COMMUNICATIONS
[11] a) P. J. Perry, V. H. Pavalidis, J. H. Hadfield, I. G. C. Coutts, J. Chem.
Soc. Perkin Trans. 1 1995, 1085; b) P. J. Perry, V. H. Pavalidis, J. H.
Hadfield, Tetrahedron 1997, 53, 3195.
[12] M. A. Rizzacasa, M. V. Sargent, J. Chem. Soc. Perkin Trans. 1 1988,
2425.
[13] G. A. Kraus, T. O. Man, Syn. Commun. 1986, 16, 1037.
[14] a) E. Nakamura, Tetrahedron Lett. 1981, 22, 663; b) D. Horne, J.
Gaudino, W. J. Thompson, Tetrahedron Lett. 1984, 25, 3529.
[15] The Lemieux Johnson protocol was not applicable because under
those conditions the only product observed was the benzaldehyde: R.
Pappo, D. S. Allen Jr., R. U. Lemieux, W. S. Johnson, J. Org. Chem.
1956, 21, 478.
[16] Behar and co-workers independently described the synthesis of a
related compound by an analogous route; see ref. [8c].
[17] T. Imamoto, N. Takiyama, K. Nakamura, T. Hatajima, Y. Kamiya, J.
Am. Chem. Soc. 1989, 111, 4392.
withnucleophilic spirocyclization. Unfortunately, even after
screening a large number of possibilities, this strategy was not
successful. When projected electrophilic cyclizations were
attempted by using halonium equivalents such as NBS, NIS,
NCS, or iodine in the presence of sodium bicarbonate,
oxidative demethylation and quinone formation occurred.
Similar results were observed when various epoxidations of
the furanoid ring were attempted.[2] An important constrain-
ing factor was the electron richness of the pentamethoxy-
naphthalene moiety present in 3. This pattern lent itself to
ready pairwise oxidative demethylations to produce ring A or
ring B quinones, withsubsequent deactivation of the furan
double bond. Furthermore, no reaction occurred when metal-
based reagents suchas Pd(OAc) 2, Ti(OAc)3, Re2O7,[3] and
HgII salts were explored to activate the furan double bond for
nucleophilic attack. Even after extensive experimentation, we
were unable to carry out the transformation 3 !4. In
substance, we were unable to overcome the combination of
nonreactivity of the furanoid moiety to some reagent combi-
nations, and the high vulnerability of the pentamethoxynaph-
thalene structure to others.
[18] R. A. Hill, G. W. Kirby, G. J. O×Loughlin, D. J. Robins, J. Chem. Soc.
Perkin Trans. 1 1993, 1967.
[19] T. Siu, D. Qin, S. J. Danishefsky, Angew. Chem. 2001, 113, 4849;
Angew. Chem. Int. Ed. 2001, 40, 4713.
The one successful oxidation which targeted the furan ring
and did not compromise the integrity of the pentamethox-
ynaphthalene moiety, arose from the action of osmium
tetroxide on 5,[4] which gave a diastereomeric mixture of 6
(Scheme 2, 50 60%). The product was difficult to separate
and could not be satisfactorily characterized by means of
1H NMR spectroscopy. Our decision to go forward withthis
material was based largely on a supportive mass spectrum.
Deprotection of the benzyl ether exposed the C10a phenolic
function, again as a poorly characterized mixture of diaster-
eomers 7.
The Total Synthesis of Heliquinomycinone**
Tony Siu, Donghui Qin, and Samuel J. Danishefsky*
In the preceding communication,[1] we described the
assembly of intermediate 3, which was envisioned as a
substrate for an oxidative dearomatization-spiroketalization
sequence (3 !4, Scheme 1), en route to heliquinomycinone
(2), the aglycone of the naturally occurring helicase inhibitor
O
Withtriol 7 in hand, all that remained to reach hydro-
quinonoid versions of 2 was acid-induced spiroketalization
(Scheme 3). Remarkably, this seemingly attainable goal could
not be accomplished. The hydroxy group at the pre-C3'
benzylic position was unexpectedly vulnerable.[5] An attempt
at a spirocyclization under Mitsunobu-type conditions was
unsuccessful and instead led to the transformation of diaster-
eomers 7 into 9.[6]
In retrospect, this result reflects the ease of formation of a
quinone-methide-like heterolysis product, presumably medi-
ated by the strong electron-donating nature of the five
methoxy groups on the naphthalene system. Various protec-
tions of C3 and C3' in the hope of favoring the desired
spirocyclization were not productive.
HO
O
10
O
O
OH
OH
CO2Me
1'
1
O10a
9'
4'
MeO
O
2
4
3
3'
R1
R2
O
OMe
O
R1
=
=
HO
OH
R2 = OH
R2 = OH
1
2
heliquinomycin
R1
heliquinomycinone
heliquinomycin (1). A large number of reagent combinations
were used in attempts to bring about the conversion of 3 into a
product of the type 4 (Scheme 1). In particular, we were
seeking electrophiles that could be introduced concomitantly
A chance observation proved to be critical in solving the
problem. Exposure of diastereomers 6 to air, in the presence
of triethylamine/methanol led to oxidation at C3', thus
forming the a-hydroxyketone (Scheme 4).[7] Subsequent de-
benzylation gave 10 as a 1:1 mixture of anomers. It was hoped
that the presence of the ketone would prevent bond formation
between the ∫C10a∫ phenol and ∫C3'∫ (except for that arising
from a presumably reversible hemicacetal link). However, the
feasibility of spirocyclization in 10, adjacent to a ketone
linkage, was by no means certain.
[*] Prof. S. J. Danishefsky,[] T. Siu, Dr. D. Qin
Department of Chemistry, Columbia University
Havemeyer Hall, New York, NY, 10021 (USA)
[ ] Laboratory for Bioorganic Chemistry
Sloan Kettering Institute for Cancer Research
1275 York Ave., New York, NY, 10021 (USA)
Fax : (1)212-772-8691
[**] This work was supported by the National Institutes of Health (Grant
numbers: AI 16943 and HL25848). The authors thank Dr. Makoto
Chino of Nippon Kayaku Co., Ltd. for kindly providing a sample of
heliquinomycin, and Yashuiro Itagaki of Columbia University for
high-resolution mass spectral analyses.
Under Mitsunobu conditions,[6] the desired cyclization was
achieved and mixture 10 afforded compounds 11 and 12 in a
1:1 ratio after removal of the silyl ethers. These products were
Angew. Chem. Int. Ed. 2001, 40, No. 24
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4024-4713 $ 17.50+.50/0
4713