, 50, 767. (c)
Curran, D. P.; Sun, S. Tetrahedron Lett. 1993, 34, 6181 and references
cited therein. See also ref 19, pp 935 ff.
(19) For a detailed discussion of the Felkin-Anh model, see: Eliel,
E. L.; Wilen, S. H. Stereochemistry of Organic Compounds; J ohn Wiley
& Sons: New York, 1994; pp 875-880.

Synthesis of (()-Chlorovulone II
J . Org. Chem., Vol. 63, No. 5, 1998 1671
Sch em e 6
Sch em e 7a
Key: (a) 2-methoxypropene, cat. PPTS, CH2Cl2, rt; (b) TBAF, THF, rt.
Sch em e 8
relative electronegativity of C (more electronegative) vs
Si (less electronegative) atoms already suggest that an
O-Si bond will be more highly polarized than an O-C
bond; therefore, a greater extent of partial negative
charge will be present on the silyl ether oxygen relative
to the MIP ether oxygen. Fragmentation should thus be
favored in the silyl ether series. The above hypothesis
was confirmed by the simple calculation summarized in
Scheme 9.21 Single point UHF-MNDO on MM+ struc-
tures of cis oxy radical intermediate 32 suggest that the
electron density on the protected oxygen doubles in going
from MIP acetals to silyl ethers. Thus, the C-C bond
about to rupture is weaker in the silyl substrate, and a
greater proportion of fragmentation product is observed.
Notice also that the nature of the O-blocking group has
no effect on the electronic properties of the oxy radical
center. Identical electronic properties are are calculated
for trans diastereomer 33. The possibility that the rate
of fragmentation of oxy radical intermediates may be
sterically accelerated, in the sense that more sterically
compressed molecules may fragment more readily, was
briefly explored by estimating the steric energy content
(MM+ force field) of structures 34 and 35, computation-
ally better tractable congeners of 32-33. As seen in
Figure 2, the MIP series of compounds is more sterically
encumbered than the silyl series,22 while cis diastereo-
mers are slightly more energetic than the trans isomers.
If steric acceleration of radical fragmentation were opera-
tive, one would expect a greater extent of fragmentation
in the MIP series and among the cis disatereomers of the
presumed radical intermediates. This is the exact op-
posite of what is observed experimentally. At this time,
therefore, we do not believe that steric effects play a
significant role here.
Sch em e 9
of the nature of the O-protecting group present at the
R-position of the aldehyde. The above observation may
be rationalized as follows. Consider the electronic makeup
of the C-C bond that undergoes fragmentation in
presumed radical intermediates 32 and 33. This bond
is highlighted in Scheme 9. Hyperconjugative dispersal
of nonbonding electrons from the protected O atom into
the C-C antibonding σ orbital, σ*C-C, should facilitate
C-C bond fragmentation, since it would lower the bond
order between the two C atoms and thereby weaken the
bond. Consequently, increasing negative charge concen-
tration on the protected oxygen would be expected to
accelerate the rate of bond breaking. A similar argument
has been put forth to rationalize the greatly enhanced
rate of oxy-Cope reactions, and anionic variants thereof,
relative to plain Cope processes.20 Considerations of
The conversion of the mixture of alcohols 26a /27a to
enone 3 was effected as shown in Scheme 10. Ley-type
oxidation with N-methylmorpholine N-oxide (NMO) and
(20) Cf., e.g., (a) Evans, D. A.; Golob, A. M. J . Am. Chem. Soc. 1975,
97, 4765. (b) Evans, D. A.; Baillargeon, D. J .; Nelson, J . V. J . Am. Chem.
Soc. 1978, 100, 2242. (c) Koreeda, M; Luengo, J . I. J . Am. Chem. Soc.
1985, 107, 5572.
(21) Calculation were carried out with the Hyperchem 4.0 package,
available from Hypercube, Inc., Ontario, Canada.
(22) Recall that an Si-O bond is rather long compared to a C-O
bond: approximately 1.64 Å vs 1.43 Å. Because steric interactions are
inversely dependent on the 6th power of the distance between groups,
the steric interactions between the silyl subunit and the rest of the
molecule are not as severe as those created by an MIP unit.

1672 J . Org. Chem., Vol. 63, No. 5, 1998
Ciufolini and Zhu
Sch em e 11
stereochemical course suggests a prevalent chelation-
controlled addition of the enolate into the R-alkoxy
ketone.24 The poor diastereoselectivity observed in this
reaction is probably a consequence of the generally weak
ability of Li+ organometallics to engage in chelation-
controlled carbonyl additions.25 On the other hand,
acetals (cf. the MIP group in 36) are well recognized for
their strong coordinating/directing ability in metalation
reactions by organolithium agents.26 This directing abil-
ity may partially overcome inherent inadequacies of
lithium enolates.
Removal of the MIP group in the mixture of 37/38 gave
diol 40, whose secondary hydroxyl unit was activated by
conversion to mesylate 41 in preparation for acetal
hydrolysis. This step was performed with 80% aqueous
AcOH under conditions that induced simultaneous â-
elimination of mesylate to afford enone 3. It should be
noted that direct acid hydrolysis of 37/38 or of 40 gave
enone 42 via â- elimination of the sensitive tertiary OH.
Creation of a mesylate was thus necessary to ensure the
correct regioselectivity of enone formation.
F igu r e 2. Steric energy content (MM+, kcal/mol) of the
conformers of cis and trans diastereomers of radical intermedi-
ates 34 and 35. For the sake of clarity, hydrogen atoms and
lone pairs are not shown.
Sch em e 10a
A P r a ctica l Syn th esis of (()-Ch lor ovu lon e II
An even more efficient alternative for the assembly of
3 may be envisioned on the basis of a new strategic idea:
a 1,2-addition of the enolate of ethyl acetate into 4-cy-
clopentene-1,3-dione, 43. This commercially available,
if moderately expensive ($14/g or $1.4/mmol), building
block seems to have received very little attention as an
educt for prostanoid synthesis.27 Indeed, the compound
has found primary use in Diels-Alder reactions,28 con-
jugate addition,29 and Knoevenagel chemistry,30 but we
were unable to find a recorded example of enolate
addition to 43. A potential difficulty with such an
operation might be base-promoted enolization of the
â-diketone (Scheme 11). However, the resultant anion
44 displays antiaromatic character by virtue of its
cyclopentadienone nature. This should both disfavor
enolization and promote 1,2-addition of excess enolate to
the remaining, reactive carbonyl. Dianion 45 thus
formed would still produce enone 3 upon acidic workup,
Key: (a) TPAP/NMO, CH2Cl2, 100%; (b) LDA, EtOAc, -78 °C,
81%; (c) 10% AcOH/H2O 92%; (d) MsCl, Et3N, CH2Cl2, 0 °C; (e)
80% AcOH/H2O, 75% of 3d ,e.
(24) Cf., e.g.: Still, W. C.; Schneider, J . A. J . Org. Chem. 1980, 45,
3375.
(25) Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1984, 23, 556.
(26) Gschwend, H.; Rodriguez, H. R. Org. React. 1979, 26, 1.
(27) Cf., e.g.: Tolstikov, G. A; Miftakhov, M. S.; Danilova, N. A.;
Vel’der, Y. L.; Shirikhin, L. V. Zh. Org. Khim. 1989, 25, 438.
(28) Cf. (a) Lu, H.-J .; Llinas-Brunet, M. Can. J . Chem. 1984, 62,
1747. (b) Cardinale, G.; Laan, J . A. M.; Ward, J . P. Recl. Trav. Chim.
Pays-Bas 1987, 106, 62. (c) Paquette, L. A.; Vannucci, C.; Rogers, R.
D. J . Am. Chem. Soc. 1989, 111, 5792. (d) Houwen-Claassen, A. A.
M.; Klunder, A. J . H.; Kooy, M. G; Steffann, J .; Zwanenburg, B.
Tetrahedron 1989, 45, 7109.
catalytic tetrapropylammonium perruthenate (TPAP)23
furnished ketone 36, which combined with the enolate
of EtOAc leading to a roughly 2.5:1 mixture of diaster-
eomers 37 (major) and 38 (minor). The stereochemical
assignment of these compounds rests on the facile lac-
tonization of minor diastereomer 38 to give 39, whereas
the major product resisted lactonization. The observed
(29) Cf. Lepitskaya, M. A.; Manukina, T. A.; Pivnitskii, K. K. Zh.
Org. Khim. 1986, 22, 872.
(23) Ley, S. V.; Norman, J .; Griffith, W. P.; Marsden, S. P. Synthesis
1994, 639.
(30) Cf. Takahashi, K.; Namekata, N.; Takase, K.; Takeuchi, A.
Tetrahedron Lett. 1987, 28, 5683.

Synthesis of (()-Chlorovulone II
J . Org. Chem., Vol. 63, No. 5, 1998 1673
Sch em e 12a
Sch em e 13a
Key: (a) from EtOAc + LDA, THF, -78 °C, 86%; (b) Cl2 gas,
CCl4, then Et3N, 77%; (c) DHP, CSA, CH2Cl2, rt, 100%.
rendering 43 still serviceable for the ultimate production
of the desired subgoal.
Treatment of 43 with 1.0 equiv of ethyl acetate enolate
cleanly furnished enone 3 in 86% yield (Scheme 12).
Excess enolate must be avoided in this reaction, in order
to suppress formation of the double addition product, 46.
This material, which seemingly was obtained as a single
diastereomer within the limits of 1H NMR spectroscopy,
represented as much as 30% of the product when 43 was
treated with 2 equiv of the enolate. The stereochemistry
of 46 is tentatively assigned as cis, by analogy with
precedent.8 The observed formation of 46 rules against
the intervention of anion 44, ergo enolization of 43, since
no further nucleophilic addition would be possible to
dianion 45.
Regioselective introduction of a chlorine atom into 3
was accomplished by standard treatment with excess Cl2
gas in CCl4, followed by Et3N (Scheme 12). Subsequent
experiments indicated that it is possible to advance
chloroenone 47 into the synthesis without protection of
the tertiary alcohol. However, several technical problems
arose as a consequence, details of which will be discussed
shortly. This necessitated protection of the tertiary OH
group as a THP ether to give 48. A minor problem with
this protecting group was the introduction of a new
stereogenic center (the “anomeric” site) without stereo-
control. Although diastereomers were formed as a con-
sequence, no significant difficulties were encountered
during scrutiny of the NMR spectra of this intermediate
and subsequent derivatives thereof.
Reduction of 48 with DIBAL furnished aldehyde 50, a
good substrate for Wittig installation of the lower side
chain. However, the cyclopentenone moiety underwent
1,4-reduction to a varying extent under such conditions,
rendering the yield of this step unsatisfactory. This
problem was corrected by Luche reduction31 of 48 to
allylic alcohol 49, as a 1:1 mixture of diastereomers,
followed by DIBAL reduction to aldehyde 50. Fortu-
nately, no protection of the secondary allylic alcohol in
50 was necessary during subsequent manipulations. By
contrast, it was difficult to reduce compound 47 to an
analog of aldehyde 50 lacking a THP protecting group
at the tertiary alcohol. Severe difficulties were also
observed upon reoxidation of the secondary alcohol in an
unprotected variant of compound 51. These hurdles
vanished when the labile tertiary hydroxy unit was
blocked as a THP ether.
Key: (a) NaBH4, CeCl3‚7H2O, MeOH, 0 °C, 88%; (b) DIBAL,
THF, -78 °C, 79%; (c) Ph3P+(CH2)5CH3Br-, BuLi, THF, DMPU,
-40 °C, 80%; (d) PDC, DMF, 83%; (e) LDA, 53, THF, 82%; (f)
TsOH, MeOH rt, 88%; 21% overall from 43 over nine steps.
phorane in THF/N,N’-dimethylperhydropyrimidin-2-one
(“DMPU”) gave the Z-olefin 51,32 which was oxidized to
enone 52 by PDC in DMF.33 The use of DMPU was
essential for maximum efficiency and Z selectivity.
Condensation of the lithium enolate of 52 with 1.0 equiv
of known8 aldehydo ester 5334 under conditions substan-
tially similar to those described by Yamada8 afforded
chlorovulone II THP ether (54) as a single geometric
isomer in 82% yield. Finally, removal of the THP group
in 54 with TsOH in MeOH provided totally synthetic 1
(Scheme 13).
In summary, quantities of chlorovulone II are now
efficiently available in nine steps from commercial 4-cy-
clopentene-1,3-dione and in an overall yield of 21%.
Other clavulones should be accessible by minor modifica-
tions of our strategy. Ring-labeled anologues of 1 that
might be needed for biological studies are available
through a combination of our carbonyl-ene-like reaction
and aldehyde-radical cyclization. This latter step has
been shown to proceed with significant stereoselectivity.
Mechanistic and synthetic studies of these radical pro-
cesses continue in our laboratory.
Exp er im en ta l Section 35
1-P h en yl-3-[(2-m eth oxy-2-p r op yl)oxy]-5-m eth oxyh exa -
1,5-d ien e (19). Silica gel (1 g) was added to a solution of
cinnamaldehyde (6.74 g, 51 mmol), 2-methoxypropene (30 mL,
6 equiv), CH2Cl2 (60 mL), AcOH (10 mL, 2 mol %), and Yb(fod)3
(100 mg, 2 mol %). The mixture was stirred at rt for 18 h,
(32) For a review of the Wittig reaction see: Schlosser, M. Top.
Stereochem. 1970, 5, 1.
(33) Corey, E. J .; Schmidt, G. Tetrahedron Lett. 1979, 20, 399.
(34) Reference 8 describes a route to 53 that involves three steps
from δ-valerolactone: (i) methanolysis, (ii) Swern oxidation of the
hydroxy ester, and (iii) Wittig reaction. A more practical two-step
alternative involving a Schreiber ozonolysis (Claus, R. E., Schreiber,
S. L. Organic Syntheses; Wiley: New York, 1990; Collect. Vol. VII, p
168. Schreiber, S. L.; Claus, R. E.; Regan, J . Tetrahedron Lett. 1982,
23, 3867) is detailed in the Experimental Section.
Conversion of 50 to chlorovulone II proceeded unevent-
fully. Treatment of with n-hexylidenetriphenylphos-
(31) Luche, J . L. J . Am. Chem. Soc. 1978, 100, 2226.

1674 J . Org. Chem., Vol. 63, No. 5, 1998
Ciufolini and Zhu
and then it was diluted with ether (200 mL), washed (satu-
rated aqueous NaHCO3), dried (Na2SO4), and concentrated to
give 14.1 g (100% yield) of 19 as a light yellow oil. 1H (C6D6):
7.28-7.01 (5H), 6.53 (d, 1H, J ) 16.0), 6.29 (dd, 1H, J ) 16.0,
7.1), 4.82 (m, 1H), 3.97 (d, 1H, J ) 1.7), 3.84 (d, 1H, J ) 1.7),
3.18 (s, 3H), 3.15 (s, 3H), 2.70 (dd, 1H, J ) 13.5, 6.2), 2.44 (dd,
1H, J ) 13.5, 7.0 ), 1.39 (s, 3H), 1.36 (s, 3H). 13C (C6D6): 161.3,
138.1, 133.3, 130.3, 129.1, 127.9, 127.1, 101.4, 83.7, 70.6, 54.7,
49.5, 44.2, 26.5, 25.8. IR: 1663, 1615, 1504, 1234, 1213, 1157,
1067, 1032. MS: 277 (M + H)+, 245, 205, 187, 173, 73.
HRMS: calcd for C17H25O3 (M + H)+ 277.1804, found 277.1800.
1-P h en yl-3-[(2-m eth oxy-2-p r op yl)oxy]-5,5-d im eth oxy-
6-br om oh ex-1-en e (21). A solution of 19 (10 g, 36.2 mmol)
in CH2Cl2 (30 mL) was added into a mixture of NBS (6.5 g,
36.5 mmol), MeOH (20 mL), CH2Cl2 (120 mL), and powdered
NaHCO3 (0.5 g) at -78 °C. After 30 min, the mixture was
washed (saturated aqueous NaHCO3), dried (Na2SO4), and
concentrated to give 12.2 g (87%) of 21, as a colorless oil. 1H
(C6D6): 7.41-7.21 (5H), 6.52 (d, 1H, J ) 16.0), 6.30 (dd, 1H, J
) 16.0, 8.1), 4.45 (m, 1H), 3.55 (s, 2H), 3.25 (s,2H), 3.23 (s,
3H), 3.21 (s, 3H), 2.25 (dd, 1H, J ) 15.1, 5.9), 2.15 (dd, 1H, J
) 15.1, 7.0), 1.44 (s, 3H), 1.37 (s, 3H). 13C (C6D6): 137.1, 132.3,
130.3, 128.7, 127.7, 126.6, 101.3, 100.6, 69.1, 49.6, 48.6, 48.4,
38.6, 32.7, 26.0, 25.4. IR: 1254, 1212, 1150, 1067, 1019. MS:
m/z 357 [(M + H)+ - MeOH, 81Br], 355 [(M + H)+ - MeOH,
79Br], 299, 297, 169, 167, 73. HRMS: calcd for C17H24O379Br
[(M + H)+ - MeOH] 355.0909, found 355.0906.
2-[(2-Meth oxy-2-pr opyl)oxy]-4,4-dim eth oxy-5-br om open -
ta n a l (24a ). Ozonized oxygen was bubbled through a cold
(-78 °C), well-stirred mixture of 21 (6.65 g, 17.2 mmol),
powdered NaHCO3 (0.75 g), MeOH (10 mL), and CH2Cl2 (100
mL) until the mixture turned blue. Me2S (4.5 mL) was added,
and then the reaction was warmed to rt, washed (saturated
aqueous NaHCO3), dried (Na2SO4), and concentrated. Chro-
matography (10% EtOAc/hexane) gave 4.3 g (80%) of 24a as a
colorless oil. Rf ) 0.25. 1H (C6D6): 9.50 (d, 1H, J ) 3.0), 4.05
(dt, 1H, J ) 5.6, 3.0), 3.60 (d, 1H, J ) 11.0), 3.46 (d, 1H, J )
11.0), 3.20 (s, 3H), 3.15 (s, 3H), 3.11 (s, 3H), 2.19 (dd, 1H, J )
15.2, 5.6), 2.13 (dd, 1H, J ) 15.2, 5.6), 1.36 (s, 3H), 1.29 (s,
3H). 13C (C6D6): 202.9, 101.7, 100.4, 72.8, 49.6, 48.6, 48.5, 34.1,
33.4, 25.3, 24.5. IR: 1733, 1254, 1213, 1074, 1053. MS: m/z
283 [(M + H)+ - MeOH, 81Br], 281 [(M + H)+ - MeOH, 79Br],
251, 249, 223, 221, 73. HRMS: calcd for C10H18O479Br [(M +
H)+ - MeOH] 281.0388, found 281.0386.
cis- (26a ) a n d tr a n s-2-[(2-Meth oxy-2-p r op yl)oxy]-4,4-
d im eth oxycyclop en ta n -1-ol (27a ) a n d 3,3- Dim eth oxy-5-
[(2-m eth oxy-2-p r op yl)oxy]p en n ta n a l (28a ). A benzene
(450 mL) solution of 24a (3.6 g, 11.5 mmol), n-Bu3SnH (4.0 g,
13.8 mmol), and AIBN (10 mg) was refluxed for 2 h. The
solution was then cooled and concentrated. Chromatography
(20% EtOAc/hexane) gave 1.67 g (62%) of a mixture of alcohol
26a and 27a (10:1 ratio, 1H NMR), Rf ) 0.15, and 0.41 g (15%)
of aldehyde 28a , Rf ) 0.44, both as colorless oils. 26a (major
diastereomer). 1H (C6D6): 4.20 (m, 2H), 3.08 (s, 3H), 3.00 (s,
3H), 2.99 (s, 3H), 2.35 (m, 2H), 1.93 (m, 2H), 1.26 (s, 6H). 13C
(C6D6): 107.2, 101.1, 77.6, 76.4, 49.1, 49.0, 48.7, 41.6, 41.4,
25.7, 25. IR: 3466, 1254, 1219, 1143, 1080, 1046. MS: 203
[(M + H)+ - MeOH], 185, 113, 73. HRMS: calcd for C10H19O4
[(M + H)+ - MeOH] 203.1283, found 203.1281. 28a : 1H
(C6D6): 9.65 (t, 1H, J ) 2.9), 3.42 (t, 2H, J ) 6.6), 3.01 (s, 3H),
2.93 (s, 3H), 2.57 (d, 2H, J ) 2.9), 1.93 (t, 2H, J ) 6.6), 1.18 (s,
6H). 13C (C6D6): 199.8, 101.0, 100.4, 56.5, 48.6, 48.5, 48.1, 35.7,
24.7. IR: 1719, 1268, 1213, 1164, 1116, 1039. MS: 205 [(M
+ H)+ - CHO], 161, 145, 113, 73. HRMS: calcd for C10H21O4
[(M + H)+ - CHO] 205.1440, found 205.1436.
Eth yl 2-(1-Hyd r oxy-3-oxocyclop en t-4-en yl)a ceta te (3).
(a ) F r om Alcoh ols 26a /27a . A CH2Cl2 (10 mL) solution of
the mixture of alcohols (1.23 g, 5.3 mmol) was added to a slurry
of TPAP (50 mg, 0.03 equiv), NMO (770 mg, 1.25 equiv), and
coarsely powdered molecular sieves (0.5 g) in CH2Cl2 (40 mL)
at rt. After 1 h, the dark mixture was poured into hexanes
(120 mL), causing precipitatation of Ru compounds. The solid
was filtered off, and the filtrate was concentrated to give 1.22
g (100%) of ketone 36, pale yellow oil. A a solution of this
ketone in THF (3 mL) was added into a cold (-78 °C) solution
of the enolate of ethyl acetate [preformed at -78°C in THF
(10 mL), under Ar, from BuLi (2.5 M in hexane, 2.6 mL, 6.5
mmol), i-Pr2NH (0.8 mL, 6.3 mmol), and EtOAc [0.5 mL, 5.3
mmol)]. After 20 min, the mixture was warmed to rt,
quenched (saturated aqueous NaHCO3), and extracted with
ether. The combined extracts were dried (Na2SO4) and
concentrated to give 1.37 g (81%) of a mixture of esters 37 and
38 as a colorless oil, a solution of which in THF (10 mL) and
10% aqueous AcOH (5mL) was stirred at rt for 4 h, and then
it was neutralized (saturated aqueous NaHCO3) and extracted
with CH2Cl2. The combined extracts were dried (Na2SO4) and
concentrated to give 0.98 g (92%) of diols 40 as a colorless oil.
These diols were treated with MsCl (0.3 mL, 4.3 mmol) and
Et3N (1.5 mL) in CH2Cl2 (15 mL) at 0 °C, and after 10 min
the solution was washed (saturated aqueous NaHCO3), dried
(Na2SO4), and concentrated. The resulting mesylates 41 were
dissolved in 80% aqueous AcOH (4 mL) at rt, and after 2 h,
the solution was neutralized (saturated aqueous NaHCO3), and
extracted (CH2Cl2). The combined extracts were dried (Na2SO4)
and concentrated to give 0.55 g (75%, over two steps) of enone
3, colorless oil.
(b) F r om 4-Cyclop en ten e-1,3-d ion e, 43. A THF (10 mL)
solution of 43 (3.0 g, 31.0 mmol) was introduced into a cold
(-78 °C) solution of ethyl acetate enolate prepared as described
above from EtOAc (3.1 mL, 31.0 mmol). After 20 min, the
mixture was warmed to rt, quenched (saturated aqueous
NaHCO3), and extracted with ether. The combined extracts
were dried (Na2SO4) and concentrated to give 4.8 g (86%) of 3
as a colorless oil. 1H: 7.52 (d, 1H, J ) 5.8), 6.17 (d 1H, J )
5.8 ), 4.22 (q, 2H, J ) 7.1), 2.80 (d, 1H, J ) 16.5), 2.70 (d, 1H,
J ) 16.5), 2.62 (d, 1H, J ) 18.4), 2.49 (d, 1H, J ) 18.4), 1.29
(t, 3H, J ) 7.1). 13C: 206.2, 171.5, 164.2, 133.8, 76.5, 61.3,
48.9, 43.8, 14.1. IR: 3452, 1719, 1213, 1026. MS: 185 (M +
H)+, 167, 139. HRMS: calcd for C9H13O4 (M + H)+ 185.0814,
found 185.0812. Anal. Calcd for C9H12O4: C, 58.69;, H, 6.57.
Found: C, 58.82; H, 6.54.
Eth yl 2-(4-Ch lor o-1-h yd r oxy-3-oxocyclop en t-4-en yl)-
a ceta te (47). Chlorine gas was bubbled through a CCl4 (15
mL) solution of 3 (1.4 g, 7.6 mmol) at 0 °C, until the Cl2 color
persisted. After 20 min, Et3N (5 mL) was added. The mixture
was warmed to rt, diluted with ether (120 mL), washed (water,
then brine), dried (Na2SO4), and concentrated to give 1.3 g
(77%) of 47 as a colorless oil. 1H: 7.47 (s, 1H), 4.23 (q, 1H, J
) 7.2), 2.85 (d, 1H, J ) 16.7), 2.77 (d, 1H, J ) 18.4), 2.72 (d,
1H, J ) 16.7), 2.64 (d, 1H, J ) 18.4), 1.30 (t, 3H, J ) 7.2). 13C:
197.4, 171.5, 156.9, 137.1, 73.8, 61.6, 48.2, 43.8, 14.2. IR:
1726, 1594, 1296, 1206, 1019. MS: 221 and 219 (M + H)+,
203 and 201. HRMS: calcd for C9H12O435Cl (M + H)+
219.0424, found 219.0421. Anal. Calcd for C9H12O4Cl: C,
49.44; H, 5.07. Found: C, 49.35; H, 5.06.
Eth yl 2-(1-(2-Oxa n yloxy)-4-ch lor o-3-h yd r oxycyclop en t-
4-en yl)a ceta te (49). Dihydropyran (0.5 g, 5.6 mmol) was
added to a CH2Cl2 (15 mL) solution of 47 (1.1 g, 5.1 mmol)
and camphorsulfonic acid (“CSA”, 30 mg, 0.13 mmol) at 0 °C.
The mixture was warmed to rt, and after 1 h Et3N (0.1 mL)
was added to neutralize CSA. The mixture was diluted
(CH2Cl2, 20 mL), washed with water, dried (Na2SO4), and
concentrated to give 48 (1.6 g, 100%, colorless oil). A cold (0
°C) MeOH (10 mL) solution of this compound and CeCl3‚7H2O
(1.9 g, 5.1 mmol) was treated with powdered NaBH4 (0.3 g).
After 15 min, the reaction was quenched (saturated aqueous
NH4Cl) (CAUTION: H2 gas evolved) and extracted with ether.
The combined extracts were dried (Na2SO4) and concentrated.
Chromatography (50% EtOAc/hexanes) gave 1.37 g (88%) of
alcohols 49 as a colorless viscous oil. Rf ) 0.62. 1H NMR of
(35) Exp er im en ta l p r otocols. NMR spectra (ppm, δ, 1H ) 250;
13C ) 62.5 MHz, 25 °C, J in Hz) were recorded in CDCl3, unless
otherwise indicated. FTIR spectra (cm-1) were obtained from thin films
on NaCl plates. Low- and high-resolution mass spectra (m/e) were
obtained in the CI (CH4) mode. Microanalyses were carried out by
Galbraith Laboratories, Inc., Knoxville, TN. All sensitive reactions were
carried out under Ar. Reagent and solvent were used as received,
except CH2Cl2, i-Pr2NH, Et3N: distilled from CaH2; EtOAc and
2-methoxypropene: distilled; MeOH: dried over 4 Å molecular sieves;
NBS: recrystallized from H2O; THF: freshly distilled from Na/Ph2CO.

Synthesis of (()-Chlorovulone II
J . Org. Chem., Vol. 63, No. 5, 1998 1675
one of the four diastereomers: 6.03 (s, 1H), 4.85 (t, 1H, J )
3.7), 4.60 (m, 1H), 4.26 (d, 2H, J ) 3.1), 4.03 (q, 2H, J ) 7.1),
3.81 (m, 1H), 3.40 (m, 1H), 2.77 (dd, 1H, J ) 15.0, 6.1), 2.66
(d, 1H, J ) 12.3), 2.48 (d, 1H, J ) 12.3), 2.15 (dd, 1H, J )
15.0, 3.5), 1.41-1.70 (m, 6H), 1.17 (t, 3H, J ) 7.1). 13C of one
of the four diastereomers: 170.6, 143.0, 132.6, 94.6, 85.0, 76.1,
62.7, 60.6, 45.6, 45.2, 31.6, 25.3, 19.8, 14.2. IR: 3418, 1726,
1629, 1206, 1109, 1019. MS: 307 and 305 (M + H)+, 289 and
287, 269, 205. HRMS: calcd for C9H22O535Cl (M + H)+
305.1156, found 305.1151. Anal. Calcd for C9H21O5Cl: C,
55.17; H, 6.95. Found: C, 55.04; H, 6.95.
Et3N (6.4 mL, 2 equiv) and Ac2O (4.3 mL, 2 equiv) at 0 °C for
1 h, and then it was washed (saturated aqueous NaHCO3),
dried (Na2SO4), and concentrated (CAUTION: its is advisable
to test the solution for residual peroxides) to give 2.8 g (92%)
of methyl 5-oxopentanoate, colorless liquid. A solution of this
material (0.7 g, 5.4 mmol) and (formylmethylene)triphenylphos-
phorane (3.5 g, 11.3 mmol) in toluene (30 mL) was heated at
60 °C for 3 h, and then it was cooled and poured into petroleum
ether (200 mL). The precipitate was filtered off, the filtrate
was concentrated, and the residue was chromatographed (30%
EtOAc/hexane) to give 0.7 g (84%) of 53 as a colorless oil. Rf
) 0.31. 1H: 9.51 (d, 1H, J ) 7.7), 6.82 (dt, 1H, J ) 15.7, 6.7),
6.13 (ddt, 1H, J ) 15.7, 7.7, 1.0), 3.68 (s, 3H), 2.35 (m, 4H),
1.86 (quintet, 2H, J ) 7.4). 13C (C6D6): 192.9, 173.0, 156.0,
133.9, 51.4, 33.3, 31.9, 23.4. IR: 1729, 1696, 1636, 1204, 1171,
1118. MS: 157 (M + H)+, 125. HRMS: calcd for C8H13O3 (M
+ H)+ 157.0865, found 157.0864.
(()-Meth yl 7-(3-Ch lor o-5-h yd r oxy-5-oct-2-en yl-2-oxo-
cyclop en t-3-en ylid en e)h ep t-5-en oa te (Ch lor ovu lon e II,
1). A THF (1 mL) solution of 52 (104 mg, 300 µmol) was added
to a cold (-78 °C) solution of LDA [preformed in THF (0.5 mL)
at -78 °C from i-Pr2NH (45 µL, 340 µmol) and BuLi (2.5 M in
hexane, 150 µL, 370 µmol), phenanthroline indicator]. After
30 min, a THF (1 mL) solution of compound 53 was introduced.
The mixture was warmed to -30 °C, quenched with saturated
aqueous NaHCO3, and extracted with ether. The combined
extracts were dried (Na2SO4) and concentrated to give 54 in
82% yield. A solution of 54 (84 mg) and TsOH (10 mg) in
MeOH (1 mL) was stirred for 1 h at rt, and then it was diluted
with ether (10 mL), washed (saturated aqueous NaHCO3),
dried (Na2SO4), and concentrated to give (()-chlorovulone II,
1, as a colorless oil, in 88% yield. 1H: 7.21 (s, 1H), 7.07 (d,
1H, J ) 11.8), 6.76 (ddt, 1H, J ) 15.1, 11.8, 1.7), 6.28 (dt, 1H,
J ) 15.1, 7.1), 5.57 (m, 1H), 5.24 (m, 1H), 3.69 (s, 3H), 2.80
(dd, 1H, J ) 14.5, 7.5), 2.68 (dd, 1H, J ) 14.5, 8.1), 2.36 (t,
2H, J ) 7.4), 2.31 (q, 2H, J ) 6.9), 2.00 (br q, 2H J ) 7.1),
1.82 (quintet, 2H J ) 7.4), 1.31-1.23 (m, 6H), 0.88 (t, 3H, J )
6.4). 13C: 187.9, 173.9, 153.9, 147.9, 138.2, 135.2, 135.0, 134.3,
126.2, 122.0, 77.8, 51.9, 36.9, 33.5, 33.0, 31.7, 29.3, 27.6, 24.0,
22.7, 14.2. IR: 1736, 1709, 1629, 1284, 1198, 1164, 1052, 1038.
MS: 383 and 381 (M + H)+, 365 and 363, 333 and 331.
HRMS: calcd for C21H30O435Cl (M + H)+ 381.1833, found
381.1831. Anal. Calcd for C21H29O4Cl: C, 66.22; H, 7.67.
Found: C, 66.08; H, 7.65.
4-(2-Oxa n yloxy)-2-ch lor o-4-oct-2-en ylcyclop en t-2-en -1-
on e (52). DIBAL-H (1.5 M in toluene, 6.3 mL) was added to
a THF (20 mL) solution of alcohols 49 (1.3 g, 4.3 mmol) at -78
°C. After 2 h, MeOH was added dropwise to the cold solution
until no more bubbling was apparent (CAUTION: flammable
gases released). The mixture was partitioned between water
(60 mL) and ether (200 mL). The combined extracts were dried
(Na2SO4) and concentrated to give 0.9 g (79%) of aldehyde 50
as a colorless oil. Without further purification, this aldehyde
in THF (5 mL) was added to a cold (-40 °C) slurry of
hexylidenetriphenylphosphorane [preformed at 0 °C from
CH3(CH2)5Ph3P+ I- (2.9 g, 6.6 mmol) and BuLi (2.5 M in
hexane, 2.7 mL) in THF (20 mL), followed by addition of
DMPU (10 mL) after 10 min]. After 15 min, the Wittig
reaction was quenched (saturated aqueous NaHCO3, 10 mL)
and extracted with ether. The combined extracts were con-
centrated, and the residue was treated with petroleum ether
(40 mL), causing formation of a precipitate (phosphine oxide),
which was filtered off. The filtrate was concentrated to give
0.9 g (80%) of olefin 51 as a colorless oil. A DMF (3 mL)
solution of this olefin was added to a DMF (5 mL) solution of
PDC (2.0 g, 5.4 mmol) at rt. After 2 h, the mixture was diluted
(ether, 150 mL), washed (H2O), dried (Na2SO4), and concen-
trated. Chromatography (10% EtOAc/hexanes) gave 0.7 g
(83%) of 52 as a colorless oil, Rf ) 0.43. 1H NMR of one of the
two diastereomers: 7.47 (s, 1H), 5.57 (m, 1H), 5.33 (m,1H),
4.69 (t, 1H, J ) 2.1), 3.89 (m, 1H), 3.42 (m, 1H), 3.00 (d, 1H,
J ) 18.5), 2.59 (m, 2H), 2.54 (d, 1H, J ) 18.5), 2.01 (m, 2H),
1.52-1.81 (m, 6H), 1.28 (m, 6H), 0.87 (t, 3H, J ) 6.7). 13C of
one of the two diastereomers: 197.5, 158.9, 136.9, 134.4, 122.0,
95.3, 80.6, 80.4, 63.1, 45.4, 37.2, 31.4, 29.0, 27.4, 25.1, 22.5,
19.9, 13.9. IR: 1729, 1603, 1297, 1198, 1105, 1018. MS: 329
and 327 (M + H)+, 311 and 309. HRMS: calcd for C18H28O335Cl
(M + H)+ 327.1727, found 327.1724. Anal. Calcd for
Ack n ow led gm en t. We gratefully acknowledge the
National Institutes of Health (CA-55268), the National
Science Foundation (CHE 95-26183), and the Robert A.
Welch Foundation (C-1007) for their generous support
of our research program. M.A.C. is a Fellow of the
Alfred P. Sloan Foundation, 1994-1998.
C
18H27O3Cl: C, 66.14; H, 8.33. Found: C, 66.02; H, 8.34.
Meth yl 7-Oxoh ep t-5-en oa te (53). Ozonized oxygen was
bubbled into a cold (-78 °C) mixture of cyclopentene (2 mL,
23 mmol), NaHCO3 (300 mg), MeOH (6 mL), and CH2Cl2 (30
mL) until a blue color persisted, and then N2 was bubbled
through the solution until the blue color disappeared. The
reaction mixture was diluted with benzene (150 mL) and
concentrated to about 65 mL. This solution was treated with
J O972073F