A Novel Strategy for the Convergent Synthesis of 1,3,5,…-Polyols: Enone Formation, Asymmetric Dihydroxylation, Reductive Cleavage, Hydride Addition

Abstract

Asymmetric dihydroxylation of a,b-unsaturated ketones provided a,b-dihydroxyketones with up to 100% ee. The C a -O bond of these intermediates or their bis-TMS ethers, acetonides, phenylborates or orthoformiates was cleaved with SmI 2 , affording b-hydroxyketones. The latter can be reduced to furnish syn-or anticonfigured 1,3-diols of any desired configuration. 1,3,5,…-Polyols without substituents at C 2 , C 4 , C 6 ,… are the core structural feature of the polyol/polyene macrolide antibiotics 1 (Scheme 1). 1 This family of compounds comprises more than 200 members. Those representatives of which not only the constitution but also the configuration is known have received considerable attention in the synthetic community. 1,2 This is due to a desire of developing methodology for making such molecules in a stereocontrolled manner. In a longer perspective, the synthetic interest in polyol/polyene antibiotics 1 is motivated by the wish to understand the correlation (if any) between their 3D structure and their biological activity.Preferably, extended stretches 2 of 1,3,5,…-polyols should emerge from a convergent strategy (Scheme 1). In state-of-the-art syntheses of 2,4,6,…-substituted 1,3,5,…-polyol building blocks, such convergency is typically due to the inclusion of b-hydroxyketone intermediates and their construction through 'complex aldol additions'. 3 bHydroxyketone intermediates 3 have been used en route to extended polyols 2, too; they arose from the aldol addition of enolates 5 to aldehydes 4 2e,g,4 or from a different approach. 5 Here we communicate a novel way of making polyols 2 via b-hydroxyketones 3: by the a-defunctionalization of a,b-dihydroxyketones 6. The latter stem from the asymmetric dihydroxylation 6 ('AD') of enones 7, which, in turn, result from Horner-Wadsworth-Emmons olefinations of aldehydes 8 or from analogous Wittig reactions.We started with the AD of the a,b-unsaturated ketone 9a (Scheme 2). Trying to accelerate conversion, we utilized the so-called improved 7 rather than standard procedure, 8 i. e., employed 1 mol% K 2 OsO 2 (OH) 4 instead of 0.2 mol% and 5 mol% (DHQD) 2 PHAL instead of 1 mol%. Even so this reaction required 2.5 days at 0°C to furnish diol aS,bR-10a 9 in 89% yield and with 100% ee 10 (after extractive work up and purification by flash chromatography on silica gel 11 ). Hence, the reaction rate was somewhat lower and stereocontrol somewhat better than in the not too many known ADs of a,b-unsaturated ketones. [12][13][14][15] Scheme 1 Convergent approaches to 'skipped polyol' building blocks 2 of polyol/polyene macrolide antibiotics 1 via type-3 b-hydroxyketone intermediates.Downloaded by: University of Liverpool. Copyrighted material.

Multigramme synthesis and asymmetric dihydroxylation of a 4-fluorobut-2E-enoate

“…It was thought that the so-called “improved procedure” [24], which uses higher ligand/oxidant loadings (1 mol % osmium/5 mol % ligand) might be required to allow the reactions to proceed in acceptable yields and enantioselectivities [25]. Fig.…”