Catalytic Enantioselective Total Syntheses of Bakkenolides I, J, and S: Application of a Carbene-Catalyzed Desymmetrization
Eric M. Phillips, John M. Roberts, and Karl A. Scheidt
“The bakkanes are a large class of sesquiterpene natural products containing a characteristic cis-fused 6,5-bicyclic core”. They possess a wide variety of biological activity for example antifeedant effects, platelet aggregation inhibition and presumably some activity against various cancer cell lines. Some total syntheses were published to date but this one catched my eye because of the nice methodology presented here. As you might know, NHC (N-heterocyclic carbene) catalyzed reactions can be used in analogy to nature’s TPP-catalyzed aldol reactions, e.g. in the Strecker reaction. Further examples are the use of NHC’s as ligands in metathesis reactions, Suzuki- and Buchwald-cross couplings or, as presented here, in an enantioselective synthesis of β-lactones.
It’s a rather short synthesis but with two cool key steps presented separately. First the three guys which were synthesized:
As you can see with the core of Bakkenolide S in hand the remaining two are easily made.
The synthesis starts off with a Tsuji-Trost reaction giving them the allylic alcohol which was oxidized with BAIB in the presence of TEMPO to the unsatured aldehyde (why didn’t they use manganese dioxide?). This was cyclised to the β-lactone employing the group’s own chemistry with a good yield and excellent enantio- and diastereoselectivity.
The mechanism looks like this:
As in the Strecker reaction the NH-carbene (in situ produced with Hünig’s base) attacks the aldehyde and forms after loss of the α-proton an unsaturated enolate. This is re-protonated with enol formation and reformation of the positively charged NH-ligand. Subsequent enantio- and diastereoselective aldol reaction gave the tertiary alcohol which reacts with the strongly activated ketone to give the β-lactone under catalyst recovery. NICE…
With the key intermediate in hand the group removed the lactone in the presence of silica gel to give the olefin and carbon dioxide. Dioxolane formation was followed by stereoselective boronation/oxidation to the alcohol followed by deprotection of the ketone and TBS ether formation. Wittig reaction to the terminal olefin and isomerization with Crabtree’s catalyst gave the trisubstituted internal alkene.
Reduction of the alkene, de-silylation and DMP-oxidation then furnished the ketone shown. Deprotonation was accomplished with LDA, the resulting enol reacted with Mander’s reagent and the methyl ester transesterified with propargyl alcohol. The prepended isomerization of the terminal olefin proved to be necessary because direct reduction under various conditions didn’t produce the expected product.
The following step presents again a nice methodology which I will present to you separately.
Originally the group planned to produce the δ-lactone via a Conia-ene reaction (http://www.organic-chemistry.org/namedreactions/conia-ene-reaction.shtm) but this attempt was unsuccessful. Nevertheless by reacting the propargyl ester with Mn3+ the lactone was formed in very good yield with excellent diastereoselectivity.
Reduction of the ketone and subsequent isomerization of the lactone then produced Bakkenolide S.
The mechanism of the lactone formation might be this one:
First a SET oxidation by manganese to give the strongly stabilized radical which reacts after rotation of the ester group with the alkyne moiety to give the 5-exo-dig radical.
Further info about this kind of reactions can be found here: Chem. Rev. 1996, 96, 339-363
To the end, ester formation with the corresponding acid chloride gave Bakkenolide I and J.
Overall a nice synthesis in which a lot of interesting methodology was employed. If you’re interested in further reactions catalyzed by this NHC’s have a look in the references.
THX for reading my stuff J
I received a question on the isomerization step so here’s the mechanism for this transformation:
The TBAF acts as a base and deprotonates the alcohol. This undergoes a retro aldol reaction followed by bond rotation of the latone and reverse aldol reaction to give the final product.