The Funk Synthesis of (-)-Nakadomarin A

The Z alkene of Nakadomarin A (3) suggested to Raymond L. Funk an approach (Org. Lett. 2010, 12, 4912. DOI: 10.1021/ol102079z) based on ring-closing alkyne metathesis. The efficient assembly of 3 he reported illustrates the power of convergent design in target-directed synthesis.

A practical limit on applications of alkyne metathesis is the requirement for internal alkynes, necessitating methyl capping of a terminal alkyne. In an alternative approach, Professor Funk took advantage of the long-known (J. Chem. Soc. 1954, 3201. DOI: 10.1039/JR9540003201) equilibration of a terminal alkyne 4 to the internal alkyne 5. Homologation of 5 with the phosphonate 6, followed by condensation with the ketone 7, then delivered the furan 8.

The assembly of the other half of 1 began with the commercial alcohol derived by reduction of D-pyroglutamic acid. Protection gave 9, that on hydride addition and dehydration was converted to 10. One-carbon homologation with the Vilsmeier-Haack reagent proceeded with the expected regiocontrol. This set the stage for the triply-convergent assembly of 14, first reductive amination of the aldehyde 11 with 12, then acylation of the resulting secondary amine with 13. The nucleophilic 14 was condensed with the aldehyde 8 to give an enone (not illustrated). Exposure of the enone to InCl₃ initiated an elegant cascade cyclization, first of the enamide in a conjugate sense to the enone, then Friedel-Crafts addition of the resulting N-stabilized carbocation to the furan, to deliver 15. The pendant silyloxymethyl group exerted the hoped-for diastereocontrol, allowing the direct construction of the central tetracycle of 3. Hydrolysis and decarboxylation completed the assembly of the diyne 1.

Initially, it was found that exposure of 1 to an Mo catalyst delivered 2 in only modest yield. As an alternative, they employed the technically more challenging tungsten-based Schrock catalyst. Later, they found that the recently-developed Fürstner Mo protocol also worked well.

The amide 2 could readily be carried on to the triene 18. With the first generation Grubbs catalyst (G1), kinetic ring-closing metathesis of 18, to complete the assembly of (-)-Nakadomarin (3), could be effected without jeopardizing the existing Z alkene.