Enantioselective Synthesis of Oasomycin A, Part III: Fragment Assembly and Confirmation of Structure

Abstract

Dedicated to Professor Y. Kishi on the occasion of his 70th birthday.Herein we address the total synthesis of the natural product oasomycin A by assembly of the C1-C12, C13-C28, and C29-C46 subunits, whose syntheses have been described in the preceding Communications.[1]The synthesis plan (Scheme 1) incorporates a speculative late-stage macrolactonization of the linear seco acid precursor to form a 42-membered lactone that upon global deprotection would provide the natural product. Since oasomycin A is known to rearrange to the oasomycins D and E under basic conditions, [2] an acid-mediated global deprotection was obligatory. It was our intention to assemble the requisite seco acid by using an aldol addition of the C1-C28 ketone I to the C29-C46 aldehyde II with a concomitant installation of the C29 stereocenter, followed by a stereoselective reduction of the C27 ketone.The assembly of ketone I through a Kocienski-Julia olefination [3] of the C13-C28 aldehyde III with C1-C12 fragment IV was undertaken first (Scheme 2). Sulfone 1 was selectively deprotonated with KHMDS and treated with aldehyde 2[3] to afford the coupling product 3 a as a 7:1 mixture of E/Z isomers (57 % yield). In addition, a significant amount of a by-product was consistently formed in 15-25 % yield in this and related olefinations. This by-product with the general structure 3 b (Scheme 2) may be rationalized by a Brook rearrangement of the Julia intermediate followed by alkoxide attack on the sulfur center. All efforts to suppress this side reaction were unsuccessful. [4] With both the C1-C28 and C29-C46 subunits in hand, we addressed the aldol coupling which would provide the oasomycin A skeleton. The logic behind the selection of an aldol addition to form the C28 À C29 bond was based on the fact that the diastereoselectivity of this reaction should be reinforced by resident chirality in both reaction partners: the C25 stereocenter on the enolate [Eq. (1)], [5] and the C31 stereocenter on the aldehyde fragment [Eq. (2)].[6] Although Scheme 1. Assembly of oasomycin A subunits.

An Amidinohydrolase Provides the Missing Link in the Biosynthesis of Amino Marginolactone Antibiotics

“…Only for desertomycin has the absolute configuration been experimentally established. [23,25] Comparison of this with the configuration predicted from the PKS structure ( Figures S1-S3) showed exact agreement, except for the configuration at C-30, which is reversed from that predicted.…”

An Amidinohydrolase Provides the Missing Link in the Biosynthesis of Amino Marginolactone Antibiotics

“…Work is in progress to determine the mechanisms involved. Only for desertomycin has the absolute configuration been experimentally established . Comparison of this with the configuration predicted from the PKS structure (Figures S1–S3) showed exact agreement, except for the configuration at C‐30, which is reversed from that predicted.…”

Total Synthesis of (−)-Marinisporolide C

“…As shown by the research groups of Yonemitsu, Roush, and Evans, the presence of a α substituent in a α,β-unsaturated seco acid creates a bulky environment, which decreases the efficiency of the macrolactonization step. With the aim to increase this efficiency, we decided to incorporate a pre-organization element through the prior formation of the spiroketal before the macrolactonization step (Scheme ).…”