The Nakada Synthesis of (+)-Ophiobolin A
Ophiobolin A (3) showes nanomolar toxicity toward a range of cancer cell lines. A central feature of this sesterterpene, isolated from the rice fungus Ophiobolus miyabeanus, is the highly-substituted eight-membered ring. A key step in the synthesis of 3 described (Chem. Eur. J. 2013, 19, 5476, ; Angew. Chem. Int. Ed. 2011, 50, 9452, ) by Masahisa Nakada of Waseda University was the acid-mediated cyclization of 1 to 2.
The preparation of 1 began with the enantioselective hydrolysis of 4 to the monoester 5. Selective reduction followed by protection gave 6, that was carried on to 8. The ethoxyethyl group was selectively removed, and the alcohol was carried on to an iodide (not illustrated) that was condensed with the lactone 9 to give 1.
The cyclization of 1 could jeopardize the stereogenic center adjacent to the masked carbonyl, so eight diastereomers were possible. Careful optimization led to a preparatively useful yield of the desired product 2. Hydroboration gave 10, that was carried on to the aldehyde 11.
The cyclopentanone 15 was prepared from the enantiomerically-enriched epoxide 12. Opening with vinyl magnesium bromide followed by exposure to the second-generation Grubbs catalyst gave the diol 13, that was selectively protected, leading to 14. The derived bromohydrin was a mixture of regioisomers and diastereomers, from which, after oxidation, 15 dominated. Generation of the boron enolate from 15 in the presence of 11 gave the aldol product, that could be dehydrated with the Burgess reagent. Reduction with Raney nickel set the stereogenic center adjacent to the ketone.
Metathesis to close the eight-membered ring was not trivial. Finally it was found that 17 could be induced to cyclize, at elevated temperature using the second-generation Hoyveda catalyst. Routine functional group manipulation then completed the synthesis of (+)-Ophiobolin (3).
Some years ago, Neil E. Schore of the University of California, Davis showed (Tetrahedron Lett. 1994, 35, 1153. ) that the opening of Sharpless-derived epoxides such as 12 with vinyl nucleophiles was unexpectedly flexible. One set of conditions gave the expected inversion, but alternative conditions led to opening with clean retention (or double inversion) of absolute configuration.