The Lee Synthesis of (−)-Crinipellin A

Abstract

The crinipellins are the only tetraquinane natural products. The enone crinipellins, including crinipellin A 3, have anticancer activity. Hee-Yoon Lee of the Korea Advanced Institute of Science and Technology (KAIST) envisioned (J. Am. Chem. Soc. 2014, 136, 10274) the assembly of 2 and thus 3 by the intramolecular dipolar cycloaddition of the diazoalkane derived from the tosylhydrazone 1. The initial cyclopentene was prepared from commercial 4 following the Williams procedure. Conjugate addition of the Grignard reagent 5 in the presence of TMS-Cl led to the silyl enol ether 6. Regeneration of the enolate followed by allylation gave 7. The preparation of the racemic ketone was completed by ozonolysis followed by selec­tive reduction and protection. Addition of hydride in an absolute sense led to separa­ble 1:1 mixture of diastereomers. Reoxidation of one of the diastereomers delivered enantiomerically enriched 8. A few steps later, after coupling with 10, the sidechain stereocenter was set by Sharpless asymmetric epoxidation. Oxidation of 11 gave the aldehyde, that was converted to the alkyne 12 by the Ohira protocol. Addition of the Grignard reagent 13 gave the allene 14 as an inconse­quential 1:1 mixture of diastereomers. Deprotection then led to the tosylhydrazone 1. The transformation of 1 to 2 proceeded by initial formation of the diazo alkane 15. Intramolecular dipolar cycloaddition gave 16, that lost N2 to give the trimethylene–methane diradical 17. The insertion into the distal alkene proceeded with remarkable stereocontrol, to give 2 as a single diastereomer—in 87% yield from 1. Direct α-hydroxylation of the ketone derived from 2 gave the wrong diastereo­mer, and hydride addition to 18 reduced the wrong ketone. As an alternative, the enantiomerically-pure sulfoximine anion was added to the more reactive ketone, and the product was reduced and protected to give 19. Allylic oxidation converted the alkene to the enone, and heating to reflux in toluene reversed the sulfoximine addi­tion, leading to 20. Epoxidation of 20 followed by α-methylenation delivered the enone 21, that proved to be particularly sensitive. Eventually, success was found with TASF. With a similarly sensitive substrate, Douglass F. Taber of the University of Delaware observed (J. Am. Chem. Soc. 1998, 120, 13285) that TBAF in THF buffered with solid NH4Cl worked well.