Synthesis of Undecachlorosulfolipid A: Re-evaluation of the Nominal Structure
Christian Nilewski, Nicholas R. Deprez, Thomas C. Fessard, Dong Bo Li, Roger W. Geisser, and Erick M. Carreira
This time I will present to you a real beasty of a molecule. It belongs to the class of the chlorosulfolipids which gained some synthetic interest in the last few years. Their biological profiles combined with the low supply of material drives researchers to produce some quantities for biological evaluation. Nevertheless from a synthetic point of view the development of asymmetric halogenation reactions is a growing field which can be further examined during the synthesis of such complex tasks.
The most complex member of the chlorosulfolipids looks like this:
As can be seen from the structure it contains no less than 15 stereocenters of which 9 are contiguous and 9 are chiral chlorine atoms. In the light that only a handful of tactics for asymmetric chlorination reactions are known to date this is an extremely challenging task.
We start off with two precursor molecules whose syntheses are described in the supporting material which I highly recommend for reading because it contains a lot of useful information about NMR analysis of chiral chlorine bearing carbon atoms.
The first intermediate stems from commercially available (S)-1,2,4-butanetriol 1 which was protected as the acetonide, coupled under Mitsunobu conditions with phenyltetrazolylsulfide which in turn was oxidized to the sulfone. Julia-Kocienski olefination gave a mixture of E/Z-isomers which were isomerized to the major E-isomer under photolytic conditions to give 4.
The more reactive g,d-olefin was dichlorinated with tetraethylammonium trichloride with a d.r. of about 1.8 / 1 which can be further enhanced after epoxidation because of the easier separation of diastereomers. DiBAl-H reduction of the ester was followed by acetylation of the alcohol, stereoselective Sharpless dihydroxylation, and regioselective epoxide formation upon treatment with triflic anhydride. The overall yield of this sequence is only 9 % but during these five steps four of the fifteen stereocenters are formed.
5 was subjected to acetonide cleavage conditions and the resulting diol was protected as the bis-TBS ether of which the primary alcohol was again set free with HF – pyridine to give 6. Oxidation of the terminal alcohol to the aldehyde and Wittig olefination with phosphonium salt 12 gave 13 whose double bond was again dichlorinated to give after deacetylation compound 14.
Fragment 12 stems from two commercially available building blocks. On one hand ethyl lactate was protected and selectively reduced to aldehyde 7 while on the other hand propanediol was monoprotected and iodinated under Appel conditions to give iodide 8. Diastereoselective alkylation with the lithium reagent derived from 8 then furnished alcohol 9 which in turn was benzylated, and converted after selective monodeprotection into iodide 11 which gave Wittig salt 12 in the presence of triphenylphosphine.
The second half of the molecule derived from pentanediol. Monoprotection, Ley oxidation and dichlorination with NCS produced aldehyde 15. Next asymmetric alkynylation under Carreira’s conditions gave 16 with excellent enantioselectivity. Semireduction of the alkyne and hydroxy directed epoxidation of the trans-alkene necessitated DMP oxidation of the alcohol because no suitable conditions for selective epoxide opening could be identified. Thus ZrCl4 mediated epoxide opening of 16a and stereoselective reduction of the ketone gave diol 17.
To the end the diol was protected as an acetonide, the TBDPS group removed, the terminal alcohol oxidized, and reacted with Still-Gennari modified HWE reagent A to yield 18.
Going on with the synthesis ester 18 was successively reduced and the resulting allylic alcohol exposed to Sharpless asymmetric epoxidation reaction conditions. The epoxide was then regioselectively opened to give diol 19. Acetonide formation and debenzylation was followed by Mitsunobu coupling and oxidation to yield sulfone 21.
Fragments 14 and 21 were coupled under Julia-Kocienski conditions after prior DMP oxidation of the terminal alcohol of 14. Subsequently the epoxide of the intermediate was opened with PPh3Cl2 to give alcohol 22. Stereoselective dichlorination of the double bond then gave 23.
To the end 23 was debenzylated and the resulting alcohol used as a handle to introduce a double bond with Martin sulfurane. TBS removal and selective esterification with palmitoyl chloride gave protected Undecachlorosulfolipid 25.
Conversion of 25 into Undecachlorosulfolipid was then accomplished first by sulfate introduction with SO3 in DMF followed by acetonide cleavage with TFA.
With a few µg of Undecachlorosulfolipid A in hand the group compared the analytical data of their lab work to the reported data and noticed that the compounds were not identical. Especially the assignment of the ester bearing hydroxy group caused some problems. It was assumed that the stereochemistry should be R instead of S.
Well… nevertheless congratulations to the group to get this synthesis to work. Hopefully there will be a full account of this work showing all their tactics.