Study for Closure of the B Ring

We chose methyl groups to protect the pyrimidine ring of 12 because they were easily introduced by treatment of 10b with excess MeOH. For 

Addition of acetylene 9 to aldehyde 12 afforded 85% of propargylic alcohol 13 as a mixture of diastereomers. Hydrogenation reduced the triple bond and both double bonds and cleaved the Cbz groups providing 94% 

We therefore turned our attention to deprotection of the dimethoxy-pyrimdine. Very harsh conditions are typically required and our situation was further complicated by the polarity of the product, which would make purification troublesome. The use of 6 M hydrochloric acid15 to cleave the methyl groups was therefore appealing because excess reagent could 

The spectral data of 24 are very similar to those of the right half of cylindrospermopsin (1), while those of 25 are quite different. For instance, the absorption of the side chain methine hydrogen H7 of 24 at 8 4.70 (d, 1, / = 4. 0 Hz) is identical to that of cylindrospermopsin and very different from that of 25 at 8 4.44 (d, 1, J = 6.8 Hz). Therefore, we concluded that 24 has the same stereochemistry as cylindrospermopsin. The spectral data for 25 was later used to support the structural assignment of 7-epicylindrospermopsin (2) in which the side chain hydrogen absorbs at 8 4.50 (d, 1, J = 6.6 Hz).5 

The stereochemistry at C7 was introduced by the hydrogenation of the ketone. Although this hydrogenation was very selective for the desired stereoisomer, the process provided no information to confirm or refute the tentative stereochemical assignment at this center, which remained incorrectly assigned until Weinreb completed a stereocontrolled synthesis of 7-epicylindrospermopsin in 2001.7a 

---