A mercury-catalyzed transetherification cyclization leading to fused cyclic polyethers

Article abstract of DOI:10.1016/S0040-4039(00)01658-0

Dihydropyran 6 serves as a key intermediate in a route to fused cyclic polyethers. The dihydropyran was first converted to 2-hydroxy-1-γ-methoxyallyltetrahydropyrans, 10. Mercury trifluoroacetate-catalyzed transetherification cyclization generated new bic

Full text of DOI:10.1016/S0040-4039(00)01658-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 9509–9513
A mercury-catalyzed transetherification cyclization leading
to fused cyclic polyethers†,‡
Derek S. Tan§ and Stuart L. Schreiber*
Department of Chemistry and Chemical Biology, Howard Hughes Medical Institute,
Institute of Chemistry and Cell Biology, Harvard University, 12 Oxford St., Cambridge, MA 02138, USA
Received 31 August 2000; revised 20 September 2000; accepted 21 September 2000
Abstract
Dihydropyran 6 serves as a key intermediate in a route to fused cyclic polyethers. The dihydropyran
was first converted to 2-hydroxy-1-g-methoxyallyltetrahydropyrans, 10. Mercury trifluoroacetate-cata-
lyzed transetherification cyclization generated new bicyclic dihydropyrans, 12, ready for additional rounds
of synthesis. Attempted cyclization under stoichiometric conditions resulted in isolation of a-mercurial
acetal intermediates, 11, that could also be converted to the dihydropyran by treatment with ethyl vinyl
ether. © 2000 Published by Elsevier Science Ltd.
Keywords: polyether; dihydropyran; mercury; cyclization; diversity-oriented synthesis.
Target-oriented syntheses of members of the fused cyclic polyether class of natural products
have received much attention due to the biological potency and structural complexity of these
molecules.1 Their signature structural element, a series of fused cyclic ethers having regular
trans-syn-trans stereochemistry, has stimulated numerous iterative routes to these molecules.2–5
As a central discovery element in chemical genetic research6 we have been using diversity-ori-
ented syntheses of small molecules for use in both phenotypic and protein-binding screens.7–9
Polycyclic compounds are attractive in diversity-oriented synthesis because they provide a rigid
three-dimensional structure for the display of diverse building blocks. An efficient and iterative
route to fused cyclic polyethers would allow this class of molecules to be featured in diversity-
oriented synthesis, particularly if the reaction pathway were amenable to branching, where
different skeletal arrays of polyethers would be generated in a single synthesis.7
* Corresponding author. Fax: +1 (617) 495-0751; e-mail: sls@slsiris.harvard.edu
†
Dedicated to Professor Harry H. Wasserman, whose research, teaching and humanity continue to inspire many
generations of chemists. We feel fortunate to be among this group.
‡
Adapted in part from Tan, D. S. Ph.D. Dissertation, Harvard University, 2000.
Present address: Memorial Sloan–Kettering Cancer Center, Laboratory for Bioorganic Chemistry, 1275 York Ave.,
§
RRL 1301, Box 106, New York, NY 10021, USA.
0040-4039/00/$ - see front matter © 2000 Published by Elsevier Science Ltd.
PII: S0040-4039(00)01658-0

9510
Target-oriented synthesis of the polycyclic natural products in this class has generally
depended upon convergent approaches, however, structures with only two to four rings would
be attractive for a diversity-oriented synthesis, making a linear, iterative approach feasible.
Diversity could arise from both the number and size of the rings, the coupling of structurally
diverse building blocks to the rings using diverse coupling reactions, and, notably, the systematic
alteration of stereochemistry at the ring junctions.
With these considerations in mind, we developed the general route shown in Fig. 1. The
common intermediate is a dihydropyran, 1. Dihydropyrans have served as key intermediates in
iterative routes reported by Rainer and Allwein,2,3 and Bowman and McDonald.5b This type of
starting material could undergo stereoselective epoxidation at either face of the olefin to provide
either stereoisomer at C2. If desired, stereochemistry at C1 could be inverted with the
appropriate nucleophile, followed by introduction of the C1 sidechain. In addition to allowing
various ring sizes, substituents might be included or introduced on the sidechain to increase
structural diversity further. The presence of appropriate functionality on the C1 sidechain
should allow cyclization directly to a new dihydropyran that could undergo additional cycles of
synthesis. The presence of an olefin in the C1 sidechain might allow even non-cyclized products
to undergo further rounds of synthesis. Thus, the ideal route would require only three to four
steps per cycle, with the potential to introduce diversity at every step.
Figure 1. Dihydropyran-based general strategy for iterative synthesis of fused cyclic polyethers
An efficient and related route has been reported by Rainer and Allwein, involving transacetal-
ization–elimination at the cyclization step.3 With a view towards diversity-oriented synthesis, we
elected to develop a new cyclization reaction that proceeds under mild, neutral conditions
tolerant of a wide range of functional groups. Mercury(II)-mediated transetherification (X=
OMe) seemed an attractive candidate.10
A first generation synthesis of cyclization precursors 10a–c followed substantial literature
precedent (Fig. 2). The C6 hydroxyl of
-glucal was protected with a triisopropylsilyl group11 as
D
a mimic for the alkyldiisopropylsilyl solid support linker in use at the Harvard Institute of
Chemistry and Cell Biology.12 Benzylation at both C3 and C4 provided protected dihydropyran
6.11 Epoxidation with dimethyldioxirane proceeded with the expected b-stereoselectivity.13
Allylation at C1 provided 8a, along with a limited amount of the C1 epimer 8b, which was
carried along separately.2,14 Oxidation of 8a, followed by reduction with L-Selectride, provided
mannose-type intermediate 8c. Ozonolysis and Wittig olefination with methoxymethyl-
triphenylphosphorane provided the desired cyclization substrates 10a–c.4

9511
Figure 2. Synthesis of g-methoxyallyl cyclization precursors 10a–c from protected
D-glucal
Initial attempts at cyclization of 10a were carried out with a stoichiometric amount of
mercuric acetate. However, only the a-mercurial acetal, 11a, was isolated from the reaction
mixture (Fig. 3). Use of mercuric trifluoroacetate led to the corresponding a-mercurial acetal
11b. In intermolecular mercury-catalyzed transetherifications, ethyl or butyl vinyl ether is used
as the solvent (Fig. 4). Presumably, this is required not only to drive the overall equilibrium
toward the desired vinyl ether product, but also to promote elimination of the stable a-mercurial
acetal intermediate.
Figure 3. Cyclization of g-methoxyallyl substrates with stoichiometric and catalytic mercury(II)

9512
Figure 4. Elimination of the a-mercury acetal intermediate is equilibrium controlled
Thus, we were gratified to find that treatment of a-mercurial acetal intermediate 11b with an
excess of ethyl vinyl ether led to the desired bicyclic dihydropyran 12a. This two step process
could also be carried out in a single step using butyl vinyl ether as the reaction solvent.
Ultimately, by performing the reaction with 5 mol% mercuric trifluoroacetate in refluxing THF,
we were able to isolate 12a directly in good yield. Two methyl acetal side products were each
isolated in 5% yield, presumably arising from reaction of 12a with the liberated methanol,
catalyzed by uncomplexed trifluoroacetic acid present during the reaction. As expected, a trace
of the a-mercurial acetal was also observed in the crude product. Cyclization of the cis
glucose-type substrate 10b under these optimized conditions provided bicycle 12b with similar
high efficiency. However, cyclization of the cis mannose-type substrate 10c yielded only 10% of
the desired dihydropyran bicycle 12c with the major products being the corresponding bicyclic
methyl acetals. Future versions of this reaction might eliminate these side products by neutral-
ization of either the methanol or free trifluoroacetic acid present in the reaction.
We have demonstrated an intramolecular mercury-catalyzed transetherification cyclization as
part of a potential diversity-oriented synthesis of fused cyclic polyethers. To our knowledge, this
is the first example of an intramolecular mercury-catalyzed transetherification cyclization. The
cyclization reaction of cis and trans glucose-type substrates is efficient and mild and should
tolerate a wide range of functional groups in diversity-oriented synthesis. Further development
of the complete synthetic route and adaptation to solid phase synthesis remain to be explored.
Supplementary material. Complete experimental procedures and analytical data for all com-
pounds (PDF, 18 pages) are available on the Internet at ‘http://www-schreiber.
chem.harvard.edu/home/research results.html’ and ‘http://preprint.chemweb.com/orgchem/
–
0009004’.
Acknowledgements
D.S.T. was supported by NDSEG and Roche Organic Chemistry fellowships. The ICCB is
sponsored by the National Cancer Institute, Merck & Co., and Merck KGaA. The Harvard
University Mass Spectrometry Facility is funded by grants from the NSF (CHE-9020043) and
NIH (S10-RR06716). The Harvard University NMR Facility is supported by the NSF (BIR95-
22677, CHE88-14019), NIH (1-S10-RR11319-01, 1-S10-RR04870-01), and Howard Hughs
Medical Institute (HHMI). S.L.S. is an Investigator at the HHMI.
References
1. Alvarez, E.; Candenas, M.-L.; Pe´rez, R.; Ravelo, J. L.; Mart´ın, J. D. Chem. Rev. 1995, 95, 1953–1980.
2. Rainier, J. D.; Allwein, S. P. J. Org. Chem. 1998, 63, 5310–5311.

9513
3. Rainier, J. D.; Allwein, S. P. Tetrahedron Lett. 1998, 39, 9601–9604.
4. Simart, F.; Brunel, Y.; Robin, S.; Rousseau, G. Tetrahedron 1998, 54, 13557–13566.
5. For other recent examples, see: (a) Hori, N.; Matsukura, H.; Nakata, T. Org. Lett. 1999, 1, 1099–1101. (b)
Bowman, J. L.; McDonald, F. E. J. Org. Chem. 1998, 63, 3680–3682. (c) Mori, Y.; Yaegashi, K.; Furukawa, H.
J. Org. Chem. 1998, 63, 6200–6209. (d) Leeuwenburgh, M. A.; Overkleeft, H. S.; van der Marel, G. A.; van
Boom, J. H. Synlett 1997, 1263–1264. (e) Evans, P. A.; Roseman, J. D.; Garber, L. T. J. Org. Chem. 1996, 61,
4880–4881.
6. Schreiber, S. L. Bioorg. Med. Chem. 1998, 6, 1127–1152.
7. Schreiber, S. L. Science 2000, 287, 1964–1969.
8. Tan, D. S.; Foley, M. A.; Shair, M. D.; Schreiber, S. L. J. Am. Chem. Soc. 1998, 120, 8565–8566.
9. Tan, D. S.; Foley, M. A.; Stockwell, B. R.; Shair, M. D.; Schreiber, S. L. J. Am. Chem. Soc. 1999, 121,
9073–9087.
10. Watanabe, W. H.; Conlon, L. E. J. Am. Chem. Soc. 1957, 79, 2828–2833.
11. Gordon, D. M.; Danishefsky, S. J. J. Am. Chem. Soc. 1992, 114, 659–663.
12. Tallarico, J. A.; Depew, K. M.; Pelish, H.; Athanasopoulos, J.; Lindsley, C.; Westwood, N.; Sternson, S. M.;
Stavenger, R. A.; King, R. W.; Schreiber, S. L.; Shair, M. D.; Foley, M. A. unpublished results. See:
‘http://iccb.med.harvard.edu/’.
13. Halcomb, R. L.; Danishefsky, S. J. J. Am. Chem. Soc. 1989, 111, 6661–6666.
14. For C-allylation with allyltributylstannane: Evans, D. A.; Trotter, B. W.; Coˆte´, B. Tetrahedron Lett. 1998, 39,
1709–1712.
.