Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 20, 2000 4983
Scheme 3a
scrutinized by in situ 1H NMR monitoring in C6D6. The
cyclization does not require high dilution; the yield does not suffer
when the monomer concentration is raised from 0.001 to 0.02
M.
With access to EE-25 secure, we addressed the introduction
of the remaining four stereogenic centers and the completion of
the synthesis of 1A. The esters in EE-25 were converted
stereospecifically to the E,E-diene 28 (DIBAL-H reduction to 26,
CBr4/PPh3 13 to bromide 27, and Superhydride reduction to 2814).
Initial hydroboration/oxidation with Me2S‚BH3 gave a mixture
of products, suggesting that the level of substrate control of the
crucial hydroboration event was low. We turned to an asymmetric
hydroboration of 28 using the hindered monoisopinocampheyl
borane15 derived from (+)-R-pinene. Following H2O2 oxidation
the desired (R,R)-diol 29 was obtained in 58% yield, along with
a small amount of the unsymmetrical (R,S)-diol (∼10-15%). The
identity of 29 was first made by comparison with the reported
1H NMR data for tetramethylcylindrocyclophane A (29), which
Moore and co-workers had prepared by treatment of natural 1A
with diazomethane. This assignment was secured by a single-
crystal X-ray structure determination.
All four methyl ethers in 29 were cleanly removed by fusion
with excess methyl magnesium iodide at 160 °C16 under vacuum
to provide (-)-cylindrocyclophane A (1A) in ∼60% yield. No
other byproducts were observed in this remarkable demethylation
reaction in which both secondary benzylic alcohols were main-
tained. The structure of 1A was confirmed by its melting point
a (q) PDC, CH2Cl2, 96%. (s) NaH, benzene, cat. 15-crown-5, 55%. (t)
DIBAL, CH2Cl2, 100%. (u) CBr4, PPh3, CH2Cl2. (v) LiBHEt3, THF, rt,
91% overall from 26. (w) IpcBH2, THF, -20 °C to rt; H2O2, NaOH,
58%. (x) MeMgI, neat, 160 °C, 1 h, 60%.
(276-278 °C; lit. 276-278 °C1b), optical rotation ([R]RTD ) -20°;
1
lit. [R]RT ) -20° 1b), HRMS (FAB), and H NMR (in CD3OD
D
and in DMSO-d6 1b) data.
the benzylic alcohol that was observed with Pd/C) and then
oxidized (PDC) to the arylaldehyde 22 or directly oxidized (PDC)
to the 4,5-dehydro aldehyde 23. Each of the phosphonoester
aldehydes 22 or 23 was a potential candidate to play the role of
the bifunctional monomer 2.
In conclusion the synthesis of (-)-cylindrocyclophane A (1A)
was achieved by the use of an efficient double Horner-Emmons
macrocylic dimerization reaction. It is interesting that this process
did not require high dilution and that it was more stereoselective
when the less rigid saturated monomer 22 was used instead of
the olefinic analogue 23. The clean perdemethylation of the tetra-
O-methyl ether 29 by the action of MeMgI at 160 °C warrants
the use of these improbable, yet trivially implemented, conditions
for the demethylation of many anisole derivatives.17
With monomer synthesis completed, we began macrocyclic
dimerization studies. When aldehyde 23, containing the E-4,5-
alkene, was subjected to the Masamune olefination conditions
(LiCl, DBU, CH3CN, 0.01 M),11 macrocyclic dienes 24 were
obtained in 53% yield but as a mixture of EE-24, EZ-24, ZZ-24
(∼2:4:1 ratio). Although the yield was acceptable, the enoate
isomer distribution was unsatisfactory for further manipulations.
When we examined the cyclization of the saturated phosphono
ester aldehyde 22 under the same conditions, only a single
stereoisomer, EE-25, was formed (Scheme 3); however, the yield
was only 15%. This remarkable effect of the carbon chain
structure on the stereoselectivity of the olefination reaction
prompted us to screen various reaction conditions for the
macrocyclization. Sodium hydride in benzene containing a
catalytic amount of 15-crown-5 ether gave the most favorable
results.12 Macrocycle EE-25 was formed in 55% yield and to the
exclusion of the Z,Z-isomer, even when the reaction was
Acknowledgment. This investigation was supported by a grant
awarded by the DHHS (CA-76497). We thank Drs. Maren Pink and Victor
J. Young of the University of Minnesota X-ray Crystallographic Labora-
tory for their determination of the structure of 29 and Professor Amos B.
Smith, III, for providing us with a copy of the manuscript of the
accompanying contribution prior to submission.
Supporting Information Available: Experimental procedures for
preparation of and characterization data for all new compounds and X-ray
structural information for compound 29 are included as Supporting
Information (PDF). This material is available free of charge via the
JA000429Q
(13) Axelrod, E. H.; Milne, G. M.; Van Tamelen, E. E. J. Am. Chem. Soc.
1970, 92, 2139.
(14) (a) Krishnamurthy, S.; Brown, H. C. J. Org. Chem. 1980, 45, 849.
(b) Nonacidic workup was required at this step to avoid isomerization of the
alkenes.
(15) Brown, H. C.; Jadhav, P. K.; Mandal, A. K. J. Org. Chem. 1982, 47,
5074-5083.
(16) (a) Mechoulam, R.; Gaoni, Y. J. Am. Chem. Soc. 1965, 87, 3273,
wherein reference is made to the earlier unpublished use and recommendation
of this method for cleavage of dimethyl resorcinols by Professor G. Ourisson.
(b) Meerwein, H. Houben-Weyl, Methoden der Organische Chemie; Mu¨ller,
E., Ed.; George Thieme Verlag: Stuttgart, 1964; Vol. VI, pp 160-164.
(17) For example, demethylation of sensitive ArOMe-containing intermedi-
ates in recent vancomycin syntheses was solved by the use of AlBr3/EtSH,
conditions that were not successful in the case of conversion of 29 to 1A.
(10) Baltzly, R. J. Am. Chem. Soc. 1952, 74, 4586.
(11) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.; Masamune,
S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183.
(12) To probe whether the enoates comprising the mixture of 24 were likely
to be isomerizing under the reaction conditions, we performed the following
pair of control experiments. Diene-dienoate ZZ-24 was exposed at rt to 3,5-
(MeO)2PhCHO and MeO2CCH2PO(OMe)2 in the presence either of (i) NaH
and 15-C-5 in C6D6 or of (ii) DBU and LiCl in D3CCtN. The intermolecular
olefination proceeded smoothly, but there was no observable change in the
isomeric integrity of ZZ-24, even after each mixture was incubated for 27 h.
It is therefore unlikely that the diene-dienoates 24 (or, we presume, 25) are
equilibrating under either of these sets of Horner-Emmons olefination
conditions. The mixture of dienoate isomers appears result from a kinetically
rather than thermodynamically controlled event.
Hoye, Thomas R.
