Synthesis of (S)-2-hydroxy-4-oxohex-5-enal

Question

I am currently working on a total synthesis strategy, but I was told that a key step in my total synthesis probably wouldn't work, so now I am trying to figure out a way around it. The compound that I need to make is (S)-2-hydroxy-4-oxohex-5-enal 4(see structure, aldehyde is protected as a thioacetal). I first considered an asymmetric aldol reaction between 2 and 3, but since the proton on the tertiary carbon of 2 is probably the most acidic and the other ketone is likely to undergo conjugate addition, the chances of this reaction happening in the right direction are small. I have been considering starting from (R)-glycidol and then do a ring-opening with an enolate, but literature seems to indicate that enolates do not readily attack epoxides. Does anyone have a suggestion for a direction?

Answer 1

I agree with @Waylander that the synthesis of (S)-2-hydroxy-4-oxohex-5-enal (7) by asymmetric means may be problematic. Why not start with what Mother Nature gives you---glucose. Diacetone glucose (1; DAG) is readily prepared and commercially available. The conversion of DAG to unsaturated acetonide 2 has been accomplished. The authors^[1] removed the $\ce{C3-OH}$ by a modified Barton radical deoxygenation. Subsequent selective hydrolysis of the $\ce{C5-C6}$ acetonide (aq. HOAc) and bis-deoxygenation via $\ce{Me2NCH(OMe)2}$. Treatment of unsaturated acetonide 2 with $\ce{MeOH/H+}$ leads to a presumed mixture of unsaturated acetals 4. Inversion of the $\ce{C2-OH}$ stereochemistry via mesylation and potassium superoxide displacement^[2] affords 4. Other oxygen nucleophiles may be considered. Ethylene glycol acetal 5 should be favored entropically over 4 given the bis-functionality of ethylene glycol. Any equilibrium may be favored by distillation or absorption (molecular sieves) of methanol. Finally, selective allylic oxidation of diol 5 with $\ce{MnO2}$ affords the enone 6. Hydrolysis of the acetal moiety of 6 leads to the target compound 7. This hydrolysis may well be problematic with β-elimination looming. Alternatively, the route 4 → 8 → 7 may prove superior give the mild Fetizon conditions for cleavage of the dithiane group of 8.

1. F. E. Ziegler and M. A. Sarpong, Tetrahedron, 2003, 59, 9013.
2. F. E. Ziegler, C. A. Metcalf III, A. Nangia and G. Schulte, J. Am. Chem. Soc., 1993, 115, 2581.

Answer 2

After consultation with former colleagues, we think your best option is to react 2 with an chiral enamine of 3 under mild Lewis acid catalysis (something like Sn(OEt)4)