Progress Towards the Total Synthesis of Mitomycin C

Abstract

Mitomycin C is a clinically used antitumor natural product isolated from <i>Streptomyces caespitosus</i>. The biological activity and synthetic challenge of mitomycin C has generated broad interest in its chemical synthesis. However, due to its compact array of reactive functionalities, the total synthesis of mitomycin C has been achieved only twice since its isolation in 1958.

Our approach to mitomycin C centers on the Brønsted acid catalyzed aza-Darzens reaction to install the aziridine, which is responsible for DNA alkylation. Prior work reported by our group described the concise synthesis of an advanced intermediate lacking only C10 of the mitomycin C backbone. This route utilized an aminomercuration of an alkynyl amine followed by quinone addition, resulting in an unprecedented oxidative ketalization.

This dissertation describes the preparation of several advanced intermediates en route to mitomycin C using novel pathways. In the first approach, the last carbon of the mitomycin C skeleton (C10) is successfully installed on an <i>N</i>,<i>O</i>-ketal intermediate using a lithium-halogen exchange. The installation of C10 was also accomplished using a methyl-substituted alkynyl amine for the aminomercuration. This resulted in an intriguing change of reaction outcome. Additionally, a functionalized C10 intermediate was prepared from the alkynyl amine using a halonium promoted cyclization. In each of the preceding cases, the key bond formed was the N4-C9a bond, and in the first two cases, the aminomercuration reaction led to concomitant formation of the C9-C8a bond.

The second key disconnection investigated was the N4-C4a bond. The key aminoquinone intermediate was successfully constructed under mild coupling conditions. Attempts to add the last two carbons of the mitomycin skeleton provided an unexpected change in regioselection. Further studies to achieve proper regiocontrol in aminoquinone formation are also described.